Between worlds: understanding ritual cave use in later prehistory 9783319990217, 9783319990224, 3319990225

Table of contents :
Between Worlds: Bridging the Divide between Method and Theory in Understanding the Ritual Use of Caves in Prehistory --
Part I: Theoretical Frameworks for Caves as Dynamic Spaces --
Caves as Vibrant Places: A Theoretical Manifesto --
Do Caves Have Agency? --
Animate Caves and Folded Landscapes --
Familiar Caves and Unfamiliar Landscapes: Approaching Cave-Use During the 1st Millennia in Britain --
How to Detect Ritual in Middle Bronze Age Central Italy? A Contextual Approach at Pastena Cave --
Pan Rituals of Ancient Greece Revisited --
The Watery Way to the World of the Dead: Underwater Excavations (Old and New) at the Cave of Han-sur-Lesse, Belgium --
Part II: Innovative Digital Capture Techniques in Cave Archaeology --
The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French Heritage Site --
From Macro to Micro: Multi-scalar Digital Approaches at the Sculptor's Cave, NE Scotland --
How a Community Digital Heritage Project has Helped to Imagine the Circumstances of Pictish symbols in the Wemyss Caves, Scotland --
Reverberant Spaces Preferentially Selected by Prehistoric Artists: Ritualistic Use of Sound Underscored by Choice of Subject Matter.

Citation preview

Lindsey Büster Eugène Warmenbol Dimitrij Mlekuž Editors

Between Worlds

Understanding Ritual Cave Use in Later Prehistory

Between Worlds

Lindsey Büster • Eugène Warmenbol Dimitrij Mlekuž Editors

Between Worlds Understanding Ritual Cave Use in Later Prehistory

Editors Lindsey Büster School of History, Classics and Archaeology University of Edinburgh Edinburgh, UK

Eugène Warmenbol Centre de Recherches en Archéologie et Patrimoine Université Libre de Bruxelles (ULB) Bruxelles, Belgium

Dimitrij Mlekuž Department of Archaeology Faculty of Arts University of Ljubljana Ljubljana, Slovenia

With thanks to Rachael Kershaw (University of Bradford) for preparation of illustrations.

ISBN 978-3-319-99021-7    ISBN 978-3-319-99022-4 (eBook) https://doi.org/10.1007/978-3-319-99022-4 Library of Congress Control Number: 2018957129 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Cover image: Terrestrial laser scanner in operation at the Sculptor’s Cave, north-east Scotland (Photograph: L. Demay). This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Contents

1 Between Worlds: Bridging the Divide Between Method and Theory in Understanding the Ritual Use of Caves in Later Prehistory����������������������������������������������������������������������������������    1 Lindsey Büster, Eugène Warmenbol, and Dimitrij Mlekuž Part I Theoretical Frameworks for Caves as Dynamic Spaces 2 Caves as Vibrant Places: A Theoretical Manifesto��������������������������������    9 Agni Prijatelj and Robin Skeates 3 Do Caves Have Agency?��������������������������������������������������������������������������   29 Rick Peterson 4 Animate Caves and Folded Landscapes������������������������������������������������   45 Dimitrij Mlekuž 5 Familiar Caves and Unfamiliar Landscapes: Approaching Cave Use During the First Millennia in Britain������������������������������������   67 Sam Wilford 6 How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach at Pastena Cave����������������������������������������������   87 Letizia Silvestri, Katia Francesca Achino, Micaela Angle, Maurizio Gatta, and Mario Federico Rolfo 7 Pan Rituals of Ancient Greece Revisited������������������������������������������������  113 Nektarios-Petros Yioutsos 8 The Watery Way to the World of the Dead: Underwater Excavations (Old and New) at the Cave of Han-sur-Lesse, Belgium����������������������������������������������������������������������  137 Christophe Delaere and Eugène Warmenbol

v

vi

Contents

Part II Innovative Digital Capture Techniques in Cave Archaeology 9 The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French Heritage Site����������������������������������������������  165 Albane Burens, Laurent Carozza, Raphaelle Bourrillon, Stephane Petrognani, Pierre Grussenmeyer, Samuel Guillemin, François Lévêque, Vivien Mathé, Yves Billaud, Aurélie Brodard, Pierre Guibert, Stéphane Jaillet, Olivier Jest, and Mathieu Koehl 10 From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave, North-East Scotland ����������������������������������������  199 Lindsey Büster, Ian Armit, Adrian Evans, Tom Sparrow, Rachael Kershaw, and Andrew S. Wilson 11 How a Community Digital Heritage Project Has Helped to Imagine the Circumstances of Pictish Symbols in the Wemyss Caves, Scotland ��������������������������������������������������������������  221 Joanna Hambly, Marcus Abbott, and Mike Arrowsmith 12 Hear Here: Prehistoric Artists Preferentially Selected Reverberant Spaces and Choice of Subject Matter Underscores Ritualistic Use of Sound����������������������������������������������������  251 Steven J. Waller Index������������������������������������������������������������������������������������������������������������������  265

Contributors

Marcus Abbott  York Archaeological Trust, York, UK Katia  Francesca  Achino  Quantitative Archaeology Laboratory, Department of Prehistory, Autonomous University of Barcelona, Bellaterra, Spain Micaela Angle  Soprintendenza Archeologia del Lazio e dell’Etruria Meridionale, Rome, Italy Ian  Armit  School of Archaeology and Ancient History, University of Leicester, Leicester, UK Mike  Arrowsmith  Centre for Archaeology, Technology and Cultural Heritage, University of St Andrews, Fife, UK Yves Billaud  UMR 5204 EDYTEM, CNRS, Le Bourget du Lac, France Raphaelle Bourrillon  TRACES-UMR 5608, CREAP, Toulouse, France Aurélie Brodard  UMR 5060 IRAMAT CRPAA, CNRS, Pessac, France Albane  Burens  Environmental Geography Laboratory, CNRS, UMR 5602 GEODE, Toulouse, France Lindsey  Büster  School of History, Classics and Archaeology, University of Edinburgh, Edinburgh, UK Laurent  Carozza  Environmental Geography Laboratory, CNRS, UMR 5602 GEODE, Toulouse, France Christophe  Delaere  Centre de Recherches en Archéologie et Patrimoine, Université Libre de Bruxelles (ULB), Bruxelles, Belgium Adrian  Evans  School of Archaeological and Forensic Sciences, University of Bradford, Bradford, UK Maurizio Gatta  Department of Archaeology, University of York, York, UK

vii

viii

Contributors

Pierre  Grussenmeyer  ICube Laboratory, CNRS, UMR 7357, Photogrammetry and Geomatics Group, INSA Strasbourg, Strasbourg, France Pierre Guibert  UMR 5060 IRAMAT CRPAA, CNRS, Pessac, France Samuel  Guillemin  ICube Laboratory, CNRS, UMR 7357, Photogrammetry and Geomatics Group, INSA Strasbourg, Strasbourg, France Joanna Hambly  School of History, University of St Andrews, Fife, UK Stéphane Jaillet  UMR 5204 EDYTEM, CNRS, Le Bourget du Lac, France Olivier  Jest  ICube Laboratory, CNRS, UMR 7357, Photogrammetry and Geomatics Group, INSA Strasbourg, Strasbourg, France Rachael Kershaw  School of Archaeological and Forensic Sciences, University of Bradford, Bradford, UK Mathieu  Koehl  ICube Laboratory, CNRS, UMR 7357, Photogrammetry and Geomatics Group, INSA Strasbourg, Strasbourg, France François  Lévêque  UMR 7266 LIENSs, CNRS, Université de La Rochelle, La Rochelle, France Vivien Mathé  UMR 7266 LIENSs, CNRS, Université de La Rochelle, La Rochelle, France Dimitrij  Mlekuž  Department of Archaeology, Faculty of Arts, University of Ljubljana, Ljubljana, Slovenia Centre for Preventive Archaeology, Institute for the Protection of Cultural Heritage of Slovenia, Ljubljana, Slovenia Rick Peterson  Archaeology, School of Forensic and Applied Sciences, University of Central Lancashire, Preston, UK Stephane Petrognani  UMR 7041 ArScAn, Ethnologie Préhistorique, MAE René Ginouvès, Nanterre, France Agni Prijatelj  Department of Archaeology, Durham University, Durham, UK Mario Federico Rolfo  Department of History, Culture and Society, University of Rome Tor Vergata, Rome, Italy Robin Skeates  Department of Archaeology, Durham University, Durham, UK Letizia Silvestri  Department of Archaeology, Durham University, Durham, UK Tom  Sparrow  School of Archaeological and Forensic Sciences, University of Bradford, Bradford, UK Steven  J.  Waller  Rock Art Acoustics, 1952 Sonoma Lane, Lemon Grove, CA, USA

Contributors

ix

Eugène  Warmenbol  Centre de Recherches en Archéologie et Patrimoine, Université Libre de Bruxelles (ULB), Bruxelles, Belgium Sam  Wilford  Florida Department of State, Bureau of Archaeological Research, Tallahassee, FL, USA Andrew S. Wilson  School of Archaeological and Forensic Sciences, University of Bradford, Bradford, UK Nektarios-Petros  Yioutsos  Department of Ancient History and Archaeology, National Kapodistrian University of Athens, Athens, Greece

About the Editors

Lindsey  Büster  is a Teaching Fellow in European Iron Age Archaeology  at the University of Edinburgh. With a PhD in later prehistoric roundhouses, her research interests include ritual and domestic life in later prehistoric Europe and complex later prehistoric funerary practices, which she is exploring through excavations at the Covesea Caves in north-east Scotland. Eugène Warmenbol  is Professor of North-Western European Archaeology at the Université Libre de Bruxelles. He is a specialist of the Bronze and Iron Ages, with interests in cave occupation and the dynamics of long-distance exchange. Formerly, he has spent time excavating in Egypt and Syria and has organised several exhibitions on Ancient Egypt. Dimitrij  Mlekuž  is a Research Associate and Lecturer at the University of Ljubljana and is Heritage Officer at the Institute for the Protection of Cultural Heritage of Slovenia. He completed a PhD on the Neolithic landscapes of the East Adriatic, and his main research interests are landscape archaeology, remote sensing and the Neolithic societies of south-eastern Europe.

xi

Chapter 1

Between Worlds: Bridging the Divide Between Method and Theory in Understanding the Ritual Use of Caves in Later Prehistory Lindsey Büster, Eugène Warmenbol, and Dimitrij Mlekuž

1.1  Worlds Apart: Caves in the Prehistoric Psyche Caves are unique and enigmatic places; and they make for rewarding and challenging archaeological sites. Whether features of karst or coast, they burrow into the earth, taking up position halfway between the visible above-ground world and the invisible world below. Even today they continue to attract visitors, with so-called ‘show caves’ drawing large numbers of people to view their towering caverns and otherworldly geological formations. If caves excite the imagination in us, then for prehistoric people, with very different views of the world—in which the veil between this world and the next was likely to have been much thinner—they truly were the entrances to the underworld. Located at the edge of inhabited landscapes— between land and sea, between above ground and below, between inside and outside, and between the sacred and the profane—they were liminal places in every sense and would have required careful physical and spiritual management. No wonder, then, that much of the prehistoric activity that took place in caves appears to have been highly ritualised and to have involved communication between worlds. Such activity often presents itself in (amongst other things) the deposition L. Büster (*) School of History, Classics and Archaeology, University of Edinburgh, Edinburgh, UK e-mail: [email protected] E. Warmenbol Centre de Recherches en Archéologie et Patrimoine, Université Libre de Bruxelles (ULB), Bruxelles, Belgium e-mail: [email protected] D. Mlekuž Department of Archaeology, Faculty of Arts, University of Ljubljana, Ljubljana, Slovenia Centre for Preventive Archaeology, Institute for the Protection of Cultural Heritage of Slovenia, Ljubljana, Slovenia e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_1

1

2

L. Büster et al.

of human remains or votive objects, and the creation of images on the cave fabric itself, but it may also have included less tangible elements such as song, dance and music. Since performance often lies at the heart of ritual, it may be that these latter elements were in fact the primary focus of activity, with the deposits recorded by archaeologists simply representing the material residues of more complex, intangible rites.

1.2  The Archaeology of Caves: Challenges and Opportunities Caves were, and continue to be, focal places in the landscape. The multi-period nature of many cave sites (intermittent and episodic if not continuous)—including those in this volume—demonstrates their persistent importance as points of reference in settlement and ritual landscapes. Continual visitation, however, displaces and fragments the archaeological record; things are moved, reused, tidied-up, added to and taken away; sometimes deliberately, sometimes unintentionally. Embedded in  local cosmology, the locations and stories associated with caves would have been passed down (at the very least) through folklore and oral tradition, so that their significance continued independently of physical visitation; indeed, some caves may have been places to actively avoid! This notoriety also made caves early targets for investigation: by antiquarians, cavers and tourists. Many early collections are lost, fragmented or, at best, survive without detailed contextual information (cf. Dowd 2015, Chap. 2). In somewhat ironic juxtaposition to their early focus for investigation, caves are often challenging places to visit, not least to conduct fieldwork to the modern standards enjoyed by more conventional above-ground archaeological sites. Karst systems can be extensive, narrow, dark and claustrophobic, whilst sea caves are often isolated and dangerous places at high tide. Their study requires meticulous planning; nimble work; minimal, robust and portable equipment; and often the help of others outside of the archaeological discipline (cavers, mountaineers, local guides, etc.). Caves do not give up their secrets easily. In contrast to many other kinds of archaeological site, the very nature of caves— their fabric, or anti-fabric, their negative voids—are integral to any interpretation of the ways in which they were used and perceived. The natural and the cultural are indivisible; each exists in a reflexive and symbiotic relationship with the other. This symbiosis is expressed as the ‘agency’ or ‘vibrancy’ of caves (Peterson, Chap. 3, this volume; Prijatelj and Skeates, Chap. 2, this volume), acting both on people and on the archaeological material which they lay down. Rockfalls break bones and fragment artefactual material; roof erosion buries what was once exposed on cave floors and erases what was once traceable on walls. In karst environments, speleothems grow like living organisms, whilst flowstone enmeshes people and things into the fabric of the cave itself (Prijatelj and Skeates, Chap. 2, this volume). In sea caves, wind and salt create conditions for preservation not witnessed elsewhere; bodies and artefacts do not behave as they should.

1  Between Worlds: Bridging the Divide Between Method and Theory…

3

1.3  Into the Light: Resurgence in Cave Archaeology Both the theoretical and methodological challenges of caves saw them previously overlooked in archaeological narratives, though recent years have witnessed a refocus and a resurgence in cave research on a global scale. Volumes such as Caves in Context (Bergsvik and Skeates 2012) and Sacred Darkness (Moyes 2012) have synthesised the current state of knowledge in a diverse range of contexts, with the latter including ethnographic studies which recognise the continued importance of these places in the worldviews of many communities around the world today. Others, such as Dowd’s Archaeology of the Caves of Ireland (2015) and Underground Archaeology (2016), seek to incorporate the archaeology of caves into long durée regional narratives, acknowledging them as integral to the ways in which past societies conceptualised and utilised the above-ground landscapes around them. More general volumes such as The Archaeology of Darkness (Dowd and Hensey 2016) have noted the experiential nature of caves, in which senses traditionally given primacy (e.g. sight, in a modern Western context at least) give way to heightened aural and haptic tenacity. Indeed, more recent approaches to understanding caves have drawn heavily on other disciplines; not just anthropology but psychology too (e.g. Montello and Moyes 2012). Caves are unique places that require bespoke theoretical frameworks.

1.4  B  ridging Divides: Theoretical Challenges and Technological Applications How do we make sense of these complex and ever-changing environments, and how do we capture this dynamism for display and dissemination to others? Furthermore, if each cave environment is unique, how do we compare them? How do we build a bigger picture? And how do we incorporate old and understudied assemblages into new narratives? Caves are often unsuited to conventional recording methods. They come in odd shapes and sizes; undulating walls mould deposits to their contours. How do we interpret this record when the matrices and the spaces within which we work are constantly changing? Furthermore, how do we communicate the dynamic nature of these places in two dimensions? Is it even possible, or do caves deserve more than this? In recent archaeoacoustic studies (e.g. Fazenda et al. 2017; Reznikoff 2008; Waller 1993), which acknowledge that sound—whether natural or artificially created—likely played an important role in the activities which took place in caves, negative space is equally (if not more) important than that which is tangible. Even that which we perceive as tangible may not be so in caves. Images carved into or applied to walls are not only difficult to photograph but emerge from the darkness— partially visible, partially hidden—and then disappear once again into the gloom (Pettitt 2016). How do we incorporate studies of the ephemeral and the intangible into more conventional archaeological narratives?

4

L. Büster et al.

This volume was borne from a conference session held at the 20th Annual Meeting of the European Association of Archaeologists in Istanbul in September 2014. Against a resurgence in cave archaeology, the development of ever-more innovative data capture technologies, and following consideration of the ritual use of caves in Early Medieval Europe (Bersgvik and Dowd 2018), Between Worlds seeks to bring together researchers utilising new theoretical and methodological frameworks in understanding these enigmatic spaces. Traditionally, there has been a degree of polarisation between arts- and science-based disciplines, and thus little collaboration between archaeological theorists and technical specialists. Archaeology, as an inherently interdisciplinary field, has perhaps been more forward-­thinking than others in this respect, but there still remains a divide between theory and practice, particularly when it comes to digital technologies. The unique environments and challenges presented by caves provide an excellent opportunity to demonstrate the potential benefits of collaboration within archaeology. Indeed, innovative digital capture techniques arguably represent the only way in which these dynamic three-dimensional spaces, and their more intangible characteristics, can be adequately recorded, understood and disseminated. They also have an important role to play in the heritage management of these fragile and vulnerable environments, which in many regions are major contributors to the tourist industry. Can we go further than this however? Can we use digital capture technologies to unlock aspects of our theoretical understandings of cave environments, and the ways in which they were used and perceived by prehistoric communities? Avenues of thought which were previously invisible to us? This volume sets out to explore such themes. The first part examines new ways of thinking about and experiencing caves, not only in terms of the ways in which they impacted upon the minds and bodies of those who encountered them, but also on the post-depositional and taphonomic processes which affected and manipulated the deposits left behind. Agni Prijatelj and Robin Skeates begin by summarising the unique nature of caves and, using a range of illustrative case studies, set out a ‘theoretical manifesto’ for the study of these enigmatic spaces. Rick Peterson explores this in greater detail and asks whether caves can be considered to have their own agency, in his examination of the dynamic and active nature of cave environments and the blurring between animate and inanimate worlds. Dimitrij Mlekuž continues with this theme but takes us into the surrounding environment, seeing caves as animate parts of ‘folded landscapes’. Sam Wilford picks up this thread, integrating caves into wider landscape studies and demonstrating the idiosyncratic nature of caves, and their status within prehistoric societies as transitional sites at the edges of inhabited landscapes, between familiar and unfamiliar worlds. Letizia Silvestri and colleagues follow by taking us on a site-specific tour of the Pastena Cave system, demonstrating the complex site-formation processes at play, and the detailed contextual analysis required to unpick the specific nature of the ritual activities which took place there. Bringing this section to a close, Nektarios-Petros Yioutsos explores a less tangible aspect of cave rituals: the central role that music must undoubtedly have played in the ‘Pan Rituals of Ancient Greece’; a theme which is picked up later in the volume.

1  Between Worlds: Bridging the Divide Between Method and Theory…

5

The second part explores new ways of investigating dynamic cave environments, with particular focus on digital capture technologies. Christophe Delaere and Eugène Warmenbol demonstrate the challenges of working within dynamic subterranean and subaquatic environments, particularly in the contextualisation of archival material. Albane Burens and colleagues then demonstrate the innovative, non-invasive, multi-scalar approaches used to record Bronze Age rock art at the cave of Les Fraux in south-western France; techniques which not only serve to protect it for future generations, but which are providing new insights into the bodily gestures and tactile relationships which existed between rock and artist. Lindsey Büster and colleagues continue this theme, demonstrating how the integration of different scanning methods can create rich interactive models for use by archaeologists, heritage professionals and members of the public, and allow ‘virtual visitation’ to these often vulnerable and difficult-to-access sites. Public engagement lies at the heart of the work by Joanna Hambly and colleagues at the Wemyss Caves. Here, photogrammetric recording of the carvings in this unique series of caves in Fife has been used to empower local community groups in the management of a once-threatened heritage asset and promote the caves as a catalyst for social regeneration. Finally, Steve Waller returns to the more intangible aspects of caves through the use of archaeoacoustic modelling, building on a growing body of evidence which suggests that the aural characteristics of these underground chambers were integral to the pattern and character of the activities which took place inside them.

1.5  Between Worlds: Caves—Past, Present and Future Caves are liminal places: located at the edge of the known world, they are portals to gods and ancestors, where physical and spiritual journeys are made. Caves preserve the vestiges of ephemeral rites otherwise wiped clean in the bustling everyday world of the living; through them the past speaks to us anew, revealing the complex and intangible ways in which communities understood (and continue to mediate) their place in the world. Caves straddle time. In the present, caves force us to rethink the confines of our discipline, to question traditional divisions between archaeologist and geologist, between archaeologist and caver, and between archaeologist and community. Edges become fuzzy and blurred. In caves, art and science, theory and practice, collide, as we seek ever-more innovative ways to find the tangible in the intangible, to measure the immeasurable, and to disseminate the inaccessible and the ‘out of bounds’. In many ways, then, caves lie between worlds: between past and present (and future), between disciplines, between sub-disciplines, and between the specialist and the non-specialist. This volume sets out the current state-of-the-art in cave archaeology. But momentum is strong, and as new challenges are met, and innovative ways to overcome them developed, this is unlikely to remain the status quo for long. New tools (both theoretical and methodological) promise new answers to new questions and have significant implications not just for cave archaeology but for the wider archaeological discipline beyond.

6

L. Büster et al.

References Bersgvik, K. A., & Dowd, M. (Eds.). (2018). Caves & ritual in Medieval Europe, AD 500–1500. Oxford: Oxbow Books. Bergsvik, K. A., & Skeates, R. (2012). Caves in context: The cultural significance of caves and rockshelters in Europe. Oxford: Oxbow Books. Dowd, M. (2015). The archaeology of caves in Ireland. Oxford: Oxbow Books. Dowd, M. (2016). Underground archaeology. Oxford: Owbow Books. Dowd, M., & Hensey, R. (Eds.). (2016). The archaeology of darkness. Oxford: Oxbow Books. Fazenda, B. M., Scarre, C., Till, R., Pasalodos, R. J., Guerra, M. R., Tejedor, C., Peredo, R. O., Watson, A., Wyatt, S., Benito, C.  G., Drinkall, H., & Foulds, F. (2017). Cave acoustics in prehistory: Exploring the association of Palaeolithic visual motifs and acoustic response. The Journal of the Acoustical Society of America, 142(3), 1332–1349. Montello, D. R., & Moyes, H. (2012). Why dark zones are sacred: Turning to behavioral and cognitive science for answers. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual use of caves (pp. 385–398). Boulder, CO: University Press of Colorado. Moyes, H. (Ed.). (2012). Sacred darkness: A global perspective on the ritual use of caves. Boulder, CO: University Press of Colorado. Pettitt, P. (2016). Darkness visible. Shadows, art and the ritual experience of caves in Upper Palaeolithic Europe. In M. Dowd & R. Hensey (Eds.), The archaeology of darkness (pp. 11–23). Oxford: Oxbow Books. Reznikoff, I. (2008). Sound resonance in prehistoric times: A study of Paleolithic painted caves and rocks. The Journal of the Acoustical Society of America, 123(5), 4137–4141. Waller, S. (1993). Sound reflection as an explanation for the content and context of rock art. Rock Art Research, 10(2), 91–101.

Part I

Theoretical Frameworks for Caves as Dynamic Spaces

Chapter 2

Caves as Vibrant Places: A Theoretical Manifesto Agni Prijatelj and Robin Skeates

2.1  Introduction In this chapter, we offer an alternative way of thinking about places in general, and caves1 in particular, as ‘vibrant’. This term acknowledges caves to be animate, hybrid places, constructed by various human and non-human agencies; ‘made up of matter, meaning and memory’ (Harmanşah 2015, p. 1); and embodying different temporalities. To illustrate the global applicability of these ideas, we draw upon a wide range of examples of archaeological and ethnographic research on human-­ cave entanglements, including our own research in the central Mediterranean region. More specifically, in the context of this volume, we also indicate how our approach can enhance understandings of caves as liminal places and ritual spaces in later prehistoric Europe and beyond.

2.2  Caves as Vibrant Places The idea of caves as vibrant places builds upon contemporary theoretical debate over relational persons, human and non-human agencies and network approaches. Central to all of these is an emphasis on minds, bodies, materials and things— enmeshed in various actions and in a constant state of flux, in a process of continuous creation, regeneration and transformation within distinct systems and dynamic networks. These ideas have recently been developed within the theoretical frameworks of material culture studies (e.g. Miller 2005), actor-network-theory (ANT)

 The term ‘cave’ is used here to include both caves and rockshelters, unless otherwise specified.

1

A. Prijatelj (*) · R. Skeates Department of Archaeology, Durham University, Durham, UK e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_2

9

10

A. Prijatelj and R. Skeates

(e.g. Latour 2005), political ecologies and anthropologies (e.g. Bennett 2010; Descola and Pálsson 1996; Haraway 1991; Ingold 2013; Latour 2004), human agency (e.g. Robb 2010), symmetrical archaeology (e.g. Olsen 2010; Olsen et al. 2012; Witmore 2007) and various other strands of interpretative archaeology (e.g. Alberti et  al. 2013; Boivin 2010; Conneller 2012; Fowler 2013a; Hodder 2012; Knappett 2005). What these discourses share is their negation of the traditional dichotomies of nature/culture, organism/environment, mind/body, subject/object, past/present, domestic/ritual and academic/indigenous. Instead, current approaches develop the idea that the listed binaries tend to have blurred or fluid boundaries. They also build on the premise that the traditionally perceived ‘passive or receptive’ elements of the pairs, such as nature, environment, objects, things and materials, actually have their own agency (e.g. Knappett 2005; Knappett and Malafouris 2008; Latour 2005; Olsen 2010; Olsen et al. 2012). In addition, they emphasise the tendency of the listed elements to interconnect on various levels to form dynamic, ever-changing systems and networks (e.g. Ingold 2000; Jones and Alberti 2013; Knappett 2011; Witmore 2007). In line with this current thinking, we reject traditional, de-personalised accounts of the archaeology of caves, centred primarily on questions of chronology, the typologies of artefacts and the scientific potential of buried human remains (e.g. Budja 1993; Merlatti 2001; Pergar 2007). We also call for the abandonment of the frequently employed portrayal of caves and rockshelters as a blank stage upon which the economic, social and symbolic are enacted. In other words, we argue that the biographies and meanings of such places cannot simply be thought of in terms of being uni-directionally ascribed by people—a way of thinking frequently found in the archaeological literature on the ‘human use of caves’, in which they are interpreted as dwelling places, animal pens, workshops, storage facilities, hiding places, ossuaries, theatres of ritual and so on (e.g. Bergsvik and Skeates 2012; Bonsall and Tolan-­Smith 1997; Dowd 2015, 2016; Gullì 2014; Mavridis and Jensen 2013; Moyes 2012a; Whitehouse 1992). Instead, we understand caves as participant, vibrant and hybrid places, made up of matter, meaning and memory, and entangled within a meshwork of human and non-human agencies, characterised by multiple symmetric and fluid connections between people, animals and plants, materials, things, supernatural entities, places and landscapes (cf. Bjerck 2012; Mlekuž 2012; Moyes 2012b, p. 5; Prijatelj forthcoming, 2019; Skeates 2007). In doing so, we are neither dismissing empirically grounded cave archaeology nor advocating the development of highly speculative archaeological interpretations of the place of caves in ancient religious thought and action (cf. Brady and Prufer 2005); instead, we are advocating a rapprochement of social theory, political ecology and scientific techniques in new approaches to caves. Caves must be observed through the prism of non-human as well as human agencies. Dobres and Robb (2005, p. 162) acknowledge that ‘social reproduction and culture change… depend fundamentally on the nexus of agency and materiality’. Social reality, then, rather than being intentionally imprinted by humans on their environment (and therefore created exclusively via human agency), emerges through various processes, relations and connections that occur between distinct nodes

2  Caves as Vibrant Places: A Theoretical Manifesto

11

within societal networks consisting of human and non-humans alike, all of whom possess a capacity to act. This non-anthropocentric, relational and symmetrical view of agency, which emphasises the mutually constitutive connections between humans and non-humans, presupposes the existence of material, multiple and collective agencies, and draws primarily upon ANT (e.g. Latour 2004, 2005) and various theories of assemblage (e.g. Bennett 2010; Fowler 2013b; Lucas 2012). ANT—rather than being concerned with identifying the agent whose intentionality triggers the act in question—views agency as a relational process that emerges from the connections between various nodes. From this perspective, we can explore distinct forms of sociality associated with caves by observing various ways in which caves are structured as networks or assemblages. More specifically, we can approach them as living, pulsating, mobile and open confederations of distinct entities and vibrant materials of various sorts, including water, stalactites and stalagmites, stone, minerals, dung, ash, people, animals, things, bones, plants, bacteria and chthonian spirits, to name but a few (Fig. 2.1). The term ‘vibrant places’ pays homage to Jane Bennett’s (2010) Vibrant Matter, which acknowledges the active participation of non-human forces in events. Bennett reflects on—amongst other issues—the failure of the vital electrical power grid during the 2003 blackout, which left 50 million people in North America paralysed (ibid., p.  24–28); the pulsating force of trash in landfills, generating poisonous streams of chemicals (ibid., p. 6); and the tangible ability of edible matter to influence our moods, dispositions and decisions (ibid., p. 39–52). By presenting such

Fig. 2.1  Caves as vibrant places and pulsating assemblages of distinct entities and vibrant materials (Collage: A. Prijatelj)

12

A. Prijatelj and R. Skeates

powerful examples, she advocates the vitality of matter, expressed as various types of non-human agency, and, in the process, introduces the political ecology of things. The latter should, according to Bennett (ibid., p. x), acknowledge the political voice of non-human actants and in this way ‘promote greener forms of human culture and more attentive encounters between people-materialities and thing-materialities’. We believe that more attentive encounters are needed between archaeologists and place-materialities, given the vibrant, hybrid and continuously changing nature of places. As locales of lived experience and social memory, places are ‘deeply historical, culturally contingent and politically contested sites of human engagement’ (Harmanşah 2015, p. 3). In addition, places are throbbing meshworks of various human and non-human agencies that co-operate or are in conflict with  and enhance or confound each other yet, at the same time, also act as a whole with a vital force that is distinct from the sum of each materiality considered alone. In the case of caves, their hybridity and vibrancy is co-shaped and co-produced by the many different geological, pedological, biological, social and cultural processes in play at these sites. It is essential to note that the vibrancy of many of the materials and processes involved exist independently of people, yet can significantly influence their relationships with particular caves—coming to the fore especially during ritual performances. Think, for example, of the precipitation of calcium carbonate and the consequent growth of stalagmites and stalactites in caves or the formation of shallow water pools. These formations can, over a geological timescale of thousands of years, significantly alter the interiors of caves, as well as actively influence people’s relationships with particular sites. Thus, in the cave complex of Grotta Scaloria in south-east Italy, Neolithic people ritually deposited ceramic storage and serving vessels below stalactites, on truncated stalagmites and adjacent to an artificially cut water pool in the deepest section of the complex, and the vessels eventually became fused to these features by calcite formations (Tinè and Isetti 1975–1980). Similarly, at Grutta 1 de Longu Fresu in Sardinia, a human skull was placed against the back wall of the cave and eventually merged with it into a hybrid, due to the deposition of a thick layer of calcite (Skeates et al. 2013) (Fig. 2.2). These numerous distinct materialities exert different powers. Some parts of hybrid, complex and volatile clusters may co-exist, cohabit and co-operate, while others may be in conflict with each other and cause or undergo friction. There is a further difference between the human and the diverse non-human and material agencies involved. For example, the potent physicality of a cave and the topology of its stone often directly influenced European Upper Palaeolithic artists who touched and marked the cave walls: with images concentrated around natural shafts, animals were depicted as if issuing from or disappearing into recesses at the ends of passages, and suggestive natural outlines and reliefs on wall surfaces were ‘completed’ with other parts of animal bodies (Clottes 2012, p. 20–21). The story of cave art thus serves as a reminder that the social reality of distinct places, rather than being imprinted by human agency onto the environment, emerges through various interactions between people, materials and places and therefore through the interplay of various human and non-human agencies. Hein Bjerck (2012, p. 48) is also thinking in a similar way when he describes caves as ‘connections to realms beyond the

2  Caves as Vibrant Places: A Theoretical Manifesto

13

Fig. 2.2  Longu Fresu, Sardinia: (a) special features, 1 hole at back of cave, 2 human skull, 3 alignment of stones, 4 wall painting (Photograph: J.  Veitch); (b) skull coated by calcite flowstone (Photograph: R. Skeates)

reach of humans’. He emphasises the fact that most caves do not have a definite termination; they have points beyond which the human body cannot pass but which smaller animals and supernatural powers can negotiate. One of his examples is Brusteinshola, a Bronze Age painted cave in northern Norway. Here, the cave painters may have consciously drawn the attention of ritual initiates to its ‘osmotic membrane’ by positioning a group of stickmen at the ‘end’ of the cave, on a flagstone and on the roof just above, at the point where the roof height decreases to 20–30 cm. Another evocative example concerns Triglavca rockshelter in Slovenia (Čok 2012; Prijatelj 2018). According to ethnographic records, a fertility rite at Triglavca took place every autumn until the 1830s. It involved a nightly visit to a rockshelter by a group of five villagers from Prelože, namely, an elder, who led the ceremony, and four maidens. At the site, the group placed grains of wheat, rye and buckwheat in a small pool on top of a stalagmite in which water dripping from the stalactite above collected. According to local lore, the two dripstones, and the water dribbling from one and collecting in a pool on the other, as well as the seeds, symbolised the

14

A. Prijatelj and R. Skeates

union of a man and a woman (Fig. 2.3). The ritual act of the elder placing the grains into the pool was accompanied by incantations by the maidens, calling for fertility. A few days later, the elder would return to the rockshelter alone, during the day, to collect the cereals which had sprouted in the meantime. On his way back to the village, he would secretly plant the young shoots in the devotees’ ploughed and sown fields, once again uttering an incantation for abundant and fertile crops (Čok 2012, p. 21–22). While different materialities clustered within caves and rockshelters interact with each other, they also act as a whole. More specifically, as an assemblage, they exert distributive agency which Bennett (2010, p. 35) compares to the Chinese concept of shi, described by her as the ‘energy, propensity, trajectory, or élan inherent to specific arrangements of things’ and also as ‘the dynamic force emanating from a spatio-temporal configuration rather than from any particular element within it’. The distributive agency of a particular place can be obvious or subtle. It can be at the very background of our perception. Alternatively, it can be very dramatic. And so, if we return to Longu Fresu, Middle Neolithic rites of passage drew upon not only the agency of the skull but also the distributive agency of an assemblage of other human remains, a cave painting, a deposited greenstone axe, a possible boundary line of stones, the cave walls, springs, darkness, diminutive light and shadows,

Fig. 2.3  Dripstone with a water pool at Triglavca rockshelter, a site of fertility rituals until 1830 (Photograph: A. Prijatelj)

2  Caves as Vibrant Places: A Theoretical Manifesto

15

and an inaccessible hole at the back of the cave (into which some of the bones were pushed). Places are also complex assemblages of different temporalities, with traces of the past inscribed in their materialities and with such past residues living on in the present. The importance of the time component for place (which is not something purely physical but rather a hybrid mix of matter, time, meaning and memory) is illustrated by Casey (1996, p. 26), who defines place as ‘more an event than a thing’. Place is thus in a constant process of becoming, from one moment to another moment, from one event to another event, and from one action to another action. Since the relationship between time and place is symmetrical, the inverse also applies: places are in a continuous process of formation by distinct events or actions, while all events and actions are, on the other hand, emplaced, therefore existing in particular places, rather than in the abstract space of a placeless vacuum (ibid., p. 35). Let us think of the material temporalities of caves for a moment. In their physicality, we can recognise the long-term rhythms of slow-moving geological processes, acting on a timescale of several hundred to several thousand years and being materialised in distinct geomorphological forms such as speleothems, cave conduits, gour pools or éboulis. Over the medium term, we might think of the ritualisation of caves through repeated ritual performances, where past, present and future come together but where time can also appear to stand still. There are also much shorter temporal flows, such as seasonal and yearly cycles of transhumance, for instance, which are embodied in interchanging deposits of humified and burnt dung layers within rockshelters that have been used as pens. Lastly, the short-term oscillations include singular events, such as the engraving of the name of a visitor in a cave wall, or the one-time use of a rockshelter as a refuge. The materiality of caves, then, is a hybrid assemblage not only of matter, meaning and memory but also of different temporalities. In practice, in order to understand the relationship between different temporal rhythms, caves and their mediation of the social, we need a multi-­ scalar (rather than unilinear) approach: one that will acknowledge different temporal dimensions of change through time (e.g. Robb and Pauketat 2013). Furthermore, by engaging with ‘deep genealogies of places without prioritising the horizons of meanings that have been acquired since their creation’ (Harmanşah 2015, p. 18), we can avoid the trap of viewing caves and rockshelters as buried in the distant past. Instead, we should regard them as dynamic, animate places that are part and parcel of contemporary, living landscapes. While continuities (and discontinuities) may be acknowledged by local people living in and with the landscape, they are too frequently brushed aside by archaeologists. One exception is Hamish Forbes, who, in writing on the Greek landscape of Methana, acknowledges ‘… whereas archaeologists generally consider past places as removed from the present and no longer part of the contemporary landscape, people in traditional societies who are integrated with their landscapes view history as part of a long process which includes the present’ (Forbes 2007, p. 4). Like landscape narratives, archaeological discourse on caves and rockshelters has generally detached them from the recent past and present—reducing them to pockets of (pre) history thatare of particular interest to scholars conducting research at particular

16

A. Prijatelj and R. Skeates

sites. In Slovenia, for example, few syntheses of caves and rockshelters have adequately addressed their use during antiquity or the Medieval period, not to mention the Modern period (for exceptions, see Hrobat Virloget 2015; Prijatelj 2018). Yet many caves have continued to serve as natural, cultural, social, economic, ritual, political, historical and archaeological places right up to and including the present. Trhlovca rockshelter, for example, was used as a temporary dwelling place from prehistory to the end of the Medieval period (Osole 1977; Leben 1988), and the archaeological material recovered has led to its designation as a heritage site. But few visitors learn that it was also associated with fertility rituals and divinations performed by local people from the villages of Lokev and Prelože, lending it significant place-power in the recent past (Čok 2012; Čok and Placer 2010). We must, then, think critically about cave narratives as constructed—generally separately—by scholars and local people. When concerned with archaeological accounts, we tend to trace the development of theoretical discourses and modern scientific techniques that have influenced archaeological thought on particular sites and to present narratives on caves and rockshelters that end with scientifically conforming, air-brushed discussions of their distant past (Bjerck 2012, p. 55–57). But we must also pay greater attention to the indigenous experiences, knowledge and belief systems that have shaped and reflected connections between local people and distinct places. In arguing this, we believe that archaeological discourse and indigenous knowledge represent two different perspectives on place-power that should be viewed as complementary rather than incompatible (cf. Ingold 2000, p.  14). Archaeological debates on caves and rockshelters should therefore also encompass issues of memory, myth, identity, storytelling, local knowledge production and contemporary engagement with the places in question. This ties in with recent calls and incentives for archaeology to include amongst its objectives the study of the recent past (e.g. Harmanşah 2015; Rathje and Murphy 2001; Schnapp et  al. 2004) and builds upon recent studies that attest to material entanglements of the past and present (e.g. Hodder 2012, p. 94–101; Olsen et al. 2012, p. 136–156). Archaeological discourse on caves also needs to address the heritage policies governing these sites and their management as heritage sites. Which sites, and what aspects of their natural and human biographies, are selected to be (or not to be) protected and presented for the public? How and why? In practice, this theoretical agenda, with its goal of conveying the richness, variety and complexity of ways in which caves mediate the social, calls for multiscalar and interdisciplinary research. Local-scale archaeological research is essential, identifying and examining an array of different agencies at individual caves and rockshelters, and in their deposits. This demands the use of a wide variety of complementary techniques. But caves must also be studied and understood in context (Bergsvik and Skeates 2012), particularly on a regional scale where more complete narratives of human-place entanglements can be explored through time and space. Archaeologists, historians, ethnographers and environmental scientists should collaboratively undertake field interviews and guided visits with

2  Caves as Vibrant Places: A Theoretical Manifesto

17

local people, ­systematic cave surveys, phenomenological experiments in and around caves, archaeological excavation, and the full spectrum of specialist postexcavation studies. Critical reflexivity is essential throughout. This has been widely employed in the social sciences (e.g. May and Perry 2011), including in archaeology (e.g. Bender et al. 2007), and consists of three levels of reflection. On the first level, reflexivity is an acknowledgement that one’s narrative is a product of specific social and historical contingencies. On the second level, it is a transparent description of the particular research methods and theories used, since the selection of these inevitably influences the construction of one’s interpretations. And, on the third level, it is an acknowledgement that one cannot achieve complete objectivity and removal from one’s subject matter. For example, when it comes to interpreting Pre-Colombian Mesoamerican cave rituals, Keith Prufer and James Brady (2005, p. 407) note that ‘the atheistic or nontheistic personal outlook of most archaeologists may contribute to the dismissal of values that they do not share’. One inescapable consequence of adopting a reflexive stance is the acknowledgement that places, including caves and rockshelters, are themselves inherently political—contested, appropriated and controlled. Think about sacred caves, for example. Power relations are inherent in the religious rituals performed in them. Tensions may surround the maintenance of secrecy about their locations. And different religious groups may fight for control of them. A host of other political tensions can also surround the designation, protection and touristic presentation of caves, such as Mammoth Cave in the USA (West 2010), as natural and cultural heritage sites. Likewise, caves and rockshelters in Slovenia have been political places over the centuries (Prijatelj 2011). The most recent, painful and contested example of this phenomenon is the use of karstic shafts as locales of post-war execution between May 1945 and January 1946, in the aftermath of the Second World War (Ferenc 2005). The topic remains (like other post-war communist crimes in Yugoslavia) highly divisive in Slovenian society and politics, despite official state recognition of the executed as political victims in the years following the declaration of Slovenian independence in 1991. A prehistoric example is the manipulation of power through cave rituals by Bronze and Iron Age groups from the hillfort of Škocjan (see Mlekuž, Chap. 4, this volume) (Fig. 2.4). Within the surrounding landscape, several caves with long-term prehistoric ritual use have been documented. Most notably, at the bottom of the deep vertical shaft of Mušja jama (Fig. 2.5), ritual deposits were identified consisting of hundreds of intentionally broken and partially melted bronze objects, including more than 240 spearheads, 17 socketed axes, 11 swords, 3 helmets, knives and sickles, bracelets, rings, fragments of bronze vessels, burnt animal bones and charcoal. The typology of the objects, dated between the twelfth and eighth centuries BC, indicates that their provenance stretched from Greece in the south to the Pannonian plain in the north and the Italian peninsula in the west. The importance of people living at Škocjan (also indicated by their wealthy grave goods) was clearly associated with their performance, maintenance and control of cave rituals within a

18

A. Prijatelj and R. Skeates

Fig. 2.4  Reimagining of a Late Bronze Age cave ritual adjacent to the Mušja jama shaft (Illustration: I. Rehar, Škocjan Caves Regional Park archive)

sacred landscape intimately related to the awe-inspiring natural karstic phenomena of the sinking river, the vast and complex Škocjan cave system, and deep vertical shafts (Leben 1990; Turk 2002). But the significance of caves can also be challenged: their uses performed, transformed and reinterpreted by different participants—archaeologists included. It is worth admitting, then, for example, that despite the Seulo Cave Project’s published emphasis on research results (Skeates et al. 2013), in practice the project faced a series of logistical and political challenges (limited funding, short field seasons, disturbed deposits, insufficient technical expertise, dependence on local informants, political corruption and so on), which inevitably constrained the quality of the published findings.

2.3  The Place-Power of Caves In this second section, we explore caves in more depth, using the language of phenomenology and the work of Edward Casey (1993, 1996) to think about four aspects of place-power that have rarely been considered in relation to cave narratives, despite their acute relevance to the ritualisation of caves.

2  Caves as Vibrant Places: A Theoretical Manifesto

19

Fig. 2.5  A large mound of rock rubble at the foot of Mušja jama, a 50 m deep vertical shaft where prehistoric votive offerings have been recorded (Photograph: J. Hanc, Škocjan Caves Regional Park archive)

2.3.1  Physicality Human relationships with caves are grounded in material. We come to know, perceive, experience, use, remember and imagine caves through and in relation to our material bodies, and our material bodies interact with the physicality of these particular places. Various basic spatial dimensions help us physically relate to and experience the constrained spaces of caves and rockshelters: above/below, infront/behind, to the right/to the left and close/distant (e.g. Casey 1993, 1996; Kant 1928; Tilley 2004; Tuan 1977; cf. Zubrow 2012). Sight is often dominant in our perception of place; yet hearing, smell, touch and balance also effectively contribute to our making sense of the world around us (e.g. Rodaway 1994; Skeates 2010; Tilley 2004). When sight is restricted—as in the dark zones of caves, for example—these other senses can come to the fore in our perceptions of a particular place. A diversity of embodied experiences can also be expected, according, for example, to the mental and p­ hysical

20

A. Prijatelj and R. Skeates

development and health of our bodies (cf. Casey 1996, p. 22). Similarly, one’s sense of place at a particular cave is contingent upon the wider cultural landscape and one’s social and historical context. The physicality of caves and rockshelters themselves also plays an important part in the process of emplacement (cf. Casey 1996, p.  24). As demonstrated by countless archaeological examples, the physical properties of caves represent one of the decisive factors in the selection of these places, and in their use for specific human activities, including dwelling and ritual performances. Places used for dwelling generally include rockshelters and the front areas of caves with large entrances, which create a relatively well-lit internal space and have a horizontal walking surface. Places utilised for various rituals, on the other hand, are frequently—although not always (e.g. Díaz-Andreu and García 2012)—located in deeper sections of caves. Here, darkness and artificial illuminations, restricted sound and air currents, restraints on movement and a heightened or suppressed sensorial experience— sometimes combined with extended isolation, fear, pain and/or disorientation—certainly impacted on people’s physical and spiritual experiences and perceptions, whether or not mediums, hypnotic music and the consumption of hallucinogens were used to induce a transcendental experience or altered state of consciousness (e.g. Clottes and Lewis-Williams 1998; Lewis-Williams 2002; Ustinova 2009a, b). Human engagements with the darkness of caves offer particular food for thought. As Marion Dowd and Robert Hensey (2016, p. xi) point out in the preface to their volume The Archaeology of Darkness, ‘Certain religious practices evident in the archaeological record have, at their core, an interaction with darkness’. In the same volume, Paul Pettitt (2016) argues that cave surfaces, art and shadows created by the constantly moving light of simple torches and stone lamps interacted to play a significant part in the ritual animation and perception of European Palaeolithic art in the dark zones of deep caves such as those at Lascaux, Altamira and El Castillo. He also points out that hand stencils reflect the importance of physical touch in the prosaic and ritual exploration of the wall surfaces of dark caves. Robin Skeates (2016a) likewise emphasises the way in which the particular effects of light and darkness mediated between people’s sensory experiences, cave architecture, material symbols and the supernatural during the course of ‘enlightening’ rites of passage navigated through later prehistoric caves in Seulo, Sardinia, such as Grutta 1 de Longu Fresu and Grutta de is Janas. Human engagements with different forms of water in cave rituals are equally provocative. Graeco-Roman oracle rituals in numinous caves sometimes involved priests drinking sacred water, as at Klaros, or breathing in toxic vapours, as at Hierapolis, in order for the medium to be enlightened through contact with, or possession by, a deity (Friese 2013; Ustinova 2009b). The Classic Maya also undertook pilgrimages to caves and sinkholes containing water pools, which they regarded as animate portals to the underworld. In Belize, at a time of climatic drying, they ceremonially petitioned the gods and ancestors for plentiful rain and maize, frequently leaving offerings of water jars (Lucero and Kinkella 2015; Moyes et  al. 2009). During the Bronze and Iron Ages in Europe, thousands of artefacts were similarly deposited in and around watercourses flowing through caves such as that of

2  Caves as Vibrant Places: A Theoretical Manifesto

21

­ an-­sur-­Lesse in Belgium (Delaere and Warmenbol, Chap. 8, this volume) and the H Pertosa Caves in Italy (Larocca 2010). And other culturally specific practices involving ‘abnormal’ forms of water can be envisaged for later prehistoric cave cults in Italy, including depositing storage vessels in the hot and steamy Grotta di San Calogero, collecting water dripping from stalactites in Grotta Scaloria, and drinking water from a pool at Pozzi della Piana (Whitehouse 1992, p. 132–133).

2.3.2  Interaction Such relationships are mutual: we act on caves, and they act back on us (Mlekuž 2012). It is precisely through these mutual interactions that caves become places: being transformed (sometimes ritualised) from purely physical natural locales into vibrant cultural assemblages of matter, meaning and memory. This reversibility of subject and object, and the action occurring between the body and the world around it, is, in fact, the central ontological thesis of much phenomenological writing (e.g. Casey 1993, 1996; Eyles 1985; Malpas 1999; Merleau-Ponty 1968): to touch the cave’s wall is to feel its touch on our hands; to look at the cave is to sense its returning gaze; and to be in the cave is for us to be part of it and for it to be part of us. Such a relationship between the body and the surrounding world is possible, according to phenomenological ontology, due to two factors. The first is the existence of blurred and fluid boundaries between the body and the world (e.g. Haraway 1991, 2008; Merleau-Ponty 1968). The second is—in the words of phenomenological poetics— the ability of the world to mirror our actions and therefore to return our gaze, to touch us back or to echo our listening. In the more prosaic language of social theory, this ability of the world to ‘act back’ is described by the term ‘material agency’. Of course, there is also an element of asymmetry in the relationship between the subject and the world: neither caves nor rockshelters are sentient. On the other hand, they, like us, possess agency, since they affect us through our bodies and provoke our senses, and consequently actively contribute to the structuring of our perceptions and ultimately our consciousness (cf. Peterson, Chap. 3, this volume). A few archaeologists have already begun to think along these lines in relation to caves. Drawing on the work of the environmental psychologist James Gibson (1986), Dimitrij Mlekuž (2012) has considered the ‘affordances’ provided by particular caves in the karst region of north-east Italy and western Slovenia, such as shelter for hunters against rain or wind, as shelter for herders and their sheep against the midday sun, and as passage for a shaman or mourners to the underworld. Mlekuž (ibid.) argues that, through the performance of such tasks, humans, animals and caves mutually constitute each other. In order for a relationship between a person and a cave to be established, there first needs to be movement. This can be either movement of the human body, which can take the form of various actions (such as food preparation, a burial or a votive offering) and relate the human body to a particular place, or movement within the ­ emories, human body in the form of sensorial experiences, emotions, thoughts and m

22

A. Prijatelj and R. Skeates

which again establishes a connection between the body and cave. If one strips such interactions down to their most basic level, one can, for example, get into the cave, stay in the cave, move within the cave, or move between the cave and other places (cf. Casey 1996, p.  21–24). Various types of movement and motion (or a lack thereof), then, build spatial, physical and cognitive relationships between person and place, enabling, amongst others, the reciprocal relation of place to be sensed and senses to be placed (Basso 1996). Imagine, for example, small groups of people moving through the artificial hypogeum of Ħal Saflieni in Temple Period Malta during the course of funerary rites of passage: walking, stooping, crawling and climbing through doorways, along pathways and via steps, deep into and up out of this animate underworld (Skeates 2008, p. 222–224).

2.3.3  Connecting Threads The place-power over us is not always equally strong, since it depends on the strength of the threads that connect a person and a cave (e.g. Casey 1993, 1996; Donohoe 2014; Eyles 1985; Malpas 1999). On the one hand, the threads between caves used as dwelling and/or ritual places and people were probably strong. As discussed by Casey (1996, p. 39), dwelling places can be seen as locales with particularly strong and durable connections to people, due to the great density of everyday actions occurring at such places. With this in mind, we can understand—in a way that the authorities never could—the distress of cave-dwelling communities when evicted from their ‘primitive’ and ‘insanitary’ cave houses at places such as the Petra caves in Jordan or Ghar il-Kbir in Malta, to be resettled in standardised accommodation (Bienkowski 1985; Messina 1989). Durable threads between people and places also emerge through the uniqueness and strength of a singular experience, as, for example, in the case of a ritual visit to a dark cave zone. By contrast, the threads connecting people and temporary shelters (or ‘inter-places’) were probably weaker. According to Casey (1996, p. 39), these are intermediate places that have weaker, less durable threads since they are traversed or visited only briefly by those moving from one particular place to another and therefore engage with them less frequently or less intensely.

2.3.4  Gathering Caves and rockshelters also ‘gather’ (cf. Casey 1996, p.  24–26, 38–39). This is another essential trait that adds to their vibrancy and place-power. Due to their morphology, caves act as containers and can, for example, gather various people, animals, things and deposits (some ritual), thus creating a meshwork of different nodes and agencies that can weaken, intensify or transform each other (e.g. Bennett 2010; Knappett 2011). Often, caves hold stuff together in a particular configuration

2  Caves as Vibrant Places: A Theoretical Manifesto

23

which ‘allows for certain things—people, ideas, and so forth—to overlap with, and sometimes to occlude others as they recede or come forward together’ (Casey 1996, p. 25). Think, for example, of the artificial caves dug into the soft sandstone of Kinver Edge in Staffordshire, UK, which were used as homes until the 1960s (Willetts 2010). When we look at one such home, within the Holy Austin Rock complex (Fig. 2.6a), the physicality of natural features, most notably the physicality of the sandstone—its solidness, redness, texture, cracks and folds—seamlessly flows into and merges with the physicality of cultural features, including a chimney that is built of bricks of a red hue similar to the colour of the sandstone itself and a window that is adjusted to the topography of the rock wall. Now, imagine stepping into one such home (Fig. 2.6b): the sight of the tiled floor and a wooden table covered with a white cloth, the sensation of warmth and the crackling of burning coal in the cast iron stove, and all other amenities necessary for cosy life would, we believe, occlude the notion of a natural place and foreground the power of the caves’ cultural features as a home. In addition to their ability to gather various material things, caves can also gather and hold in place thoughts, emotions, ideas and experiences. This ability stems from the relationship between place and memory (e.g. Casey 2000; Donohoe 2014; Jones 2007). Places not only impress themselves upon memory but are also impressed upon by memory (Donohoe 2014). Caves, then, have the ability to act as memory anchors and to hold the past in place. This contributes significantly to their vibrancy and active participation in building a sense of belonging, and to the preservation of cultural identities and traditions of both individuals and communities (e.g. Chapman

Fig. 2.6  Rock houses at Kinver, UK; (a) one of the houses from the outside (Photograph: I. Bryan); (b) young visitors in one of the restored houses, now part of the National Trust site (Photograph: © National Trust)

24

A. Prijatelj and R. Skeates

1998; Jones 2007; Proshansky et al. 1983). Caves such as Grotta Regina Margherita in central Italy (Skeates 2016b), which were repeatedly visited in the Middle Bronze Age (for the purpose of depositing the bodies of recently deceased persons and accumulating ancestors’ bones amongst spectacular stalactites which gradually incorporated them) certainly had this potential.

2.4  Conclusion Cave archaeology has come a long way since the beginning of the 1990s, particularly in terms of the range of interpretative approaches and ideas now available to scholars. However, we see no room for complacency and plenty of opportunity for rethinking the significance of caves in the past and present. Here, with the help of a range of theories, we have represented caves as animate, vibrant places that act on us and actively participate in the construction of social reality, and as pulsating assemblages of various human and non-human agencies that interact with social structures on different timescales. These ideas are widely applicable, particularly when it comes to thinking about ritual performance in caves. Indeed, we regard them as the basis for a new way of thinking about caves that can finally take us beyond their ‘human uses’. In practice, this will require us to develop an approach to cave archaeology that is more interdisciplinary, works across a wider range of spatial and temporal scales, and re-employs scientific techniques to tackle theoretically informed issues.

References Alberti, B., Jones, A. M., & Pollard, J. (Eds.). (2013). Archaeology after interpretation: Returning materials to archaeological theory. Walnut Creek, CA: Left Coast Press. Basso, K. F. (1996). Wisdom sits in places: Notes on Western Apache landscapes. In S. Feld & K. F. Basso (Eds.), Senses of place (pp. 53–90). Santa Fe, NM: School of American Research Press. Bender, B., Hamilton, S., & Tilley, C. (2007). Stone worlds: Narrative and reflexivity in landscape archaeology. Walnut Creek, CA: Left Coast Press. Bennett, J. (2010). Vibrant matter: A political ecology of things. Durham, NC: Duke University Press. Bergsvik, K. A., & Skeates, R. (Eds.). (2012). Caves in context: The cultural significance of caves and rockshelters in Europe. Oxford: Oxbow Books. Bienkowski, P. (1985). New caves for old: Beduin architecture in Petra. World Archaeology, 17(2), 149–160. Bjerck, H. B. (2012). On the outer fringe of the human world: Phenomenological perspectives on anthropomorphic cave paintings in Norway. In K. A. Bergsvik & R. Skeates (Eds.), Caves in context: The cultural significance of caves and rockshelters in Europe (pp. 48–64). Oxford: Oxbow Books. Boivin, N. (2010). Material cultures, material minds: The impact of things on human thought, society, and evolution. Cambridge: Cambridge University Press.

2  Caves as Vibrant Places: A Theoretical Manifesto

25

Bonsall, C., & Tolan-Smith, C. (Eds.). (1997). The human use of caves. Oxford: Archaeopress. Brady, J. E., & Prufer, K. M. (2005). Introduction. In J. E. Brady & K. M. Prufer (Eds.), In the maw of the earth monster: Mesoamerican ritual cave use (pp. 1–17). Austin, TX: University of Texas Press. Budja, M. (1993). Neolitizacija Evrope: Slovenska perspektiva/Neolithisation of Europe: Slovenian aspect. Poročilo o raziskovanju paleolita, neolita in eneolita v Sloveniji, 21, 163–193. Casey, E. (1993). Getting back into place: Toward a renewed understanding of the place-world. Bloomington, IN: Indiana University Press. Casey, E. (1996). How to get from space to place in a fairly short stretch of time: Phenomenological prolegomena. In S. Feld & K. F. Basso (Eds.), Senses of place (pp. 13–52). Santa Fe, NM: School of American Research Press. Casey, E. (2000). Remembering: A phenomenological study (2nd ed.). Bloomington, IN: Indiana University Press. Chapman, J. (1998). Objectification, embodiment and the value of places and things. In D. Bailey (Ed.), The archaeology of value: Essays on prestige and the process of valuation (pp. 106– 130). Oxford: Archaeopress. Clottes, J. (2012). Ritual cave use in European Paleolithic caves. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual uses of caves (pp. 15–26). Boulder, CO: University Press of Colorado. Clottes, J., & Lewis-Williams, J. D. (1998). The shamans of prehistory: Trance and magic in the painted caves. New York & London: Harry N. Abrams. Čok, B. (2012). V siju mesečine: ustno izročilo Lokve, Prelož in bližnje okolice. Ljubljana: Inštitut za arheologijo ZRC SAZU, Inštitut za slovensko narodopisje ZRC SAZU, Založba ZRC. Čok, B., & Placer, L. (2010). Trhlovca = lokavska Triglavca. Studia Mythologica Slavica, 13, 371–378. Conneller, C. (2012). An archaeology of materials: Substantial transformations in early prehistoric Europe. London: Routledge. Descola, P., & Pálsson, G. (1996). Nature and society: Anthropological perspectives. London: Routledge. Díaz-Andreu, M., & García, B. C. (2012). Acoustics and Levantine rock art: Auditory perceptions in La Valltorta gorge (Spain). Journal of Archaeological Science, 39, 3591–3599. Dobres, M.-A., & Robb, J.  E. (2005). ‘Doing’ agency: Introductory remarks on methodology. Journal of Archaeological Method and Theory, 12(3), 159–166. Donohoe, J. (2014). Remembering places: A phenomenological study of the relationship between memory and place. Lanham, MD: Lexington Books. Dowd, M. (2015). The archaeology of caves in Ireland. Oxford: Oxbow Books. Dowd, M. (2016). In search of darkness: Cave use in Late Bronze Age Ireland. In M. Dowd & R. Hensey (Eds.), The archaeology of darkness (pp. 63–74). Oxford: Oxbow Books. Dowd, M., & Hensey, R. (2016). Background and acknowledgements. In M. Dowd & R. Hensey (Eds.), The archaeology of darkness (pp. xi–xii). Oxford: Oxbow Books. Eyles, J. (1985). Senses of place. Warrington: Silverbrook Press. Ferenc, M. (2005). Prikrito in očem zakrito: prikrita grobišča 60 let po koncu druge svetovne vojne. Celje: Muzej novejše zgodovine. Forbes, H. A. (2007). Meaning and identity in a Greek landscape: An archaeological ethnography. Cambridge: Cambridge University Press. Fowler, C. (2013a). Dynamic assemblages, or the past is what endures: Change and the duration of relations. In B. Alberti, A. M. Jones, & J. Pollard (Eds.), Archaeology after interpretation: Returning materials to archaeological theory (pp. 235–256). Walnut Creek, CA: Left Coast Press. Fowler, C. (2013b). The emergent past: A relational realist archaeology of Early Bronze Age mortuary practices. Oxford: Oxford University Press.

26

A. Prijatelj and R. Skeates

Friese, W. (2013). ‘Through the double gates of sleep’ (Verg. Aen. 6.236.): Cave-oracles in Graeco-­ Roman antiquity. In F. Mavridis & J. T. Jensen (Eds.), Stable places and changing perceptions: Cave archaeology in Greece (pp. 228–238). Oxford: Archaeopress. Gibson, J. J. (1986). The ecological approach to the visual perception. Hillshade, NJ: Lawrence Erlbaum Associates. Gullì, D. (Ed.). (2014). From cave to dolmen: Ritual and symbolic aspects in the prehistory between Sciacca, Sicily and the central Mediterranean. Oxford: Archaeopress. Haraway, D.  J. (1991). Simians, cyborgs, and women: The reinvention of nature. New  York: Routledge. Haraway, D. J. (2008). When species meet. Minneapolis, MN: University of Minnesota Press. Harmanşah, Ö. (2015). Place, memory, and healing: An archaeology of Anatolian rock monuments. London: Routledge. Hodder, I. (2012). Entangled: An archaeology of the relationships between humans and things. Malden, MA: Wiley-Blackwell. Hrobat Virloget, K. (2015). Caves as entrances to the world beyond, from where fertility is derived. The case of SW Slovenia. Studia Mythologica Slavica, 18, 153–163. Ingold, T. (2000). The perception of the environment: Essays on livelihood, dwelling and skill. London: Routledge. Ingold, T. (2013). Making: Archaeology, anthropology, art and architecture. London: Routledge. Jones, A. (2007). Memory and material culture. Cambridge: Cambridge University Press. Jones, A. M., & Alberti, B. (2013). Archaeology after interpretation. In B. Alberti, A. M. Jones, & J. Pollard (Eds.), Archaeology after interpretation: Returning materials to archaeological theory (pp. 15–35). Walnut Creek, CA: Left Coast Press. Kant, I. (1928). The critique of judgement (1964 reprint). Oxford: Clarendon Press. Knappett, C. (2005). Thinking through material culture: An interdisciplinary perspective. Philadelphia, PA: University of Pennsylvania Press. Knappett, C. (2011). An archaeology of interaction: Network perspectives on material culture and society. Oxford: Oxford University Press. Knappett, C., & Malafouris, L. (2008). Material agency: Towards a non-anthropocentric approach. New York: Springer. Larocca, F. (2010). Le Grotte dell’Angelo a Pertosa: Il sistema sotterraneo e il giacimento archeologico. Pertosa: Fondazione Musei Integrati dell’Ambiente. Latour, B. (2004). Politics of nature: How to bring the sciences into democracy. Cambridge, MA: Harvard University Press. Latour, B. (2005). Reassembling the social: An introduction to actor-network-theory. Oxford: Clarendon Press. Leben, F. (1988). Novoodkrite prazgodovinske plasti v jamah na Krasu. Poročilo o raziskovanju paleolita, neolita in eneolita v Sloveniji, 16, 65–76. Leben, F. (1990). Arheološki ambient v porečju Reke in na Divaškem pragu. In G.  Cuscito, E. Halupca, & L. Foscan (Eds.), Reka – Timav: Podobe, zgodovina in ekologija kraške reke (pp. 321–351). Ljubljana: Mladinska knjiga. Lewis-Williams, J. D. (2002). The mind in the cave: Consciousness and the origins of art. London: Thames & Hudson. Lucas, G. (2012). Understanding the archaeological record. Cambridge: Cambridge University Press. Lucero, L. J., & Kinkella, A. (2015). Pilgrimage to the edge of the watery underworld: An Ancient Maya water temple at Cara Blanca, Belize. Cambridge Archaeological Journal, 25(1), 163–185. Malpas, J.  (1999). Place and experience: A philosophical topography. Cambridge: Cambridge University Press. Mavridis, F., & Jensen, J. T. (2013). Stable places and changing perceptions: Cave archaeology in Greece. Oxford: Archaeopress. May, T., & Perry, B. (2011). Social research and reflexivity: Content, consequences and context. Los Angeles, CA: SAGE.

2  Caves as Vibrant Places: A Theoretical Manifesto

27

Merlatti, R. (2001). Il complesso protostorico della Grotta delle ossa a San Canziano del Carso. Atti e Memorie della Società Istriana di Archeologia e Storia Patria, 49, 7–85. Merleau-Ponty, M. (1968). The visible and the invisible. Evanston, IL: Northwestern University Press. Messina, A. (1989). Trogloditismo medievale a Malta. Melita Historica, 10(2), 109–120. Miller, D. (Ed.). (2005). Materiality. Durham, NC: Duke University Press. Mlekuž, D. (2012). Notes from the underground: Caves and people in the Mesolithic and Neolithic karst. In K. A. Bergsvik & R. Skeates (Eds.), Caves in context: The cultural significance of caves and rockshelters in Europe (pp. 199–211). Oxford: Oxbow Books. Moyes, H. (Ed.). (2012a). Sacred darkness: A global perspective on the ritual use of caves. Boulder, CO: University Press of Colorado. Moyes, H. (2012b). Introduction. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual use of caves (pp. 1–11). Boulder, CO: University Press of Colorado. Moyes, H., Awe, J. J., Brook, G. A., & Webster, J. W. (2009). The ancient Maya drought cult: Late Classic cave use in Belize. Latin American Antiquity, 20(1), 175–206. Olsen, B. (2010). In defense of things: Archaeology and the ontology of objects. Walnut Creek, CA: AltaMira Press. Olsen, B., Shanks, M., Wemoor, T., & Witmore, C. (2012). Archaeology: The discipline of things. Berkeley, CA: University of California Press. Osole, F. (1977). Trhlovca pri Divači. Varstvo spomenikov, 21, 173–175. Pergar, M. (2007). Mala Triglavca 1979–1985: Opredelitev keramičnega gradiva mlajšega prazgodovinskega horizonta. Arheološki vestnik, 58, 39–63. Pettitt, P. (2016). Darkness visible: Shadows, art and ritual experience of caves in Upper Palaeolithic Europe. In M. Dowd & R. Hensey (Eds.), The archaeology of darkness (pp. 11–23). Oxford: Oxbow Books. Prijatelj, A. (2011). History of cave archaeology in Slovenia: Politics, institutions, individuals, methods and theories. Cave and Karst Science, 38(3), 137–143. Prijatelj, A. (2018). The form and fabric of Late Antique and medieval cave use in Slovenia. In K. A. Bergsvik & M. Dowd (Eds.), Caves & ritual in Medieval Europe, AD 500–1500. Oxford: Oxbow Books. Prijatelj, A. (forthcoming, 2019). Vibrant places beneath the surface: Caves and rockshelters in the Kras region, SW Slovenia. Unpublished PhD thesis, Durham University. Proshansky, H. F., Fabian, A. K., & Kaminoff, R. (1983). Place-identity: Physical world socialization of the self. Journal of Environmental Psychology, 3, 57–83. Prufer, K. M., & Brady, J. E. (2005). Concluding comments. In J. E. Brady & K. M. Prufer (Eds.), In the maw of the earth monster: Mesoamerican ritual cave use (pp. 403–411). Austin, TX: University of Texas Press. Rathje, W. L., & Murphy, C. (2001). Rubbish! The archaeology of garbage. Tucson, AZ: University of Arizona Press. Robb, J. E. (2010). Beyond agency. World Archaeology, 42(4), 493–520. Robb, J. E., & Pauketat, T. R. (2013). Theorizing scale and change in human history. In J. Robb & T. R. Pauketat (Eds.), Big histories, human lives: Tackling problems of scale in archaeology (pp. 3–33). Santa Fe, NM: School for Advanced Research Press. Rodaway, P. (1994). Sensuous geographies: Body, sense and place. London: Routledge. Schnapp, J., Shanks, M., & Tiews, M. (2004). Archaeologies of the modern. Modernism/Modernity, 11, 1–16. Skeates, R. (2007). Religious experience in the prehistoric Maltese underworld. In D.  A. Barrowclough & C.  Malone (Eds.), Cult in context: Reconsidering ritual in archaeology (pp. 90–96). Oxford: Oxbow Books. Skeates, R. (2008). Making sense of the Maltese Temple period: An archaeology of sensory experience and perception. Time and Mind, 1(2), 207–238. Skeates, R. (2010). An archaeology of the senses: Prehistoric Malta. Oxford: Oxford University Press.

28

A. Prijatelj and R. Skeates

Skeates, R. (2016a). Experiencing darkness and light in caves: Later prehistoric examples from Seulo in central Sardinia. In M.  Dowd & R.  Hensey (Eds.), The archaeology of darkness (pp. 39–49). Oxford: Oxbow Books. Skeates, R. (2016b). Caves in context: The place of caves in Middle Bronze Age southern Lazio, Italy. https://www.dur.ac.uk/archaeology/research/projects/all/?mode=project&id=734. Accessed 24 May 2016. Skeates, R., Gradoli, M.  G., & Beckett, J.  (2013). The cultural life of caves in Seulo, central Sardinia. Journal of Mediterranean Archaeology, 26, 97–126. Tilley, C. (2004). The materiality of stone: Explorations in landscape phenomenology. Oxford: Berg. Tinè, S., & Isetti, E. (1975–1980). Culto neolitico dele acque e recenti scavi Nella Grotta Scaloria. Bullettino di Paletnologia Italiana, 82, 31–70. Tuan, Y.-F. (1977). Space and place: The perspective of experience. London: Edward Arnold Publishers. Turk, P. (2002). Archaeology. In B.  Peric (Ed.), The Škocjan caves regional park (pp.  86–97). Škocjan: Park Škocjanske jame. Ustinova, Y. (2009a). Cave experiences and ancient Greek oracles. Time and Mind, 2(3), 265–286. Ustinova, Y. (2009b). Caves and the ancient Greek mind: Descending underground in the search for ultimate truth. Oxford: Oxford University Press. West, P. (2010). Trying the dark: Mammoth Cave and the racial imagination, 1839–1869. Southern Spaces. http://www.southernspaces.org/contents/2010/west/print/1a.htm. Accessed 23 March 2010. Whitehouse, R. (1992). Underground religion: Cult and culture in prehistoric Italy. London: Accordia Research Centre, University of London. Willetts, M. (2010). Living on the edge: The social and economic culture of the rock house communities in Kinver from 1700–1963. Unpublished MPhil thesis, University of Birmingham. Witmore, C. (2007). Symmetrical archaeology: Excerpts of a manifesto. World Archaeology, 39(4), 546–562. Zubrow, E. B. W. (2012). Caves and spatial constraint: The prehistoric implications. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual use of caves (pp. 371–383). Boulder, CO: University Press of Colorado.

Chapter 3

Do Caves Have Agency? Rick Peterson

3.1  Introduction This chapter draws on previous research around memory and human remains in caves, particularly on Neolithic artefacts and human remains from caves in Britain (Peterson 2013). In that paper, it was argued that long-term social memory was constructed at cave burial sites, among other places, through a combination of human actions and transformations to place. Central to this process was both the embodied agency of groups of mourners and the feedback received from objects, such as decaying corpses or cave spaces. It was suggested that the passage of time was indexed by both changes to caves and also by changes made by caves. Mourners had agency, but so did the recently deceased body being mourned and the distinctive natural formation where the burial took place. This analysis took for granted that theory around agency could be usefully applied, not only to living human actors but also to the dead and to the caves they were buried in. The second part of that assumption raised a number of substantial questions around what precisely we mean when we discuss the agency of non-human (or formerly human) actors. This chapter will review, therefore, the theoretical background to studies of agency in archaeology and related disciplines. In particular, it will focus on what we imply when we suggest that a geological structure, such as a cave, has agency. Recent wider research in cave archaeology has drawn upon the idea that the physical properties of cave systems can be said to ‘act’ upon people in various ways. Bjerck (2012) analysed the placement of Bronze Age anthropomorphic rock art in Norwegian caves in terms of the profound sensory and bodily experiences created by being in dark and restricted places. In this example, the response of human actors to a particular set of sensory cues seems to have led to particular parts of certain R. Peterson (*) Archaeology, School of Forensic and Applied Sciences, University of Central Lancashire, Preston, UK e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_3

29

30

R. Peterson

caves becoming ‘sacred’ or ritualised and therefore to have become appropriate places for the creation of rock art. In a somewhat similar case from the Slovenian Karst, Mlekuž (2011) draws on Gibson’s (1986) concept of ‘affordance’ to discuss how the bodies of Neolithic sheep and shepherds were created through repeated encounters with caves. Affordances are the specific properties that a cave possesses in relation to a particular human or animal user. Mlekuž analyses cave affordances by drawing on the concept of ‘affect’ as defined by Brian Massumi in his introduction to Deleuze and Guttari (1987, p. xvi). Massumi’s definition describes affect as pre-cognitive bodily responses to a physical encounter with another body. These are separate from conscious or emotional response and are held to underpin both consciousness and emotional responses to other bodies. Mlekuž (2011, p. 3–4) broadens this definition to look at how the constricted space of caves influences affect between human and animal bodies, and between bodies and the cave space itself. The cave space thus becomes a key part of the creation of new kinds of bodies associated with Neolithic domesticity and pastoralism. Both Bjerck and Mlekuž discuss the way in which cave spaces act upon living bodies. In her study of earlier Neolithic cave burials from Yorkshire, Leach (2008, p. 39) has identified a potentially important way in which caves and cave systems act upon the dead. At Cave Ha 3, near Settle, four individuals were buried within an actively forming tufa deposit while their bodies were still articulated. Leach (ibid., p. 51) suggests that this was a deliberate choice and cites other known examples of Early Neolithic votive deposition associated with tufa (Davies and Lewis 2004, p. 8). She argues (Leach 2008, p. 51) that the petrifying properties of tufa springs were actively incorporated into the burial rites at some caves, either to commemorate particular individuals or to hold them apart from the more collective rites used on other Early Neolithic burials in the region. In either case, it could be argued that the agency of the cave system was being deliberately invoked. Research such as this opens up a wider range of questions around the agency of caves and other natural places. What exactly do we mean when we talk about a cave having agency? Is agency even an appropriate term to use to describe something inanimate like a rock formation? Agency as a descriptive term is both powerful and fluid, but it has its origins in theory about human interaction and human social institutions. In applying it to the three examples above, we are presupposing a definition of agency which makes it a property of both inanimate objects and living subjects. We also need to consider how we differentiate between the dead and the living body. Similarly, we would need to question any distinction we might make between inhabited place and Cartesian space. Many different formulations of agency are possible, which would give different weights to these different properties. The aim of this chapter is to review some of the wider social theory about agency and, in particular, to compare theories of practice. By comparing theories of practice about agency, it is possible to move towards using agency theory, in the broadest sense, in thinking about the human use of caves.

3  Do Caves Have Agency?

31

3.2  Defining Agency 3.2.1  Structuration Theory Theory about agency in archaeology was initially developed by Barrett (1988) as an attempt to sidestep a dichotomy between two different perceptions of the archaeological record. Barrett drew extensively on both the structuration theory of Giddens (1984) and the work of Bourdieu (1977). He sought to move archaeological analysis away from studying patterns in artefacts (whether these were conceived of as analogous to texts or to fossils) and to find a methodology for thinking about the way that relationships between people were structured (Barrett 1988, p. 8–10). In Giddens’ early writings on the subject (e.g. Giddens 1979, p. 2–3), structuration is presented as a way of creating a theory of action in the social sciences. For Giddens (ibid., p. 51–53), theories of individual human action developed in philosophy had little impact on a sociology which was concerned almost exclusively with the analysis of large-scale social structures. Therefore, the theory of structuration can be seen as an attempt to overcome the disjunction between structuralist and Marxist accounts of modern human society: social determinist analyses which give a disproportionate weight to large-scale impersonal forces and a decontextualised philosophy of human intentions and actions. Giddens (1979, p. 56; 1984, p. 5) developed a ‘stratification’ model of the agent for this purpose. As we shall see throughout this chapter, geological metaphors of depth and stratification are an important common theme in theory about agency more broadly. Giddens’ stratification model begins with motivation of the action, moves on to rationalisation of the action, and then to reflexive monitoring of the action. However, it is important to realise that these three layers cannot be worked out independently of the surrounding structure, which Giddens characterises as both the unintended conditions of the action and its unintended consequences. Without this surrounding structure, there is nothing to motive the action, no context within which to rationalise it and no comparative standard for the reflexive monitoring of the action. Nested within this model of agency is another metaphorical set of layers to define human actions and their functioning in a specific way and within a particular set of conditions. Giddens’ (1984, p. 7) definition of the way an agent acts is grounded in a tripartite model of human consciousness, ultimately derived from psychoanalytic theory. At the lowest level are unconscious motives, above this is the realm of practical consciousness, and above that is discursive consciousness. These two models together provide a methodology for working out a recursive relationship between individuals’ thoughts and actions, and the structures which govern their society. Structuration theory analyses four important components to provide a model which aims to describe both large-scale institutions and individual human action. These are human action, social structure, power and time. The preceding paragraphs have described the analysis of human action. For Giddens (1984, p.  4), human beings are fundamentally ‘knowledgeable actors’. The interplay of practical and

32

R. Peterson

discursive consciousness ensures that people are able to carry out their routine lives. The conditions and consequences of actions within the stratification model of the agent ensure that knowledgeable actors understand wider social structures and how to use and influence them to achieve their own ends. The analysis of social structures (Giddens 1984, p.  25) also follows from the implications of the stratification model of the agent. In structuration theory, all institutions are made up of the actions of knowledgeable actors. They form the unacknowledged conditions of actions and are created from their unacknowledged consequences. Giddens (e.g. ibid., p. 16–34) analyses many different definitions of social institutions, but, for our purposes, the key part of structuration theory as it relates to social structures is his concept of the ‘duality of structure’ (Giddens 1979, p. 69). This states that social structures are both the medium within which actions take place and are created from the outcomes of these actions. Therefore, social institutions cannot be reified as something abstracted from human action (Giddens 1984, p. 34). This, as Barrett (1988, p. 8) has noted, is a very powerful argument for the archaeological utility of the duality of structure. The analysis of time and memory within structuration theory is similarly linked to the stratification model of the agent. Actions take place over time, they are influenced by the memory of past actions, and they will have consequences for future actions. However, what is important here is not the chronometric time of empirical measurement but the lived experience of time. This leads Giddens (1984, p. 35–36) to consider yet another stratigraphic system of classifications around the human experience of time. Day-to-day experience is regarded as essentially repetitive (‘reversible’ in the terminology Giddens borrows from Levi-Strauss). The life of any one individual, however, has a clear directionality arising from memory and bodily changes. Institutions, in Giddens’ view, have their own form of reversible time. The duality of structure discussed above operates within time; existing structures constrain actions and are created from them through time. This combination of the repetitive nature of reversible time and the directionality of individual lifespans is a fundamental part of the workings of the duality of structure. The analysis of power within structuration theory (e.g. Giddens 1979, p. 88–91) assumes that social rules and conventions are not neutral; they always serve someone’s or some group’s ends. However, as they are made up of the actions of people in society, they are also always open to being reworked and renegotiated as they are created and perpetuated. According to Giddens (1984, p. 15–16), power is implicated in all kinds of action; it is not something intrinsic to certain types of behaviour, such as domination or resistance. The duality of structure implies that resources can both constrain and enable actors in different ways and to different degrees. From the perspective of the archaeological analysis of agency, then, the central contribution of structuration theory is to place the emphasis of study on human action and bodily experience. The creation of the duality of structure (Giddens 1984, p. 25–26) from memory traces links both of these aspects.

3  Do Caves Have Agency?

33

3.2.2  Habitus With the notable exception of Mizoguchi (1993), archaeological studies drawing on pure structuration theory have been relatively rare. The early work of Bourdieu (especially Bourdieu 1977) has therefore been an important point of reference for the  integration of agency into archaeological work. For Barrett (1988, p.  27), Bourdieu’s work offered a corrective to a perceived lack of focus on the material world in Giddens’ writing, whereas for Gardner (2004, p. 7), Bourdieu allowed a more nuanced grasp of problems of subordination and domination. Bourdieu’s work is explicitly concerned with developing a theory of practice (Bourdieu 1977, 1990). Of particular interest here is the concept of ‘habitus’, the analysis of daily routines of everyday life. Habitus is the unconscious knowledge of what constitutes appropriate behaviour used by people to get through their day-to-­ day life. As such, it is both generally unarticulated—practical consciousness in Giddens’ (1984, p.  7) terms—and extremely culturally specific (Bourdieu 1977, p.  72). However, while Giddens is careful to point to a variable and permeable boundary between practical and discursive consciousness, Bourdieu (1985, p. 13) also stresses the ‘creative, active and inventive capacities of habitus’. Habitus is an example of Bourdieu’s belief that theory cannot be developed in philosophical isolation. It is necessary to apply theory to practice in order to develop and understand it (Bourdieu 1985, p. 12–14). Therefore, Bourdieu articulates the details of what habitus is by the use of concrete examples. Although his discussion of the relationship between structures and habitus draws on a wide range of philosophical positions about consciousness (Bourdieu 1977, p.  73–95), it does not require a specific, stratified model of how consciousness works for its effectiveness. In other works, Bourdieu (1985, p. 14) has talked about the need to find a way of discussing agency whilst getting away from the ‘philosophy of consciousness’. Bourdieu (1984, p.  170) moves away from a top-down view of structures by studying the taken-for-granted routines of daily life. In his phraseology, habitus becomes not only the ‘structuring structure’ but also the ‘structured structure’. That is to say, that social structures and institutions are both created from and reinforced by the actions of habitus. In both Giddens’ and Bourdieu’s work, the apparent circularity of this argument is overcome by invoking the action of memory. Bourdieu (1977, p. 87) discusses the concrete example of the way in which learning within the family underpins the way learning is experienced in school, which in turn underpins the way learning is experienced in later life. For archaeological applications of the concept of habitus, and particularly for cave archaeology, the most positive effect of Bourdieu’s insistence on theory as practice is its engagement with the material world. In numerous detailed examples, the structures developed in habitus are real physical structures. Whether it is on the French proletarian dining table (Bourdieu 1984, p. 193–200) or within the Kabyle house of Algeria (Bourdieu 1977, p. 89–91), what is being analysed are relationships between people, which are mediated through material culture and how that material culture is deployed in space.

34

R. Peterson

3.2.3  Environment and Agency Classical structuration theory ties agency very firmly to particular stratified models of human consciousness, and therefore caves, as inanimate objects, can’t plausibly be described as having this kind of agency. Caves do not have discursive or practical consciousness; they are not, therefore, in Giddens’ terms, agents. Bourdieu’s pragmatic focus on theory as practice and his focus on space and materiality offer a more promising programme for the study of the interaction between living people and caves. However, there is no sense within Bourdieu’s work that the Kabyle house, for example, has habitus as distinct from the Kabyle women and men who inhabit it. Building on the material perspective developed in the review of Bourdieu’s work, other strands of theory about agency can also be worked into a discussion of cave archaeology. Ingold’s essay Building, dwelling, living develops a detailed argument to dissolve the distinction between the cultural world of human agents with consciousness and the natural environment (2000, p.  172–188). Interestingly, Ingold (ibid., p. 172) begins by characterising the traditional divide between culture and the environment as a distinction between intentionally motivated human agency, the ‘social domain’, and the ‘ecological domain’ of relationships between animals and the environment. Ingold’s (ibid., p. 173) stated project is to find a ‘new ecology’ which rethinks the way in which both animals and humans are perceived as interacting with their environment. The problematic issue for Ingold is intentionality; what people do is intentionally motivated, while what animals do is apparently not. However, both Giddens (1984, p. 3–14) and Bourdieu (1977, p. 72) devoted considerable space to discussing non-intentional agency in people. Ingold (2000) focuses his argument on intentionality, discursive consciousness in Giddens’ terms, presumably as a rhetorical device to stress the synthesising power of his ‘new ecology’. Ingold (2000, p. 174–181) develops this ‘new ecology’ by resolving a perceived distinction between the structures made by animals, for example, spider webs, beaver lodges and ape nests, and structures designed and constructed by people. The former are usually described as the results of biological imperatives encountering particular ecological conditions and the latter as the results of intentional human design (ibid., p. 181). Ingold draws on Heidegger (1971, p. 145–161 particularly) to develop a ‘dwelling perspective’, to unify the description of both of these kinds of making and, therefore, of both human and animal agency. In the dwelling perspective, in common with structuration theory and habitus, agency is developed through the passage of time and through memory (Ingold 2000, p.  186). Both humans and other animals dwell within environments which have been fashioned by previous activity. Their actions are structured by those environments, and they in turn create new kinds of environment. Therefore, for Ingold (ibid., p. 187–188), agency does not operate in isolation from its surroundings. Both human and animal agents respond to pre-existing structures and conditions in similar ways, and create and modify these environments by their actions. The recursively organised connection between structure and agency has been embodied and materialised in a way which allows us to see no fundamental difference between the

3  Do Caves Have Agency?

35

building of beaver lodges and the building of houses. The dwelling perspective therefore suggests that we can ascribe agency to non-human animals. In other writing on this theme, Ingold (1993) elaborates on the role of temporality in the dwelling perspective. He coins the neologism ‘taskscape’ to describe an array of related activities, analogous to ‘landscape’ as an array of related features (ibid., p. 158). Taskscape is the material equivalent of Giddens’ structures; it is both the medium within which actions take place and is created from the outcomes of those actions. One of the important contributions which Ingold makes to this debate is to find a similarly embodied description of the way in which the passage of time is experienced. As we have seen, time and memory are a crucial part of the recursive cycle that lies at the heart of structuration theory, habitus and the dwelling perspective. Ingold (ibid., p. 159) uses the term ‘temporality’ to describe a conception of time—derived from the phenomenological understanding of Merleau-Ponty (1962, p. 416–421)—which is neither tied to specific models of human consciousness nor calibrated to an external constant. When people or other animals do things, they make time pass. Temporality is the time of the participant. Therefore, from Ingold, we can gather both a continuity of broad themes around agency, the recursive duality of structure and agency, and the importance of an experiential approach to time. However, what is novel in this work is the extension of Bourdieu’s concerns with objects and space, to see the material basis of both structure and time. The second major contribution of the dwelling perspective is to see that agency is not limited to human consciousness.

3.2.4  Art with Agency Although Ingold’s arguments are persuasive, they do not necessarily resolve the problem of environments with agency. Caves, as conventionally described, are inanimate objects, and therefore they do not spin webs or build nests. Accepting for the moment—although we shall return to this point later—the description of caves as inanimate, then further discussion is needed on the nature of the agency of inanimate objects. There is an extensive literature on the agency of such objects. An excellent review from an archaeological perspective is provided by Alberti and Bray (2009). As they (ibid., p.  339–340) point out, archaeological approaches to this problem have largely followed either the work of Gell (1998) or Latour (2005). The distinction between these two approaches can be characterised, following Alberti and Bray (2009, p. 340), as lying between those, such as Gell, who maintain a distinction between the intrinsic agency of living subjects and the ascribed agency of passive objects, and those, such as Latour, who would describe the agency of both people and things in the same way. For Gell (1998, p. 16), agency is what allows us to distinguish between ‘happenings’ (something which happens because of the consequences of natural physical laws) and ‘actions’ (which are caused by prior human intentions). Agency is therefore strongly identified not merely with human consciousness but also with deliberate human intentions.

36

R. Peterson

Despite this, Gell (1998, p. 13–17) goes on to develop a thesis which ascribes extremely powerful agency to art objects. In order to follow this argument, it is necessary to introduce two further pieces of terminology. Gell (ibid., p. 13) sidesteps the debate about how precisely we should define art, by developing the concept of the ‘index’. An index is any object which has the property of permitting people to make a ‘causal inference’. Gell (ibid.) uses the example of the smile, which we use to infer friendliness. This kind of inference is not a matter of axiomatic truth nor is it a linguistic convention; rather, it is a kind of synthesis based on the informal probabilities of previous experience. Gell (ibid., p. 14–15) borrows the term ‘abduction’ from semiotics to describe this kind of inference. An index, therefore, is any object which permits abductions to be made from it. However, in order to focus his argument more tightly on the agency of objects, Gell (ibid., p. 15–16) further restricts the definition of an index to stipulate that it must permit an abduction and that some kind of social agency must lie behind the creation of the index. The important point which Gell (ibid., p. 17) makes in defining this agency is that it does not have to be part of a ‘philosophically defensible’ system of thought about agency. It merely has to be an example of a way in which agency is or has been thought about by people. Gell (1998, p. 19) uses the example of ‘vehicular animism’—the modern western habit of ascribing personalities and power to cars—to illustrate the way in which this agency can be perceived to belong to non-human agents. This habit of mind is regarded by Gell as fundamentally irrational but is nevertheless an example of a commonly held belief about agency. It is therefore a potential object of study, but it throws up a significant paradox when considered in the light of the definition of agency cited above, in which human ‘action’ is contrasted with natural ‘happenings’. To resolve this paradox, Gell (ibid., p. 20–21) suggests that we classify agents into two groups: ‘primary agents’, which are intentional beings, and ‘secondary agents’, which are objects through which primary agents distribute their agency. Gell (ibid., p.  140–141) draws on the idea of distributed personhood, especially Wagner (1991), to stress that secondary agents are not in any sense inauthentic but rather that they are the distributed material aspects of primary agency. For Gell, objects have agency as distributed parts of the people who have made and used them. Any other agent encountering such an object is able, through a process of abduction, to infer things about the primary agent. Such an object ‘embodies intentionalities’ (Gell 1996, p. 36).

3.2.5  Actors and Networks While Gell’s characterisation of the agency of inanimate objects is very persuasive, this is not a completely helpful solution to the problem for cave archaeologists. It does not seem to provide us with a clear and obvious methodology that we should use if we want to identify what it was about a cave which may have led people in the past to ascribe agency to it. Gell’s characterisation of object agency is an example

3  Do Caves Have Agency?

37

of what Pels (1998, p.  94) would describe as ‘animist’ object agency, in which things have life because an external soul or spirit is perceived as inhabiting them. Pels (ibid., p. 95) contrasts this with ‘fetishist’ object agency, in which an object’s power comes from the very nature of the materials of which it is comprised. This internal, materialistic perspective leads us to consider a whole range of different approaches to object agency, which have three things in common. These are a fundamental critique of the distinction between active subjects and passive objects, a focus on the relationships between objects and people, and a ‘flat ontology’ which does not prioritise one kind of agent or structure over another. Examples of this kind of thinking include the ‘relational realist’ archaeology proposed by Fowler (2013, p. 20–67), the ‘symmetrical archaeology’ described in Shanks (2007), and in the broader social sciences, the ‘assemblage theory’ associated with De Landa (2006). All of these writers draw to a greater or lesser extent on the ‘actor-network-­ theory’ developed by Latour (2005) and colleagues. This provides us with a very different way of thinking about the agency of inanimate objects to that proposed by Gell. At the heart of actor-network-theory is Latour’s (ibid., p. 70–74) critique of the distinction between active subjects and passive objects. In contrast to Gell’s (1998, p.  20–21) categorisation of primary and secondary agents, Latour (2005, p.  46) declares ‘an actor is what is made to act by many others’. He adopts the technical term ‘actant’ to describe this property of making a difference, which can apply equally to people, animals or objects. The ‘actant’ is introduced in a way which does not require it to possess any kind of consciousness or intentionality: Kettles ‘boil’ water, knives ‘cut’ meat, baskets ‘hold’ provisions… if action is limited a priori to what ‘intentional’, ‘meaningful’ humans do it is hard to see how a hammer could act… any thing that does modify a state of affairs by making a difference is an actor… the question to ask about any agent is simply the following: Does it make a difference in the course of some other agent’s action or not? (ibid., p. 71)

For Latour (2005, p.  72), objects as actants allow, afford, permit, encourage, block or forbid actions. They don’t themselves ‘have agency’, but then, neither is there a separate category of humans with intentions who possess agency. Agency exists in the network of relationships between the actants. Relationships between actants are central to actor-network-theory, precisely because Latour (2005, p. 75–76) does maintain that there is a significant difference between subjects and objects. Actor-network-theory can potentially analyse all networks that involve human social agency; it is not supposed to prejudge which of the actants in these networks has priority, but the presence of human subjects is necessary for the analysis to take place (ibid., p. 78). It is not, therefore, ‘symmetrical’ in quite the way in which the  term is used by Fowler (2013, p.  30). Latour (2005, p. 171–172) is also sceptical about the usefulness of both the highly localised study of actors’ agency and the study of wider social structures. This leads to a concern with metaphorical ‘flattening’: a focus on the descriptive analysis of the connections between actants, without making a priori assumptions about the kind of actant they are.

38

R. Peterson

Theory of practice in actor-network-theory draws on the analogy of map-­making; connections should be described and followed wherever they lead, without disjunctures or jumps between different scales of analysis (Latour 2005, p. 174). This focus on local connectedness needs, however, to recognise that the local situations which are being described are not created from nothing. Latour (ibid., p. 194–195) replaces the ‘structure’ of structuration theory with actual, physical things. Places and objects are actants in the network and they act, in a somewhat similar way to Gell’s (1998, p. 14–16) indexes, to connect the network described in the local present with other previously existing networks. Places and objects are also embedded in different times (Latour 2005, p. 200–201); in any given interaction, some of them will be ancient and others newly created. The third major practical implication of actor-­ network-­theory is that the descriptions it furnishes are necessarily incomplete. Latour (ibid., p.  246) is clear that we should resist the temptation to ‘fill in the blanks’, to assume that some ineffable essence of society is held between the connecting strands of the network. From the perspective of actor-network-theory, caves can be considered as actants. They are places and objects which would be drawn into the network of connections associated with any human use of them. As physical and temporal spaces, they would also act to link older and more recent networks together. Each of the different discussions of agency which have been considered makes a different kind of contribution to the study of caves as agents. Ultimately, the question for us is not which of these formulations of agency are right—or even which is the most persuasive as an abstract description—but which provides us with a workable programme of research.

3.3  Towards an Archaeology of Active Caves 3.3.1  How Do Caves Act? The first point which arises from the preceding discussion is the danger of reifying ‘agency’ as a social force: something which is somehow above or behind the actions of people and things. Following the lead of Ingold (2007), who sought to shift from the study of ‘materiality’ to the study of the ‘properties of materials’, we should transpose the opening question of this chapter from ‘do caves have agency?’ to ‘how do caves act?’ Throughout this chapter we have been concerned with theories of practice for this reason. How do different thinkers describe the ways in which people and things act? The first common thread we can see is that how people and things act is strongly influenced by what they are like. They act in the way they do because of the properties they have. For Giddens (1984, p. 4), the property which matters is the stratified model of consciousness; the specific interplay between practical and discursive consciousness. In Bourdieu’s (1977, p.  72) work, a similar role is played by the

3  Do Caves Have Agency?

39

c­ ulturally specific ‘habitus’ of any individual. Both of these concepts are embodied, material and also historically contingent (Rorty 1989, p.  30–43). Ingold (2007, p. 14) gives us a similar way of thinking about why things act. The properties of materials are parts of processes and relations; they are experienced, and once again they are embodied and historically contingent. Similarly, for Gell (1996, p.  36), things which can act are things which embody intentionalities. They have the property of being visibly altered or placed in a way which makes them the kind of object which is comprehensible to an observer as part of the distributed agency of another person. In Latour’s work (2005, p. 71), the way that both people and objects make a difference is governed by their material form. They allow, encourage, facilitate or block actions, depending on what they are made from.

3.3.2  Properties of Caves The way that caves act will, therefore, be structured by their physical form. If we think about Ingold’s (2000, p. 186) definition of the dwelling perspective as relating to something which affects and is affected by its environment, then we can broaden this insight beyond the living non-human animals which form the bulk of his examples. Both the caves cited from Bjerck (2012) and Mlekuž (2011) at the start of the chapter and the decomposing bodies described by Leach (2008) affect their environment and are affected by it. Caves act in the way they do, at least in part, because they are constricted spaces. However, across later prehistoric Europe, we also have evidence for the importance of other properties of caves. They are dark, underground spaces, often difficult to access and might be summed up as sharing the property of separateness (see e.g. discussions in Dowd and Hensey 2016). At Glencurran Cave, Co. Clare, Ireland, excavations by Dowd (2009, p.  98) showed evidence for the Late Bronze Age deposition of lithic and shell beads in association with neonatal remains of humans and wild and domestic animals. This practice took place at least 50 m into the cave, in the zone of permanent darkness, and much deeper into the system than the Middle Bronze Age burials from the same site (ibid., Fig. 10.2). The deposit is interpreted by Dowd (ibid., p. 98) as a single complex event, gaining at least part of its meaning and power from its location in the dark zone of the cave. Sites in Scotland also indicate the way in which the physical separateness of caves could be both drawn upon by and act on people and objects. At High Pasture Cave, Skye, passages appear to have been deliberately modified in the Middle Iron Age to create a stairway down into an extensive cave system (Birch et al. 2005). At the Sculptor’s Cave, Covesea, there is evidence for ritual activity in both the Late Bronze Age and Late Iron Age. This is a site which is inaccessible at high tide and  would have required very specific knowledge of such  natural phenomena to approach at all (Armit et al. 2011, p. 254–255; Büster et al., Chap. 10, this volume; Ian Armit pers. comm.).

40

R. Peterson

Caves also act in more physical ways; water, mud and scree all flow through them, often substantially moving any objects within the cave, as  for example at Pontnewydd Cave in north Wales (Aldhouse-Green et al. 2012, p. 255). There are good examples of how the active properties of caves and their associated fluvial systems were drawn upon in later prehistory. At the Trou de Han, Han-sur-Lesse, Belgium, large-scale ritual deposition of metalwork took place in the Late Bronze Age in two specific locations of the underground sections of the River Lesse (Delaere and Warmenbol, Chap. 8, this volume; Warmenbol 2014, p.  69–73). Further south, in Charente, France, Manem (2012, p. 142–149) discusses the consistent association between cave deposition of pottery in the Middle Bronze Age Duffaits Culture and river submergences. At both Perrats Cave and Duffaits Cave, there was deliberate deposition of groups of pot sherds made by the same potter, and this practice was shared with the nearby enclosure site at Fouilloux. All three sites are within 800 m of the submergence points of four substantial rivers. Manem (ibid., p. 146–149) suggests that, in the case of the Duffaits Culture, we are witnessing a specific depositional practice focused on karstic aquifer features. In these cases, we see practices which draw on two interconnected kinds of separation—of being both underground and underwater—to move artefacts away from the everyday world. Caves also have the ability to trap and petrify material in speleothem deposits. In addition to the tufa-encrusted material described by Leach (2008, p. 51), prehistoric human bone has been reported from within flowstone at Carsington Pasture Cave, Derbyshire (Barnatt and Edmonds 2002, p.  117), and from Longu Frescu Cave, Sardinia (Skeates 2016, p. 41; see also Prijatelj and Skeates, Chap. 2, this volume). Detailed evidence of the interaction of human and cave agency around speleothems is provided in Whitehouse’s (2015) review of the cult of ‘abnormal water’ in Italian Neolithic caves. In Puglia, at Grotta Scaloria, and probably also at Grotta di Porto Badisco, pottery vessels were placed on broken stalagmite surfaces, apparently to collect the precipitating water from the stalactites above. These vessels in turn became petrified parts of the remodelled stalagmites (ibid., p. 57–58). Therefore, in the Italian Neolithic, we can see direct evidence of the interplay between human and material agency in caves through time. This material evidence of the role of time brings us back to our consideration of theories of practice about how people and things act. Alongside the structure of a caves’ physical form, the second common thread in all of the theories of agency we have considered is a concern with the passage of time and memory.

3.3.3  Caves, Time and Memory In all the theories about agency we have considered, the passage of time is fundamental to the recursively organised nature of action, the ‘duality of structure’ as described by Giddens (1979, p. 69). In the work of both Giddens (1984, p. 35–36) and Bourdieu (1977, p. 87), what is important is the lived experience of time. In essence, people act as a consequence of their memories and create a framework for

3  Do Caves Have Agency?

41

future actions by remembering the consequences of past actions. With the dwelling perspective, Ingold (2000, p. 186) develops an environmentally embodied version of this recursive ‘duality of structure’. Central to this is the definition of temporality as the ‘time of the participant’ (Ingold 1993, p. 159); when things act they experience time passing. We can strengthen this insight by drawing on the way that Gell (1998, p.  13–15) defines an index. Temporality is not simply experienced, it is embodied and materialised in the scars and traces that these actions leave on the world (Peterson 2013, p. 273–274). Latour (2005, p. 200–201) points out that both places and objects are embedded in different times. They contain within themselves the histories of their previous manufacture or use. In actor-network-theory all of these connections are potentially informative, and they should all be followed. It follows that we should analyse the ways in which caves act through this embodied and material experience of time. Caves act as they do because people have previous embodied experience to draw on when they encounter them. They are also able to act through the indexes and traces that are left in them. For example, the Romano-British cemetery established in chamber 4 at Wookey Hole (Hawkes et al. 1978) indicates repeated use of one particular place within the cave. A minimum of 28 individuals were placed on a spit of cave sediment at the back of chamber 4 during the later third century AD. Enough skeletal material remained in situ to show that the bodies were not placed in grave cuts and that waterborne silts accumulated around them as they lay on the surface (ibid., p. 26–28). The types and frequency of grave goods at Wookey Hole can be paralleled at other late third-century cemeteries in south-western England; in many ways these were entirely typical burials for the period (ibid., p. 30–31). Presumably they drew upon a habitus of social responses to death which had developed across the region (and province) as part of the transition to inhumation burial in the third century AD. However, they did so in such a way that responded both to a particular cave environment and to what appears to have represented around 100 years of cemetery use (ibid., p. 29). This time depth would have been indexed by the presence of bodies in various stages of decomposition and covered to different degrees by periodic silting from the flooding of the River Axe. The location of the cemetery at the back of the deepest accessible part of the system, 200  m from the entrance, was probably, as Hawkes et  al. (ibid., p.  30) suggest, driven by a belief that the point of emergence of the Axe from the rock in chamber 4 was spiritually significant. Within this model, the River Axe was itself more than a passive natural phenomenon to be referenced; it was an active contributor to the ongoing funerary activity at the site. We can see the impact of these interactions particularly in the longer-term fate of parts of these bodies after the Roman period. Flooding events within the cave moved the crania from many of these burials and caused them to be redeposited at different locations nearer to the cave entrance (Hawkes et al. 1978, p. 25). The cave removed the skulls and collected them in a new deposit. This is an action which was only possible because of the past geological history of the streamway, in combination with the hydrodynamic properties of particular skeletal elements; the disarticulated crania in particular would float (Nawrocki et al. 1997). It was also only possible, however, because of the past decision to place the Romano-British bodies on the

42

R. Peterson

sediment surface, rather than burying them, and to place them close to the emergence of the Axe in chamber 4. Caves, then, do act. We can describe their actions in terms which draw upon agency theory, to give ourselves a powerful theory of practice to interpret those actions. It may be objected, however, that we have stretched some of these definitions almost to breaking point. In the example given above, all of the actions ascribed to the cave can also be described as the result of geochemical or biological processes which would have taken place regardless of any further human intervention. Did the cave then ‘act’? To answer this question, we can return to Gell (1998, p. 19) and his insight that we do not need to describe ‘philosophically plausible’ agency for something to have acted. Caves undoubtedly have acted, and people, who shared their material, temporal and embodied world, would have perceived them to have had agency. Acknowledgements  Thanks to Lindsey Büster, Dimitrij Mlekuž and Eugène Warmenbol for the invitation to speak at the 2014 European Association of Archaeologists conference in Istanbul and for the further invitation to submit this chapter to the present volume. Particular thanks are due to Lindsey who kindly read my contribution in Istanbul when I was unfortunately unable to be physically present. The arguments in this chapter have benefitted from extensive discussions about agency and caves with many colleagues, but I should particularly thank Vicki Cummings, Julia Roberts, David Robinson and Duncan Sayer. Thanks are also due to Josh Cameron who introduced me to the detail and implications of the Wookey Hole sequence.

References Alberti, B., & Bray, T. (2009). Animating archaeology: Of subjects, objects and alternative ontologies. Cambridge Archaeological Journal, 19(3), 337–343. Aldhouse-Green, S., Peterson, R., & Walker, E. (2012). Neanderthals in Wales: Pontnewydd and the Elwy Valley caves. Oxford: Oxbow Books. Armit, I., Schulting, R., Knüsel, C. J., & Shepherd, I. A. G. (2011). Death, decapitation and display? The Bronze and Iron Age human remains from the Sculptor’s Cave, Covesea, north-east Scotland. Proceedings of the Prehistoric Society, 77, 251–278. Barnatt, J., & Edmonds, M. (2002). Places apart? Caves and monuments in Neolithic and earlier Bronze Age Britain. Cambridge Archaeological Journal, 12(1), 113–129. Barrett, J.  (1988). Fields of discourse: Reconstituting a social archaeology. Critique of Anthropology, 7(3), 5–16. Birch, S., Wildgoose, M., & Kozikowski, G. (2005). Uamh an Ard Achadh (High Pasture Cave), Strath, Isle of Skye 2004 (NGR NG 5943 1971). The preliminary assessment and analysis of late prehistoric cultural deposits from a limestone cave and associated surface features. Unpublished Data Structure Report, West Coast Archaeological Services. Bjerck, H. B. (2012). On the outer fringe of the human world: Phenomenological perspectives on anthropomorphic cave paintings in Norway. In K. A. Bergsvik & R. Skeates (Eds.), Caves in context: The cultural significance of caves and rockshelters in Europe (pp. 48–64). Oxford: Oxbow Books. Bourdieu, P. (1977). Outline of a theory of practice. Cambridge: Cambridge University Press. Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste. Cambridge, MA: Harvard University Press.

3  Do Caves Have Agency?

43

Bourdieu, P. (1985). The genesis of the concepts of ‘habitus’ and ‘field’. Sociocriticism, 2(2), 11–24. Bourdieu, P. (1990). The logic of practice. Cambridge: Polity Press. Davies, P., & Lewis, J.  (2004). A Late Mesolithic/Early Neolithic site at Langley’s Lane, near Midsomer Norton, Somerset. Pastoralism, 49, 7–8. De Landa, M. (2006). A new philosophy of society: Assemblage theory and social complexity. London: Continuum. Deleuze, G. & Guttari, F. (1987). A thousand plateaus: Capitalism and schizophrenia (B. Massumi, Trans.). Minneapolis: University of Minnesota Press. Dowd, M. (2009). Middle and Late Bronze Age funerary and ritual activity at Glencurran Cave, Co. Clare. In N. Finlay, S. McCartan, & C. Wickham-Jones (Eds.), Bann flakes to Bushmills: Papers in honour of Peter C. Woodman (pp. 86–96). Oxford: Oxbow Books. Dowd, M., & Hensey, R. (2016). The archaeology of darkness. Oxford: Oxbow Books. Fowler, C. (2013). The emergent past: A relational realist archaeology of Early Bronze Age mortuary practices. Oxford: Oxford University Press. Gardner, A. (2004). Introduction: Social agency, power and being human. In A.  Gardner (Ed.), Agency uncovered: Archaeological perspectives on social agency, power, and being human (pp. 1–15). London: UCL Press. Gell, A. (1996). Vogel’s net: Traps as artworks and artworks as traps. Journal of Material Culture, 1(1), 15–38. Gell, A. (1998). Art and agency: An anthropological theory. Oxford: Clarendon Press. Giddens, A. (1979). Central problems in social theory: Action, structure and contradiction in social analysis. London: Macmillan. Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Cambridge: Polity Press. Gibson, J. (1986). The ecological approach to visual perception. London: Routledge. Hawkes, C., Rogers, J., & Tratman, E. (1978). Romano-British cemetery in the fourth chamber of Wookey Hole Cave, Somerset. Proceedings of the University of Bristol Spelaeological Society, 15(1), 23–52. Heidegger, M. (1971). Poetry, language, thought (A.  Hofstadter, Trans.). New  York: Harper & Row. Ingold, T. (1993). The temporality of the landscape. World Archaeology, 25(2), 152–174. Ingold, T. (2000). The perception of the environment: Essays in livelihood, dwelling and skill. London: Routledge. Ingold, T. (2007). Materials against materiality. Archaeological Dialogues, 14(1), 1–16. Latour, B. (2005). Re-assembling the social: An introduction to actor-network-theory. Oxford: Clarendon Press. Leach, S. (2008). Odd one out? Earlier Neolithic deposition of human remains in caves and rock shelters in the Yorkshire Dales. In E. Murphy (Ed.), Deviant burial in the archaeological record (pp. 35–56). Oxford: Oxbow Books. Manem, S. (2012). The Bronze Age use of caves in France: Reinterpreting their functions and the spatial logic of their deposits through the chaîne opératoire concept. In K.  A. Bergsvik & R. Skeates (Eds.), Caves in context: The cultural significance of caves and rockshelters in Europe (pp. 138–152). Oxford: Oxbow Books. Merleau-Ponty, M. (1962). The phenomenology of perception (C.  Smith, Trans.). London: Routledge & Kegan Paul. Mizoguchi, K. (1993). Time and the reproduction of mortuary practices. World Archaeology, 25(2), 223–235. Mlekuž, D. (2011). What can bodies do? Bodies and caves in the Karst Neolithic. Documenta Praehistorica, 38, 97–108. Nawrocki, S., Pless, J., Hawley, D., & Wagner, S. (1997). Fluvial transport of human crania. In W. Haglund & M. Sorg (Eds.), Forensic taphonomy: The post-mortem fate of human remains (pp. 529–552). Boca Raton: CRC Press.

44

R. Peterson

Pels, P. (1998). The spirit of matter: On fetish, rarity, fact, and fancy. In P. Spyer (Ed.), Border fetishisms: Material objects in unstable spaces (pp. 91–121). London: Routledge. Peterson, R. (2013). Social memory and ritual performance. Journal of Social Archaeology, 13(2), 266–283. Rorty, R. (1989). Contingency, irony and solidarity. Cambridge: Cambridge University Press. Shanks, M. (2007). Symmetrical archaeology. World Archaeology, 39(4), 589–596. Skeates, R. (2016). Experiencing darkness and light in caves: Later prehistoric examples from Seulo in Central Sardinia. In M.  Dowd & R.  Hensey (Eds.), The archaeology of darkness (pp. 39–49). Oxford: Oxbow Books. Wagner, R. (1991). The fractal person. In M. Strathern & M. Godelier (Eds.), Big men and great men: Personifications of power in Melanesia (pp. 159–173). Cambridge: Cambridge University Press. Warmenbol, E. (2014). Le "Trou de Han" à Han-sur-Lesse. In C. Frébutte (Ed.), Coup d'oeil sur 25 ans de recherches archéologiques à Rochefort, de 1989 à 2014 (pp. 68–81). Namur: Institut du Patrimoine Wallon. Whitehouse, R. (2015). Water turned to stone: Stalagmites and stalactites in cult caves in prehistoric Italy. Accordia Research Papers, 14, 49–62.

Chapter 4

Animate Caves and Folded Landscapes Dimitrij Mlekuž

4.1  Introduction Why were caves so appealing to prehistoric people? Why do we encounter unusual assemblages and evidence of ‘special’ activities in caves? In order to find out, we must first descend into the cave; after all, descent into the cave—or katábasis (Bernabé 2015)—allows the brave access to superior knowledge, possession of extraordinary objects, and contact with the dead and the gods. Caves can only be reached by first crossing the landscape; after all, caves are places within a landscape. This journey is already part of the descent into the cave. The practice of walking weaves together landscape, establishes relations to the other places, and prepares our bodies through numerous encounters along the way. A cave is a rupture in the fabric of the world, a place where the innards of the landscape become accessible. We must then face the cave: its materiality, its quality of textures, its shapes and forms, its chill and its darkness; the unusual, the distinctive and the individuality of the place. Plunging into the darkness, we must make sense of numerous conflicting feelings and emotions. It puts our human presence in perspective; it hints at powers much greater than human, of other (inhuman) agencies that might lurk in the dark. It forces us to create at least a semblance of safety, of order and of culture in this utterly alien, chaotic and wild place. Only in this way might we return with our prize. Will we be the same, when we return?

D. Mlekuž (*) Department of Archaeology, Faculty of Arts, University of Ljubljana, Ljubljana, Slovenia Centre for Preventive Archaeology, Institute for the Protection of Cultural Heritage of Slovenia, Ljubljana, Slovenia e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_4

45

46

D. Mlekuž

4.2  What Makes Places Special? What makes places special? Are special, sacred places or landscapes just ‘social constructs’ or sites of competing discourses? Social constructivism approaches see materiality as constituted solely by human representation and language. Thus the object of study is the structure of symbolic meaning (or cultural representation) understood, as the anthropologist Clifford Geertz (1973, p. 44) puts it, as ‘a set of control mechanisms—plans, recipes, rules, instructions… for governing behavior’. Places, caves, forests, paths, monuments, etc. are just anchors for ‘the symbolic qualities of landscape, those which produce and sustain social meaning’ (Cosgrove and Jackson 1987, p. 96). Meanings are therefore ‘invented’ and not ‘natural’. They are plural and contested. Meaning cannot exist independently of the human mind, because language cannot exist independently in the world (Anderson and Harrison 2010). The world is out there, but representations of the world are not. The truth is not out there, simply put. The landscape is therefore made, not encountered; places are not inherently valuable nor do they carry innate meaning. As Laurajane Smith (2006, p. 3) reminds us, Stonehenge is basically ‘a collection of rocks in a field’. What makes it meaningful and special are ‘cultural processes and activities that are undertaken at and around them and of which they become a part’ (ibid.). This is a classic Cartesian divide, separating the world and its meaning. Once this divide is established, there can be no sense of how meaning and values may emerge from practices and events in the world, no sense of the ontogenesis of sense and no sense of how real the made-up can be. Matter is just a surface to be interpreted or written upon by something other than itself. But what if nature is not passive: an inert substrate on which meaning is projected or inscribed? If the nature of matter is generative—if it conceives and construes itself through an involved representation or differentiation of itself—then why must we presume that meaning is alien to its identity or this process (Kirby 1997, p. 115; Thrift 2008, p. 129)? Humans, with their desires and plans, are clearly not the only active things in the world; in fact, often we may be very small players in much bigger systems and complexes. The world beyond humans is not a meaningless one, made meaningful only by humans. Meanings, purposes, intentions, functions and significance emerge in a world beyond the human in ways that are not fully exhausted by our attempts to define and control them. The landscape houses emergent loci of meanings: meanings that do not necessarily revolve around, or originate from, humans (Kohn 2013, p. 72). Rather than passive signifiers of human cultural codes or norms, rethinking matter in terms of its own signifying power, ‘irreducible to representations’ (Bryant 2001, p.  23), enables us to acknowledge the world as comprising heterogeneous active partners intersecting one another. This approach introduces all kinds of new actors, forces and entities. Without hard boundaries, membranes or containers to divide them, humans and things continuously enter into coercive, interdependent and integrative relationships, or what Jane Bennett (2010, p. 32) calls a ‘swarm of

4  Animate Caves and Folded Landscapes

47

vitalities’ that animates both non-human things and their human companions. The world becomes a more-than-human assemblage: a tangled web, shaped and constituted through the material practices through which meanings, images, desires and power—the power to act, to imagine, to define, to impose and to resist—circulate. Meaning is not a foundation upon which all else rests; meaning emerges from certain kinds of material assemblages and can no longer be divorced from the matter that it used to objectify. Meaning is thus understood as a weaving of material bodies that can never be cleanly or clearly cleaved into a set of named, known and represented identities. Humans, non-humans, nature, culture, subjects and objects: all of these things ‘are one and the same essential reality’ in these assemblages (Deleuze and Guattari 1983, p. 5). They exist on a single plane of being. Meanings of the world emerge in manifold actions and interactions (Thrift 1996, p. 6). This approach redirects attention from representation towards the material composition and conduct of representations; to the world itself and practices of its rearrangement (Dewsbury et al. 2002, p. 438).

4.3  Immersive Landscapes A landscape is no longer a static visual frame where meaning is projected. Rather, a landscape is an assemblage which binds together the material presence of stone, earth, weather, water, trees, plants, animals and people, all with their own powers, desires and agencies (Greenhough 2010; Ingold 2008; Whatmore 2002, 2006). Landscapes are alive with their own vitality and power; they are fluid as the weather and as water, and with forests, mountains, rocks and caves, are continually undergoing change as they are entangled and folded together in a continual process of making (Edensor 2010, p.  244). Landscape is thus a force, an energy and a process; landscapes are present and alive in their ongoing processes of becoming (Rose and Wylie 2006). These qualities come into being as human (and non-human) bodies move around the landscape and encounter them. When we move, we are immersed in the currents and energies of a landscape in formation, continually experimenting with the body’s capacity to affect and to be affected. A landscape is felt, not just seen (Greenhough 2010, p. 43; Lash 2006, p. 324). This highlights the transpersonal, pre-personal relations between material bodies and things, and provides a broader, more-than-human concept of relations and sociality. It prompts us to consider how different configurations of things and bodies combine to form affective fields that are distributed across the landscape. Affect is independent of conscious perception, language and emotion. Affect is found in intensities that pass from body to body (human or non-­ human), in resonances that circulate around and between them, and sometimes become attached to them. Affect is about being hit by a sensation; it is about the force and intensity of an encounter. It is about the vibrancy of matter. Affect allows us to think of matter as always in formation, as always becoming, through forces and sensations (Massumi 2002). Thus, affective fields are ‘temporary configurations

48

D. Mlekuž

of energy and feeling’ (Conardson and Latham 2007, p. 238), a ‘quality of environmental immersion, sensed through bodies’ (McCormack 2008, p. 413). These visceral engagements constitute the context within which the world is comprehended and experienced. They are the foundation for the ‘embodied unconscious’; as Alphonso Lignis (1998, p.  59) says, ‘the sense and recognisability of things… do not lie in conceptual categories in which we mentally place them but in their positions and orientations which our postures address’. Representations do of course happen; we all lock down affects into representational schemes. However, affect allows us to be ‘in the middle of things’, where things are generative and activating. Bodies are plunged into the imminent affective field, in which sherds of representation interweave with others to create stronger intensities (Edensor 2010, p. 236).

4.4  Folded Landscapes And then we encounter the cave: a strange opening into the world. Caves gather the landscape around themselves, as a zone of connection between our lives on the surface, the earth below and the sky above. The relationship between landscape and cave is not a binary opposition between the ‘outside’ of the landscape and ‘inside’ of the cave. Caves are spaces that are different and separated from a landscape, but at the same time, they are integral parts of it. There is a whole spectrum of different shapes and forms of caves, from cliffs and rockshelters to shafts and hidden underground cavities, making clear distinctions between caves and the rest of the landscape arbitrary. The relationship between the landscape and the cave is best described as a ‘fold’. The fold is a flexible figure that allows distinction (after all, a cave is a cave) while maintaining continuity (caves are part of a landscape). The fold, as used by Deleuze (2006a, b), is neither figure nor ground but contains aspects of both. The fold announces that the inside is nothing more than a fold of the outside: ‘The outside is not a fixed limit but a moving matter animated by peristaltic movements, folds, and foldings that together make up an inside: they are not something other than the outside, but precisely the inside of the outside’ (Deleuze 2006b, p. 96–97). An interior is not an autonomously grown inside but merely a doubling of the outside. A fold creates uncertain boundaries instead of defined borders. These uncertainties create the potential multiplicity of folding and unfolding, a re-reading of a process of becoming. All that appears to be solid is but a momentary glimpse of a universal process of folding: ‘the fold is the general topology of thought, “inside” space is topologically in contact with the “outside” space and brings the two into confrontation at the limit of the living present’ (ibid., p. 118–119). Caves are places where the landscape is folded into itself, places where the underground is folded into the landscape. Folding blurs the boundaries between inside/outside, solid/void and space-to-space thresholds; folding redefines notions of spatial connections and

4  Animate Caves and Folded Landscapes

49

separation. Caves connect sky, landscape and the underworld; at the same time, they are labyrinths leading nowhere, only into themselves, eternally folded. Caves render visible the architectural powers of nature: powers that shaped the landscape. The landscape can be best revealed in a cave, where its innards are exposed. As Deleuze (2006b, p.  27–28) says, ‘For ages there have been places where what is seen is inside: a cell, a sacristy, a crypt, a church, a theatre, a study, or a print room’; this is also true of the cave. Deleuze (ibid., p. 28) notes that crypts— in the same way as caves—fold in on themselves, simultaneously expressing the ‘autonomy of the inside’ and the ‘independence of the façade’: an inside without an outside or an outside without an inside, depending on how you approach it. Landscapes contain caves in the same way as early basilicas contained a crypt (a ‘second’ or ‘lower’ church under the choir); crypts of especially renowned saints have inspired mass pilgrimages. So, beyond the association of caves with rot and death, they are a projection of subterranean architectural power that itself created the landscape (Lambert 2002, p. 44–45). When we enter the cave (cautiously) and when we become enfolded by darkness and chill, it becomes obvious that this is a strange place. The cave opens up only when we move along and push through its passages and its passages push back. As Tim Ingold (2007, p. 53) notes, moving through the cave is nothing like walking upon solid ground, with the earth beneath your feet and the sky above, with all-­ round vision and hearing. In a cave, we are fully enclosed within the earth and shut up in a medium that affords movement only along its cracks and crevices and that insulates us from sensory contact with our surroundings. Caves are dark, there is no light. This engagement with darkness is a key aspect of engaging with caves (Montello and Moyes 2012). To be able to see in a cave we must rely on technology, lamps or torches. This changes the way we use our bodies, as holding lamps or torches does not leave our hands free to climb, crawl or slither along. The light is flickering, illuminating only a small area at a time and casting long, moving shadows. The result is moving and ever-shifting lightscapes: ‘changing landscapes of light and darkness’ that determine ‘the appearance of the world’ as they cast ‘shadows in the relationship between things, persons and light’ (Bille and Sørensen 2007, p. 267). Movement requires us to conquer fears of the dark and encounter these ‘unseeable’ dark spaces, relying on shifting senses of enclosure and space. It forces us to scale vertical passages that require strength and skill to climb, to enounter narrow passages which require us to crawl and to squeeze through, to climb narrow ledges above chasms and to overcome clefts in the rock. This engagement with danger plays ‘tricks’ on the body (Cant 2003, p.  70–72). The effects generated by the coalescence of the pervasive dark, shifting lightscapes, chill, silence, the feel of water and moisture, the texture and shape of the cave, the sounds and gestures of moving human bodies, and a closeness to others, penetrate the body, enfolding it into affective atmosphere: ‘the emergent and fluid dimension of how place is sensed and experienced’ (Adey 2009, p. 5). This affective atmosphere provides a context which sets the tone for the experience and practice of engaging with the cave, an environment which literally ‘gets into’ the individual (Brennan 2004, p. 2). It is also

50

D. Mlekuž

a contagious environment, acting upon other bodies to maintain collective disposition; affects that bind us together and which bind us together with the environment (Edensor 2010, p. 236–237).

4.5  Haunted, Spectral, Uncanny, Sublime, Abject The affective atmosphere of caves can be described as ‘haunting’. Haunting is something attached to a place, as in the haunted house or ghost town. It speaks of a sense of uneasiness and of the feeling of other presences at the edge of awareness. It is about ‘the power of things to dwell with us and their power to haunt us (and we them)’ (Thrift 1999, p. 312). The idea of haunting explains why we often return to a place because of the place itself: the senses, feelings, desires and attachments that remain inside or upon us and/or others, and that also reside within the place. It draws on the affects it had on the participants, and the ongoing legacy of unusual places and events cemented within bodies. This suggests that human bodies, identities and places are constituted and constitutive, involving a vast array of shifting, ongoing and unpredictable processes, practices and discourses, some of which require tremendous effort to develop, uphold or maintain. Nigel Thrift (1999, p. 316–317) sees these ‘hauntings’ as an entry into an ecology of place, ‘a rich and varied spectral gathering’. ‘Spectral gatherings’ describe the assembly of many different things, human and non-human alike: ‘a diverse range of actors and forces, some of which we are aware, some not, and some of which may be just on the edge of awareness’ (Anderson and Harrison 2010, p. 12). The spectral presence suggests the presence of something which we cannot explain, do not expect, do not understand or struggle to comprehend. Spectral presence may be understood as a name for convoluted experiences of uncanny agencies. It is about ‘the unsettling of self, the haunting taking-place of place, the unhinging of past and present… an irreducible condition…’ (Wylie 2007, p. 172–173). There is an unsettling discomfort in confronting haunting places. This unease can be described in different ways, using interrelated concepts of uncanny, sublime or abject. All revolve around the theme of boundary blurring, destabilisation, breakdown and the destruction of boundaries and concepts which make the world intelligible and meaningful. The uncanny feeling—unheimlich (or unhomely)—is a sense of alienation and disorientation. The taken-for-granted is no longer familiar; it becomes strange, scary and haunted. It is a ‘species of the frightening that goes back to what was once well known and had been long familiar’ (Freud 2003, p. 123). The uncanny is something which seems familiar but which should not feel familiar. Freud’s definition is taken from Schelling, for whom unheimlich ‘is the name for everything that ought to have remained… secret and hidden but has come to light’ (ibid., p. 623). The cave, where the familiar landscape is folded into itself, where the monstrous landscape innards become exposed and tangible, is a perfect stage for the uncanny. The cave is an unhomely home of something mysterious; a womb and a tomb at the

4  Animate Caves and Folded Landscapes

51

same time. Being in a cave is like not yet being born, yet buried alive. And for Freud, ‘… the idea of being buried alive by mistake is the most uncanny thing of all’ (ibid., p. 241). The uncanny is thus closely associated with place, with its affective atmosphere that touches upon this amorphous unease. Uncanny can be described as an affect, a felt experience, a shiver down the spine that disturbs the body, resulting in a departure from the everyday; disturbing embodied experience and disrupting received expectations about the self in the world (Morgain 2012, p. 545). This encounter of ontological disruption is not entirely dissimilar to Kant’s (2000, p. 521–523) concept of sublime, described as experience of something existing beyond the physical world, split between the mundane and the ethereal, dwarfing humanity and threatening our reasoning and understanding. Sublime raises feelings of the insignificance of the self, the feeling of humility as we are overwhelmed by some big and powerful object (Berleant 1993; Hitt 1999). There is something otherworldly about the sublime. It puts humanity in perspective and gives us a brief insight into the possibility of something larger. We are confronted not with some socially constructed phenomenon but with the material experience of a natural world that ‘surprises and resists human desires and ambitions’ (Entrikin 2009, p. 222). The sublime is rooted in the unknown; it allows us to see for a moment beyond the human universe, which previously seemed stable and sure. Lyotard (1989, p. 199) describes the sublime as encountering the ‘inexpressible’ and ‘unpresentable’. Encounters with aspects of caves—that we strive to describe as great, powerful, vast, infinite, rugged, dark, gloomy and massive—cause uneasiness and confusion and destabilise the boundaries created by human beings to make sense of the world. And yet this uneasiness is also an aesthetic experience; paradoxical feelings and emotions that Edmund Burke (Phillips 1990, p. 123) calls ‘pleasant terror’ and ‘delightful horror’. When this distinctive, aesthetic relationship with an environment becomes unbearable—when it is no longer pleasant and delightful, when it becomes threatening—it becomes abject. For Julia Kristeva (1982, p. 3), those feelings of sheer terror and unease—‘the twisted braid of affects and thoughts’—are encapsulated in abjection. Our physical body—our skin—is the border that separates us from the outside world. Each being, then, engages in a constant process of ingesting into the body and secreting (or abjecting) from the body in order to live. The abject refers to human reaction to a threatened breakdown in meaning caused by the loss of distinction between subject and object or between self and other. Horror and repulsion of the abject are a source of protection to maintain this boundary. A symbolic order is a set of representations that separates us from direct contact with the chaotic world, with wilderness and with nature. Language, naming and formulation of meaning are forms of boundary creation. Symbolic order creates a safe bubble of culture; it makes the world meaningful, manageable and intelligible. Abject things are those that highlight the fragility of the symbolic order and that exist on the other side of the border which separates the living subject from that which threatens its extinction. But the abject is always there, inviting the self to take up the place of abjection: the place where meaning collapses. Kristeva emphasises the attraction, as well as the horror, of the undifferentiated. Abject is an affective

52

D. Mlekuž

horror caused by the breakdown of any distinction between subject and object, and of any distinction between ourselves and the material world, where we are confronted with an original state before binary oppositions between self and other, subject and object, and culture and nature were put in place. The subject, constructed during the process of creating a meaning of the world through representations, constantly faces abjection. The primary example of what causes such a reaction is the corpse, which traumatically reminds us of our own materiality. However, other phenomena can evoke the same reaction: an open wound, excrement, sewage or a cave, as a dark rupture in the continuity of landscape. Abject refers to the raw body within; blood and guts, water and flowstone, and to the powers of the earth, chaotic nature, sublime non-human powers and meaningless otherness which threatens order and culture.

4.6  ‘Rotten Nature’ and the Safe Bubble of Culture The non-human subterranean realm of caves testifies to something ambiguously creative or destructive. Stones (durable and hard) can be plastic and malleable and were shaped into fantastically formed passages. Caves are filled with weird, almost organic, shapes in the form of stalagmites and stalactites, testifying to strange creative forces and the vitality of non-organic life. Who built it? When? It was obviously built in some distant past, by some power which created the strange shapes and forms of its voids and planned by some incomprehensible alien agency. The affective atmosphere testifies to the haunting presence of something alien. This is not a stable, self-regulating, nurturing ‘Mother Nature’ which provides a safe stage for humanity, a source of materials and a canvas upon which meaning and symbolic order can be projected. This nature is one that is inherently powerful and controlling in itself: different, dangerous, strange, wild, scary, haunting, horrifying, discordant, ambiguous and chaotic. It is a nature full of promiscuous and obscene vitality, it is chaotically creative and has destructive forces; it is an unnatural nature. This is the nature which Lacan, in his later works, describes as ‘antiphusis’, anti- or counter-nature—internally plagued by ‘rottenness’ (pourriture), by decay and by defect—out of which culture (as antiphusis) bubbles forth in the form of images and signifiers that provide a place for human existence (Johnston 2012, p. 39–40; Žižek 2008, p. 56). Within this ‘rotten nature’, humans are not sovereigns of being but are beings with no privileged place. In this light, ‘natural places’—defined by Richard Bradley (2000) as places that gain significance in the minds of people—obtain a much more sinister appearance, as places where the bubbling chaos of nature can be encountered and felt. Bodies are submerged within the affective field, where sherds of representation collide with other stronger intensities and threaten established representational schemes, and where a continuous effort must be made to contain this and maintain a safe boundary between symbolic order and this bubbling chaos. But once representations are there, they fold back in on  the world, changing its configuration and its shape.

4  Animate Caves and Folded Landscapes

53

Natural places become meaningful, cultural constructs. Representations are not codes to be broken or illusions to be dispelled; representations are performative in themselves (Dewsbury et al. 2002, p. 438). They are activities that create meaning: a safe bubble for humans. Performances and representations of the haunted emerge as ways of approaching the experience of the ultimate alternative, the erosion of certainty and ways of making sense, and even in anticipating unexpected and indeterminable happenings. Notions of the sacred and the numinous are ways of approaching the subject of the uncanny, the sublime or the abject. ‘Pleasant terror’ can be related to mysterium tremendum, the feeling of terror before the sacred described by Otto (1926). The transcendent, sacred and numinous for Kristeva (1982, p. 57–58) really represents our effort to cover over the breakdowns (and subsequent reassertion of boundaries) associated with the abject. These practices are often centred on the body and the body politic: ritual, sacrifice, and the transformation of bodies and things. Thus, by way of abjection, societies create a safe bubble of meaning in order to remove its threatening, dangerous, unpredictable and ultimately meaningless nature. For Kristeva, rituals reflect an effort to delineate the boundaries that surround the abject and protect the symbolic order from the pollution or chaos embodied by the abject. As she (ibid., p. 17) notes, ‘the various means of purifying the abject—the various catharses—make up the history of religions’. Ritual is thus a technology for creating and maintaining symbolic order and for keeping it separated from meaningless chaos. In this way, it can be considered as ‘the basic social act’ (Rappaport 1979, p. 174). It is an act that creates meaning, shared representations of the world, and an ordered symbolic universe: the ‘Big Other’, where lie norms, expectations, desires, prohibitions, regimes of representation, guarantee of meaning, and other things that constitute human subjectivity and social order. Ritual is first of all a set of practices which offer a heightened sense of involvement through various performative technologies and which allow us to immerse in the abject in order to be protected from it. It is where mystical experiences can be brought forth and animated through the power of movement, body postures, repetitive movements, schedules of recall and spatial juxtapositions (Thrift 2008, p. 66). Focus on these practices exposes a critical circularity to the body’s interaction with the environment: by generating it, it is in turn moulded by it. Ritual performance both organises space and produces a ritualised body through the interaction of the body with a structured and structuring environment. ‘It is in the dialectical relationship between the body and a space structured according to mythic-ritual oppositions that one finds the form par excellence of the structural apprenticeship which leads to the embodying of the structures of the world, that is, the appropriating by the world of a body thus enabled to appropriate the world’ (Bourdieu 1977, p. 89).

Van Gennep (1960) associated the demarcation of cultural ‘passages’ with movement and progression in space. The process of rites of passage comprises passing through the boundary of the abject, cathartic trauma of facing the abject, and ultimately a return to society with a sense of renewed hope, rejuvenation or rebirth. Ritual thus organises space and creates religiously and socially sanctioned places.

54

D. Mlekuž

Mircea Eliade (1959) understood ritual as a delineation of a sacred place and the ‘regeneration’ of time. Victor Turner (1975, p. 69) specifically discussed the creation of ‘ritualised space’, while Jonathan Z. Smith (1982, p. 63) focused on the ritual performance of demarcating a ‘controlled environment’ in order to harness (at least temporarily) the unpredictability and chaos of nature. Space and time are structured through the physical movements of bodies projecting organising schemes into the world; these schemes become cemented in bodies. Hence, through a series of physical movements and ritual practices, we construct an environment organised according to representational schemes based on oppositions and delimitation of the sacred and profane (Bell 2009, p. 99). Ritual might, however, be less concerned with defining and proclaiming sacredness than with raising the boundary between sacred and profane, following the Latin origin of the term sacred from sacrâre (‘to set apart, consecrate’). Ritual practices separate caves from the rest of the landscape; previously just folds in a landscape, they become demarcated, assigned to the domain of the sacred and incorporated within the symbolic order. Once separated and isolated, they become the realms of the uncanny, sublime, abject and sacred. However, to protect itself from the abject—from the breakdown of boundaries—society must immerse itself in it, through the impure process of catharsis (Kristeva 1982, p.  13). This incorporates the bodily experience of a cave; immersion in its affective fields and experience of the abject through bodily performances that result in altered states of consciousness, visual and auditory hallucinations, disturbed perceptions of time, perceptual distortion, changes of body image, and changes in the feeling of control and emotional expression. These changes temporarily erode the distinction between subject and object or between self and the world, allowing for full experience of the abject. Some elements of this hallucinatory experience might be ‘hardwired’ in the human brain (Lewis-Williams 2002), whereas other major elements derive from past experience and shared symbolic order. These experiences are often represented as a journey to the absolute ‘Other’: into the world of darkness and of the dead, contact with the divine, and revelation of ultimate truth or hidden reality (Bernabé 2015, p. 30). An archaeology of caves is thus an archaeology of bodies and environments that focuses on the topics of relations, contacts, movements and encounters, and which reveals intimate relationships between the underground materiality of caves and the symbolic order. This mapping of enmeshments between humans and the world can provide re-evaluation of the elemental vitality and agency of the inhuman, and its role in the human social order. Caves are places in a landscape where people can come into direct contact with alien, inhuman nature: where the weirdness, power and horror of nature can be felt. It is dangerous; it comes too close; it can break, dissolve or negate existing concepts, representations and ideas. It can shatter the symbolic order. It requires significant effort to contain. But if we are successful, it is an unlimited source of creativity, vitality and power; it offers all sorts of alien wisdom, insight and imagination. Human interaction with caves can be seen as attempts at domestication of this alien power.

4  Animate Caves and Folded Landscapes

55

4.7  The Škocjan Landscape I want to explore the relationship between caves and landscape using a case study from the Škocjan Caves, Slovenia. The Škocjan Caves are an exceptional limestone cave system which lies on the eastern edge of the Karst (Kras in Slovenian and Carso in Italian): an extensive limestone plateau covering around 800 km2 in the hinterland of the Gulf of Trieste, on the border between Italy and Slovenia. Here, a unique karstic landscape has been formed from the dissolution of limestone, which has been shaped into a bizarre series of caves, dolines, shafts, dry valleys and outcrops. There is no surface water; extensive and complex cave systems were formed by disappearing streams and are connected to the surface by numerous shafts. The Škocjan Caves are the largest cave complex in the Karst and, together with Kačna jama (which is part of the same complex), are more than 20 km long. They lie within one of the world’s largest known underground river canyons, cut into the limestone bedrock by the River Reka (Habič et al. 1989). In a relatively small area around the point at which the River Reka dramatically sinks underground, there are an unusually large number of archaeological sites (more than 30), most of them caves and rockshelters (Turk et al. 2016). The majority of the sites were explored in the decades before World War I, and thus archaeological finds were mostly discovered by speleologists. Archaeologists only joined later, if at all (Turk 2002). Although most of the evidence is fragmentary and collected in a rather unsystematic way, it clearly shows that this is an exceptional landscape. But what makes it so peculiar? Airborne laser scanning (ALS) or LiDAR (light detection and ranging) of the area reveals a very rugged and weathered landscape. ALS allows us to systematically cover large areas and to detect and record the ground beneath the vegetation canopy (Opitz 2012). ALS records landscapes in an indiscriminate way: every place, every feature, every trace and every square metre are, in principle, treated with the same attention and resolution. This means that all features, created either by humans or by natural processes, are recorded in the same way (Mlekuž 2012, p. 92). ALS is very suitable for recording landscapes like Škocjan, as it treats the landscape as a whole, as the result of a number of different agencies and processes (ibid., p. 95–96). It shows that the patterns and forms which we see were not arbitrarily imposed on the landscape; the surface of the Karst was not a passive object of human agency: this craggy, rugged, irregular landscape was shaped by non-­ human agencies. As such, it was full of possibilities for alliances and collaborations that recognised the inherent patterns that the stony landscape already held. It seems that non-human agencies in fact dwarf human efforts to change the landscape. The main feature of the landscape is the gorge of the River Reka, carved into the limestone bedrock. The gorge ends in two large collapsed dolines, where the Reka sinks underground into the Škocjan Caves, beneath a sheer rock wall. The river then flows through an underground canyon and channels which expand into massive chambers. The river emerges after around 30 km into the springs of Timavo and runs into the Gulf of Trieste.

56

D. Mlekuž

The chemical dissolution of carbonate rocks exposed underground cavities and expanded cracks in the limestone, creating a large network of sinkholes which act as sediment traps and structure the texture of vegetation and human activities. Here, the landscape surface is merely a thin membrane over the underworld. The underworld is palpable, visible on the surface, as airborne laser scanning of the area clearly shows. This is a landscape of dolines, sheer drops, collapsed caves, shafts, chimneys, sinkholes, abysses, walls, caves, canyons, roofless caves and dry valleys. It is difficult to isolate the caves from the landscape. There is hardly outside or inside, surface or underworld. The landscape is folded into itself, it is merely a series of folds: a relative position, full of shadows, obscurities and edges that all give way once you round the corner and make it over the crease. This folded landscape is the product of chaotic creative and destructive forces of enormous magnitude that can be seen and felt at certain moments. After torrential autumn rainfalls, caves cannot accommodate the resultant volumes of water, which can flood the caves and collapsed dolines for up to 100 m but can also drain in a matter of hours (Kranjc and Mihevc 1988, p.  210). Other processes work on a longer-­term scale, for example, the gradual dissolution of limestone on the surface, which has caused the formation of the numerous collapsed dolines and has exposed previously hidden caves on the surface in the form of roofless caves (Mihevc 2007). The caves are also full of speleothems—strange organic-like forms, obscene almost biological shapes—a visual reminder of the wild, untamed vitality and creativity of the material world. People have visited these places since the Upper Palaeolithic. Most of the cave sites concentrate in the dramatic setting around the Mala and Velika Dolinas or in the gorge of River Reka. Since the Neolithic and Copper Age, the caves in Škocjan were visited only periodically for burials and ritual practices. The assemblages and depositional sequences here are in striking contrast to caves elsewhere on the Karst, which were used mainly as sheep pens and shelters (Mlekuž 2005). One of the key sites is Tominčeva Cave, a large chamber with a wide entrance in the north wall of the Velika collapsed doline, overlooking the entrance to the Škocjan Caves. It has very perilous access that requires climbing skills and can be completely submerged during the catastrophic floods of the River Reka. In a shallow recess at the front of the cave, bones of around ten individuals were discovered and dated to the Copper Age (Battaglia 1942; Marchesetti 1889). Pottery and animal bones in the same layer suggest communal participation in rituals associated with the burial or special treatment of the bodies. Numerous isolated human bones from Karst caves point to an operational sequence in the treatment of dead bodies that originated in special places like Tominčeva Cave, traversed the landscape and led to final deposition in other caves around the Karst. Other finds, such as a copper axe and a dagger hilt-­ plate, suggest ritual deposition of prestigious objects at Tominčeva. These artefacts are more common as votive deposits in watery contexts (Turk 2002, p. 88). It seems that we are witnessing the negotiation of this folded landscape and the undertaking of a dangerous descent to the cave, in order to face the ancestors and perform sacrifices, returning with the bones of the ancestors and the experience of another world.

4  Animate Caves and Folded Landscapes

57

In the Late Bronze Age, we see a development of the landscape around Škocjan. Dynamics of sovereignty led to the construction of material markers of territoriality and tenure; especially visible are monumental structures such as ramparts, enclosures and barrows. The prehistoric Karst was organised into a network of small communities that emerged around scarce areas of arable land. Central hillforts dominated the main agricultural area, while satellite hillforts were located next to the smaller fields, with pastures in the hinterland. Most structures in the landscape appear to be connected to pastoralism and the use of other resources in the landscape. Territorial boundaries are often marked by stone barrows, erected in exposed locations (Slapšak 1995, p. 97–80). In Škocjan, a hillfort was erected on a plateau between the Mala and Velika Dolinas, just above the dramatic place where  the River Reka enters the Škocjan Caves (Figs. 4.1 and 4.2). The hillfort is surrounded by the steep cliffs of the canyon and a sunken valley, and on one side is protected by a defensive wall. A second satellite hillfort was erected in Gradišče, a few 100 m away, again on the edge of a steep cliff. The hillforts were surrounded by numerous cemeteries, the largest of which is in Brežec and comprises 325 urn graves. Older graves from the eleventh and tenth centuries BC contain rich finds assemblages, including some of the first iron artefacts (axes, blades) in this area, almost 200 years before the widespread use of iron in central Europe (Ruaro Loseri et al. 1977). Graves containing exceptional assemblages of grave goods testify to the special status of the Škocjan community within the political and power dynamics of the prehistoric Karst.

Fig. 4.1  A view over the Mala and Velika collapsed dolines towards the village of Škocjan, the site of a prehistoric hillfort

58

D. Mlekuž

Fig. 4.2  Hill-shaded high-resolution topographic data derived from airborne laser scanning of the Škocjan area

Permanent stone structures, such as barrows, cairns and low stone walls, were built in the landscape, creating new patterns (Fig.  4.3). Ordered, cultivated landscapes emerged, with evidence of enclosed farms, enclosures for animals, walled tracks and cairn fields. Fields and field systems are more than just places where crops grow and where animals graze, and they are more than just boundaries and borders in a landscape. Built structures and visible traces mark temporal dimensions of tenure and genealogical ties to the land and important places. During daily and seasonal movement and tasks in the landscape, material traces mark the deep history of the settling and work of ancestors. People and communities become interwoven with the long-term history of places. The lives of people were closely connected with landscapes through daily and seasonal engagement with fields, by walking on paths and by working in meadows and pastures. The Škocjan landscape became an inhabited landscape: a mosaic of places with names, connected through paths and tracks; an ordered landscape and a true cultural landscape domesticated by agricultural diagrams in stone. The previously folded landscape was unfolded into different conceptual categories by the actions of humans. Nature was brought into the system of social relations. The Škocjan landscape was enclosed by a boundary wall which can be traced for more than 20 km around the Škocjan hillfort. This is a unique phenomenon on the Karst. This massive drystone wall, clearly visible on the ALS (Fig. 4.2), runs mostly on the ridges of hills and encircles the territory of the community, demarcating the

4  Animate Caves and Folded Landscapes

59

Fig. 4.3  The prehistoric cultural landscape of Škocjan

safe bubble of a cultural landscape. Beyond this lay the terrain that pastoralists crossed and re-crossed, along with the ancient routes of transhumance which formed daily and seasonal ebbs and flows. And beyond these boundaries was the liminal, untamed, wild edge of the cultivated world; ambiguous, dangerous, the unheimlich (or unhomely) domain of the sacred, expelled and kept outside the tamed landscape. Differentiation, separation and demarcation lie at the heart of the relationship between sacred and profane: ‘the sacred thing is par excellence, that which the profane should not touch, and cannot touch with impunity’ (Durkheim 1995, p. 55). And since the sacred flows into everything that approaches it, it must be kept within its own bounds. Precisely because the sacred is a source of enormous power, but also of instability and danger, it is only any good if it is kept at a distance. This separation takes continuous effort to maintain; to keep the folds apart, so that they are not unfolded into each other and into chaos. Two places, both deep caves located at the edge of the territory, emerge as focal points for the Škocjan community during the Late Bronze and Early Iron Age, where the power of the sacred was harnessed. This required, however, facing the abject. Mušja jama (Jama na Prevali II, Grotta delle Mosche, Fliegenhöhle) is a cave located on a ridge above the Škocjan hillfort (Fig. 4.4). It has a very unassuming entrance, just an opening in the ground that expands into a vertical abyss around 50  m deep, with voluminous subterranean chambers (Mihevc 2016). Without advanced climbing equipment and skill it is impossible to descend into it. In the

60

D. Mlekuž

Fig. 4.4 Mušja jama

mound of rubble at the bottom of the abyss, a large number of bronze objects and a few iron objects have been found. Most are broken, twisted, cut, bent and folded beyond recognition; others are partly melted due to intense exposure to fire. Various weapons dominate the assemblage. Especially numerous are spearheads, accompanying socketed axes, swords, helmets, greaves, armour and fragments of bronze vessels. According to the reports, burnt animal bones were also found among the rubble. The objects have been dated to between 1200 and 700 BC and are roughly contemporary with the Brežec cemetery mentioned earlier. Most objects have obvious associations with central and northern Italy, the central Alps, the west Balkans, the Carpathian area and a few even with the Aegean (Teržan 2016). The other cave is Okostna jama (Skeletna jama, Jama na Prevali I, Grotta delle Ossa, Knochenhöhle), located in the immediate vicinity of Mušja jama (Fig. 4.5). Under a large stone cliff is a very narrow entrance, which opens into a 21 m deep abyss. At the bottom is a steep floor and progressively low ceiling which continues to a depth of around 60 m. In winter, wisps of fog can be seen emerging from the entrance. Two trenches were excavated in the rubble layer at the lower end of the cave, where numerous disarticulated human and animal bones were found, together with bronze objects, mainly spears (Szombathy 1913). These metal objects are contemporaneous with the Mušja jama assemblage. The remains of least 11 human bodies were identified from the numerous fragments: mostly skulls and long bones. Eight individuals were determined to be male (their ages ranging from 18 to 40 years old), three individuals could be female and one fragment could belong to a younger person. Much more common were animal remains, with cattle the most common, followed by sheep and goat; the most

4  Animate Caves and Folded Landscapes

61

Fig. 4.5  Okostna jama

c­ ommon wild animal was wolf (Szombathy 1913, p. 188–189). According to analysis, neither human nor animal bodies were butchered or dismembered. Instead, they appear to have been thrown into the cave ‘whole’ and were presumably sacrificed (Riedel 1977). Crucial to the determination of sacrifice as a motive are the notions of giving, on the one hand, and separation, on the other hand. Separation is important; sacrificial offerings must be physically or metaphorically removed from the human world on the surface in order to have the desired effect (Pedley 2006, p. 80). These sacrifices progressed through a series of operations on bodies and things. Ritual burning transformed the artefacts into substances, and killing transformed persons into things. They were then thrown into the cave, rendering them invisible and inaccessible to humans. The structured deposits in both caves are obviously organised according to some semiotic system. The burnt and destroyed metal objects (mainly weapons) and possible burnt animal sacrifice in Mušja jama contrast with unmodified human and animal bodies and metal objects at Okostna jama, where another obvious opposition between sacrificed domestic herd animals and wild predators occurs. These sacrifices should, therefore, be understood as performances that established the meaning of particular places through the manipulation of things and bodies. Sacredness was enacted through bodily movement, which established spatial and conceptual relations between places. John Barrett (1994, p. 9) suggests a prime feature of emerging sacred landscapes may have been procession; setting out from the world of humans along paths and arriving at the caves: places of sacrifice, places of experience and places for taming the abject. Ritual movement was the lifeblood of landscapes; through movement, relations within a landscape were established, and

62

D. Mlekuž

borders and places were established (Connelly 2011). The Škocjan landscape was a dynamic place, full of movement, of choreography and of marching. Procession is a release of energy. Ritualised movement changes the body—from stretching of muscles, increased heart rate and sweating—which, through the articulations of the body in encounter, fixes ‘symbolic’ thought as affect, mood, emotions and feelings. As Nigel Thrifts (2008, p. 68) says, ‘nature observes and writes us, bumping intensities into our thoughts’. These two cave sanctuaries, located at the edge of territories (Fig. 4.6), indicate the importance of boundaries. They were set up as focal points of mediation; these ‘passages between two worlds’ (de Polignac 1995, p. 8) marked boundaries which were not only political borders between neighbouring communities but also boundaries between the sacred and profane, between the mortal and the divine, between this world and that beyond, and between nature and culture (ibid., p. 30). Experiences of the abject on the edge of the territory were rituals that maintained boundaries and separated folds between the world of humans and the world of the sacred. Both caves, with their unassuming entrances that open into deep shafts, show that the boundary—the membrane between surface and underworld—is extremely thin; that the otherworld is literally everywhere and that it can open up anywhere. The separation between worlds takes continuous effort, enacted through a sacred calendar, with periodic processions and festivals. This continuous effort of ­separation

Fig. 4.6  Relation of Mušja and Okostna jama to the surrounding landscape, to Škocjan hillfort and to the boundary wall

4  Animate Caves and Folded Landscapes

63

allowed the community to harness the powers of place of the supernatural. The prosperity of the Škocjan community, their power and fertility depended on harnessing and controlling the power of this sacred place. The importance of Škocjan diminishes after the eighth century BC. The landscape was still settled, but there is less evidence of prosperity. Deposition in both caves ceases after 700 BC, rich burials are no longer present, and most of the cemeteries are abandoned (Turk 2002, p. 91–94). The separation between sacred and profane—carefully established through ritual practices—appears to have broken. It seems that the effort to keep the folds of the human and the sacred world separate failed. Once these boundaries were broken, the landscape folded back into itself, with no clear separation between sacred and profane. There remained, however, a memory of past efforts and isolated attempts to keep these worlds separate: the burial, for example, of an individual in Okostna jama in the fifth century BC (Merlatti 2001, p. 16; Szombathy 1913, p. 175). This indicates new practices, requiring the technology and skill to climb into the cave and to bring participants into much closer contact with the abject. There are also isolated finds, such as an early Republican helmet and an Aucissa fibula in Mušja jama, reflecting memories of the centuries-old practice of sacrificing bronze objects (Teržan 2016, p. 471; Turk 2002, p. 91–94). Caves in the Mala and Velika Dolinas that were not used during the Late Bronze Age and Early Iron Age become places for sacrifice or burials (Turk 2002, p. 91–94). The abject emerged everywhere and had to be tamed everywhere it emerged. A fourth- or fifth-century bronze helmet found deep in the Škocjan Caves system indicates that the entrance to the Škocjan Caves just below the hillfort became a place of sacrifice once again. Even within the walls of the hillfort, burnt animal bones and isolated objects from the Okroglica shaft testify to the presence of sacrifice and efforts to tame the dangerous abject within the settlement. Ritual landscapes emerge through landscape practices, involving not only human agents but also other forms of agency. Here, the agency of caves and the underworld of these places was so powerful and dangerous that it took enormous effort to carve out places for humans, to separate the world of humans and the sacred, and to unfold the landscape into conceptual categories. Yet, these places are the source of vitality and power for human communities (though they must actively be kept separate through sacrifice and the experience of abject). They are, however, too powerful to be entirely domesticated and brought wholly under the human symbolic order.

4.8  Conclusions We have returned from the cave. And such journeys can be revealing but often also fatal. The landscape is full of force, energy and process. These qualities come ­especially into being in places such as caves. Caves are places where landscape is folded into itself and where innards are exposed: ‘where what is seen is inside’

64

D. Mlekuž

(Deleuze 2006b, p. 27–28). Caves are places where non-human meaning bubbles forth in chaotic affective atmospheres that can best be described as spectral and haunting. Caves are places where this excess of non-human meaning must be brought under control; otherwise, it can turn destructive: it threatens to break down the boundaries between the body and the world, between object and subject, and between culture and nature. To explore and confront the unknown and the unintelligible, people have developed ways to impose boundaries, in order to maintain structure in a chaotic, ambiguous world. Ritual is a creative way of dealing with threat; it actively creates meanings, identities and social relations through the experience of facing the unknown. However, such performances in caves are not isolated from the rest of the landscape. The things, substances and bodies that enter caves travel from somewhere else, and some of them return, establishing relations of meaning between places, things and activities. The cave becomes a node in the landscape, bound to the rest of the landscape through ritual practices that serve also to keep the unknown safely enclosed.

References Adey, P. (2009). Holding still: The private life of an air raid. M/C Journal, 12(1). http://journal. media-culture.org.au/index.php/mcjournal/article/viewarticle/112. Accessed 23 July 2017. Anderson, B., & Harrison, P. (2010). The promise of non-representational theories. In B. Anderson & P. Harrison (Eds.), Taking-place: Non-representational theories and geography (pp. 1–36). Burlington, VT: Ashgate. Barrett, J. C. (1994). Fragments from antiquity: An archaeology of social life in Britain, 2900– 1200 BC. Oxford: Blackwell. Battaglia, R. (1942). Indagine sull’età dei resti umani rinvenuti nelle caverne e nel Castelliere di S. Canziano Del Timavo. Atti del Museo Civico di Storia Naturale di Trieste, 15(1), 1–50. Bell, C. (2009). Ritual theory, ritual practice. Oxford: Oxford University Press. Bennett, J. (2010). Vibrant matter: A political ecology of things. Durham, NC: Duke University Press. Berleant, A. (1993). The aesthetics of art and nature. In S. Kemal & I. Gaskell (Eds.), Landscape, natural beauty and the arts (pp. 228–243). Cambridge: Cambridge University Press. Bernabé, A. (2015). What is a katábasis? The descent to the netherworld in Greece and the Ancient Near East. Les Études Classiques, 83, 15–34. Bille, M., & Sørensen, T. F. (2007). An anthropology of luminosity: The agency of light. Journal of Material Culture, 12, 263–284. Bourdieu, P. (1977). Outline of a theory of practice. Cambridge: Cambridge University Press. Bradley, R. (2000). An archaeology of natural places. London: Routledge. Brennan, T. (2004). The transmission of affect. New York, NY: Cornell University Press. Bryant, L. (2001). The democracy of objects. Ann Arbor, MI: Open Humanities Press. Cant, S. G. (2003). ‘The tug of danger with the magnetism of mystery’: Descents into ‘the comprehensive, poetic-sensuous appeal of caves’. Tourist Studies, 3(1), 67–81. Conardson, D., & Latham, A. (2007). The affective possibilities of London: Antipodean transnationals and the overseas experience. Mobilities, 2(2), 231–254. Connelly, J. B. (2011). Ritual movement through Greek sacred space: Towards an archaeology of performance. In A. Chaniotis (Ed.), Ritual dynamics in the Ancient Mediterranean (pp. 314– 346). Stuttgart: Franz Steiner Verlag. Cosgrove, D., & Jackson, P. (1987). New directions in cultural geography. Area, 19, 95–101.

4  Animate Caves and Folded Landscapes

65

de Polignac, F. (1995). Cults, territory, and the origins of the Greek city-state. Chicago, IL: University of Chicago Press. Deleuze, G. (2006a). Foucault. New York, NY: Bloomsbury Academic. Deleuze, G. (2006b). The fold. New York, NY: Bloomsbury Academic. Deleuze, G., & Guattari, F. (1983). Anti-Oedipus: Capitalism and schizophrenia. Minneapolis, MN: University of Minnesota Press. Dewsbury, J.-D., Harrison, P., Rose, M., & Wylie, J. (2002). Enacting geographies. Geoforum, 32, 437–441. Durkheim, E. (1995). The elementary forms of religious life. New York, NY: Free Press. Edensor, T. (2010). Aurora landscapes: Affective atmospheres of light and dark. In K. Benediktsson & K. A. Lund (Eds.), Conversations with landscape (pp. 227–240). Burlington, VT: Ashgate. Eliade, M. (1959). Cosmos and history. New York, NY: Harcourt. Entrikin, J.  N. (2009). Afterword: ‘The unhandselled globe’. In D.  Cosgrove & V. della Dora (Eds.), High places: Cultural geographies of mountains, ice and science (pp.  216–226). London: I. B. Tauris. Freud, S. (2003). The uncanny. London: Penguin. Geertz, C. (1973). The interpretation of cultures. New York, NY: Basic Books. Greenhough, B. (2010). Vitalist geographies: Life and the more-than-human. In B. Anderson & P. Harrison (Eds.), Taking-place: Non-representational theories and geography (pp. 37–54). Burlington, VT: Ashgate. Habič, P., Knez, M., Kogovšek, J., Kranjc, A., Mihevc, A., Slabe, T., Šebela, S., & Zupan, N. (1989). Škocjanske Jame speleological revue. International Journal of Speleology, 18(1–2), 1–42. Hitt, C. (1999). Toward an ecological sublime. New Literary History, 30(3), 603–623. Ingold, T. (2007). Lines: A brief history. London: Routledge. Ingold, T. (2008). Bindings against boundaries: Entanglements of life in an open world. Environment and Planning A, 4, 1796–1810. Johnston, A. (2012). Reflections of a rotten nature: Hegel, Lacan, and material negativity. Filozofski vestnik, 33(2), 23–52. Kant, I. (2000). Critique of the power of judgment. Cambridge: Cambridge University Press. Kirby, V. (1997). Forgetting flesh. London: Routledge. Kohn, E. (2013). How forests think: Toward an anthropology beyond the human. Berkeley, CA: University of California Press. Kranjc, A., & Mihevc, A. (1988). Poplavni svet ob Notranjski Reki. Geografski zbornik, 28, 193–218. Kristeva, J. (1982). Powers of horror. New York, NY: Columbia University Press. Lambert, G. (2002). The non-philosophy of Gilles Deleuze. New York, NY: Continuum. Lash, S. (2006). Life (vitalism). Theory, Culture & Society, 23(2–3), 323–349. Lewis-Williams, J. D. (2002). The mind in the cave: Consciousness and the origins of art. London: Thames & Hudson. Lignis, A. (1998). The imperative. Bloomington, IN: Indiana University Press. Lyotard, J.-F. (1989). The sublime and the avant-garde. In A. Benjamin (Ed.), The Lyotard reader (pp. 196–211). Oxford: Blackwell. Marchesetti, C. (1889). Ricerche preistoriche nelle caverne di S. Canziano presso Trieste. Bolletino della Sociate Adriatica di Scienze Naturali in Trieste, 11, 1–19. Massumi, B. (2002). Parables for the virtual. Durham, NC: Duke University Press. McCormack, D. (2008). Engineering affective atmospheres on the moving geographies of the 1897 Andrée expedition. Cultural Geographies, 15, 413–430. Merlatti, R. (2001). Il Complesso Protostorico Della Grotta Delle Ossa a San Canziano Del Carso. Atti e Memorie della Societa Istriana di Archeologia e Storia Patria, 49, 7–85. Mihevc, A. (2007). The age of karst relief in west Slovenia. Acta Carsologica, 36(1), 35–44. Mihevc, A. (2016). Speleološke značilnosti jame na prevali (Mušja jama)/Charatteristiche speleologiche della grotta jama na Prevali (Grotta delle Mosche). In B. Teržan, E. Borgna, & P. Turk (Eds.), Depo iz Mušje Jame pri Škocjanu na krasu/Il ripostiglio della Grotta delle Mosche presso San Canziano del Carso (pp. 15–22). Ljubljana: Narodni Muzej Slovenije.

66

D. Mlekuž

Mlekuž, D. (2005). The ethnography of the Cyclops: Neolithic pastoralists in the eastern Adriatic. Documenta Praehistorica, 32, 15–51. Mlekuž, D. (2012). Messy landscapes: Lidar and the practices of landscaping. In R. S. Opitz & D. Cowley (Eds.), Interpreting archaeological topography: Lasers, 3D data, observation, visualisation and applications (pp. 102–116). Oxford: Oxbow Books. Montello, D. R., & Moyes, H. (2012). Why dark zones are sacred: Turning to behavioral and cognitive science for answers. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual use of caves (pp. 385–398). Boulder, CO: University Press of Colorado. Morgain, R. (2012). On the use of the uncanny in ritual. Religion and Society, 42(4), 521–548. Opitz, R. S. (2012). An overview of airborne and terrestrial laser scanning in archaeology. In R. S. Opitz & D. Cowley (Eds.), Interpreting archaeological topography: Lasers, 3D data, observation, visualisation and applications (pp. 13–31). Oxford: Oxbow Books. Otto, P. R. (1926). The idea of the holy: An enquiry into the non-rational factor in the idea of the divine. Oxford: Oxford University Press. Pedley, J. (2006). Sanctuaries and the sacred in the Ancient Greek world. Cambridge: Cambridge University Press. Phillips, A. (Ed.). (1990). Edmund Burke: A philosophical enquiry into the origin of our ideas of the sublime and beautiful. Oxford: Oxford University Press. Rappaport, R. A. (1979). Ecology, meaning and religion. Richmond, CA: North Atlantic Books. Riedel, A. (1977). I resti animali della grotta delle Ossa (Škocjan). Atti del Museo Civico di Storia Naturale di Trieste, 30(2), 125–208. Rose, M., & Wylie, J. (2006). Animating landscape. Environment and Planning D: Society and Space, 24, 475–479. Ruaro Loseri, L., Steffé de Piero, G., Vitri, S., & Righi, G. (1977). La necropoli di Brežec Presso S. Canziano Del Carso: Scavi Marchesetti 1896–1900. Trieste: Civici Musei di Storia ed Arte. Slapšak, B. (1995). Možnosti študija poselitve v arheologiji. Ljubljana: Slovensko Arheološko Društvo. Smith, J. Z. (1982). Imagining religion: From Babylon to Jonestown. Chicago, IL: University of Chicago Press. Smith, L. (2006). Uses of heritage. London: Routledge. Szombathy, J.  (1913). Altertumsfunde aus höhlen bei St. Kanzian im Osterreichischen Künstenlande. Mitteilungen der Prähistorischen Kommission der Kaiserlichen Akademie der Wissenschaften, Wien, 2(2), 127–190. Teržan, B. (2016). Škocjan: At the meeting point of worlds. In B. Teržan, E. Borgna, & P. Turk (Eds.), Depo iz Mušje Jame pri Škocjanu na krasu/Il ripostiglio della Grotta delle Mosche presso San Canziano del Carso (pp. 431–486). Ljubljana: Narodni Muzej Slovenije. Thrift, N. (1996). Spatial formations. London: Sage. Thrift, N. (1999). Steps to an ecology of place. In J. Allen, D. Massey, & P. Sarre (Eds.), Human geography today (pp. 295–321). Cambridge: Polity Press. Thrift, N. (2008). Non-representational theory: Space, politics, affect. London: Routledge. Turk, P. (2002). Arheologija. In B. Pelc (Ed.), Park Škocjanske Jame (pp. 86–97). Škocjan: Park Škocjanske Jame. Turk, P., Hrobat, K., & Bratina, P. (2016). Katalog arheoloških najdišč Škocjana in Okolice/ Catalogo dei siti archeologici dell’area di San Canziano. In B. Teržan, E. Borgna, & P. Turk (Eds.), Depo iz Mušje Jame pri Škocjanu na krasu/Il ripostiglio della Grotta delle Mosche presso San Canziano del Carso (pp. 55–62). Ljubljana: Narodni Muzej Slovenije. Turner, V. W. (1975). Ritual as communication and potency: An Ndembu case study. In C. E. Hill (Ed.), Symbols and society: Essays on belief systems in action (pp. 58–81). Athens: University of Georgia Press. van Gennep, A. (1960). The rites of passage. Chicago, IL: University of Chicago Press. Whatmore, S. (2002). Hybrid geographies: Natures, cultures and spaces. London: Sage. Whatmore, S. (2006). Materialist returns: Practising cultural geography in and for a more-than-­ human world. Cultural Geographies, 13, 600–609. Wylie, J. (2007). The spectral geographies of W. G. Sebald. Cultural Geographies, 14, 171–188. Žižek, S. (2008). Nature and its discontents. SubStance, 117/37(3), 37–72.

Chapter 5

Familiar Caves and Unfamiliar Landscapes: Approaching Cave Use During the First Millennia in Britain Sam Wilford

5.1  Introduction In Britain, caves were key components of early archaeological research. During the nineteenth century, work by the likes of William Boyd Dawkins and William Buckland highlighted the multi-period significance of cave use throughout England and Wales (see White 2016). Later, at the beginning of the twentieth century, Herbert Balch (1928), who worked extensively in the Mendip Hills and Arthur Raistrick (1939), who excavated caves across the north of England, argued that caves were significant places for many Iron Age and Roman Iron Age communities (800 BC–AD 410). More recently however, caves have been frequently overlooked in studies of first millennia landscapes. Current research has often focused on analysis of individual site assemblages or has discussed cave use as part of broader chronological and regional syntheses (e.g. Branigan and Dearne 1992; Bryant 2011; Leach 2015). As a result, caves have become increasingly isolated from contemporary landscapes. Despite this, excavation and reanalysis of assemblages from several sites has begun to demonstrate complex social attitudes towards caves during the Iron Age and Roman Iron Age. Human use of sites, such as High Pasture Cave, Skye (Birch et al. 2005); the Sculptor’s Cave, Covesea (Armit et al. 2011; Büster et al., Chap. 10, this volume); and Read’s Cavern, Mendip (Marcucci and Kerns 2011), is characterised by the deposition of rich assemblages of material, including human and animal remains, metal and bone artefacts, and craftworking debris, which suggests that some caves were sites of complex, orchestrated acts of votive deposition. Whilst such studies highlight the potential significance of caves as ritual spaces during the first millennia, they also raise important questions as to why certain caves were chosen, what motivated acts of cave use, and how these sites fit within contemporary archaeological frameworks. S. Wilford (*) Florida Department of State, Bureau of Archaeological Research, Tallahassee, FL, USA © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_5

67

68

S. Wilford

This chapter, as part of wider research into the significance of underground spaces in Britain during the first millennia BC and AD, will outline the nature and landscape context of cave use across the limestone uplands of southern Cravendale, North Yorkshire, between c. 200 BC and AD 200. The period covers the end of the British Iron Age, known as the Late Iron Age (c. 400 BC–AD 43), and the beginning of the Roman Iron Age (AD 43–410), following the Roman conquest of southern Britain in AD 43. Notably, in southern Cravendale, material recorded in caves suggests that they were used for a complex series of activities, including the structured deposition of artefacts, and human and animal remains, as well as for bone working and metalworking. Discussion will therefore focus on contextualising this cave use, and will explore what may have motivated the use of these sites and how they may have been connected to broader concepts of landscape.

5.2  Caves, Landscapes and Topography There are two important factors which must be considered when contextualising the human selection and use of caves: the nature of a cave’s morphology and ‘sensescape’, or ‘cavescape’, and the site’s position within the landscape (Skeates 2008; Whitehouse 2001). This chapter will focus primarily on the landscape context of caves, namely, their relationship with contemporary archaeological sites and other natural locales. Recent research has begun to highlight the complex relationship between cave use and landscape. Work in southern America has demonstrated that caves were fundamental components of extensive ritual landscapes. For example, in their analysis of Kayuka Naj Tunich in Belize, Moyes et al. (2016) identified a series of enclosures beneath the cave which were contemporary in date. As such, it was argued that the cave and the enclosures were part of a sacred landscape used for display during ascension rites of local elites in the early first millennium AD (ibid., p. 46). Such work demonstrates the need to consider caves as part of landscapes, rather than as static, isolated features. This is also true when considering the topographic situation of caves. Generally, most inland caves are located on, or near, the base of cliff-faces and rocky ridges situated on valley and upland edges. As a result, some caves can often be difficult to access, resulting in unique physical and mental stresses which would have created heightened and unusual sensory experiences (Barnatt and Edmonds 2002). Studies on the perception of place in Anglo-Saxon England have highlighted the potential significance that topographic features may have had as marking boundaries and routeways. For example, in their discussion of long-distance travel during the Anglo-Saxon period, Reynolds and Langlands (2011) have argued that features such as forests and upland promontories served as points in mental maps belonging to those who were moving through unfamiliar landscapes. The liminal nature of these features, on the edge of landscape, may also have meant that these places were considered as boundaries (Jones and Semple 2012). Prominent hills, cliffs and upland ridges were often selected as sites for burial grounds, chosen because of their

5  Familiar Caves and Unfamiliar Landscapes: Approaching Cave Use During the First…

69

visibly liminal nature (Semple and Williams 2007; Williams 2006, p.  211). As places connected to the dead, hills—as well as antecedent features such as prehistoric burial mounds and chambered tombs, which were often found on ‘topographical edges’—became feared places, imbued with social memory (Semple 1998). In many Anglo-Saxon charters, these features served as natural boundaries of ownership, acting as visible references demarcating land (Semple 2013, p.  175). Similarly, wet places, such as rivers, bogs and marshes, were also seen as dangerous boundaries. Because of this, areas of crossing, including natural promontories, bridges and fords, appear to have served as foci for votive deposition (Lund 2010). Whilst direct comparisons cannot be made between the Anglo-Saxon perception of topographic edges as liminal spaces and the current data, such material shows that many locations where caves are situated were imbued with memory and meaning. Evidence from Anglo-Saxon England, as well as from studies of cave use in areas such as South America, demonstrate the need to contextualise caves within their cognitive landscapes. By comparing the location of caves to other contemporary archaeological sites and natural features, it is possible to understand the broader nature of caves as part of complex social landscapes.

5.3  The Human Use of Caves in Southern Cravendale The following section will outline the evidence for use of caves in the southern Cravendale district of North Yorkshire and will explore what may have influenced the selection and use of these sites. The discussion will outline the nature of Iron Age and Roman Iron Age archaeology in the region, before exploring the character and location of cave use. Through this, analysis will focus on the position of caves in the landscape, in relation to topography and contemporary settlement. Southern Cravendale, taken here to include the valleys of Chapel-le-Dale, Ribblesdale, Malhamdale, Littondale and southern Wharfedale, is an upland node of carboniferous limestone located on the southern edge of the Yorkshire Dales National Park in the north of England (Fig. 5.1). Characterised by steep-sided karst glacial valleys, or dales, the region boasts more than half of Britain’s known caves (Brook et al. 1988; Waltham and Lowe 2013). Recent estimates suggest that there are at least 1600 caves, active potholes and sinks in the highly soluble karst, located on exposed limestone scars and horizontal pavements found on valley edges (Waltham and Lowe 2013, p. 1). The large number of caves in Cravendale has meant that the region is historically important for the development of antiquarian interests in both caving and cave archaeology. Notably, some of the earliest systematic excavations of caves in Britain took place during the 1830s in Victoria and Dowkerbottom Caves in the south of the district. As a result of this interest, caves were important parts of early understandings of Iron Age and Roman Iron Age landscapes. Writing in 1939, in his overview of the Iron Age of West Yorkshire, Arthur Raistrick (1939, p. 112) argued that ‘Caves must be considered along with the adjacent enclosures and not as isolated inhabited sites divorced from the land’. Despite this, recent work has tended to focus on spe-

70

S. Wilford

Fig. 5.1  Location and principal topography of southern Cravendale, North Yorkshire

cialised aspects of cave use, namely, early prehistoric utilisation (Gilks 1973, 1989, 2005; Gilks and Lord 1985). Nevertheless, there is ample evidence of cave use during later prehistory and in the Roman Iron Age, which needs to be contextualised within contemporary archaeological narratives (e.g. Dearne and Lord 1998; King 1970; Lord and Howard 2013).

5.3.1  I ron Age and Roman Iron Age Archaeology in the Yorkshire Dales Compared to landscapes further south in England, archaeology of the Iron Age and Roman Iron Age in southern Craven dale is poorly understood and is characterised by a limited artefactual record, which has severely affected the visibility of archaeological sites. Because of these limitations, caves have been treated as key sites in past syntheses, which has resulted in several well-studied assemblages. Nevertheless, new programmes of excavation are beginning to reveal the complex and long-term nature of settlement in the region, in which cave use can be contextualised. For example, excavation of sites as part of large-scale landscape projects, such as those around Chapel House Wood, Upper Wharfedale (Martlew 2011) and Upper Swaledale (Laurie et  al. 2011), has identified dense concentrations of enclosures associated with field systems dating throughout the first millennia. Pollen records from the region also indicate periods of land clearance beginning in the mid-first millennium BC and the establishment of mixed farming practices near Ingleborough (King and Simpson 2011), in the east of the region, and in Swaledale towards the north (Fleming 1998, p. 137; Laurie 2004, p. 89).

5  Familiar Caves and Unfamiliar Landscapes: Approaching Cave Use During the First…

71

Across southern Cravendale, 21 caves have been identified as having been used sometime during the Late Iron Age and early Roman Iron Age; all are located towards the southern edge of the district. Activity appears to have focused on and near the Giggleswick (n  =  5) and Attermire Scars (n  =  5) at the entrance to Ribblesdale, east and west of Settle, respectively. Further activity is also recorded in three caves at the entrance to Chapel-le-Dale, two caves in Malhamdale, and four caves near Littondale (Fig.  5.2). Finds from these sites suggest that during this period caves were used for a variety of different purposes, including limited domestic activities, treatment of the dead, the intentional deposition of artefacts and craftworking (summarised in Table 5.1). Because of the limited material assemblage, cave use dating to the Late Iron Age was identified primarily through stray finds of locally diagnostic material, underneath Roman Iron Age contexts, in 13 caves. Finds include conical-shaped limestone loom weights, recorded in four caves; and Giggleswick-type bone toggles, thought to date to the turn of the millennium, also recorded in four caves; as well as local-type carinated pottery sherds found in Victoria Cave on Attermire Scar, east of Settle (see below). Finds of loom weights and bone toggles indicate possible limited domestic activity taking place in caves before the Roman Iron Age. Fragments of disarticulated human remains recorded at the entrances to four caves in the region also demonstrate the possible funerary significance of certain caves during the Late Iron Age. Excavations of the entrance to North End Pot, a vertical cave located on the edge of Chapel-le-Dale, recovered a fragment of juvenile skull which dated to between 310 and 130 cal. BC (Gilks and Lord 1993).

Fig. 5.2  Location of caves containing Late Iron Age and early Roman Iron Age material (n = 21). Each cave is numbered and referenced in Table 5.1

Base of scar

Cliff-­face Limestone plateau

Cliff-­face Base of scar Limestone outcrop Scree slope

1928–1932

1930; 2012 1850–1863

1985–1988 1935–1938 1990s 1838–1868 1870–1878 1977–1978 1996 Attermire Cave (10) 1870; 1920s; 1930–1931 Sewell’s Cave (11) 1932–1933 Scoska Cave (12) None (stray find) Spider Cave (13) None (stray find) Ravenscar Cave 1973–1983 (14)

Cliff-­face

Base of scar

Scree slope Cliff-­face

Cliff-­face

Location Scree slope Base of scar

Date excavated 1925–1932 1928–1932

Cave name (site number) Kinsey Cave (1) Lesser Kelcoe Cave (2) Greater Kelcoe Cave (3) Fairy Hole (4) Dowkerbottom Cave (5) North End Pot (6) Jubilee Cave (7) Chapel Cave (8) Victoria Cave (9) ✓

✓

✓

✓

✓ ✓

✓

✓

✓ ✓

✓ ✓

✓

✓

✓ ✓ ✓

✓

✓

✓

✓

✓

✓

✓

✓

✓ ✓

✓

Bone objects Pottery ✓ ✓ ✓

✓

✓

✓

✓

Archaeological material (c. 200 BC–AD 200) Human Metalworking remains evidence Metal objects ✓ ✓

Table 5.1  Overview of material found in caves in Cravendale

✓

✓

✓ ✓

✓

✓

Other finds ✓ ✓

72 S. Wilford

Date excavated 1873 1893 1953 1938 1870 None (stray find) 1950s

Cliff-­face

Location Scree slope Limestone outcrop Cliff-­face Base of scar Base of scar Limestone plateau ✓

✓ ✓ ✓

Archaeological material (c. 200 BC–AD 200) Human Metalworking remains evidence Metal objects

Other finds include animal bonework and stonework; see Wilford (2016) for more detail

Dead Man’s Cave (21)

Cave name (site number) Cave Ha (15) Calf Hole Cave (16) Langscar Cave (17) Fern Cave (18) Wet Cave (19) Doe Pot (20) ✓

Bone objects Pottery ✓ ✓ ✓

Other finds

73

74

S. Wilford

Fragments of a culturally smashed femur found at the entrance to Ravencliffe Cave on Giggleswick Scar were also likely deposited sometime between 511 and 376 cal. BC (Lord and Howard 2013, p. 245). Similar instances of smashed human bone have also been reported from Iron Age contexts outside Victoria Cave, Settle (ibid.). Unlike the generally limited evidence of activity in caves during the Late Iron Age, Roman Iron Age activity is characterised by rich artefactual assemblages containing metalwork and bonework, ceramics and coins, much of which dates to AD 50–150. Extensive assemblages of typologically diagnostic material have been recorded in Victoria, Attermire and Jubilee Caves on Attermire Scar (discussed below), as well as in Greater and Lesser Kelcoe Caves, Kinsey and Sewell Caves near Giggleswick Scar, west of Settle. Finds from these caves include an array of brooch types, bonework, bronze and iron tools, as well as coins and ceramics dating to the first and second centuries AD. In their excavation of Dowkerbottom Hole, located on a limestone pavement on the edge of Littondale, Denny and Farrer (Denny 1859; Farrer 1857; Farrer and Denny 1865) recovered a similarly rich assemblage of Roman Iron Age material. Finds included Dragonesque and Polden Hill-type brooches, thought to date to the late first century AD, coins belonging to the reigns of seven emperors, and an extensive corpus of ceramics (Branigan and Dearne 1991, p. 86). Whilst there are no human remains dated to the Roman Iron Age recorded in these caves, the unique nature of the artefactual assemblages in the region may indicate the votive use of certain sites. Amongst the assemblage at Dowkerbottom were 14 bone spoon brooches that are geographically restricted to the northern Pennines and thought to have been symbolically linked to regional identity (Eckardt 2014, p. 144). Spoon fibulae are decorated bone spoons that have a central perforation in the bowl and generally date to the first few centuries AD (ibid.). These brooches, along with stylistically similar metal brooches (namely, Dragonesque fibulae), have been recorded in six other caves along Attermire and Giggleswick Scars, which may indicate contemporary or near-contemporary deposition (Dearne and Lord 1998, p. 93; Eckardt 2014, p. 137; King 1970, p. 49). There is also evidence that spoon brooches and metal fibulae were being made in the same caves used for artefact deposition. Excavations at Dowkerbottom and Victoria Caves have recovered bone spoon brooches in various stages of production, including blanks (Dearne and Lord 1998, p. 101). Furthermore, finds of iron, bronze and galena slag from Victoria and Attermire Caves, as well as crucibles and bronze fibulae blanks, also demonstrate the use of caves for metalworking (see below). Roman Iron Age material found in caves also has implications for how we may approach earlier Iron Age evidence in the region. Because of the limited artefactual record during the Iron Age, the introduction of more robust Roman material culture has resulted in a bias in the increase of visible archaeological activity during the first and second centuries AD. Finds of early forms of Roman brooches, along with coins and ceramics of first and second century AD date, may therefore form part of existing Iron Age activity taking place in caves. This is further demonstrated by finds of Roman artefacts, such as metalwork and coinage, in caves that also contained objects including pre-Roman bone toggles and loom weights, as at Victoria Cave (Dearne and Lord 1998, p. 135) and Greater and Lesser Kelcoe Caves on Giggleswick Scar (Branigan and Dearne 1992, p. 92–94).

5  Familiar Caves and Unfamiliar Landscapes

75

Overall, the material evidence suggests that multiple caves in Cravendale were used during the late first and early second centuries AD.  Whilst many of these appear to have been used earlier in the Late Iron Age, extensive assemblages of material indicate intensive periods of contemporary deposition of stylistically similar artefacts focused in several caves. This implies that activity in multiple caves may have been linked and, as a result, motivated by similar processes.

5.3.2  Location and Topography of Cave Use To examine the relationship between caves and contemporary settlement, data was obtained through searches of the Yorkshire Dales Historic Environment Record, as well as published excavation reports. From this, a preliminary model of settlement during the Iron Age and Roman Iron Age was constructed in order to contextualise the caves under study. It is important to note, however, that there has been limited excavation in the region, and many of the sites have been identified morphologically through aerial and geophysical surveys (Horne and MacLeod 2001; Laurie et al. 2011). As such, many are not firmly dated, and it is therefore difficult to ascertain the relationship between caves and specific sites, although there is a notable exception (see below). Nevertheless, such a model is useful in establishing where caves are situated in relation to concentrations of prehistoric and Roman activity within the landscape. Figure 5.3 illustrates this concentration by using kernel density analysis of all archaeological sites considered to date to the Iron Age or Roman Iron Age (see Baxter et  al. 1997 for methodology) and demonstrates that cave use was often located on the edge of areas of concentrated archaeological activity. This is partly to be expected, since caves are predominately found on steep-sided valleys; it may, however, also indicate that these sites were chosen because of their location between settled landscapes. In southern Cravendale, caves are naturally found on the rocky scars that line the edges of valleys, as well as in exposed limestone in upland areas, which may explain the focus of cave use on the edge of settled landscapes. However, Late Iron Age and Roman activity predominately appears to have been concentrated in areas that are located on, or near, valley entrances (Fig. 5.4). For ease of discussion, I will consider three different areas in the region: Littondale and Wharfedale Towards the east, in the steep-sided and less fertile valleys of Littondale and Wharfedale, four caves contained evidence of use during the Late Iron Age and early Roman Iron Age. Both Fairy and Calf Hole Caves, which appear to have been used during the Iron Age (Raistrick 1939, p. 122, 128), are found on the interface between less fertile Upper Wharfedale, and flatter and agriculturally rich Lower Wharfedale. Dowkerbottom Hole (discussed above) is also found on a limestone plateau between Littondale and Upper Wharfedale, whilst Scoska Cave, which produced a coin dating to the third century AD, is located on the edge of a small valley running through the uplands towards Ribblesdale (Hill 1907).

76

S. Wilford

Fig. 5.3  Location of caves and kernel density estimates of all archaeological sites dating to the Iron Age and Roman Iron Age (n = 226). Darker shades represent higher concentrations of activity

Fig. 5.4  Position of caves containing material dating to the Iron Age and Roman Iron Age. Discussion will focus on the location of caves in three areas: 1 Littondale and Wharfedale, 2 Malhamdale, 3 Ribblesdale and Chapel-le-Dale

5  Familiar Caves and Unfamiliar Landscapes

77

Northern Malhamdale In the uplands between Littondale and Ribblesdale in the west, Langscar and Chapel Caves were used during the Iron Age (Lord and Howard 2013, p. 246). Both caves are located near Malham Tarn and are close to the modern road, as well as the old ‘Monk’s Way’, which was a route running from Littondale to Ribblesdale during the Monastic period (Raistrick 1947, p. 70). Ribblesdale and Chapel-le-Dale To the west of the region, activity in caves appears to have been concentrated in 11 caves along the Giggleswick and Attermire Scars, which lie at the interface between Ribblesdale and more fertile lowlands towards the south. The majority of these caves contain similar assemblages, which predominately date to the first and second centuries AD (see below). Overall, the majority of caves used during the Iron Age and Roman Iron Age appear to have been located near to, or on, routeways or valley entrances. This is further seen when comparing Iron Age and Roman Iron Age cave use to the natural distribution of caves in the region (Fig. 5.4). Importantly, whilst this suggests that caves may have been selected by those moving through valleys and on routeways, concentrations of activity focused on certain areas of the landscape may also reflect wider symbolic processes that motivated cave use connected with human movement.

5.3.3  Caves on the Edge? Attermire Scar and Ribblesdale Concentrations of activity at the entrance to the Ribblesdale Valley were predominately focused on two prominent limestone outcrops: Giggleswick Scar on the western flank of the valley and Attermire Scar, directly east of Settle. The discussion below will focus on the nature of cave use on Attermire Scar and explore how this use may have been related to wider experience of the landscape. Attermire Scar is a mile-long exposed carboniferous limestone shelf overlooking the entrance to Ribblesdale Valley, oriented north–south. At least four caves on the scar have presented evidence of human use during the Iron Age and Roman Iron Age: Jubilee Cave Jubilee Cave is located in a depression towards the northern end of Attermire Scar. The site, which comprises a large central chamber and a number of smaller tunnels, was excavated by Arthur Raistrick between 1935 and 1938, who recovered the disarticulated remains of ten individuals of unknown date, associated with Neolithic, later prehistoric and Roman artefacts (Challis and Harding 1975, p. 62; Raistrick 1939, p. 128). Notable finds included a conical limestone loom weight considered

78

S. Wilford

Iron Age in date, a spindle whorl made of samian ware, and a twisted copper alloy U-shaped binding identical to that found in Victoria Cave and dating to the second or third centuries AD (Raistrick 1939, p. 141). Victoria and Wet Caves Extensive deposits of Roman Iron Age and Late Iron Age activity were recorded at Victoria Cave, a complex multi-chambered cave accessed by a slope of talus sheltered by a small spur to the west. Finds of later prehistoric artefacts from the cave included local coarse pottery and a limestone conical loom weight (Dearne and Lord 1998, p. 44). In total, 58 brooches were recovered, including S-shaped fibulae dating to the late first century AD, as well as 38 coins, lead ingots, a seal box, and a ceramic assemblage dating predominately to the late first to third centuries AD (ibid., p. 53). Approximately 80 m north of Victoria Cave, a glass bead dating to the Roman Iron Age was also found in Wet Cave, which comprises a single, 12  m long passage. Attermire Cave Attermire Cave is a complex karst cave accessed by a steep climb up a talus slope on the southern end of Attermire Scar. Like Victoria Cave, the site contained a rich assemblage of Roman Iron Age material, including early forms of metalworking such as Auccisa and S-type fibulae thought to date to the early first century AD (King 1970, p. 411). One of the most unusual finds from the cave, however, was a gilded bronze chariot, dating to the late first century AD, dismantled and placed into the cave’s entrance chamber (ibid.). Along with this were a number of brooches and coins, the majority of which dated to between the late first and third centuries AD (Branigan and Dearne 1991, p. 74). The extensive and unusual assemblages found in Victoria and Attermire Caves suggest that sites along the scar may have served as a focus for the votive deposition of artefacts, at least during the early first millennium AD. Alongside this evidence, finds of iron and lead slag, tools, and openwork brooches and spoon brooch blanks in Jubilee, Victoria and Attermire Caves also suggest that these sites were used by craftworkers (Dearne and Lord 1998, p. 68). Importantly, similar evidence for the smelting of galena, azurite and malachite ore has also been reported from Attermire Camp East, an enclosure directly south of Attermire Scar (King and Simpson 2011, p. 33–34). Excavation also recovered ceramics of similar style and date to those found in Attermire Cave, suggesting that the enclosure was likely used at a similar time. Overall, the contemporary character of deposition in caves, as well as in the enclosure at Attermire Camp East, demonstrates that cave use may have been purposefully focused on Attermire Scar, at least during the first and second centuries AD.

5  Familiar Caves and Unfamiliar Landscapes

79

This is further reflected by the negative evidence of human activity in caves close by, including Sugar Loaf Hill to the south-west. Attermire Scar is visible to those moving north towards the entrance to Ribblesdale Valley, as well as east over the uplands from Malham (Fig. 5.5). As such, activity may have focused on the scar because of its prominence in the landscape and its position at the entrance to Ribblesdale.

Fig. 5.5  Attermire Scar as viewed from (a) its northern end on the Malham to Settle road, (b) from its southern end near Sugar Loaf Hill. This demonstrates the prominence of the scar in the landscape to those moving below it towards Ribblesdale

80

S. Wilford

5.4  Discussion: Moving Mnemonics From the evidence discussed above, cave use during the Late Iron Age and early Roman Iron Age appears often to have been focused on prominent natural features near valley entrances, as well as on routeways. In order to further explore why these caves were selected and who may have used them, it is necessary to examine the possible nature of first millennia society in the Yorkshire Dales. In their discussion of the later prehistoric communities of the northern Pennines, Challis and Harding (1975, p.  185)—drawing on earlier work by Stuart Piggott (1958, p.  28)—argued that the upland valleys of northern England had a limited capacity to support communities and concluded that ‘semi-nomadic pastoralism was the only practicable way of life’. This interpretation was the result of the apparently limited archaeology of the region, which was characterised by a number of small isolated field systems and enclosures. However, as we have seen, recent work has begun to question the nature of these societies in what is a marginal, upland environment. Instead of the so-called ‘Celtic Cowboy’ model, the long-term use of many fields systems and enclosures, coupled with evidence of large-scale land clearance and the practice of mixed agriculture, suggests that the nature of society was much more complex. Useful parallels may be gained from studies of Medieval communities across the Yorkshire Dales, which have shown that they lived in sporadic settlements focused on valley edges (Moorhouse 2003, p.  305–311; 2009). Consequently, routeways connecting settlements to the valley shelves were also located along valley edges, before moving down to valley bottoms by the mid-eighteenth century (Moorhouse 2009, p. 69). Importantly, the geology and geography of the individual valleys in the Yorkshire Dales dictated the type of agriculture practiced, which generally included a mix of seasonal transhumance and permanent farming (Moorhouse 2003). A similar model may therefore be relevant for communities inhabiting the Craven area during the Iron Age and Roman Iron Age. Instead of being broadly nomadic, the nature of society was heavily dependent on the geography and geology of individual dales. The more fertile and open valleys (such as southern Wharfedale) may have supported mixed agriculture and larger, more permanent settlements, whilst elsewhere, in the narrower valleys (such as Littondale), communities may have relied more heavily on seasonal transhumance. This relationship between individual dales, farming and transhumance is important in influencing the human experience of place. There has been a good deal of research demonstrating the importance of persistent experience in creating attachments to certain locales. Topophilic association with natural places is also well-­ recorded ethnographically, including in upland and marginal landscapes. In Finland, for example, features such as rock-faces, water bodies and waterfalls were often attributed symbolic significance and memory because of their prominence in the landscape (Lahelma 2005, 2007). Work by the likes of Tuan (1974) and Lokocz et al. (2011) has also shown that everyday sights, sounds and smells were fundamental in the creation of worldviews.

5  Familiar Caves and Unfamiliar Landscapes

81

For communities inhabiting North Yorkshire, the persistent experience of individual dales—characterised by their geography and geology—may have been a fundamental component of worldviews. The transitional areas between these valleys may, therefore, have been symbolically important as a zone between familiar, and more unfamiliar and irregularly perceived landscapes. In this respect, prominent topographic features located near to, or on, these interfaces may have been important places on the edge of everyday landscapes. This may be demonstrated by the concentration of activity in caves on Attermire Scar, located on the interface between the fertile lowlands extending towards Settle and Ribblesdale, and the uplands to the east. Consequently, the scar may have been situated between different economies and worldviews, as a result of contrasting geologies and geographies. Activity in caves may therefore have been motivated by the symbolic position of the scar on interfaces between valleys. In this context, acts of deposition of personal objects, of the dead, and the undertaking of craftworking in caves may have been tied to social attitudes towards certain areas of the landscape. Finds of personal objects such as brooches suggest that sites such as Attermire and Victoria Caves were used for the deposition of objects symbolic of personal identity. For example, in her work on identity in the Roman Iron Age, Eckardt (2014, p. 137–141) has argued that bone spoon brooches, which are predominately found in caves around Settle, may have been connected to specific regional identities. The concentrated deposition of artefacts in caves on the edges of these landscapes may therefore suggest that acts of votive deposition were also connected to the projection of personal identity. For communities inhabiting the valleys, scars and their associated caves may have acted as natural mnemonics bound to memory and meaning as places marking the transition from everyday experiences to unpredictable worlds. They may, therefore, have been seen as symbolically powerful places in which to project and reconnect to social identities before and after the undertaking of journeys away from regularly perceived landscapes. Alongside the votive deposition of personal objects, the use of caves for craftworking also highlights the potential symbolic importance of caves and their connection to broader social attitudes. Research on the nature of craftworking during the first millennia in Britain has often highlighted its metaphorical and social significance. Authors such as Giles (2007, 2012) and Hingley (2006) have argued that metalworking was connected to processes of transformation, liminality and control over supernatural elemental forces. Metalworking debris, as well as tools, moulds and crucibles, have been noted in a range of contexts, including in enclosure boundaries, pits and ritual wells, highlighting the integrated nature of landscapes with processes of exchange and manufacture (Hingley 1990; Moore 2006, p. 78–84). The relationship between caves—on the edge of settled landscapes—and craftworking is compelling. As we have seen, Attermire and Victoria Caves on Attermire Scar were used by metalworkers, as demonstrated by finds of moulds, bronze casts, and iron and galena slag. This activity may also have been contemporary with craftworking taking place in the enclosure at Attermire Camp East, directly below the scar. On one hand, the use of caves for craftworking demonstrates that these sites may have been used by itinerant workers who were attracted to caves because of

82

S. Wilford

their prominent location. However, by using caves on the edge of everyday landscapes, communities may also have sought to control or limit exposure to the dangerous and unpredictable forces generated during metalworking. This may be further demonstrated by the close connection of subterranean environments with the raw minerals and ores needed for craftworking. As a result, caves on valley i­ nterfaces may have served as ideal sites in which to negotiate with and control the dangerous liminal forces associated with metalworking. A similar argument may also be relevant for the funerary use of caves during the Late Iron Age. Disarticulated and articulated human remains have frequently been recorded in several natural and man-made boundaries, including settlement ditches, rivers and bogs (Giles 2009; Schulting and Bradley 2013, p.  68; Stevens 2008). Authors such as Giles (2009) have argued that such places acted as important social boundaries and cosmological spaces. Similarly, in his work on the ritual deposition of objects during the Iron Age and Roman Iron Age in the River Witham at Fiskerton, Leicestershire, Parker Pearson (2003) has suggested that activity was bound to the social perception of rivers as boundaries in the landscape. Like metalwork, human remains are often also associated with dangerous supernatural forces. In their work on the human bones recovered from the River Thames, Schulting and Bradley (2013) have suggested that the remains belonged to so-called deviants, including those who died prematurely, such as children and women who died during pregnancy or childbirth, victims of interpersonal violence, and the physically and mentally ill (Crummy 2010, p. 89; Giles 2009, p. 89). Votive deposition may therefore have been motivated by the social significance of human remains, leading communities to attempt to control supernatural forces attached to the dead. The human remains from the four caves in Cravendale were found outside the entrance or at the site ‘threshold’ which, coupled with the sites’ locations on landscape interfaces, may demonstrate a belief that caves, like rivers, acted as social and supernatural boundaries. Depositing the dead outside or on the threshold to underground spaces may therefore have been an attempt by communities to control cosmologically significant forces connected to the dead. The complex nature of cave use in certain parts of the landscape demonstrates that caves were likely used by a range of different individuals with multiple identities, including specialist craftworkers and those practising transhumance. As such, certain areas of the landscape may have been attributed a range of different meanings connected to a number of worldviews. Nevertheless, caves were chosen because of their connection to broader social concepts, including topophilic attachment to place and the metaphorical nature of groups such as craftworkers.

5.5  Conclusion For later prehistoric and Roman communities inhabiting the Yorkshire Dales, caves were a key component of lifeways and cosmologies, as places between familiar and unfamiliar worlds. Certain sites were selected because of their relationship with prominent topographic features and interfaces between valleys, making them

5  Familiar Caves and Unfamiliar Landscapes

83

natural mnemonics of transition. The votive deposition of personal objects and the use of these caves for craftworking may therefore have been tied to the symbolic importance of these sites as mediators between landscapes and worldviews. Consequently, they may also have acted as ideal spaces in which to negotiate with dangerous supernatural forces, such as those associated with metalworking. This study has highlighted the potential value of cave use in providing insights into the nature of landscapes and communities during the Iron Age and Roman Iron Age in Britain. By contextualising cave use within landscapes, it has been possible to explore the relationship between the use of multiple caves and other archaeological sites. Crucially, it has shown that caves were likely often connected to broader concepts of landscape and social attitudes including notions of identity. Furthermore, it highlights the value of caves in shedding light on complex processes that are often difficult to define, such as the nature of past cosmologies.

References Armit, I., Schulting, R., Knüsel, C. J., & Shepherd, I. A. G. (2011). Death, decapitation and display? The Bronze and Iron Age human remains from the Sculptor’s Cave, Covesea, north-east Scotland. Proceedings of the Prehistoric Society, 77, 251–278. Balch, H. E. (1928). Excavations at Wookey Hole and other Mendip caves, 1926–7. The Antiquaries Journal, 8(2), 193–210. Barnatt, J., & Edmonds, M. (2002). Places apart? Caves and monuments in Neolithic and earlier Bronze Age Britain. Cambridge Archaeological Journal, 12(1), 113–129. Baxter, M.  J., Beardah, C.  C., & Wright, R.  V. S. (1997). Some archaeological applications of Kernel density estimates. Journal of Archaeological Science, 24(4), 347–354. Birch, S., Wildgoose, M., & Kozikowski, G. (2005). Uamh an Ard Achadh (High Pasture Cave), Strath, Isle of Skye 2004 (NGR NG 5943 1971). The preliminary assessment and analysis of late prehistoric cultural deposits from a limestone cave and associated surface features. Unpublished Data Structure Report, West Coast Archaeological Services. Branigan, K., & Dearne, M. J. (1991). A gazetteer of Romano-British cave sites and their finds. Sheffield: Department of Archaeology, Sheffield University. Branigan, K., & Dearne, M.  J. (1992). Romano-British cavemen: Cave use in Roman Britain. Oxford: Oxbow Books. Brook, D., Davies, G.  M., Long, M.  H., & Ryder, P.  F. (1988). Northern caves, volume 1: Wharfedale and the north-east. Clapham: Dalesman. Bryant, A. (2011). Iron Age cave use on Mendip: A re-evaluation. In J. Lewis (Ed.), The archaeology of Mendip: 500,000 years of continuity and change (pp. 139–159). Oxford: Oxbow Books. Challis, A. J., & Harding, D. W. (1975). Later prehistory from the Trent to the Tyne. British archaeological reports (British series) 20. Oxford: Archaeopress. Crummy, N. (2010). Bears and coins: The iconography of protection in Late Roman infant burials. Britannia, 41, 37–93. Dearne, M. J., & Lord, T. C. (1998). The Romano-British archaeology of Victoria Cave, Settle: Researches into the site and its artefacts. British archaeological reports (British series) 273. Oxford: J. & E. Hedges. Denny, H. (1859). On the geological and archæological contents of the Victoria and Dowkerbottom Caves in Craven. Proceedings of the Geological and Polytechnic Society of the West Riding of Yorkshire, 4, 45–74. Eckardt, H. (2014). Objects & identities: Roman Britain and the north-western provinces. Oxford: Oxford University Press. Farrer, J. (1857). Dowkerbottom Hole. Proceedings of the Society of Antiquaries, 4, 111–112.

84

S. Wilford

Farrer, J., & Denny, H. (1865). Further explorations in the Dowkerbottom Caves in Craven. Proceedings of the Geological and Polytechnic Society of the West Riding of Yorkshire, 4, 414–422. Fleming, A. (1998). Swaledale: Valley of the wild river. Edinburgh: Edinburgh University Press. Giles, M. (2007). Making metal and forging relations: Ironworking in the British Iron Age. Oxford Journal of Archaeology, 26(4), 395–413. Giles, M. (2009). Iron Age bog bodies of north-western Europe: Representing the dead. Archaeological Dialogues, 16(1), 75–101. Giles, M. (2012). A forged glamour: Landscape, culture and identity in the Iron Age. Oxford: Windgather Press. Gilks, J. A. (1973). The Neolithic & Early Bronze Age pottery from Elbolton Cave, Wharfedale. Yorkshire Archaeological Journal, 45, 41–54. Gilks, J. A. (1989). Cave burials of northern England. British Archaeology, 11, 11–15. Gilks, J.  A. (2005). A sherd of later Neolithic Grooved Ware from Lesser Kelcoe Cave, Giggleswick, North Yorkshire, with a note of other finds of Grooved Ware from caves in the north of England. Transactions of the Hunter Archaeological Society, 22, 15–22. Gilks, J. A., & Lord, T. C. (1985). A Late Neolithic crevice burial from Selside, Ribblesdale, North Yorkshire. Yorkshire Archaeological Journal, 57, 1–5. Gilks, J. A., & Lord, T. C. (1993). A Neolithic antler macehead from North End Pot, Ingleton, North Yorkshire. Transactions of the Hunter Archaeological Society, 17, 57–59. Hill, C. (1907). Scoska Cave, Littondale. Yorkshire Ramblers’ Club Journal, 2(7), 229–235. Hingley, R. (1990). Boundaries surrounding Iron Age and Romano-British settlements. Scottish Archaeological Review, 7, 96–103. Hingley, R. (2006). The deposition of iron objects in Britain during the later prehistoric and Roman periods: Contextual analysis and the significance of iron. Britannia, 37, 213–257. Horne, P., & MacLeod, D. (2001). Unravelling a Wharfedale landscape: A case study in field enhanced aerial survey. Landscape, 2, 65–82. Jones, R., & Semple, S. (2012). Making sense of place in Anglo-Saxon England. In R. Jones & S. Semple (Eds.), Sense of place in Anglo-Saxon England (pp. 1–17). Oxford: Oxbow Books. King, A. (1970). Romano-British metalworking from the Settle district of West Yorkshire. Yorkshire Archaeological Journal, 42, 410–417. King, A., & Simpson, M. (2011). A review of the land use and settlement of the Ingleborough massif throughout the prehistoric and Romano-British periods. In R. D. Martlew (Ed.), Prehistory in the Yorkshire Dales: Recent research and future prospects (pp. 22–36). York: PLACE. Lahelma, A. (2005). Between the worlds: Rock art, landscape and shamanism in subneolithic Finland. Norwegian Archaeological Review, 38(1), 29–47. Lahelma, A. (2007). On the back of a blue elk: Recent ethnohistorical sources and ‘ambiguous’ Stone Age rock art at Pyhänpää, central Finland. Norwegian Archaeological Review, 40(2), 113–137. Laurie, T. C. (2004). Springs, woods and transhumance: Reconstructing a Pennine landscape during later prehistory. Landscape, 5(1), 73–103. Laurie, T. C., Mahaffy, N. W., & White, R. F. (2011). Co-axial field systems in Swaledale: A reassessment following recent fieldwork. In R. D. Martlew (Ed.), Prehistory in the Yorkshire Dales: Recent research and future prospects (pp. 37–59). York: PLACE. Leach, S. (2015). Going underground: An anthropological and taphonomic study of human skeletal remains from caves and rock shelters in Yorkshire. Leeds: Yorkshire Archaeological Society. Lokocz, E., Ryan, R. L., & Sadler, A. J. (2011). Motivations for land protection and stewardship: Exploring place attachment and rural landscape character in Massachusetts. Landscape and Urban Planning, 99(2), 65–76. Lord, T. C., & Howard, J. (2013). Cave archaeology. In T. Waltham & D. Lowe (Eds.), Caves and karst of the Yorkshire Dales (pp. 239–251). Buxton: British Cave Research Association. Lund, J. (2010). At the water's edge. In M. Carver, A. Sanmark, & S. Semple (Eds.), Signals of belief in early England: Anglo-Saxon paganism revisited (pp. 49–66). Oxford: Oxbow Books.

5  Familiar Caves and Unfamiliar Landscapes

85

Marcucci, A., & Kerns, C.  J. (2011). A preliminary report on the 2010 excavations at Read’s Cavern. Proceedings of the University of Bristol Speleological Society, 25(2), 165–186. Martlew, R. D. (2011). Late prehistory and the Roman Iron Age in Upper Wharfedale: Problems, potential and progress. In R.  D. Martlew (Ed.), Prehistory in the Yorkshire Dales: Recent research and future prospects (pp. 60–72). York: PLACE. Moore, T. (2006). Iron Age societies in the Severn-Cotswolds: Developing narratives of social and landscape change. British archaeological reports (British series) 421. Oxford: Archaeopress. Moorhouse, S. (2003). The anatomy of the Yorkshire Dales: Deciphering the Medieval landscape. In T. G. Manby, S. Moorhouse, & P. Ottaway (Eds.), The archaeology of Yorkshire: An assessment at the beginning of the 21st century (pp.  292–362). Leeds: Yorkshire Archaeological Society Occasional Paper 3. Moorhouse, S. (2009). The Medieval landscape of the Yorkshire Dales, North Yorkshire: Space, processes and procedures. In P. Sommer & J. Klápšte (Eds.), Medieval rural settlement in marginal landscapes/Peuplement rural dans les territoires marginaux au Moyen Âge/Mittelalterliche siedlung in ländlichen randgebieten (pp. 57–75). Turnhout: Brepols Publishers. Moyes, H., Robinson, M., & Prufer, K. (2016). The Kayuko Mound Group: A festival site in southern Belize. Antiquity, 90(348), 143–156. Parker Pearson, M. (2003). Regional social dynamics in later prehistoric eastern England. In M. Parker Pearson & N. Field (Eds.), Fiskerton: An Iron Age timber causeway with Iron Age and Roman votive offerings (pp. 189–193). Oxford: Oxbow Books. Piggott, S. (1958). Native economies and the Roman occupation of North Britain. In I.  A. Richmond (Ed.), Roman and native in North Britain (pp. 1–27). Edinburgh: Nelson. Raistrick, A. (1939). Iron Age settlements in West Yorkshire. Yorkshire Archaeological Journal, 34, 115–150. Raistrick, A. (1947). The Malham Moor mines, 1790–1830. Transactions of the Newcomen Society, 26(1), 69–77. Reynolds, A., & Langlands, A. (2011). Travel as communication: A consideration of overland journeys in Anglo-Saxon England. World Archaeology, 43(3), 410–427. Schulting, R., & Bradley, R. (2013). ‘Of human remains and weapons in the neighborhood of London’: New AMS 14C dates on Thames ‘river skulls’ and their European context. Archaeological Journal, 170(1), 30–77. Semple, S. (1998). A fear of the past: The place of the prehistoric burial mound in the ideology of middle and later Anglo-Saxon England. World Archaeology, 30(1), 109–126. Semple, S. (2013). Perceptions of the prehistoric in Anglo-Saxon England: Religion, ritual and rulership in the landscape. Oxford: Oxford University Press. Semple, S.  J., & Williams, H.  M. R. (2007). Early Medieval mortuary practices.  Anglo-Saxon studies in archaeology and history 14. Oxford: Oxbow Books. Skeates, R. (2008). Making sense of the Maltese Temple Period: An archaeology of sensory experience and perception. Time and Mind, 1(2), 207–238. Stevens, F. (2008). Elemental interplay: The production, circulation and deposition of Bronze Age metalwork in Britain and Ireland. World Archaeology, 40(2), 238–252. Tuan, Y. (1974). Topophilia: A study of environmental perception, attitudes, and values. Englewood Cliffs, NJ: Prentice-Hall. Waltham, T., & Lowe, D. (2013). The Yorkshire Dales. In T. Waltham & D. Lowe (Eds.), Caves and karst of the Yorkshire Dales (pp. 1–10). Buxton: British Caves Research Association. White, M. J. (2016). William Boyd Dawkins and the Victorian science of cave hunting. Barnsley: Pen and Sword Books. Whitehouse, R.  D. (2001). A tale of two caves: The archaeology of religious experience in Mediterranean Europe. In P. Biehl, F. Bertemes, & H. Meller (Eds.), The archaeology of cult and religion (pp. 161–168). Budapest: Archaeolingua. Wilford, S. (2016). Riddles in the dark? The human use of caves during the 1st millennia BC and AD across the British Isles. Unpublished PhD thesis, Durham University. Williams, H. (2006). Death and memory in Early Medieval Britain. Cambridge: Cambridge University Press.

Chapter 6

How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach at Pastena Cave Letizia Silvestri, Katia Francesca Achino, Micaela Angle, Maurizio Gatta, and Mario Federico Rolfo

6.1  Introduction Interpretation of the human use of caves, including those of the Italian Middle Bronze Age, has always been  strongly influenced by the intellectual and socio-­ political context in which it was produced (e.g. Tomkins 2009). Firstly, sites were considered as pastoral shelters for nomadic shepherds (Puglisi 1959). Then it became evident that such sites also held a strong symbolic meaning (Whitehouse 1992) and that they were used as burial sites as well as cultic places (Guidi 1992). The lack of a co-ordinated plan of action for the cave archaeology of Middle Bronze Age central Italy resulted in the uncontrolled undertaking of new excavations, especially between the 1960s and the 1990s, without contextualisation. The subjects of this work are often charming caves, but the study of these sites, even if preponderant, is often quite inconclusive; the interpretive patterns adopted have not been reworked even in the most recent publications (Bietti Sestieri 2010) and are L. Silvestri (*) Department of Archaeology, Durham University, Durham, UK e-mail: [email protected] K. F. Achino Quantitative Archaeology Laboratory, Department of Prehistory, Autonomous University of Barcelona, Bellaterra, Spain M. Angle Soprintendenza Archeologia del Lazio e dell’Etruria Meridionale, Rome, Italy e-mail: [email protected] M. Gatta Department of Archaeology, University of York, York, UK e-mail: [email protected] M. F. Rolfo Department of History, Culture and Society, University of Rome Tor Vergata, Rome, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_6

87

88

L. Silvestri et al.

now outdated. As a result, new steps made in theoretical and scientific discussions are almost always ignored. After the first systematic investigations between 1950 and 1970 by Antonio Maria Radmilli (1963, 1975, 1977), who identified a large number of caves with prehistoric deposits in central Italy (especially between the eastern Latium and Abruzzo regions), the key proponents of Holocene cave archaeology in this area were Giuliano Cremonesi and Renata Grifoni Cremonesi (Cremonesi 1968, 1976; Di Fraia and Grifoni Cremonesi 1996). From the second half of the twentieth century they have been carrying out several excavation campaigns in natural caves, mainly in Tuscany and Abruzzo (Fig. 6.1). They distinguished themselves for their adoption of systematic methodologies and for their critical approach to functional interpretations; their work today constitutes an essential reference point for Italian cave studies, since they are the first two scholars to have applied environmental methods to this category of archaeological site. Nevertheless, these studies are now 20–40 years old, and what once appeared as an innovation (e.g. the specific focus on soils and fauna) is now too approximate to be productively used in broad interpretive projects. Furthermore, attempts to gather information about these caves—in order to make wider interpretations—can be found only in two kinds of publication: typological handbooks (e.g. Cocchi Genick

Fig. 6.1  Distribution map of Early and Middle Bronze Age caves in central Italy (Pastena Cave is highlighted in yellow)

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

89

2002), which make use of pottery to build new chronologies or to strengthen old ones, as well as to recognise archaeological phases; and thematic articles and books which discuss cave cults and burials in a mainly descriptive way. The latter analyse signs of ritual activity and speculate on a basic level, or in a resigned way, on their meaning (e.g. Cocchi Genick 1999; Grifoni Cremonesi 1996, 2000; Guidi 1992) fertility cults to Mother Earth, and possibly rites of passage, are the general and most common explanations given for such evidence. It is, however, important to acknowledge a new approach first proposed by Bradley (2005), according to which the distinction between domesticity and ritual is a modern dichotomy of human reality. Applying such a perspective to the past is unnatural and even misleading; strong interpretive distinctions thus need to be reconsidered and overcome. Through the contextual analysis of one recently investigated case study, we aim to clarify as scientifically as possible a type of interpretation that usually belongs to the ‘non-scientific’ sphere of archaeology. In doing so we hope to reconcile these two misleadingly dichotomised sides of the coin.

6.2  Pastena Cave: History and Research Pastena Cave (Fig.  6.2), also known as Il Pertuso, Chiavica dell’Acqua or San Cataldo Cave, is located in the Cretacean limestones of the Monte Lamia-San Cataldo, part of the karst valley (polje) of Pastena (196 m a.s.l.). The mountain rises between two main depressions, one of which was occupied by a lake basin up until the eighteenth century (Seuterio 1730) and is located 4.5 km from the village of Pastena. The entrance to the cave, discovered for the first time in 1926 by Baron Carlo Franchetti, is approximately 20 m high and 25 m wide, with a magnificent

Fig. 6.2  Plan of Pastena Cave (Grotticella W2 is circled in red) (Adapted from Angle et al. 2014)

90

L. Silvestri et al.

80 m long entrance chamber (Angle et al. 2010, 2014; Biddittu et al. 2006, 2007). During the rainiest seasons, the Rio Mastro creek flows through the entrance chamber of the cave; after running 5 km from its springs, it penetrates the subsoil and emerges again at the locality of Obbuco at Falvaterra. The morphology of the cave, structured on two different levels, shows that the creek once ran along a higher route for at least 880 m. This fossil branch today constitutes a tourist route (the cave has been a tourist site since 1927) and has fascinating limestone formations such as stalactites and stalagmites of various types. One of the widest chambers holds a water basin from which local archaeology amateurs recovered a Middle Bronze Age bronze axe and dagger, deposited there after being transported by the creek; the lake was inaccessible until an artificial opening was made in the cave wall in the twentieth century for tourism (Biddittu 1987). After the sporadic recovery of archaeological remains since the 1940s (Guareschi and Morandini 1943; Segre 1946), the survey undertaken in the 1980s by Biddittu and Guidi (Biddittu 1987) was the first scientific attempt to identify the archaeological use of the cave. It became clear that the waters of the Rio Mastro, on the one hand, and the construction of the tourist route, on the other, had destroyed the majority of archaeological deposits at the site. A systematic research plan was thus felt necessary, in order to rescue the few still intact sectors of the site and to clarify the archaeological importance of the cave. This resulted in the first excavation campaigns, led by the University of Perugia in 2001 (Biddittu et al. 2006, 2007) and by the Soprintendenza Archeologia del Lazio e dell’Etruria Meridionale in 2008 (Angle et al. 2010). Several areas with traces of prehistoric human activity were identified, attributable mainly to burial use in the Neolithic and to a more complex combination of ritual practices in the Bronze Age. One of the tunnels located on the east side of the entrance chamber, for example, revealed the presence of a partial but articulated human skeleton, dated to the Middle Bronze Age (according to the ceramic remains found in the area). Meanwhile, a small area within the west wall contained heaps of burnt grain, a spindle whorl and a possible overturned vessel, again dated to the Middle Bronze Age, suggesting a specific agricultural ritual (Grifoni Cremonesi 1996; Miari 1995; Silvestri et  al. 2018). Furthermore, two Middle Bronze Age bronze weapons were discovered in the lake and were preliminarily interpreted as evidence of a possible male-/war-related cult by Biddittu (1987). The areas investigated were, however, limited in space and are likely to have been affected by several post-depositional events. Such partial yet thorough analyses could, therefore, provide only very general information about anthropogenic activity in the cave during later prehistory. For this reason, investigations were resumed by the Soprintendenza and the University of Rome Tor Vergata and are ongoing. The main research area has never previously been excavated. It is a c. 20 m wide chamber located about 20 m above the base of the cave, within the west wall of the entrance chamber. This chamber, which is completely dark, is known as Grotticella W2 (Fig. 6.3). Preliminary soundings of an adjacent area were undertaken in 2001 and 2008, but this chamber was left untouched. The archaeological deposit of the Grotticella W2 is distributed on two levels. The first consists of the current floor level, with three natural drops from west to east sloping towards the entrance, and is

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

91

Fig. 6.3  Grotticella W2 with its terraces: top left the terrace and bottom left detail of carbonised seeds, top right drawing and bottom right photograph of stone paving with upturned bowl

about 5 × 5 m wide. The second level of the deposit lies on two difficult-to-access natural terraces about 2 m higher than the current floor, located on the north-east side of the chamber. Collapsed rocks have been found in the area below the present boundaries of the terraces, suggesting that they were more extensive during the Bronze Age and that they were easier to access by visitors to the cave.

6.3  ‘The Cave in the Cave’: Description and Finds The Grotticella W2 did not suffer from as many post-depositional processes as most of the other areas investigated. Its location prevented most of the chamber from cyclical and severe flooding by the waters of the Rio Mastro c. 10 m below. The archaeological deposit seems extraordinarily well preserved, except for the entrance to the chamber, which was irremediably affected by the water flow. The presence of a 1 m thick layer of sterile fluvial silts testifies, however, to some rare and weak episodes of flooding of the entire cave. At the start of the systematic investigations, the most superficial level was covered with a thick layer of recent avifaunal remains and a few World War II items, indicating recent use of the cave as a refuge for both humans and animals. Beneath this was a sequence of layers, about 5  cm thick, with traces of combustion and ­figulina pottery of the Archaic period (seventh to sixth centuries BC), together with Bronze Age pottery. The main Bronze Age activity level, dated typologically on the basis of pottery recovered (Fig. 6.4), became evident beneath these mixed layers: a well-preserved 35  cm  deep deposit where many features and remains were found in primary

92

L. Silvestri et al.

Fig. 6.4 Some of the intact and significant ceramic material, typologically dated to the Protoapennine B strata of the Middle Bronze Age

c­ ontexts was sealed by silt layers above and by limestone veils at its base. The inner sub-horizontal parts of the chamber indicate the regular recurrence of at least four alternating burnt and paved layers. This sequence can be traced also in the collapsed zone next to the entrance slope. It consists of wide areas of thin but clearly visible layers of burnt crops and charcoal (Figs. 6.3 and 6.5), covering paved surfaces comprising small- to medium-sized stones, together with much faunal bone, more rarely human remains, and ceramic sherds. These, in turn, cover other thin burnt layers, under which was found further stone paving. Areas of reddish soil and formal hearths have also been identified (Figs. 6.3 and 6.5), as well as a small pit and, possibly, the remains of a stone structure (see Sect. 6.3.1). The terrace level presents a similar but less complex situation: a 10  cm  deep sequence of burnt seeds and charcoal layers between thin layers of sterile soil and a pit defined the stratigraphy, which terminated on the burnt natural karst surface of the western terrace (Fig. 6.3). Formal stone paving, not investigated due to the instability of the terrace, was identified on the eastern side. Figure 6.6 shows a reconstructive section of the Grotticella W2 Middle Bronze Age deposit, sealed by the thick sterile layer comprising stratigraphic units (SUs) 22–23. Although the sequence of stone and seed layers may appear less clear-cut in the deeper contexts (SUs 52–57), due to compression of the layers over time and repeated human activity, the alternation between SUs 30–34 and 48–51 is evident. Some small pits surrounded by stones and containing collapsed overturned vessels are also visible.

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

93

Fig. 6.5  Detail of the paved floors: (a) dispersed carbonised seeds; (b) reddened soil adjacent to the natural pool

6.3.1  Stone Structures As previously noted, one of the most remarkable archaeological features of this chamber are the stone floors. This is not a typical element of Bronze Age caves and so far it seems to be unique. Three different types of stone structure, in different degrees of preservation, have been identified. On the floor level, three small-sized (5–10 cm diameter) stone slabs, separated by thin layers of carbonised plant material, have been identified covering the whole surface of the sub-horizontal part of the chamber (Fig. 6.5). The sloping section closest to the entrance also has traces but has unfortunately been affected by landslide events, and they are less recognisable.

94

L. Silvestri et al.

Fig. 6.6  Reconstructive stratigraphic drawing of the Middle Bronze Age deposit in Grotticella W2: in the upper layers below the uppermost, sterile deposit, alternation between contexts containing stone paving and seeds is clearer; pits with broken ceramic vessels can be also observed

In the southern part of the sub-horizontal level, the floor comprises bigger, flat stones (10–30 cm diameter) and is apparently built in two layers, so this could possibly have constituted a raised structure. More importantly, the biggest stones of this structure seem to continue south-east, in the shape of a semi-circle. The majority of the human bones have been found here, hence the hypothesis that it might represent a dedicated burial area. The last stone structure found in Grotticella W2 was located on the eastern terrace. This consisted of flat stones laid out to cover an overturned pot (Fig.  6.3), some human bones, a bronze pin and some hare bones. Such a feature is common in cultic caves from the Neolithic and cannot be conceptually separated from that of the pits (Grifoni Cremonesi 1996, p.  332). Slight variations in structures of the same type have been found at Grotta dei Piccioni di Bolognano (Cremonesi 1976; Radmilli et al. 1978), Grotta Sant’Angelo sulla Montagna dei Fiori (Di Fraia and Grifoni Cremonesi 1996; Di Fraia and Tiberio 2008) and Grotta Mora Cavorso (Rolfo et al. 2013).

6.3.2  Hearths The peculiar layers of carbonised seeds and charcoal spread across Grotticella W2 certainly could not be natural intrusions, given their homogenous and considerable presence in the stratified deposit. The presence of at least three hearths at the site

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

95

was, therefore, not surprising: two lying on the basal level and one on the western terrace (Figs. 6.3 and 6.5). The first hearth on the basal level, located on the westernmost edge of the sub-horizontal zone, was typically delimited by burnt or blackened stones; it was also surrounded by a very compact reddish context, which lay on top of a natural, dried pond filled with soil (Fig. 6.5b). This feature probably functioned as a cooking slab. The remains of several different materials (ceramic, faunal, bronze, faience) were recovered within or immediately outside of the area of these two structures, suggesting intense utilisation or perhaps periodic reuse. The second hearth, located on the southern border of the excavated area, comprised a relatively wide (20 × 10 cm) primary deposit of ash. This hearth leant partially against a large natural stalagmite and was surrounded on the remaining sides by stones, which also prevented the structure from collapsing towards the entrance. The upper half of the skull of a young lamb was found lying on the surface of the hearth, together with an almost intact handled jug, which lay on its side. Other faunal and ceramic remains were recovered in the same area, together with a deposit of carbonised crops possibly related to one of the burnt layers. The last burnt area was identified on the western terrace. This sector appeared to have been severely compromised by the intense nesting activity of pigeons. Nonetheless, the presence of blackened stones, burnt soil and crop remains, together with the remnants of a semi-circular stone structure, allowed for reliable interpretation of the area. The homogenous distribution of charcoal and burnt grain on the western and eastern terraces (Fig.  6.3) leads to the hypothesis that the two areas were once better connected, the stalagmitic column being perhaps less invasive and the terrace floors larger than they are today.

6.3.3  Pits At least one pit, located in the middle of the basal level, was identified with certainty. It was deep and large enough to contain an overturned fragmented bowl. No other relevant features were recorded, but this structure can be compared (as with that on the eastern terrace) to several examples from the aforementioned caves and also to sites of northern and southern Italy (especially in Puglia), Slovenia (Miracle and Forenbaher 2006), France and central Europe, although not always in caves (Grifoni Cremonesi 1996, p. 316–320).

6.3.4  Limestone Pools The other structure used by Middle Bronze Age communities in the Grotticella W2 of Pastena was a small, dry pond (c. 60  cm in diameter), which was filled with anthropogenic deposits (Fig. 6.5). This natural limestone formation is common in many karst caves, and similar features are present elsewhere in Pastena Cave itself;

96

L. Silvestri et al.

most of these are still active and filled with dripping water. The microstratigraphy of the pool is very interesting because it indicates the alternation of non-­anthropogenic, anthropogenic, wet and dry layers. Certainly, the pool was already permanently dry when it came to function as the base for a cooking slab. However, in 10–12 cm of infill is a succession of six layers, two consisting of sterile clay, two of very thin karst veils, and two of clay with charcoal and burnt crops. Moreover, at the bottom of the pool, lying on top of the karst surface, were several Middle Bronze Age pottery fragments. This might indicate that the pool was used cyclically over up to three centuries during the Middle Bronze Age. The discovery of the pit, the hearths, the paving, and the seed layers is key evidence for the interpretation of the cave as a cultic site. These structures, whose functionality is neither related to production or to storage nor apparently to deposition of domestic waste, seem to result from actions linked to the deposition of human remains, special animal assemblages (associated bone groups (ABGs); Pluskowski 2012) and peculiar artefacts, including overturned pots (mostly carinated bowls and cups) each associated with a fluvial pebble. These features assume a possible symbolic meaning, since they are found in a distinct context (archaeologist’s perspective) and represent the final part of a purposeful process (performer’s perspective).

6.4  Artefacts 6.4.1  Pottery Pottery found at the site consists of about 300 fragments and 4 intact or reconstructible vessels from the Grotticella W2 and about 1000 sherds from the slope between the entrance and the footpath at the level of the Rio Mastro. Considering the limited width of the area investigated, and the relatively shallow depth of the archaeological deposit, these figures appear remarkable and suggest intense visitation to the site. All of the significant fragments (mostly rims, handles and plastic decoration on the vessel walls), together with intact vessels, can be attributed to Protoappenninico B (Cocchi Genick 2002), indicating single-phase activity at the site during the early Middle Bronze Age (though not necessarily a single episode of activity). The vessel forms range from open jars to closed jugs, cups and bowls (Fig. 6.4). The prominence of bowls and cups is significant: these are the vessel types that were deposited upside-down and left intact (one on the terrace and two or three on the floor level) (Fig. 6.3), in some cases within a dedicated structure. This is in line with the trend identified by Cocchi Genick (ibid.), which highlights the importance of pouring and drinking pots in the cultic caves of Bronze Age central Italy. This feature can be even better explained by taking into account the proximity of the Rio Mastro’s water source, as well as the possibility that the natural pool in the Grotticella was still active, at least for limited time periods. Water must have had a special role in ritual

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

97

performances at the site and may have influenced the choice of the site itself (Bradley 1998; Grifoni Cremonesi 1999). It is worth noting that a fluvial pebble was present next to every intact pot recovered in Grotticella W2; this does not seem to be a coincidence. The association of pits with intact bowls and fluvial pebbles was also a feature of the Bronze Age layers of Grotta Sant’Angelo sulla Montagna dei Fiori (Di Fraia and Grifoni Cremonesi 1996, p. 126), whilst a cattle skull was associated with a fluvial pebble at the Early Neolithic settlement of Catignano, both in the Abruzzi region. The technical quality of the pots is fair and the production is most likely local. Some of the sherds show traces of fire, but it was not possible to ascertain whether this occurred before or after their transportation to the cave, or before or after their deposition. Given the large amount of burnt crops present, and the existence of several hearths, it is likely that some of the pots (especially the jars) were used to toast seeds on site. Finally, the recovery of six spindle whorls (Fig. 6.7) in the chamber should be noted: given the context of their discovery (steep access, uncomfortable space, lack of natural light), it is hardly possible that spinning activities were undertaken in the

Fig. 6.7  Artefacts from the Grotticella W2; top clay spindle whorls, middle stone artefacts (from left to right: a drilled stone pendant, a partially drilled polished miniature stone axe and two small arrowheads) and bottom faience beads and buttons

98

L. Silvestri et al.

cave. If this was the case, it would most likely have been undertaken here for a specific reason: arguably, a ritual one. Such artefacts are, therefore, rather more likely to be related to the cultic sphere. Objects such as spindle whorls are traditionally considered as feminine gender markers (Sørensen 2013; Whitehouse 1998), since spinning is universally recognised as undertaken by women (ethnography, historical sources, burial associations with female individuals and iconography all testify to this view). It is thus possible to hypothesise a number of scenarios in the cave. All of these would have been aimed at constructing, reinforcing or symbolising the feminine identity of selected members of the group enacting the ritual performance. The deposition of spindle whorls could have been carried out by females to signify their participation in the ritual. Alternatively, non-gender-specific members of the community could have deposited the objects to celebrate female ancestors, deceased females, young women in their passage to adult life, weddings, births, and possibly even for fertility propitiation. Another possibility is a combination of the previous hypotheses, with only females performing this wide range of rites. Finally, a more traditional interpretation would assume that the spindle whorls were deposited as feminine grave goods. This cannot to be excluded, since the scattered remains of a female were found in the chamber.

6.4.2  Bronze Another infrequent occurrence recorded at the Grotticella W2 is the presence of bronze objects, focused in specific areas. Two small bronze rings, one 1.5  cm in diameter (a braid fastener or a finger ring) and another 5 cm in diameter (possibly a child’s bracelet), were found in the area of the first hearth. A broken pin, with a total length of 10  cm, was found below the stone structure on the eastern terrace, in the same area as the overturned pot and a few scattered human and faunal bones. The presence of bronze artefacts is not common in the majority of caves in central Italy, although a hoard of at least seven Middle Bronze Age bronze axes was found at Grotta Morritana (Belardelli et al. 2007, p. 111) in southern Lazio, and at Grotta Vittorio Vecchi (Belardelli and Pascucci 1996, p.  53), no more than 30  km from Pastena, two rings, a pin, a chisel, a dagger and an arrowhead were found. Unfortunately, the spatial data for this latter interesting comparative site are not published. The bronze remains from the Grotticella W2 cannot be dated typologically due to their very generic shapes, but the context of discovery is sufficiently reliable. Furthermore, such finds are in line with the Middle Bronze Age dagger and axe recovered in the inner lake during the 1980s (Biddittu 1987; Carancini 1984) and add weight to the interpretation that bronze artefacts were deliberately introduced into the cave. The deposition of metalwork in water, especially weapons such as axes, daggers and swords, is typical of the Italian Middle Bronze Age, mostly of northern Italy, e.g. in the Po River (Bettelli 1997), and the Sile and Piave Rivers

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

99

(Bianco Peroni 1994). Metal weapons are also found in central Italy, e.g. in the Pescara River in Abruzzi (D’Ercole 1997) and the Trasimeno Lake in Umbria (Bianco Peroni 1970). The deposition of the two ornaments and the tool differs from that of the dagger and the axe at Pastena. According to Bradley (1998, p. 5): The fundamental distinction [between metal depositions] is between the deposition of artefacts which could have been recovered and those which would have been difficult or impossible to retrieve. In general, that distinction corresponds to the contrast between finds which were deposited on dry land and those which were placed in water.

There are several explanations for the presence of the two weapons in the lake, bearing in mind that any kind of intentional or unintentional deposition would have occurred in the creek and only afterwards ended up in the lake. It has often been argued that single objects recovered in wet locations are to be considered as accidental losses or the product of flooding (Bradley 1998, p. 24). This is hardly the case for Pastena’s weapons: the presence of only two prehistoric artefacts in the lake indicates that the basin was not a usual collector for flooded-out objects, despite repeated floods documented by the cave sediments and by oral tradition. Moreover, the only two artefacts found in the lake are bronze weapons, dated to the Middle Bronze Age. This is further evidence for the intentionality of selective deposition of the finds. The first significant aspect of the deposit concerns them being weapons. Indeed— from a utilitarian viewpoint—axes and daggers were used for different purposes, and this may have been reflected in the different meanings assumed by the objects within the ritual performance. Nonetheless, both artefact classes have male gender connotations (Dolfini 2011). Similar to the spindle whorls of the Grotticella, the deposition of such objects in water seems to be associated with ritual performances aimed at defining, constructing or strengthening aspects of gender/status/role identity, in this case related to male individuals. If the underground place of deposition is shared by the weapons and the spindle whorls, and by the weapons and the metal objects found in the Grotticella, the choice of the dry/watery location indicates different ritual processes and different meanings conveyed through them (Bradley 1998). Spatial information, as well as stratigraphic and taphonomic data, is therefore key to elaborating reliable interpretations regarding cultural processes and behaviours. Generalisations on cave use in the past should be avoided, as several variables, when scientifically recorded, can change the meaning of similar objects and structures. It should also be noted that links to identity construction evoked by the choice of certain objects over others (e.g. spindle whorls versus weapons) might not be exclusively related to gender. This is probably only one of many meanings originally conveyed by these artefacts; it is simply easier for archaeologists to grasp, because of their frequent association with sexed skeletal remains or with later written/iconographic sources (see e.g. Robb 1994).

100

L. Silvestri et al.

6.4.3  Faience Context of retrieval is also crucial for the three glassy faience artefacts found in the Grotticella and its surroundings: two small biconical beads and a large conical button (Fig. 6.7). Unfortunately, only one of the beads came from a primary deposit within Grotticella W2; the other two were collected from the landslide below the chamber. These objects, and in particular the conical button, are typologically dated to the Middle Bronze Age (Bellintani 2000) and are consistent with the other contemporary examples in central Italy. Less than a dozen such artefacts have, however, been discovered (always singularly) in other caves, mostly in southern Lazio, e.g. Grotta dello Sventatoio (Angle et al. 1991–1992) and Grotta Vittorio Vecchi. This artefact type seems to appear in the region during the first phases of the Middle Bronze Age, after a slightly earlier presence in northern Italy. The chemical composition of the button—which includes a typically high percentage of sodium (Bellintani et al. 2005, p. 227)—testifies to Barfield’s (1978) original hypothesis of local production. The morphology of the button reflects that of earlier northern Italian examples but includes a recurring local feature in the v-sectioned central hole (Bellintani et al. 2005, p. 225). Given the scarcity of these objects (less than 100) (Bellintani et al. 2005), it is not possible to make detailed inferences about the local producers of glassy faience ornaments. Nevertheless, this scarcity could itself provide a few working hypotheses, along with chemical evidence for the high metallic content of the buttons and their production processes. These elements would, together, suggest that it might have been metallurgists, not other hyper-specialised craftsmen, who produced these objects (ibid.); however, post-depositional events would need to be explored in more depth in all the contexts in which these small artefacts were found. By identifying the taphonomic influences and impacts on these objects, we could move forward in our understanding of whether the low number of known glassy faience beads (and also of amber, stone and metal) is the product of cultural rather than natural phenomena. The sealed context of the Early Bronze Age chamber grave of Prato di Frabulino (Casi et al. 1995) is the only case in which several beads were recovered together: this would suggest that the scarcity of such materials might mainly be due to poorly-preserved contexts of discovery. Conversely, if the choice of depositing only a few beads was revealed to be a deliberate practice, new hypotheses will need to be formulated in relation to the cave finds, e.g. that the exceptional value of the material and its subsequent association with high status individuals explains their rare occurrence and/or that intentional symbolic selections were undertaken in ritual or burial contexts.

6.4.4  Stone The Bronze Age stone finds identified in Pastena Cave are not numerous but are varied and remarkable (Fig.  6.7). Unfortunately, the majority were recovered in secondary deposits from the landslide below Grotticella W2, generating doubts over

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

101

the chronological and stratigraphic attribution of the artefacts. As for the flint objects, one tool could either be prehistoric (pre-dating the Bronze Age) or a historical flint from a seventeenth- to nineteenth-century rifle. Two small Bronze Age arrowheads were also recovered (Fig. 6.7). These could have been associated with a male burial, whose existence in the chamber has not yet been confirmed. Alternatively, they could have been objects deposited at the end of a ritual performance. Such ritual processes would have been associated with the construction or reinforcement of gender roles/identities, directed towards and performed by one or more individuals, possibly males. A few flint flakes were recovered from the Grotticella, but their significance is unclear: it is unlikely that ‘everyday’ working activities were performed inside the chamber, given its aforementioned uncomfortable conditions (darkness, cramped space, gradient). This debris must rather, therefore, be related to the ritual reproduction of everyday working activities, augmented in this location with additional symbolic values (see e.g. Barfield (1986) reporting on unretouched sickle flint elements next to infant burials at the Eneolithic necropolis of Remedello). Alternatively, it may have been unintentionally lost or intentionally deposited in the chamber. Two interesting finds are a polished miniature greenstone axe (with traces of a failed piercing attempt) and a skittle-shaped pendant in soapy greystone, with traces of wear, possibly from a string, around the neck (Fig. 6.7). Neither artefact can be dated with certainty. The first accords well with the cultural and symbolic framework drawn up by Skeates (1995), who produced the first synthesis of Mediterranean perforated axe amulets from the Neolithic to the Iron Age. He identifies the occurrence of such artefacts, in similar in shapes, materials and contexts of discovery, over several millennia (hence the difficulty of attributing an exact date). With regard the regional focus of this work, he noticed the recurrence of these objects from sites located in ‘a band running across Central Italy’ (Skeates 1995, p. 281) from Tuscany and northern Lazio to Marche, northern Puglia and Campania. The Pastena discovery adds southern Lazio to the distribution and confirms the frequent association of axe pendants with caves—20% according to Skeates (ibid., p. 283)—and in general with ritual sites (an additional 62%). Pastena’s axe, however, cannot be defined as a ‘pendant’, since the drill hole is incomplete. The most likely life history of miniature perforated axes begins with an everyday stone axe (wear and use traces are often found on these artefacts). Following this, the axes likely go through a process of partial reshaping, possible repolishing and final perforation of the butt. This change in their use corresponds to a change of people’s perception of them. According to different theoretical frameworks, axe pendants have been interpreted as status markers, as healing/protection/apotropaic amulets (especially those made of greenstone) and even as gender indicators (of both sexes) (Aurino and Mancusi 2016; see Skeates 1995, p. 283–285 for a history of study). However, the most important interpretive framework, first highlighted by Skeates for the Mediterranean region, concerns the circulation of these objects over time and space and its implications for human relationships. The raw material of these artefacts does not seem to reflect qualitative preferences or specific relations with local quarries (Skeates 1995, p. 285). Their distribution, therefore, has to be related to other factors, most likely social dynamics. If the axes, once transformed,

102

L. Silvestri et al.

were kept by the owners and their family for generations, they would eventually come to be, to some extent, identified with the owners themselves. Their apparent random dispersal across the region might mean that these stone objects were given to other people as tokens of alliance, friendship or other relations, implying that, by doing this, a part of the donor was transferred to the recipient. The value of such objects would also relate to the original use of their utilitarian ‘antecedents’, as symbols of strength. In this sense, the traditional view of the miniature axes as amulets can also be upheld (ibid., p. 290). Finally, within this biographical framework, comes the only archaeologically detectable phase: that of deposition. After a symbolically meaningful life history, possibly lasting for years or centuries, these objects were too valued to be disposed of in a ‘normal’ way (e.g. after the death of the owner or breakage of the object, etc.). This led to a specific choice of depositing them in ritual sites such as caves, e.g. Grotta dello Scoglietto in Tuscany (Capasso and Piccardi 1980), Grotta Scaloria in Puglia (Tinè and Isetti 1980) and Grotta Pila in Lazio (Belardelli and Pascucci 1996), or sanctuaries. Despite the aforementioned difficulty in dating this class of artefact, it can be observed that very few axe amulets are found in burial sites dating to the Neolithic. Conversely, during the Copper Age and the early stages of the Bronze Age, over 75% of finds belong to funerary contexts, most of which consist of underground sites (caves and rock-cut hypogea; Skeates 1995, p. 295). In addition to indicating a specific change in use and perception of stone axes in the Bronze Age, this evidence can be combined with the dates of most of the artefacts recovered from Pastena Cave’s landslide deposit to suggest a Bronze Age date for the object. Unfortunately, the skittle-shaped pendant discovered in the same area does not have precise comparative examples, and its consistency with the rest of the  Middle Bronze Age remains can only be hypothesised. The symbolic value of the object is certainly as strong as that of the stone axe, given that the shape might well represent a schematic human figure. Finally, several blocks of steatite were recovered from immediately outside the entrance to Grotticella W2 and certainly appear to have slipped from it. This is very unusual for a cave context of the Bronze Age, where steatite objects have rarely been found.1 No finished artefact but various small (maximum 4–5 cm diameter) blocks of this rare raw material were found in a pile, with no trace of surrounding working activities. This could suggest the existence of some kind of hoard, voluntarily placed in the chamber. It is unclear whether this material was deposited to be later retrieved and used or abandoned for ritual reasons. For the sake of completeness, an incised steatite artefact from the landslide deposit also requires mention. This is a small pebble (2 cm long and 0.8 cm thick), with what appears to be the schematic representation of a buxom female figure: this artefact is currently under

1  Di Fraia and Tiberio (2008) do not specify the stratigraphic provenance of a steatite bead from Grotta Sant’Angelo, Teramo, Abruzzi; Delpino et al. (2009) do however present a steatite-working centre at La Puzzolente in Livorno-Tuscany, dated to the Eneolithic.

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

103

Table 6.1  Number of identified specimens (NISP) and minimum number of individuals (MNI) from the identifiable animal bones Species Ovis/capra (sheep/goat) Sus domesticus (pig) Bos taurus (cattle) Lepus sp. (hare) Small carnivores Total identified

NISP 47 29 4 13 2 95

MNI 6 4 2 1 2 15

study, but it most likely dates, according to typology, to the Neolithic (in which case, it is consistent with the date of the deposit in the cave’s eastern sector) or the Upper Palaeolithic.

6.4.5  Faunal Remains Compared to other categories of remains found at Pastena Cave, animal bones were quite rare (Table  6.1). However, zooarchaeological analysis of this dataset gave meaningful results: except for the hare (Lepus sp.)—whose remains were (perhaps significantly) located on the terrace, near the overturned bowl—the vast majority of recognised species were domesticates. Sheep/goat (Ovis aries/Capra hircus), followed by pig (Sus domesticus) and cattle (Bos taurus), were the most represented animals, with several bones showing evidence of burning and cut/butchery marks. Moreover, the pattern of body parts represented is more common in settlements than in caves (Fig. 6.8) and thus corroborates a hypothesis of meat consumption at the site, as does the average young and young-adult age of the individuals (Table 6.2). Given the difficult-to-access location and the contextual evidence described above, any meal would likely have been part of a set of ritualised activities.

6.4.6  Plant Remains Burnt seeds are undoubtedly the most common find retrieved from the site. They constituted homogenous layers between the stone paving and number several hundreds of thousands. Resulting from preliminary palaeobotanical analyses on c. 5000 grains (Table 6.3), the species ratio of various contexts appears quite constant, with an unsurprising 90% of the total comprising broad beans (Vicia faba). Distant second are the cereals, which included glume wheat (Triticum monococcum/dicoccum), free-threshing wheat (Triticum aestivum/durum) and barley (Hordeum vulgare). Two isolated grape seeds (Vitis vinifera) were recovered from the terrace.

or a

Skull&Teeth

Ovis vel Capra Sus domesticus Bos taurus Total

Young 2 2 1 5

2

Forelimbs

or a

C

av o Pa rso G ro G tta Co sten ro l tta Sa lep a Be nt’ ard A o G atri ng ro ce el tta C o de en i C ci Lu oc ni ci su lM i g C as non tig e l C er C ion ch o e io cc La ioli R C ipa ro ce tta

M

C av o Pa rso G ro C G tta o sten ro l tta Sa lep a Be nt’ ard A o a G tri ng ro ce el tta C o de en i C ci Lu oc ni ci su lM i g C as non tig e l C er C ion ch o e io cc La ioli R C ipa ro ce tta

M

C av o Pa rso G ro G tta Co sten ro l tta Sa lep a Be nt’ ard A o G atri ng ro ce el tta C o de en i C ci Lu oc ni ci su lM i g C as non tig e l C er C ion ch o e io cc La ioli R C ipa ro ce tta

a

M or

C av o Pa rso G ro C G tta o sten ro l tta Sa lep a Be nt’ ard A o a G tri ng ro ce el tta C o de en i C ci Lu oc ni ci su lM i g C as non tig e l C er C ion ch o e io cc La ioli R C ipa ro ce tta

a

M or

104 L. Silvestri et al.

100%

Ovis aries vel Capra hircus 100%

Sus domesticus

90% 90%

80% 80%

70% 70%

60% 60%

50% 50%

40% 40%

30% 30%

20% 20%

10% 10%

0% 0%

100%

Bos taurus 100%

Canis familiaris

90% 90%

80% 80%

70% 70%

60% 60%

50% 50%

40% 40%

30% 30%

20% 20%

10% 10%

0% 0%

Hindlimbs

Young/Adult 1 1

Extremities

Fig. 6.8  Body part distribution of the main animal species identified in Middle Bronze Age caves and settlements in central Italy

Table 6.2  Age categories based on MNI of domesticates (young, 6 months–1 year; young-adult, 1–2 years; adult, 2+ years)

Adult 2 1 1 4

Indet 1

1

Total  6  4  2 12

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

105

Table 6.3  Identified plant remains Vicia faba (broad bean) Indeterminate cereals Glume wheat Free-threshing wheat Hordeum vulgare (barley) Vitis vinifera (grape) Cornus mas/Olea europaea (cornel/olive) Total identified

3700 230 147 7 91 3 1 4179

Given the extraordinary number of seeds, their carbonised state and their spatial distribution, the palaeobotanical dataset of the Grotticella W2 cannot be considered as evidence of accidental overcooking. It appears, instead, that these crops were intentionally burnt and spread on the ground and on the terrace, around multiple combustion areas, for a specific purpose and in a repeated manner (at least three times). The deposition of carbonised broad beans seems to be a common feature of several funerary caves, including the nearby Grotta Vittorio Vecchi (Costantini and Costantini Biasini 2007) and Grotta Regina Margherita di Collepardo (Angle et al. in press; Silvestri et  al. 2018), together with Grotta Misa in Tuscany (Tongiorgi 1947). The association between beans and the cycle of life and death is widely recognised in the earliest literate cultures of the Mediterranean and northern Europe (de Cleene and Lejeune 2004 and references therein). Broad beans were thought to hold the souls of the dead and were often deposited close to burials or kept in the house as apotropaic charms. They were often taboo foods for certain categories of people or at certain times of the year. Such cults have ancient origins, related to the cycle of sowing and harvesting, and it is possible that this correlates with the Bronze Age evidence in funerary caves (Silvestri et al. 2017).

6.4.7  Human Remains Slightly fewer than 40 human bones have been recovered from the Grotticella W2,2 belonging to a minimum of 3 individuals. Of these, one is an adult (25% of the bones), though only one bone has been possibly identified to sex (female). Half of the remains belong to at least one infant, 50% of which represent an individual aged

2  The analysis has been carried out by Dr. Claudio Cavazzuti, Marie Curie Fellow at the Department of Archaeology, Durham University, to whom we extend our gratitude.

106

L. Silvestri et al.

between 5 and 9 years old. Two bones belong to a perinatal individual aged around 38 weeks. It can be surmised, therefore, that both adult and young/very young individuals, possibly of both sexes, were deposited within the cave. This suggests that, if they represent a selected part of the community, this selection did not depend on age and sex but rather on more complex social variables.

6.5  Interpretations of the Archaeological Evidence The Grotticella W2 and its surroundings have revealed much interesting data on the Middle Bronze Age human use of Pastena Cave. Despite its limited dimensions, its uncomfortable location, and the natural and artificial disturbances that have occurred over the course of millennia, this chamber contained an outstanding archaeological deposit rich in informative features. The exceptional preservation of the contexts allowed us to identify structures, artefacts and ecofacts: some rare or even unique in the context of central Italy’s Bronze Age caves. Combining all of these elements, it has been possible to attempt an accurate reconstruction of the cave’s use, as well as that of some of the socio-economic characteristics of the human groups that frequented the site. Firstly, three successive stone surfaces were identified, alternating with thin layers of burnt crops that covered the whole area. Another stone structure—once perhaps an elevated construction—complemented the first and, given its proximity to the majority of the recovered human bones, might yet yield a hidden burial. The last paved surface, made of flat stones, was located on one of the two terraces and yielded an inverted intact vessel, the leg of a hare, a human finger bone and a bronze pin. Two, or perhaps even three, more inverted pots (all cups and bowls) lay on the floor level, one of which was deposited in a small pit. This is likely associated with the presence of three hearths in the surrounding 30 m2 (including the terraces) and the recovery of several fine artefacts in bronze, faience, polished stone and flint, together with spindle whorls, scattered human bones and the debitage from meat-­ based meals. These elements, combined with the peculiar location of the site, indicate the unquestionable use of the cave for non-domestic purposes. Nevertheless, Bradley’s (2005) well-acknowledged theory of the constant coexistence of domesticity and cultic activity must be borne in mind (though it is perhaps not so evident and immediately applicable here as at other archaeological sites). This lack of conceptual dichotomy can be exemplified in several aspects of the archaeological record at Pastena Cave (especially in the systematically investigated Grotticella W2): firstly, in the pots recovered, whose rough manufacturing resembles those from settlements. On the one hand, these pots appear to have been used (or even reused) in activities that might appear mundane (e.g. storing and cooking), though they certainly differed in meaning. On the other hand, ‘everyday’ pots appear also to have been employed in non-everyday performances (e.g. including inverted deposition, possible smashing and intentional fragmentation). Other examples of this coexistence of sacred and profane can be identified in the ritual feasts and

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

107

repeated deposition of crops: in this case, the related, mundane subsistence activities involved in survival (i.e. the processing and consumption of meals). These activities also, however, embody a social component: that of identity legitimisation within the group. In this important aspect lies the link between mundane and non-­ mundane. In this sense, the personal ornaments found in the chamber, whose symbolic meaning has been previously explored, bear similar characteristics to the ritual meals and plant deposits. As identity markers, they belong equally to the domestic sphere, where a defined social identity is necessary for the survival of an individual and of the group. They belong also, however, to the cultic sphere, in which social identity is constructed and confirmed for active use in everyday life (Robb 1994). In this chamber, basic human needs cannot be satisfied, due to the claustrophobic space, its inaccessible position and the state of perpetual darkness. Climbing to the entrance, even with the help of a modern concrete staircase, represents a tiring physical activity. Moreover, the air becomes unbreathable after three or four people occupy the space for more than a week (considering working hours only), due to a gradual decrease in oxygen levels. It would be interesting to explore the habitability of the chamber after at least one of the hearths was lit. The hypothesis of this cave as a space for habitation, as a refuge, or even as a place where basic domestic activities took place, is therefore difficult to sustain. The Grotticella W2—if not all of the cave—must thus have been an exclusively ritual space. More complex is the interpretation of the type of ritual practices undertaken at the site. The primary burial function of the small chamber cannot be completely ruled out, since the (though limited) human bones recovered mainly comprise extremities, which are usually more likely to characterise a primary funerary deposit. However, secondary practices may also have been carried out in the Grotticella, by primarily depositing the deceased there and later moving them elsewhere (resulting in the large proportion of extremities found at the site). Other, undated tunnels and niches of the cave contained residues of possible primary deposits. It can be hypothesised, then, that deceased members of the community, or perhaps only selected members, were deposited in certain areas of the cave. Certain parts of their bodies were later transported elsewhere, in order to be honoured again and/or for use in propitiatory rituals. Given the significant amount of crops cyclically deposited, and the evidence of one or more feasts, these rituals appear to have been associated with the seeking of fertility. They can also be associated with social processes regarding the strengthening of the community’s bonds, perhaps under a leader (chief?), during a last salutation to deceased members (Parker Pearson 1999; Tarlow and Stutz 2013; Turner 1995). In caves, more than elsewhere, the earthly cycle of the seasons and the life-death cycle of human existence appear strongly correlated, and the rites dedicated to each of these aspects often appear to overlap. It has already been demonstrated that the Grotticella, as well as the cave in general, could not have been frequented over an extended period, because of its limited conditions for habitation. It is important, however, to address the issue of continuity of use. Several stratigraphic data suggest that the site was used over a protracted period of time, all within an early phase of the Middle Bronze Age (1750–1500 BC; according to pottery typology). It is, therefore, likely that the cave was used over a

108

L. Silvestri et al.

period of years or decades but for short periods only and, apparently, following the same pattern of activity (as can be expected of ritual performances). The area of the Pastena plain has never been subject to systematic survey but is an issue which must receive attention soon, given the importance of the Pastena Cave site. Since most of the original deposit has been destroyed by human action and the Rio Mastro, what is still preserved offers great potential and requires contextualisation.

6.6  C  onclusion: The Importance of Repetition in the Middle Bronze Age Cults of Central Italy Pastena Cave is a key site for the Middle Bronze Age of central Italy. The well-­ preserved context of Grotticella W2 has allowed us to identify a single-phase stratified deposit containing remarkable evidence in the form of artefacts (e.g. metal, faience and ceramic), ecofacts (especially seeds), human bones (especially extremities) and structures (e.g. hearths and paving). This deposit indicated repeated use of the site for rituals and for burial, including the deposition of structured offerings, the consumption of meals, and the undertaking of primary and secondary mortuary practices. The fact that the site was recently excavated using modern methods of retrieval and analysis has provided clearer results in comparison to those available from the majority of older cave excavations. In a broader context, Pastena Cave is part of a little-investigated region of south-­ western Lazio, which includes a good number of other caves and fewer possible settlements (Fig. 6.8). Some of the closest caves—which all bear evidence of cultic use, including burial—show evidence for different types of rituals, yet they share an important feature: repetition. For example, Collepardo Cave, which includes the deposition of around 100 individuals, shows evidence of repeated post-mortem manipulation of the bones (Skeates et  al. in prep.), whilst Mora Cavorso Cave includes at least two phases of deposition of perinatal piglets and lambs/kids (Rolfo et al. 2013, 2016; Silvestri et al. 2016). It has only been possible to observe this kind of repeated ritual action thanks to the accurate stratigraphic and taphonomic analyses carried out in the field in recent years. It is very likely that other caves published in the literature presented similar recurring patterns, now harder to detect due to the lesser importance given to such aspects in past investigations. Future research in several Middle Bronze Age caves of this region will hopefully shed further light on the topic and corroborate the hypothesis of repetition as a common ritual strategy of Middle Bronze Age communities of central Italy for dealing with and overcoming the loss associated with death, and allowing communities to move on with everyday life. Acknowledgements  We would like to thank Dr. Arturo Gnesi, Mayor of Pastena, for the support he has provided since 2012. We also thank Dr. Claudio Cavazzuti for allowing us to use preliminary data on his analysis of the human bones.

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

109

References Angle, M., Gianni, A., & Guidi, A. (1991–1992). La Grotta dello Sventatoio (S. Angelo Romano, Roma). Rassegna di Archeologia, 10, 720–721. Angle, M., Bruni, N., Cerqua, M., Riva, A., Cavazzuti, C., & Celletti, P. (2010). Nuovi scavi nella grotta di Pastena (Frosinone). In G. Ghini (Ed.), Atti del convegno sesto incontro di studi sul Lazio e la Sabina (pp. 369–380). Roma: Quasar. Angle, M., Rolfo, M. F., Fusco, I., & Silvestri, L. (2014). New investigations at the Cave of Pastena (Frosinone): Report 2012. In: G.  Ghini, Z.  Mari (Eds.), Lazio e Sabina, Scoperte Scavi e Ricerche, Atti del Decimo Incontro di Studi sul Lazio e la Sabina (pp. 205–211). Rome: Quasar. Angle, M., Rolfo, M.  F., Skeates, R., Mancini, D., & Silvestri, L. (in press). Burial in caves: Spatial, material and conceptual relations between closed and open areas of human occupation. In G. Ghini & Z. Mari (Eds.), Atti del convegno dodicesimo incontro di studi sul Lazio e la Sabina (pp. 369–380). Quasar: Roma. Aurino, P., & Mancusi, V.  G. (2016). Valore d’uso e valore d’ornamento: L’ostentazione dello status attraverso l’uso delle asce pendenti nell’Italia neolitica ed eneolitica. In N.  Negroni Catacchio (Ed.), Atti del convegno dodicesimo incontro di studi Valentano (VT) — Pitigliano (GR)  — Manciano (GR), 12–14 Settembre 2014  — ornarsi per comunicare con gli uomini e con gli dei gli oggetti di ornamento come status symbol, amuleti, richiesta di protezione ricerche e scavi (pp. 111–127). Milano: Centro Studi Preistoria e Archeologia. Barfield, L. (1978). North Italian faience buttons. Antiquity, 52(205), 150–153. Barfield, L.  H. (1986). Chalcolithic burial in northern Italy: Problems of social interpretation. Dialoghi di Archeologia, 4(2), 241–248. Belardelli, C., & Pascucci, P. (1996). Repertorio dei siti protostorici del Lazio: Province di Rieti e di Latina. Roma: Regione Lazio. Belardelli, C., Di Gennaro, F., Angle, M., & Trucco, F. (Eds.). (2007). Repertorio dei siti protostorici del Lazio: Province di Roma, Viterbo e Frosinone. Firenze: All'insegna del Giglio Editore. Bellintani, P. (2000). Bottoni conici ed altri materiali vetrosi delle fasi non avanzate della media età del Bronzo dell’Italia settentrionale e centrale. Padusa (Nuova serie), XXXVI, 95–110. Bellintani, P., Angelini, I., Artioli, G., & Polla, A. (2005). Bottoni conici e perle in glassy faience delle fasi iniziale e piena della media età del Bronzo dell’Italia Centrale Tirrenica: archeologia ed archeometria. Padusa (Nuova serie), XLI, 223–230. Bettelli, M. (1997). Elementi di culto nelle terramare. Le Terramare: La più antica civiltà padana, hrsg. von M. Bernabò Brea, A. Cardarelli und M. Cremaschi (pp. 720–725). Milano: La più antica civiltà padana. Bianco Peroni, V. (1970). Le spade nell'Italia continentale/Die Schwerter in Italien. Prähistorische Bronzefunde IV.1. München: C. H. Beck. Bianco Peroni, V. (1994). I pugnali nell'Italia continentale. Prähistorische Bronzefunde 10. Stuttgart: Franz Steiner. Biddittu, I. (1987). Insediamento dell'età del bronzo nelle Grotte di Pastena. Latium, 4, 237–239. Biddittu, I., Bruni, N., Carancini, G. L., Cerqua, M., & Riva, A. (2006). Recenti acquisizioni sulla frequentazione delle Grotte di Pastena in età pre-protostorica. In G. Ghini (Ed.), Atti del convegno secondo incontro di studi sul Lazio e la Sabina (pp. 273–282). Roma: Quasar. Biddittu, I., Bruni, N., Carancini, G. L., Cerqua, M., & Riva, A. (2007). La frequentazione delle grotte di Pastena in età preistorica e protostorica. In D.  Cocchi Genick (Ed.), Atti della XL Riunione Scientifica dell’Istituto Italiano Preistoria e Protostoria: Strategie di insediamento fra Lazio e Campania in età preistorica e protostorica volume II (pp.  683–693). Firenze: L’Erma di Bretschneider. Bietti Sestieri, A.  M. (2010). L'Italia nell'età del bronzo e del ferro: Dalle palafitte a Romolo (2200–700 a.C.). Roma: Carocci. Bradley, R. (1998). The passage of arms: An archaeological analysis of prehistoric hoard and votive deposits. Oxford: Oxbow Books.

110

L. Silvestri et al.

Bradley, R. (2005). Ritual and domestic life in prehistoric Europe. London: Routledge. Capasso, L., & Piccardi, M. (1980). La Grotta dello Scoglietto: Un probabile centro nosocomiale dell'antica età del Bronzo in Toscana. Rivista di Scienze Preistoriche, 35(1–2), 165–181. Carancini, G. L. (1984). Le asce nell’Italia continentale II. Prähistorische Bronzefunde. Abteilung IX, Band 12. München: C. H. Beck. Casi, C., D’Ercole, V., Negroni Catacchio, N. N., & Trucco, F. (1995). Prato di Frabulino (Farnese, VT): Tomba a camera dell’età del bronzo. In N. Negroni Catacchio (Ed.), Preistoria e protostoria in Etruria. Atti del primo incontro di studi. Saturnia (Manciano) — Farnese 17–19 maggio 1991. La cultura di Rinaldone. Ricerche e scavi (pp. 81–110). Milano: Centro Studi di Preistoria e Archeologia. Cocchi Genick, D. (1999). La funzione delle grotte e il significato delle acque nelle manifestazioni di culto di epoca protostorica dell’Italia medio-tirrenica. OCNUS, 7, 167–177. Cocchi Genick, D. (2002). Grotta Nuova: La prima unità culturale attorno all’Etruria protostorica. Lucca: Baroni. Costantini, L., & Costantini Biasini, L. C. (2007). Economia agricola del Lazio a sud del Tevere tra Bronzo antico e Bronzo medio. In D. Cocchi Genick (Ed.), Atti della XL Riunione Scientifica dell’Istituto Italiano Preistoria e Protostoria: Strategie di insediamento fra Lazio e Campania in età preistorica e protostorica volume II (pp. 787–801). Firenze: L’Erma di Bretschneider. Cremonesi, G. (1968). Contributo alla conoscenza della preistoria del Fucino: la Grotta di Ortucchio e la Grotta La Punta. Rivista di Scienze Preistoriche, 23, 145–204. Cremonesi, G. (1976). La Grotta dei Piccioni di Bolognano nel quadro delle culture dal neolitico all'età del bronzo in Abruzzo (vol. 2). Pisa: Giardini. de Cleene, M., & Lejeune, M. C. (2004). Compendium of symbolic and ritual plants in Europe. Ghent: Mens & Cultuur Uitgevers. Delpino, A.  M., Sarti, L., Silvestrini, M., Martini, F., Conati Barbaro, C., Muntoni, I.  M., & Volante, N. (2009). Adriatico e Tirreno a confronto: Analisi dell'occupazione territoriale tra il Neolitico finale e l'età del Rame in alcune aree campione dell'Italia centrale. Rivista di Scienze Preistoriche, 59, 115–180. D’Ercole, V. (1997). Spade dell’età del bronzo deposte nelle acque dei fiumi e dei laghi abruzzesi. In M. Pacciarelli (Ed.), Acque, grotte e Dei. 3000 anni di culti preromani in Romagna, Marche e Abruzzo (pp. 72–77). Imola: Musei Civici di Imola. Di Fraia, T., & Grifoni Cremonesi, R. (1996). La Grotta Sant'Angelo sulla Montagna dei Fiori (Teramo): Le testimonianze dal Neolitico all'Età del Bronzo e il problema delle frequentazioni cultuali in grotta. Pisa: Istituto Editoriali e Poligrafici Internazionali. Di Fraia, T., & Tiberio, D. (2008). Nuovi dati dalla Grotta Sant’Angelo di Civitella del Tronto (TE): quali culti e/o quali attività. In N.  Negroni Catacchio (Ed.), Paesaggi realie paesaggi mentali. Ricerche e scavi. Atti ottaco incontro di preistoria e protostoria in Etruria (vol. I) (pp. 477–489). Milano: Centro Studi di Preistoria e Archeologia. Dolfini, A. (2011). The function of chalcolithic metalwork in Italy: An assessment based on use-­ wear analysis. Journal of Archaeological Science, 38(5), 1037–1049. Grifoni Cremonesi, R. (1996). Osservazioni sulle buche e sulla problematica dei culti. In T. Di Fraia & R.  Grifoni Cremonesi (Eds.), La Grotta Sant’Angelo sulla Montagna dei Fiori (Teramo) (pp. 305–337). Pisa: Istituti Editoriali e Poligrafici Internazionali. Grifoni Cremonesi, R. (1999). Alcune osservazioni sui culti delle acque e sulla frequentazione delle grotte dal Neolitico all’età del Rame. OCNUS: Notebooks of the School of Specialization in Archaeological Heritage, 7, 159–165. Grifoni Cremonesi, R. (2000). Sull’interpretazione di alcuni aspetti funerari e cultuali nel neolitico abruzzese. In P. Biagi (Ed.), Studi sul Paleolitico, Mesolitico e Neolitico del bacino adriatico in ricordo di A. M. Radmilli (pp. 127–139). Trieste: Società per la Preistoria e Protostoria della Regione Friuli-Venezia Giulia. Guareschi, C., & Morandini, G. (1943). Ricerche nella Grotta di Pastena, Luppa e Pietrasecca, Lazio. Bollettino della Società Veneziana di Storia Naturale e Museo Civico di Storia Naturale, 3(1), 43–62.

6  How to Detect Ritual in Middle Bronze Age Italy? A Contextual Approach…

111

Guidi, A. (1992). Recenti ritrovamenti in grotta nel Lazio: Un riesame critico del problema dell’utilizzazione delle cavità naturali. Rassegna di Archeologia, 10, 427–437. Miari, M. (1995). Offerte votive legate al mondo vegetale e animale nelle cavità naturali dell'Italia protostorica. Atlante Tematico di Topografia Antica, 1(1), 11–29. Miracle, P. T., & Forenbaher, S. (2006). Prehistoric herders of northern Istria: The archaeology of Pupicina cave (vol. 1). Pula: Arheoloski muzej Istre. Parker Pearson, M. (1999). The archaeology of death and burial. Stroud: Sutton Publishing. Pluskowski, A. (2012). The ritual killing and burial of animals: European perspectives. Oxford: Oxbow Books. Puglisi, S. M. (1959). La civiltà appenninica. Firenze: Sansoni. Radmilli, A.  M. (1963). La preistoria d'Italia alla luce delle ultime scoperte. Firenze: Istituto Geografico Militare. Radmilli, A.  M. (1975). Culti di fertilita'della terra testimoniati in alcuni giacimenti neolitici Italiani. Capo di Ponte: Edizioni del Centro. Radmilli, A. M. (1977). Storia dell'Abruzzo dalle origini all'età del bronzo. Pisa: Giardini. Radmilli, A., Mallegni, F., & Fornaciari, G. (1978). Recenti scavi nella Grotta dei Piccioni di Bolognano (Pescara) e riesame dei resti scheletrici umani provenienti dai circoli. Atti della Societa Toscana di Scienze Naturali Residente in Pisa, 85, 175–198. Robb, J.  (1994). Gender contradictions, moral coalitions, and inequality in prehistoric Italy. Journal of European Archaeology, 2(1), 20–49. Rolfo, M. F., Achino, K. F., Fusco, I., Salari, L., & Silvestri, L. (2013). La Grotta Mora Cavorso a Jenne (Roma): I livelli dell’antica-media età del Bronzo. Rivista di Scienze Preistoriche, LXIII, 95–123. Rolfo, M. F., Achino, K. F., Fusco, I., Salari, L., & Silvestri, L. (2016). Reassessing human occupation patterns in the inner central Apennines in prehistory: The case-study of Grotta Mora Cavorso. Journal of Archaeological Science: Reports, 7, 358–367. Segre, A. G. (1946). La speleologia moderna e le esplorazioni sotterranee nell’Appennino centro-­ meridionale. Historia Naturalis, Rivista dell’Unione Italiana Naturalisti (Roma), 1(2), 47. Seuterio, M. (1730). Gli stati del Sommo Pontefice chiamati il patrimonio di San Pietro ecc. delin. da Matteo Seuterio, di S. M. Imp. geogr. in Augusta. Skeates, R. (1995). Animate objects: A biography of prehistoric ‘axe-amulets’ in the Central Mediterranean region. Proceedings of the Prehistoric Society, 61, 279–301. Skeates, R., Angle, M., Beckett, J., Cavazzuti, C., & Silvestri, L. (in prep.). Cave burial in Middle Bronze Age central Italy: A preliminary report on recent work at Grotta Regina Margherita, Collepardo (FR). Silvestri, L., Achino, K. F., Gatta, & M., Rolfo, M. F. (2017). Bioarchaeological remains as indicators of costly signalling: Two case-studies from the Middle Bronze Age of Central Italy. World Archaeology, 49(4), 1–15. Silvestri, L., Rolfo, M. F., Angle, M., Skeates, R., & Salari, L. (2018). Faunal remains and ritualization: Case studies from Bronze Age caves in Central Italy. In A. Livarda, R. Madgwick, & S. Riera Mora (Eds.), The bioarchaeology of ritual and religion (pp. 127–147). Oxford: Oxbow Books. Silvestri, L., Rolfo, M. F. & Salari, L. (2016). La fauna della media età del Bronzo di Grotta Mora Cavorso (Jenne, Lazio). In U. Thun Hohenstein, M. Cangemi, I. Fiore, & J. De Grossi Mazzorin (Eds.), Atti 7° Convegno Nazionale di Archeozoologia, Ferrara-Rovigo 22–24 Novembre 2012, Volume 12(1) (pp. 121–128). Sezione: Museologia Scientificae Naturalistica.  Sørensen, M. L. S. (2013). Gender archaeology. Chicester: John Wiley & Sons. Tarlow, S., & Nilsson Stutz, L. (Eds.). (2013). The Oxford handbook of the archaeology of death and burial. Oxford: Oxford University Press. Tinè, S., & Isetti, E. (1980). Culto neolitico delle acque e recenti scavi nella Grotta Scaloria. Bullettino di Paletnologia Italiana, 82, 31–70. Tomkins, P. (2009). Domesticity by default: Ritual, ritualization and cave-use in the Neolithic Aegean. Oxford Journal of Archaeology, 28(2), 125–153.

112

L. Silvestri et al.

Tongiorgi, E. (1947). Grano, miglio e fave in un focolare rituale dell’Età del Bronzo a Grotta Misa. Nuovo Giornale Botanico Italiano, 54, 804. Turner, V. (1995). The ritual process: Structure and anti-structure. New York: Aldine de Gruyter. Whitehouse, R. D. (1992). Underground religion: Cult and culture in prehistoric Italy. London: Accordia Research Centre, University of London. Whitehouse, R. (Ed.). (1998). Gender and Italian archaeology: Challenging the stereotype. London: Accordia Research Centre/Institute of Archaeology, University of London.

Chapter 7

Pan Rituals of Ancient Greece Revisited Nektarios-Petros Yioutsos

7.1  Introduction The use of caves either for habitation or ritual purposes dates back to the beginning of humankind. Evidence from a large number of caves prove that these breathtaking underground halls were the sites of recurring ceremonies, while the large number of cave drawings—often in extremely unreachable parts of caves—illustrates human efforts to encounter the supernatural deep underground (Betts 2003; Breuil 1933; Burenhult 2004, p.  112–117; Kuehn 1956; Whitehouse 2001).1 Archaeological research in Greece verifies the ritual use of caves since the prehistoric period (Jones 1999; Marinatos 1940–1941; Tomkins 2009). Especially during the Classical period, an expansion of ritual cave activity is evidenced in the institution of sanctuaries in many caverns devoted to various agricultural deities such as the cult of Pan and the Nymphs (Arapogianni 2000). The principal seat of Pan’s worship was Arcadia, and from there his name and cult spread over other parts of Greece. In his homeland his cult was not established inside caves but in sanctuaries high up in mountains. According to the Arcadians, the god—a spirit of the mountainous countryside—dwelled in mountains and rocks, and he was worshipped alone, having no female companions (Pausanias VIII, 24.4, 26.2; Strabo VIII, III.12).2 When his cult was transported to Attica, after the Persian Wars, the Athenians offered them a small grotto on the north-western slope of the  For example, caves Pech Merle, Les Trois Frères and Le Tuc d’Audoubert (France); Cogul, Gasulla Ravine, Minateda, Jimena de Jean-Cueva de la Graja, Tivissa-Font Vilella, Cueva de Los Letranos and Cantos de la Visera (Iberian Peninsula); Cave Bhimbetka (India); etc. 2  Pan is the god of the wild, of shepherds and flocks, of nature and of mountain wilds and rustic music, and is also the companion of the Nymphs. The god was depicted as a goat-legged man with horns and a goat’s tail, a thick beard, snub nose and pointed ears. 1

N.-P. Yioutsos (*) Department of Ancient History and Archaeology, National Kapodistrian University of Athens, Athens, Greece © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_7

113

114

N.-P. Yioutsos

Acropolis (Herodotus Histories VI, 105–106; Lucian Bis Accusatus 9, Dialogi Deorum 2.3), and thereafter his cult was connected with the cult of the Nymphs: vegetation deities who were already worshiped in caves and caverns in many places in the past, e.g. Corycian Cave on Mount Parnassus and Frini Cave in Lefkada (Pasquier 1977; Tzouvara Souli 1998).3

7.2  Pan in Athens The cult of Pan and the Nymphs in Athens was deeply connected to dance and music. Numerous votive offerings (mostly reliefs, clay plaques and statuettes)—discovered in many sanctuaries and especially grottos throughout Attica and the rest of the Greek countryside—depict the circular dance of the Nymphs, either around the musician (usually on the clay statuettes) or inside a grotto around an altar, while listening to the music played by the goat-legged god (mostly on the reliefs). These offerings represent a continuous ritual tradition that lasted for many centuries, and, despite different stylistic and thematic additions, it preserved its basic characteristics unspoiled (Figs.  7.1 and 7.2) (Amandry 1984; Edwards 1985; Feubel 1935; Romaios 1905; Tzouvara Souli 1998). The Athenians were the first to establish Pan’s cult inside a cave; a natural sanctuary away from the urban centres and the urban way of life.4 The presence of water sources, evidence for previous use, closed entrances, high altitude and liminality made caves particularly attractive as homes and sanctuaries of the Nymphs and Pan (Pierce 2006, p. 93–94).5 The grotto was not necessary when the cult existed only in Arcadia, but in the eyes of the Athenian citizen, the cave had similarities to Pan’s mountainous homeland and suited his bucolic characteristics. In other words, this was the perfect place for someone to practice Pan’s cult and express at the same time the god’s Arcadian origin, and his wild and primitive character (Figs. 7.3, 7.4, 7.5, and 7.6) (Pierce 2006, p. 4; Wickens 1986, p. 168–169). An interesting example of the New Comedy is Dyskolos (The Grouch), the only play written by Menander in the fourth century BC that survives almost in its entirety. It is set in and around the sacred cavern of Pan and the Nymphs in Phyle, Attica, where a number of festivities in honour of these deities are performed in the cave interior (and confirmed by archaeological investigation of the cave; Skias 1918). Alarmed by a dream sent by the god Pan, Sostratos’ mother has planned a 3  Also the caves of Melissani in Cephalonia, Mpoliatso in Lefkada, Polis in Ithaca, Paliambela in Vonitsa, etc. 4  Caves in Attica dedicated to Pan and the Nymphs have been discovered on the north-west slope of the Acropolis in Vari; at Dafni, Marathon and Eleusis; on Mount Penteli, Mount Parnitha, Mount Hymettus, etc. 5  All four properties were not necessarily needed for selection of a cave as sanctuary for the Nymphs, but they were influential in the selection process (Pierce 2006: 93–94). Larson (2001) considers them loci amoeni (‘pleasant spots’) and gives us a number of examples from the ancient sources that verify this characterisation.

7  Pan Rituals of Ancient Greece Revisited

115

Fig. 7.1  Votive relief depicting the dance of the Nymphs from Vari Cave, Attica (no. 2008, National Archaeological Museum, Athens; Reprinted from Thallon 1903, Pl. III)

Fig. 7.2  Votive relief depicting the dance of the Nymphs from Vari Cave, Attica (no. 2009, National Archaeological Museum, Athens; Reprinted from Thallon 1903, Pl. V)

116

N.-P. Yioutsos

Fig. 7.3 The gorge of Goura on Mount Parnitha in Attica leading to the cave of Pan (Photograph: author)

Fig. 7.4  The plateau in front of Pan’s cave on Mount Parnitha (Photograph: author)

7  Pan Rituals of Ancient Greece Revisited

117

Fig. 7.5  The southern slopes of Mount Hymettus near Vari, Attica. The arrow points to the location of the cave (Photograph: author)

Fig. 7.6  View to the south from the cave at Vari (Photograph: author)

118

N.-P. Yioutsos

sacrifice in his sacred cave to ward off the evil portent. While the play’s plot evolves, we are given information on the appropriate manner in which to worship these ­deities (lines 50–53). Ahead of the festive procession come the slaves, in order to prepare the shrine for the forthcoming religious festival (lines 419–423). They bring along the sacrificial animal (lines 393–400), cooking utensils, baskets, jugs of wine, offerings, torches and rugs for the banquet (lines 404–406, 440, 456, 964). Next, the crowd of drunken worshippers and musicians approaches, making noise and playing music, and they enter the shrine (lines 230–233, 432–434). In the cave’s interior, the animal is sacrificed (lines 197–199, 438–439). After, incense, cakes and parts of the animal are offered to the gods, and finally the rest of the meat is cooked and feasted on by the participants (lines 447–453, 546–555, 560–563). During the banquet, wine is mixed and distributed among the celebrants (lines 855–860, 901–902), and the ceremony reaches its climax with the performance of dancing rituals accompanied by music (lines 880, 909–910, 946–953; Lucian Bis Accusatus, 10; Lonsdale 1993, p. 268–272).6 It must have been an exciting experience for the celebrants who were taking part. The god’s followers would have started early in the morning for their ritual procession to the sacred cave up on the mountain.7 The difficulty in approaching such sanctuaries, always located far from cities or other inhabited places, demanded a reasonable degree of physical fitness and agility (see also Casteret 1938, p.  33, 45–47). According to Ustinova (2009, p. 67): Men born and bred to live in human society, when separated from it in an untamed landscape, would feel a spectrum of emotions from fear to bucolic delight, and would ascribe these sensations to the gods of the countryside.

Body fatigue from the difficult ascent in the wilderness, combined with emotional stimulation and intoxication from the consumption of alcohol, likely affected the participants’ senses in various ways. Deep inside the earth, the sensory impression was the result of an elaborate composition of physical influences diametrically opposed to an open-air landscape (Tomkins 2012, p. 67). Dark spaces with dim light entering from small openings force individuals to use their other senses, such as touch and hearing, to reorient themselves. Sight limitations necessitate the use of other means of light, such as small fires, torches and lamps. The flickering effect of the flames reflected on the rocky walls, in conjunction with the sometimes chaotic spatial organisation of the cave interior and the various stalagmite and stalactite formations, would have produced strange visual

 Similar ritual processions to the Nymphs (and Pan) are depicted on votive reliefs nos. 1966, 3874, 2796 and 2798 and on the votive plaque from the Nymphs’ cave in Pitsa, Corinth, no. 16464 (540 BC), in the National Archaeological Museum, Athens. 7  Theocritus, in his Idylls (I.15–20), mentions that it is not appropriate to play the syrinx at noon for fear of Pan, as his hunt is over and he is tired and rests. Noon is typically a silent and motionless part of the day, and producing noise in the realm of Pan while he is at rest is extremely risky. So the noisy approach of the celebrants to the cave sanctuary probably took place early in the morning or in the evening. 6

7  Pan Rituals of Ancient Greece Revisited

119

Fig. 7.7  The triangular opening of the cave on Mount Parnitha (Photograph: author)

effects, inspiring the visitors’ imagination (Casteret 1938, p. 215; Groenen 1997).8 Furthermore, the combination of restricted bodily m ­ ovement due to various geological phenomena (cavities, tunnel-like perspectives, fallen boulders, gaping chasms, calcite formations, etc.); the extraordinary behaviour of water (moistening, leaking, flowing or dripping); the peculiar acoustics deriving from a mixture of natural and man-made sounds echoing in the chambers; the mild temperatures; the excitement of dancing, the smell of cooking meat, fat burning and burnt incense; the taste of food and drink; and the intoxicating effects of alcohol would have created a multi-sensory experience for the celebrants (described as a ‘mental vortex’ in Ustinova 2009, p. 29). In extreme cases, such an experience could induce dislocation or distortion and could even expose the human mind to altered states of consciousness in such a way that individual’s contemporaries believed them to be seized by the gods and transformed into ‘nympholepts’ or ‘panolepts’ (Figs  7.7 and 7.8) (Borgeaud 1988, p.  88–116; Connor 1988; Ustinova 2009, p. 55–68). 8  Pausanias (I.32.7) states that the stalactites and stalagmites inside the cave sanctuary in Marathon, Attica, resemble Pan’s herd of goats.

120

N.-P. Yioutsos

Fig. 7.8  Ground plan of Pan’s cave on Mount Parnitha, Attica (Adapted from Deligiorgi Alexopoulou 1980, p. 137, Fig. 9)

The appropriate musical instruments for such ritual performances were a blown instrument, a flute (διαυλòς) or a syrinx (σύριγξ), accompanied probably by ­percussive sounds performed by idiophones, such as krotala held by the dancers, or by the clapping of the audience (ψιλῷ… τῷ κρότῳ, Lucian Bis Accusatus, 10). Representations of the dance of Pan and the Nymphs on votive reliefs confirm the use of blown instruments in combination with krotala and possibly clapping. The sound of pipes, in conjunction with the percussive noise, would have been experienced within an auditory arena (the underground chambers) with interesting sonic qualities. Music and clapping combined with the sound of the dancers’ feet on the ground would have been distorted by the cave walls, resonating in the interior and producing a sonorous noise suitable for Pan rituals.

7.3  Hearing the God Pan Pan is a deity connected with natural sounds, echoes and loud noise; sound is an important component of his cult. In the fourth century BC, Menander verifies the aforementioned statement in Dyskolos (lines 432–433), when he mentions ‘that no man should approach this god without making any noise’ (on the threatening dimension of silence, see Montiglio 2000, p. 222–223). In the Homeric Hymn to Pan, he is called φιλόκροτος (line 2), which means ‘noise lover’, an adjective that connects him to rattling aural effects (Lonsdale 1993, p. 262); the hymn continues, describing the Nymphs as ‘calling upon Pan, the shepherd God’ (lines 5–6).

7  Pan Rituals of Ancient Greece Revisited

121

Similarly, in Helen, Euripides (lines 185–190) describes Pan’s rape of a Nymph in a cave as ‘voices with shrieks beneath the rocky hollows’. According to Suda, ‘women are accustomed to celebrate Pan with clamor’ (κραυγή), and the same term is used by the scholiast on Aristophanes when he comments on a passage in Lysistrata (lines 1–5), where women approached the sanctuary of Pan to the sound of tambourines and with inarticulate cries (see also Dyskolos, lines 433–435). The god is also manifested in the echo. His connection to echo is evident in the myth of Echo and Pan. The legend exists in a number of variants, all of which emphasise the futility of desire in pursuit of an unobtainable object. The most famous version of the myth is preserved by Longus in the second century AD (Daphnis and Chloe III.23). Echo was a beautiful nymph and also an accomplished musician who fled from all males, whether men or gods, because she loved her virginity. The Arcadian god fell in love with her, but despite the god’s divine status, Echo rejected his sexual advances. In anger, the hoofed god created such a ‘panic’ that it incited a group of shepherds to attack her. Inevitably, Echo’s body was torn into little pieces and spread all over the earth. The end of the story entails the intervention of Gaia, the goddess of the earth; she collected the pieces of Echo, and the Nymph does not die. Although in pieces, she continues to live, imitating the sounds and voices of others (Borgeaud 1988, p. 79). To his followers, Pan was near but yet invisible. They never saw the goat-legged god. Pan was all sound. He was voice without bone: a disembodied voice. He could be experienced through the sound of goats’ hoofbeats, the shrill wind on the mountains, the sound of the echoing hillsides and the resonating underground chambers.9 He is ‘lord of the frivolous music and with his far-sounding flute he pours an inspired seductive melody’ on the landscape (Hymn to Pan from Epidaurus; text in Borgeaud 1988, p.  149). According to Bregman (1990, p.  11), ‘the human mind cannot consciously experience disembodied features, like sound, so he must assign them to perceived objects’. For the Ancient Greeks, echoes and all natural sounds could only belong to Pan sounding the flute and the syrinx in caves with the Nymphs.10 Mountains, glens, and all the grottos of the mountains are liable to echo. There are all sorts of complicated noises in the mountains produced by dogs and wild animals: their echoes become mixed together. So it often happens that people do not see the creatures making the noises, but hear only the disembodied voice by itself, and so say that Pan is sounding the flute and the syrinx in the caves with the Nymphs. (Apollodorus quoted by the scholiast on Euripides, Rhesus 36, trans. Borgeaud 1988, p. 92)

 In the late first century AD, Crinagoras (Anthologia Palatina, 6.253) gives a rather thorough account of the landscape around Bassai in Phigalia (Peloponnesus). While recording a gift of thanks by a hunter named Sosandros, the poet describes the caves of the Nymphs as ‘with many springs’ (ἐυπίδακες) and the cabin of Pan as ‘echoing’ (ἠχηεσσα). 10  Ιn the Orphic Hymn to the Nymphs (lines 6–10), the Nymphs are called ‘lovers of grottoes, wandering the air; you of the springs, roaming ones, dew clad/ with steps; visible, invisible, you of the glens, with many blooms; leaping with Pan on the mountains as you cry out; flowing from rocks,/ clear-voiced, buzzing like the bee, mountain haunting’. 9

122

N.-P. Yioutsos

While explaining how Pan’s cult was introduced to Athens, Herodotus (Histories VI, 105–106) describes an incident that occurred just before the Battle of Marathon in September 490 BC. He narrates the story of Pheidippides, a professional long-­ distance runner, who encountered the Arcadian god near Tegea in Peloponnesus while returning from his mission in Sparta to solicit its aid against the Persians. In the account given to the Athenians after his return, Pheidippides stated that he had met the god Pan on Mount Parthenion, above Tegea. To describe the scene of the encounter, Herodotus uses the verb περιπίπτω (‘to encounter’) rather than ὁρω (‘to see’). Pan, he continues, ‘called him by name’ (βώσαντα δέ τοὔνομα) and ‘told him to ask the Athenians’ (Ἀθηναίοισι κελεῦσαι ἀπαγγεῖλαι) why they paid him no attention in spite of his friendliness towards them. The god also referred to the fact that he had often been useful to their city in the past and would be so again in the future. The ancient historian chooses consciously to use terms which are connected to sound and oral communication, rather than visual contact. During the divine encounter, the Athenian messenger probably never saw the goat-legged god; Pheidippides heard him only (though hearing Pan was not always pleasant).11 Pan, in addition to his other powers, had the capacity to instil the most extreme fear: an irrational, blind fear that paralysed the mind and suspended all sense of judgement, i.e. ‘panic’ (πανικαί ταραχαί, πανικόν δείμα in Borgeaud 1988, p. 89, 229). Though Herodotus notes Pan’s promise to help the Athenians in the upcoming battle against the Persian invasion, he does not explain how the god actually achieved this. As a shepherd god, Pan was responsible for maintaining calmness in cattle, so that they would not scatter in fear; an ability that, in the ancient mind, extended also to soldiers. According to Borgeaud’s hypothesis (ibid., p. 94–95), the god was probably credited for the panic exhibited by the fleeing Persian soldiers after the battle (Pierce 2006, p. 41). Our earliest evidence on panic occurs towards the beginning of Rhesus, where Hector asks the guards—who had left their night posts to bring report—whether they had been disturbed by Pan (Euripides Ῥῆσος, lines 34–37). Pan’s association with natural sounds, echoes and panic in battle is also confirmed by an incident that took place at the time of the Thirty Tyrants. According to the narration of Diodorus Siculus (Library 14.32.3), during an attempt by the Thirty Tyrants to siege the stronghold of Phyle in Attica, which was occupied by a band of rebels led by Thrasybulus in 404 BC, there was ‘an uproar which men call panic’ (θορύβου τοῦ καλουμένου Πανικοῦ) that struck the army, forcing them to move their camp to another location. Polyaenus, in the second century AD (Stratagems 1.2), describes Dionysus’ strategy to use the echoing effect of the surrounding rocky slopes and hollows of the forest to amplify the loud shout produced by his army and so to induce fear in the enemy. Moreover, according to Pausanias (X.23.5–8), the same irrational and unexplained panic was also experienced by the Gauls, led by Brennos, after their defeat by the Greeks at Delphi in mid-winter AD 279; at dusk they heard  Similarly, Pausanias (I.28.4), in his account of the same incident, prefers to use the terms ἐντυχόντα (‘to meet’), φάναι (‘to speak’) and ἀγγελίᾳ (‘announcement’): terms which are all connected with sound and oral communication.

11

7  Pan Rituals of Ancient Greece Revisited

123

sounds (probably from the rocks), but as confusion fell on the army, they imagined that they heard the trampling of galloping horses and the attack of advancing ­enemies (on the priority given to aural senses in combat, see Remarque in Kahn 1999, p. 63, note 76). For the sake of Pan’s love of noise, a ritual performance to Pan and the Nymphs involved the production of various sounds (e.g. clapping and laughter in Lucian Bis Accusatus, lines 9–10). The aural qualities of caves, responding to these sound signals, would not have gone unnoticed by their followers. To approach this god, the ritual protocol demanded noise, and the resulting resonating and echoing effects of caves would have been regarded as signs of his divine presence.12 Furthermore, when there was no recorded experience of the god’s presence in a natural space, one could perform a sort of evocation ritual in order to provoke these auditory effects by using various noisemakers, such as musical instruments, clapping hands and the voice (ἀνακεκλόμεναι νόμιον θεόν in the Homeric Hymn to Pan, line 6; Dayton 1992, p. 28; Eliade 1959, p. 27). If the landscape responded to the artificially produced noise by sending sonorous sounds and echoes back, these aural features may have been regarded as indexes of Pan’s strong or weak presence (on the hallucinogenic sounds of natural spaces, see Jaynes 1976, p. 334).13 Moreover, during a Pan ritual, the participants, through their noisemaking, dancing, music and feasting, were creating a system of ritual action, in which they themselves became agents of the god’s epiphany (on art as agency, see Gell 1998).

7.4  Hearing the Past Despite Pan’s intimate connection with sound, there has been virtually no research carried out on the aural qualities of his cave sanctuaries. There have been many excavations undertaken in various ancient cave sanctuaries in Greece, but the accompanying research is mostly interested in their visual characteristics, their location, detailed visual description, artefact recording, interpretation and dating. Research is likewise evaluated more on the visual sense of the environment than on its aural dimensions (Bull and Back 2003, p. 1). There is, in fact, no reference to the possible aural phenomena (sound resonance, sound distortion, echo effects, etc.) present at these locations, nor on the numerous ways in which these may—together with other visual, olfactory and tactual features—have affected the experience of the caves’ visitors (Marinatos 1964; Sakellarakis 1983; Weller 1903; Wickens 1986).  We could describe the sound of echo as an ‘earcon’ of classical antiquity: the aural analogue of a visual icon. Earcon in computer science is a sonic event that contains special symbolic meanings. As such, the sound of an echo in the Greek countryside would have represented an ‘earcon’ denoting Pan’s presence. 13  According to Eliade (1959, p. 26–27), holy places exist naturally in the world, and they are not created but rather discovered by humans. Holy places are revealed as qualitatively different from surrounding areas by special signs. 12

124

N.-P. Yioutsos

Fig. 7.9  The shaft opening of the Vari Cave (Photograph: author)

Neither do we know, for example, what visual and aural qualities encouraged Archedimos the Nympholept to establish a sanctuary dedicated to Pan and the Nymphs inside a cave at Vari in Attica in the fifth century BC (Figs. 7.9, 7.10, 7.11, and 7.12) (Schomer and Goette 2004; Weller 1903).14 Visual senses dominate aural senses in Western culture, and, until recently, ­ocular centricity has dominated modern scientific thought (Howes 2003; Ihde 2007, p. 6–15; Synnott 1991, p. 63).15 There has, however, in the past few decades, been a turn towards the acoustic and sensual properties of the material past in such a way that we may speak of an expansion of ‘sensuous scholarship’ within the humanities (Fahlander and Kjellstroem 2010, p.  4; Samuels et  al. 2010, p.  330; Stoller in Erlmann 2004, p. 2). Within sensual archaeology, researchers try to expel the ‘tyranny of the gaze’, to understand how people in the past sensed and reacted to their natural environment and how sensory stimuli affected and shaped cultural identity, beliefs, actions and thought (Hamilakis 2011, p. 208).  Norbert Casteret (1938, p. 97–98), one of the fathers of modern speleology, describes his first visit into a cave as follows: ‘Flat on my stomach, with beating heart, I crawled into my first real cavern. It was the dry bed of an underground water-course, and I had to crawl on the soft clay bottom. The cool air, the damp soil and soundless blackness made a striking contrast with the world outside. They created an atmosphere of their own; I was entering another, a mysterious world, which scared me at the same time that it filled me with a mystical enthusiasm… From this rose a new sound, since grown familiar to my ears, the murmur of a brook in an unknown lower level. Surprise and delight riveted me to the spot’. 15  According to Kahn (1999, p. 5), ‘Sound’s life if too brief and ephemeral to attract much attention’ and ‘not until recently have there existed the conceptual and technological techniques available to sustain a full range of sounds outside the unstable environs of their own time’. 14

7  Pan Rituals of Ancient Greece Revisited

125

Fig. 7.10  Relief figure of Archedimos the Nympholept and shrine of Apollo and Hermes (?) in the interior of the Vari Cave. (Drawn by Curtius und Kaupert; reprinted from Weller 1903, p. 271, Fig. 6)

Fig. 7.11  Inscription from the interior of Vari Cave referring to Archedimos of Thera, the Nympholept, who built the sanctuary at the Nymph’s counsel (Reprinted from Dunham 1903, p. 299, Fig. 2)

Increasingly, research has been undertaken on the aural perspectives of ancient sacred grottos and other rural sacred places, offering the potential to enrich interpretations of how ancient buildings or natural spaces were used, and underlining the importance of sound as one of the determinants in their identification as places of divine presence and worship. It has been found that the majority of Palaeolithic cave art correlates with locations of optimum acoustics and that most ancient graffiti is situated in places where particular musical notes resonate (Reznikoff 2008; Scarre 1989; Waller 1993). Iegor Reznikoff (2004–2005, 2012) has explored the resonance and echo phenomena of painted caverns in France (e.g. caves Niaux in Ariege, and Le Portel and Ancy-sur-­ Cure in Burgundy), and has discovered that echoes produced by lithophones or by the human voice tend to resonate stronger near rocky wall surfaces decorated with rock art (see also Rifkin 2009, p. 594; Tomkins 2012, p. 62).16 As a result, he has discovered a link between rock art and sound, and that ‘the density of pictures in a  The same rule seems to apply to the distribution of artefacts in the interior. In several cases the principal area of prehistoric activity seems actually to have been focused in the deepest parts of the cave.

16

126

N.-P. Yioutsos

Fig. 7.12  Ground plan of the cave of the Nympholept at Vari, Attica (Adapted from Weller 1903, Pl. I)

location of a cave is proportional to the quality of the resonance of this location’ (Reznikoff 2004–2005, p. 1). In addition, he used the sound of his voice as sonar (echolocation) in the cave at Ancy-sur-Cure, moving through dark spaces and locating decorated chambers deep inside (on echolocation see Kuttruff 2009, p.  209; Wenzel et al. 1991).17 More recently, Steven Waller has conducted studies at several Neolithic cave art sites in Europe, as well as at Native American rock art and other megalithic sites around the world, using modern acoustic methods to detect resonating effects and echo phenomena (see Waller, Chap. 12, this volume). During his investigations, he discovered that many caves show a high degree of correlation between the specific location of art and areas of enhanced echo effects; images of horses, bulls, bison and deer are located in areas with high levels of sound reflection, whereas feline art is  Reverberant sound is informative about the size and layout of the surrounding environment. The idea of experiencing a place as sensation has already been supported by Feld (1996) when writing on the Kaluli, New Guinea, who have developed acute hearing for locational orientation.

17

7  Pan Rituals of Ancient Greece Revisited

127

found in regions with poor acoustics (Waller 1993, p. 363). Moreover, he believes that echoes produced by percussion, clapping or stone toolmaking can mimic the sound of hoofbeats and posits whether there is a connection between the aural properties of underground chambers and the depictions of hoofed animals.18 He has also gathered ethnographically documented legends from different cultures which connect certain acoustical phenomena with various supernatural beings (animism) and believes that: the artists may have been attempting to depict the spirits who they felt inhabited the rocks and who were responsible for the sounds. Thus, motivation for both the context (location) and the content (subject matter) of rock art can be directly derived from the phenomenon of sound reflection via the ethnographically documented animism of echoing. (Waller 2002, p. 12–13)19

‘Rock gongs’, ‘lithophones’, ‘sounding stones’, ‘roaring rocks’, ‘ringing rocks’, ‘bell rocks’, ‘percussion boulders’ or ‘rock drums’ have been associated, by a number of researchers, with rock art, and rock art research is increasingly recognising the importance of all of the senses in ancient ritual performances (Fagg 1957; Garfinkel and Waller 2012; Ouzman 2001; Waller 2012; Watson 1997). Limestone configurations in the interior of the cave at Nerja in Malaga, Spain, possess interesting sonic qualities that, according to Dams (1984), could have been used in some kind of ritual deep inside the cave. He also believes that the impressive tilted calcite formations and the presence of water in the interior may have inspired cave artists to decorate the former with elaborate symbols. At the Kupgal rock art site in Karnataka, south India, petroglyphs of individual images and scenes span a whole range of periods from prehistory to the historical era. These depictions appear to be connected with various dolerite boulders used for percussion (Boivin 2004). According to Goldhahn (2002), the Neolithic hunter-gatherer rock-engraving site at Laxforsen in Sweden likewise appears to have been intentionally associated with ‘sounding water’, and he suggests that ‘the desire to place the rock-engravings in close contact with the wild water at the rapids must have been an overriding concern of the artists who did these engravings’ (ibid., p. 35). Recent acoustic surveys at various ancient sites around the world suggest these structures possess special acoustic properties and that sound played some part in their use or was central to their significance among the communities that built them. Aaron Watson and David Keating carried out a pilot study into the acoustic properties of Stonehenge and discovered resonance in space, filtering of low frequencies, and that the stones may have been intentionally shaped for acoustic purposes (Till 2010; Watson and Keating 1999). A further six Neolithic and Iron Age constructions  I wonder whether there is a similar connection between the hoofed god Pan and the percussive sounds performed in caves during Pan rituals. 19  One of the most important features in Tibetan sacred geography is the concept of gnas; gnas translates as ‘place’, but it has a more active meaning of ‘to abide’ or ‘abode’. Landscape features are the abode of spirit forces and deities of all kinds. According to Aldenderfer (2005, p. 10–11), these entities may contribute their power to these landscape features, creating gnas-chen (‘powerplaces’), whose power literally saturates the surrounding areas. 18

128

N.-P. Yioutsos

in the United Kingdom and Ireland—dolmen structures and cairns—seem to possess resonant acoustic properties, which likely contributed to their function (Jahn et al. 1995; see also Lawson et al. 1998; Watson 2001). Interesting sound effects have also been reported at the Hal-Saflieni Hypogeum in Malta, a subterranean structure dating to the Saflieni phase (3000–2500 BC) of Maltese prehistory. Low voices create eerie, reverberating echoes, and sounds made (or words spoken) in certain places can be clearly heard across all three levels (Mifsud et al. 1999; Till 2017). Furthermore, recent acoustic surveys suggest that the extensive underground network of galleries at the ritual complex of Chav’ın de Hu’antar in the central highlands of Peru form an unusual auditory environment, which increases the disorientation induced by the labyrinth layout and appears to have been associated with sounds emitted from conch shell trumpets during ancient rituals performed at the site in the past (Abel et al. 2008; for archaeoacoustics, see also Eneix and Zubrow 2014; Scarre and Lawson 2006). In Greece, existing studies in the field of ancient acoustics are mainly focused on Greek public buildings used for ritual and theatrical performances, such as the ancient Theatre of Epidaurus in Argolis, the Odeon of Herodes Atticus in Athens, and the Nekromanteion (Oracle of the Dead) at Acheron River in Preveza, etc. The results of these investigations are primarily based on measurement methods, including the assessment of established acoustic parameters (reverberation time (RT), clarity C50 and C80 etc.), together with historical and archaeological data, leading to conclusions on their acoustic behaviour. Analysis of the acoustic and meteorological data at the ancient open-air Theatre of Epidaurus indicated no significant effects from environmental factors and confirmed the renowned exceptional acoustic characteristics of the theatre for hearing speech in all audience positions. Detailed acoustic reconstruction models (auralisation) of the Odeon of Herodes Atticus in Athens (in its original roofed form) have shown that, in its original state, it provided acoustics appropriate for musical performances; this is in contrast to the open-air theatres of the time, which were used specifically for ancient drama performances. Moreover, digital recreation of the acoustics of various enclosed spaces of Ancient Greece, such as the vaulted hall of the Nekromanteion of Acheron, has delivered intriguing results regarding their functionality and modes of communication within the interior (Tsilfidis et al. 2011; Vassilantonopoulos and Mourjopoulos 2002a, b, 2009; Vassilantonopoulos et al. 2004; Vovolis 2011).20

7.5  Hearing Cave Soundscapes Caves are understood as complex enclosed spaces comprising numerous surfaces, objects and geometries that create an acoustic arena which behaves differently from analogous open-air spaces (Gaver 1993; Lawson et al. 1998; Rodaway 1994, p. 92;  ‘Auralisation’ is the term used within acoustics for the process of producing a digital recreation of the acoustics of a space or structure.

20

7  Pan Rituals of Ancient Greece Revisited

129

on the term ‘acoustic arena’, see Barry and Satler 2007). Likewise, the underground sanctuaries dedicated to Pan and the Nymphs should not only be perceived as landscapes with certain visual characteristics but also as soundscapes: the auditory equivalent of a landscape, with important aural features (Kelman 2010; Samuels et al. 2010; Schafer 1977).21 Although these features tend to be viewed as ‘undesirable’ in modern architectural contexts, in pre-modern societies they would have been valued as significant (Cross and Watson 2006, p. 111). According to Rodaway (1994, p. 144), caves (and the countryside) have ‘hi-fi’ soundscapes, with low ambient noise, where discrete sounds emerge with clarity; conversely, ‘lo-fi’ soundscapes produce so much sonic information that little can be heard distinctly.22 ‘Hi-fi’ soundscapes, with their special acoustic components— dripping or running water, gusts of wind, reverberating and echoing effects, etc. (Witmore’s (2006, p.  274) belles noiseuses) and the presence and noise of men (voice, human movement, music, etc.)—would have produced a sonorous blend of sounds over the natural soundscape, in such a way that we could define caves as ‘naturally-formed musical instruments of gigantic proportions’ (Whalley 2008, p. 261).23 In the interior of such a gigantic ‘instrument’, the celebrants of Pan and the Nymphs (deprived of vision) needed to use their other senses, such as hearing or touch to acclimatise and move around the space (Bull and Back 2003, p. 2; Frieman and Gillings 2007; Howes 2006; see also the terms synaesthesia in Cutowic 2002 and kinaesthesia in Hamilakis 2013). This was an intense experience, in which the body resonated with the underground soundscape in order to connect with the environment and come closer to the divine (Kato 2009, p. 80). While exposed to such an environment, the mind followed cognitive processes that ‘transformed raw sensation into an awareness that had meaning’, according to cultural influences and personal experiences (on auditory spatial awareness, see Barry and Satler 2007, p. 11–13). In order to discover the ‘mindscape’ of these ancient landscapes, ‘we need to think in multi-sensory terms’ (Witmore 2006, p. 283). Instead of focusing on how ancient people viewed these sacred caves, we should explore how caves were  Corbin (1998) uses the term ‘auditory landscape’ rather than ‘soundscape’. In Sophocles’ Philoctetes (lines 1453–1462), the castaway hero’s description of the isolated cave, which has long been his home, focuses not only on its visual characteristics but also on the sound effects experienced inside: ‘…in the inmost part of my cave, my head was drenched with the lashings/ of the sound winds, and you Hill of Hermes, which often in answer to my voice, sent me back a groan as I labored under the storm’ (trans. Larson 2001, p. 45). 22  According to Barry and Satler (2007, p. 32), the level of background noise determines the quality of an acoustic arena and the reliability of its auditory channels. A silent environment creates the best auditory channel, a noisy environment the worst. 23  Compare the ritual production of noise during Pan rituals over the natural soundscape of caves to the synesthetic metaphor dulugu ganalan of the Kaluli of Bosavi, Papua New Guinea. According to Feld (1996, p. 100), for the Kaluli, the soundscape of the rainforest prevails above all sounds. In order for music to be performed and heard, the Kaluli must dulugu ganalan (‘lift-up-over-sounding’). 21

130

N.-P. Yioutsos

p­ erceived and experienced by them (Frieman and Gillings 2007, p. 5; for the term ‘mindscape’, see Rifkin 2009, p. 585). Pan and the Nymphs had an intimate connection with echo and noisemaking, and sound seems to have been an important factor in antiquity in the determination of caves as their home.24 Their worship required sacrifice, and the preparation and communal consumption of food and drink by the participants, as well as the performance of ritual dances, probably accompanied by chanting and music as depicted in votive artefacts. These underground spaces may have been selected (and even modified) to conduct and manipulate sound, and to produce various sensory effects during rituals. Recent archaeoacoustic techniques, applied to ancient sacred grottos and other constructions, have paved the way for new research into ancient cult and ritual practice, offering the potential to enrich interpretations of how ancient buildings or natural spaces were perceived as loci of divine presence and worship (Eneix and Zubrow 2014; Scarre and Lawson 2006).25 With this in mind, an acoustic survey of caves and appropriate analysis of their characteristics, combined with archaeological and musical methods, is currently underway. Such an interdisciplinary project aims to enhance our existing knowledge regarding the utilisation of sound and acoustics during ritual ceremonies in caves. The survey involves cooperation between two research teams from acoustics and archaeology, who will focus on selected examples of Greek caves from a specific geographical region (Attica in Greece), where the combined worship of Pan and the Nymphs was first organised and performed during the fifth century BC. Single channel, multi-channel and binaural acoustic measurements will be obtained in different positions in the selected caves, in order to conduct controlled experiments and recordings. Furthermore, a flexible simulation framework for auralisation and desk-based study of cave acoustics will be developed and made suitable for interdisciplinary tests under controlled laboratory conditions. These experiments will help to indicate whether the acoustics of these spaces and their response to sound sources produced during cave rituals encouraged people to believe that they were the abodes of sacred spirits. These cave environments should be revisited and remapped, with all visual and aural features recorded using sophisticated electronic instruments. Furthermore, in an attempt to understand how Pan rituals were experienced by the participants, we should start making noise! Reconstruction of the ancient musical instruments used in these ceremonies (krotala, flute, syrinx) and the recording of their sounds inside these ritual spaces, in conjunction with various human sound effects (laughing, 24  The interplay between Pan and the Nymphs’ dance and music and Εcho is described in the Homeric Hymn to Pan (lines 20–25): ‘At that hour the clear-voiced (“λιγύμολποι”) nymphs are with him/ and move with nimble feet, singing by some spring of dark water (“πύκα ποσσίν… μέλπονται”)/ while Echo wails about the mountain-top (“κορυφὴν δὲ περιστένει οὔρεος Ἠχώ”)/ and the god on this side or on that of the choirs (“ἔνθα καὶ ἔνθα χορῶν”), or at times sidling into the midst,/ plies it nimbly with his feet (“πυκνὰ ποσὶν”)….’ 25  ‘Archaeoacoustics’ refers to the study of sound in archaeological contexts. This can be done by exploring natural sounds (and acoustics) at monuments and other sites or by investigating and measuring the acoustic parameters of a place using electronic instruments.

7  Pan Rituals of Ancient Greece Revisited

131

y­ elling, singing, dancing, clapping, etc.), will help us to understand how sound behaved within the cave environment, how it affected the human senses, and how it might have led the human mind to altered states of consciousness.

References Abel, J. S., Rick, J. W., Huang, P. P., Kolar, M. A., Smith, J. O., & Chowning, J. M. (2008). On the acoustics of the underground galleries of ancient Chav’ın de Hu’antar, Peru. https://ccrma. stanford.edu/groups/chavin/publications/Acoustics08.pdf. Accessed March 2013. Aldenderfer, M. (2005). Caves as sacred places on the Tibetan plateau. Expedition, 47(3), 8–13. Amandry, P. (1984). Les culte des Nymphes et de Pan à l’antre Corycien. In L’ antre Corycien II (Bulletin de correspondance hellénique, supplement IX) (pp. 395–425). Athens: École Français d'Athènes. Arapogianni, X. (2000). Τα σπήλαια του Πανός στην Αττική. Unpublished PhD thesis, University of Thessaloniki. Barry, B., & Satler, L. R. (2007). Spaces speak, are you listening? Experiencing aural architecture. Cambridge, MA: MIT Press. Betts, E. (2003). The sacred landscape of Picenum (900–100 BC): Towards a phenomenology of cult places. In J. Wilkins & E. Herring (Eds.), Inhabiting symbols: Symbol and image in the Ancient Mediterranean (pp. 101–120). London: Accordia Research Institute. Boivin, Ν. (2004). Rock art and rock music: Petroglyphs of the south Indian Neolithic. Antiquity, 78(299), 38–53. Borgeaud, P. [trans: Atlass, K., & Redfield, J.] (1988). The Cult of Pan in Ancient Greece. Chicago, IL: University of Chicago Press. Bregman, A. S. (1990). Auditory scene analysis: The perceptual organization of sound. Cambridge, MA: MIT Press. Breuil, H. (1933). Les peintures rupestres schématiques de la péninsule Iberique (vol. IV). Lagny: Imprimerie de Lagly. Bull, M., & Back, L. (2003). Introduction: Into sound. In M. Bull & L. Back (Eds.), The auditory culture reader (pp. 1–18). Oxford: Berg. Burenhult, G. (2004). Pech-Merle: A 20,000-year-old sanctuary. In G. Burenhult (Ed.), The illustrated history of humankind: People of the past (pp. 112–117). San Francisco, CA: Fog City Press. Casteret, N. (1938). Ten years under the earth. New York, NY: Greystone Press. Connor, W. R. (1988). Seized by the nymphs: Nympholepsy and symbolic expression in classical Greece. Classical Antiquity, 7, 155–189. Corbin, A. (1998). Village bells: Sound and meaning in the nineteenth-century French countryside. New York, NY: Columbia University Press. Cross, I., & Watson, A. (2006). Acoustics and the human experience of socially-organized sound. In C.  Scarre & G.  Lawson (Eds.), Archaeoacoustics (pp.  107–116). Cambridge: McDonald Institute for Archaeological Research. Cutowic, R. E. (2002). Synesthesia: A union of the senses. Cambridge, MA: MIT Press. Dams, L. (1984). Preliminary findings at the ‘organ’ sanctuary in the cave of Nerja, Málaga, Spain. Oxford Journal of Archaeology, 3, 1–14. Dayton, L. (1992). Rock art evokes beastly echoes of the past. New Scientist, 28, 14. Deligiorgi Alexopoulou, C. (1980). Η σπηλιά, ναός του Θεού Πάνα. Hellenic Speleological Society Bulletin, XVII(1), 113–176. Dunham, M.  E. (1903). The Cave at Vari. II.  Inscriptions. American Journal of Archaeology, VII(3), 289–300. Edwards, M.  C. (1985). Greek votive reliefs to Pan and the Nymphs. Unpublished PhD thesis, New York University.

132

N.-P. Yioutsos

Eliade, M. (1959). The sacred and the profane: The nature of religion. San Diego, CA: Harcourt. Eneix, L.  C., & Zubrow, E.  B. W. (2014). Archaeoacoustics: The archaeology of sound. In Publication of proceedings from the 2014 conference in Malta. Florida: OTS Foundation. Erlmann, V. (2004). But what of the ethnographic ear? Anthropology, sound, and the senses. In V.  Erlmann (Ed.), Hearing cultures: Essays on sound, listening and modernity (pp.  1–20). Oxford: Berg. Fahlander, F., & Kjellstroem, A. (2010). Beyond sight: Archaeologies of sensory perception. In F. Fahlander, & A. Kjellstroem (Eds), Making sense of things: Archaeologies of sensory perception. Stockholm studies in archaeology 53 (pp. 1–14). Stockholm: Postdoctoral Archaeology Group, Stockholm University. Fagg, B. E. B. (1957). The cave paintings and rock gongs of Birmin Kudu. In J. D. Clark (Ed.), Proceedings of the third Pan-African Congress on prehistory, Livingstone, 1955 (pp. 306–312). London: Chatto & Windus. Feld, S. (1996). Waterfalls of sound: An acoustemology of place resounding in Bosavi, Papua, New Guinea. In S. Feld & K. H. Basso (Eds.), Senses of place (pp. 91–135). Santa Fe, NM: School of American Research Press. Feubel, R. (1935). Die attischen Nymphenreliefs und ihre vorbilder. Heidelberg: Buchdruckerei August Lippl. Frieman, C., & Gillings, M. (2007). Seeing is perceiving? World Archaeology, 39(1), 4–16. Garfinkel, P., & Waller, S.  J. (2012). Sounds and symbolism from the netherworld: Acoustic archaeology at the animal master’s portal. Pacific Coast Archaeological Society Quarterly, 46(4), 37–60. Gaver, W. W. (1993). What in the world do we hear? An ecological approach to auditory event perception. Ecological Psychology, 5(1), 1–29. Gell, A. (1998). Art and agency: An anthropological theory. Oxford: Clarendon Press. Goldhahn, J. (2002). Roaring rocks: An audio-visual perspective on hunter-gatherer engravings in northern Sweden and Scandinavia. Norwegian Archaeological Review, 35, 29–61. Groenen, M. (1997). Ombre et lumière dans l' art des grottes. Bruxelles: ULB, Centre de Recherches et d' Étudestechnologiques des Arts plastiques. Hamilakis, Y. (2011). Archaeologies of the senses. In T. Insoll (Ed.), The Oxford handbook of the archaeology of ritual and religion (pp. 208–225). Oxford: Oxford University Press. Hamilakis, Y. (2013). Archaeology and the senses: Human experience, memory and affect. Cambridge: Cambridge University Press. Howes, D. (2003). Sensual relations: Engaging the senses in culture and social theory. Ann Arbor, MI: University of Michigan Press. Howes, D. (2006). Scent, sound and synesthesia: intersensoriality and material culture theory. In C. Tilley, W. Keane, S. Küchler, M. Rowlands, & P. Spyer (Eds.), Handbook of material culture (pp. 161–172). London: Sage. Ihde, D. (2007). Listening and voice: Phenomenologies of sound. Albany, NY: State University of New York Press. Jahn, R. G., Devereux, P., & Ibison, M. (1995). Acoustical resonances of assorted ancient structures. International Consciousness Research Laboratories (ICRL) report 95(1). http://www. princeton.edu/~pear/pdfs/1995-acoustical-resonances-ancient-structures.pdf. Accessed September 2014. Jaynes, J. (1976). The origin of consciousness in the breakdown of the bicameral mind. Boston, MA: Houghton Mifflin. Jones, D. W. (1999). Peak sanctuaries and sacred caves of Minoan Crete. Jonsered: Paul Åströms Förlag. Kahn, D. (1999). Noise, water, meat: A history of sound in the arts. Cambridge, MA: MIT Press. Kato, K. (2009). Soundscape, cultural, landscape and connectivity. Sites (new series), 6(2), 80–91. Kelman, A. Y. (2010). Rethinking the soundscape: A critical genealogy of a key term in sound studies. Senses and Society, 5(2), 212–234. Kuehn, H. (1956). The rock paintings of Europe. London: Taylor & Francis.

7  Pan Rituals of Ancient Greece Revisited

133

Kuttruff, H. (2009). Room acoustics (5th ed.). New York, NY: Spon Press. Larson, J. (2001). Greek nymphs: Myth, cult, lore. Oxford: Oxford University Press. Lawson, G., Scarre, C., Cross, I., & Hills, C. (1998). Mounds, megaliths, music and mind: Some thoughts on the acoustical properties and purposes of archaeological spaces. Archaeological Review from Cambridge, 15, 111–134. Lonsdale, S. H. (1993). Dance and ritual play in Greek religion. Baltimore, MD: Johns Hopkins University Press. Marinatos, S. Ν. (1940–41). The cult of the Cretan caves. Review of Religions, 5, 132–135. Marinatos, S. (1964). Το Σπήλαιον Μελισσάνης. Archaeologiki Ephimerida, 103, 15–27. Mifsud, S., Mifsud, A., & Savona Ventura, C. (1999). The subterranean sanctuary at Hal Saflieni. In A. Mifsud & C. Savona Ventura (Eds.), Facets of Maltese prehistory (pp. 149–168). Malta: Prehistoric Society of Malta. Montiglio, S. (2000). Silence in the land of logos. Princeton, NJ: Princeton University Press. Ouzman, S. (2001). Seeing is deceiving: Rock art and the non-visual. World Archaeology, 33(2), 237–256. Pasquier, A. (1977). Pan et les Nymphes a l’antre Corycien. In: Etudes Delphiques. Bulletin de correspondance hellénique, supplément 4 (pp. 365–387). Athens: École Français d'Athènes. Pierce, N.  B. A. (2006). The archaeology of sacred caves in Attica, Greece. Unpublished PhD thesis, McMaster University. Reznikoff, I. (2004–5). On primitive elements of musical meaning. The Journal of Music and Meaning, 3(2), 1–24. Reznikoff, I. (2008). Sound resonance in prehistoric times: A study of Paleolithic painted caves and rocks. The Journal of the Acoustical Society of America, 123(5), 3603. Reznikoff, I. (2012). La dimension sonore des grottes paleolithiques et rochers à peintures. In J. Clottes (Ed.), L’ art pléistocèneen Europe. Actes du Congrès IFRAO, Tarascon-sur-Ariège, Septembre 2010, (pp. 45–56). Tarascon-sur-Ariège: Société Préhistorique Ariège-Pyrénées. Rifkin, R. F. (2009). Engraved art and acoustic resonance: Exploring ritual and sound in north-­ western South Africa. Antiquity, 83(321), 585–601. Rodaway, P. (1994). Sensuous geographies: Body, sense & place. London: Routledge. Romaios, K. (1905). Ανασκαφή επί της Πάρνηθος άντρου. Archaeologiki Ephimerida, 44, 99–158. Sakellarakis, G. (1983). Ανασκαφή Ιδαίου άντρου. Praktika Archaeologikis Etairias, 138, 415–500. Samuels, D.  W., Meintjes, L., Ochoa, A.  M., & Porcello, T. (2010). Soundscapes: Towards a sounded anthropology. Annual Review of Anthropology, 39, 329–345. Scarre, C. (1989). Painting by resonance. Nature, 338, 382. Scarre, C., & Lawson, G. (2006). Archaeoacoustics. Cambridge: McDonald Institute for Archaeological Research. Schafer, R. M. (1977). Soundscape: Our sonic environment and the tuning of the world. Rochester, VT: Destiny Books. Skias, Α. (1918). Το παρά την Φυλήν άντρον του Πανός κατά τας ανασκαφάς των ετών 1900 και 1901. Archaeologiki Ephimerida, 57, 1–28. Schomer, G., & Goette, H.  R. (2004). Die Pan-Grotte von Vari. Mainz am Rhein: Philipp Von Zabern. Synnott, A. (1991). Puzzling over the senses: From Plato to Marx. In D. Howes (Ed.), The varieties of sensory experience: A sourcebook in the anthropology of the senses (pp. 61–76). Toronto: University of Toronto Press. Thallon, I. C. (1903). The cave at Vari. III. Marble reliefs. American Journal of Archaeology, 7, 301–319. Till, R. (2010). Songs of the stones: An investigation into the musical history and culture of Stonehenge. International Association for the Study of Popular Music Journal, 1(2), 1–18. Till, R. (2017). An archaeoacoustic study of the Hal Saflieni Hypogeum on Malta. Antiquity, 91(355), 74–89.

134

N.-P. Yioutsos

Tomkins, P. (2009). Domesticity by default. Ritual, ritualization and cave-use in the Neolithic Aegean, Oxford Journal of Archaeology, 28(2), 125–153. Tomkins, P. (2012). Landscapes of ritual, identity, and memory: Reconsidering Neolithic and Bronze Age cave use in Crete, Greece. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual use of caves (pp. 59–79). Boulder, CO: University Press of Colorado. Tsilfidis, A., Vovolis, T., Georganti, E., Teubner, P., & Mourjopoulos, J. (2011). Acoustic radiation properties of Ancient Greek theatre masks. The acoustics of ancient theatres conference, Patras, September 18–21, 2011. Unpublished conference paper, Patras. Tzouvara Souli, C. (1998). Ομάδα πήλινων ειδωλίων από το σπήλαιο Ασβότρυπα στο Φρύνι της Λευκάδας. Dodoni, 27(1), 372–436. Ustinova, Y. (2009). Caves and the Ancient Greek mind: Descending underground in the search for ultimate truth. Oxford: Oxford University Press. Vassilantonopoulos, S. L., & Mourjopoulos, J. N. (2002a). A study of Ancient Greek and Roman theater acoustics. Acta Acustica united with Acustica, 89(1), 123–136. Vassilantonopoulos, S. L., & Mourjopoulos, J. N. (2002b). Ακουστική Προσομοίωση Τελετουργικών και Δημόσιων Χώρων της Αρχαίας Ελλάδας. Acoustics. Thessaloniki: Hellenic Institute of Acoustics. Vassilantonopoulos, S. L., & Mourjopoulos, J. N. (2009). The acoustics of roofed ancient odeia: The case of Herodes Atticus Odeion. Acustica, 95, 291–299. Vassilantonopoulos, S., Zakynthinos, T., Hatziantoniou, P., Tatlas, N.  A., Skarlatos, D., & Mourjopoulos, J.  (2004). Measurement and analysis of acoustics of Epidaurus theatre. Proceedings of the 2nd HELINA conference, 27–28 Sept. 2004, Thessaloniki (pp. 359–366). Thessaloniki: Hellenic Institute of Acoustics. Vovolis, Τ. (2011). Mask, actor, theatron and landscape in classical Greek theatre. The acoustics of ancient theatres conference, Patras, September 18–21, 2011. Unpublished conference paper, Patras. Waller, S. J. (1993). Sound reflection as an explanation for the content and context of rock art. Rock Art Research, 10, 91–101. Waller, S.  J. (2002). Rock art acoustics in the past, present and future. In P.  Whitehead, L.  L. Loendorf, W.  Whitehead, & W.  B. Murray (Eds.), American Indian rock art volume 26. International Rock Art Congress (IRAC) proceedings: Papers presented at the twelfth International Rock Art Congress, Ripon, Wisconsin, May 23–31, 1999 (pp.  11–20). Tucson, AZ: American Rock Art Research Association. Waller, S. J. (2012). Acoustic mapping of rock art soundscapes: Depicting echoes visibly. In E. W. Ritter, M. Greer, & P. Whitehead (Eds.), American Indian rock art, Idaho Falls, Idaho, volume 38 (pp. 181–188). Glendale, AZ: American Rock Art Research Association. Watson, A. (1997). Hearing again the sound of the Neolithic. British Archaeology, 23, 6. Watson, A. (2001). Composing Avebury. World Archaeology, 33, 296–314. Watson, A., & Keating, D. (1999). Architecture and sound: An acoustic analysis of megalithic monuments in prehistoric Britain. Antiquity, 73(280), 325–336. Weller, C. (1903). The cave at Vari. American Journal of Archaeology, VII(3), 263–288. Wenzel, E.  M., Wightman, F.  L., & Kistler, D.  J. (1991). Localization with non-individualized virtual acoustic display cues. In S. P. Robertson, G. M. Olson, & J. S. Olson (Eds.), CHI’91 conference proceedings: ACM SIGCHI conference on human factors in computing systems, New Orleans, Lousiana, 27 April–2 May 1991 (pp. 351–359). New York, NY: ACM Press. Whalley, I. (2008). HIEMPA: Hybrid instruments from electroacoustic manipulation and models of pütorino and aquascape. Organised Sound, 13(3), 259–267. Whitehouse, R.  D. (2001). A tale of two caves: The archaeology of religious experience in Mediterranean Europe. In P. Biehl, F. Bertèmes, & H. Meller (Eds.), The archaeology of cult and religion (pp. 161–167). Budapest: Archaeolingua. Wickens, J. M. (1986). The archaeology and history of cave use in Attica, Greece from prehistoric through late roman times. Unpublished PhD thesis, Indiana University. Witmore, C. L. (2006). Vision, media, noise and the percolation of time: Symmetrical approaches to the mediation of the material world. Journal of Material Culture, 11(3), 267–292.

7  Pan Rituals of Ancient Greece Revisited

Ancient Sources Aristophanes, Lysistratae Crinagoras in Anthologia Palatina 6.253 Diodorus Siculus, Library Euripides, Helen, Rhesus Herodotus, Histories Homeric Hymn to Pan Menander, Dyskolos Longus, Daphnis and Chloe Lucian, Bis Accusatus, Dialogi Deorum Orphic Hymn to the Nymphs Pausanias, Description of Greece Polyaenus, Stratagems Sophocles, Philoctetes Strabo, Geography Theocritus, Idylls

135

Chapter 8

The Watery Way to the World of the Dead: Underwater Excavations (Old and New) at the Cave of Han-sur-Lesse, Belgium Christophe Delaere and Eugène Warmenbol

8.1  Introduction The well-known cave of Han-sur-Lesse (municipality of Rochefort, province of Namur) is a major tourist attraction in southern Belgium. The River Lesse flows through it, after disappearing underground in the Gouffre de Belvaux. It reappears in the Salle d’Armes and re-emerges at the Trou de Han. The majority of the archaeological finds from the cave were made between 1963 and 1983, during subaquatic excavations in the Lesse, between the Galerie des Petites Fontaines (where daylight is lost, across the Tournant du Jour) and the Galerie de la Grande Fontaine (across the Trou de Han, where daylight reappears). Finds from the two galleries themselves have also been important and numerous (Fig. 8.1). Finds in the river were made as early as 1959, but systematic investigations only began in 1963, under the direction of Marc Jasinski of the Centre de Recherches Archéologiques Fluviales (CRAF); CRAF is still active in Han-sur-Lesse after half a century of discoveries and is now under the direction of one of the authors (C. D.). There are few sites like Han-sur-Lesse (Warmenbol 2009), and very few have been explored in the same fashion (with notable exceptions, such as the Grotte di Pertosa, in Campania, Italy; Larocca 2010). There may have been some visitation to the cave as early as the Upper Palaeolithic, but activity is not conspicuous until the end of the Neolithic (c. 2900/2800 BC). The Late Bronze Age (c. 1000/900 BC) is seemingly the best-represented period, with the richest material (in both quantity and quality). The cave was also used at different times in the Second Iron Age (e.g. in the fourth century BC) and at the end of the Roman period (primarily during the third century AD). More recent use began in the

C. Delaere (*) · E. Warmenbol Centre de Recherches en Archéologie et Patrimoine, Université Libre de Bruxelles (ULB), Bruxelles, Belgium e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_8

137

138

C. Delaere and E. Warmenbol

Fig. 8.1  The site from the Salle du Dôme to the Trou de Han (Centre de Recherches Archéologiques Fluviales; CAD: C. Delaere); contour lines at 0.5 m spacing

8  The Watery Way to the World of the Dead: Underwater Excavations …

139

second half of the seventeenth century and continues today, the caves having become a major focus for tourism and being more accessible than ever.

8.2  The Older Finds 8.2.1  The Galerie de la Grande Fontaine The Galerie de la Grande Fontaine (a ‘dry’ gallery) was first excavated in the early twentieth century. It is from here that Edouard de Pierpont extracted the so-called pilier stratigraphique (a 3 m × 1 m × 1 m column), which served as testimony to his excavations here in 1902. It is kept in the cellars of the Musée archéologique in Namur and was examined by Pierre-Paul Bonenfant in 1981–1982, with the intention of sampling the stratigraphy for chemical and physical analyses not available almost a century before (Bonenfant 1982, 1984). The samples extracted from the archaeological layers encased within the column were, in fact, only radiocarbon-­ dated 25 years later, and confirmed the importance of the Late Neolithic layers of the gallery and those of the Second Iron Age (Fig.  8.2) (Van Strydonck and Warmenbol 2012). Re-examination of the finds by Edouard de Pierpont, and some of his successors (Mariën 1982), has shown that the gallery was most probably not inhabited in the Late Neolithic but was used for burial, as are so many of the caves of southern Belgium. Several radiocarbon dates on human bone and teeth appear to confirm this (André 2013; Warmenbol 2015, p. 109).

8.2.2  The Late Neolithic Most of the finds of this period come from the bottom of the Lesse, in front of the Galerie des Petites Fontaines and the Galerie de la Grande Fontaine. As previously noted, this gallery was a burial place, and the accompanying grave goods are quite prestigious, including a set of ten bone spoons of a type only found in this region (Fig. 8.3) (Warmenbol 2013a). These artefacts are, like those of the Bronze Age, in very good condition, and the main question is whether or not they were deliberately deposited in the river. As will become apparent, the latest excavations—which show that the course of the river was very much altered between the sixteenth and the twentieth centuries—indicate that we must be cautious in our interpretations. The over-representation of some materials and the under-representation of others suggest that the finds from the river do not correspond with a domestic assemblage (Pleuger 2011), which would be of a quite different nature. Stone artefacts seem to be over-represented, more specifically axes in polished stone and arrowheads of various types, together with a dagger in exotic Grand-Pressigny flint. About 60

140

C. Delaere and E. Warmenbol Atmospheric data from Reimer et al (2009); OxCal v3.10 Bronk Ramsey (2005); cub r:5 sd:12 prob usp[chron]

Sequence 10 4188±23BP 8C 4095±30BP 8B 4245±30BP 8A 4180±35BP 7 4130±30BP 6 3415±30BP 5B 2965±25BP 5 2935±35BP 4 2195±25BP 3B 2280±30BP 3A 2145±25BP 2 2080±25BP 1B 2140±25BP 4000CalBC

3000CalBC

2000CalBC

1000CalBC

CalBC/CalAD

Calibrated date

Fig. 8.2  Sequence of radiocarbon dates from the pilier stratigraphique, taken from the Galerie de la Grande Fontaine (Van Strydonck and Warmenbol 2012, Fig. 3)

arrowheads are known—some (perhaps deliberately) broken—including 4 magnificent barbed and tanged examples (Fourny 1995). Four axes in polished flint are also known, as are numerous additional fragments of similar axes. We may add to these three further small axes in other stones and the butt of an axe of Puy type in jadeite or omphacite, most probably quarried on Monte Viso in north-western Italy (analysis by Michel Errera). Clearly, ‘exotic’ materials were deposited in the Lesse (or on its banks) at the end of the Neolithic, as they were time and again throughout the cave’s use. Neolithic pottery has also been discovered, though it has not yet been studied in detail; at first sight, however, it does not appear to represent anything out of the ordinary and is of local manufacture.

8  The Watery Way to the World of the Dead: Underwater Excavations …

141

5 cm Fig. 8.3  Newly discovered tenth Late Neolithic spoon (of which only the bowl remains) from Han-sur-Lesse (Drawing and CAD by A. Stoll, CReA-Patrimoine, ULB). These spoons may well have been made from human bone

Much of the assemblage—particularly that found (predominantly in 1902–1904) in the Late Neolithic layers of the Galerie de la Grande Fontaine (Fig. 8.4)—is reminiscent of the ‘classical’ grave goods of this period. The ten splendid bone spoons are of a type known only as grave goods, for example, at Martouzin-Neuville (Namur) and Vaucelles (Namur). The antler axe sheath found with them is also a typical funerary offering, with or without the axe, with parallels again from Vaucelles (Namur) and from Anseremme Abri du Grogneau (Namur) (Mariën 1981). It

142

C. Delaere and E. Warmenbol

Fig. 8.4  Section through the pilier stratigraphique from the Galerie de la Grande Fontaine (Drawn by M. A. De Spiegeleire; CAD: A. Stoll, CReA-Patrimooine, ULB)

8  The Watery Way to the World of the Dead: Underwater Excavations …

143

appears, then, that, as early as the Late Neolithic, at least some parts of the caves at Han-sur-Lesse were spaces dedicated to the dead (or to ‘Death’ itself), with the river (the water) perhaps serving as an interface between worlds (Sohn 2009).

8.2.3  The Bronze Age As stated earlier, most of the finds were recovered from the bed of the River Lesse, with accumulations in front of the Galerie des Petites Fontaines and the Galerie de la Grande Fontaine (both of which produced truncated Late Bronze Age stratified layers). The objects appear often to have been found not far from where they entered the river, and they seem to have been deposited deliberately, as part of well-known wider Late Bronze Age practices across north-western Europe. Again, however, the latest excavations call for caution in our identification of primary contexts, whether wet (the river) or dry (the banks); in this regard caves represent very particular environments for study. Objects in copper alloy (mainly bronze), together with weapons, tools and ornaments, are over-represented; most apparently intact, but others deliberately broken or dismantled (Warmenbol in press). The assemblage forms by far the most important group of metal objects from the Bronze Age in Belgium. Among the weapons are 3 swords, 4 chapes, 4 spearheads and about 50 arrowheads; and among the tools are 24 axes, around 15 knives (Warmenbol and Van Strydonck 2011, 2012), 3 sickle blades, about a dozen chisels, a gouge and a saw (Fig. 8.5). Ornaments are represented by more than 200 pins, around 100 bracelets (of all sizes), and belt ornaments, together with many beads and buttons, and a great quantity of small rings (which may or may not be parts of ornaments). Divers also found a number of gold ornaments, which are extremely rare elsewhere in Belgium and the southern Netherlands. These may also have been deliberately broken-up into small pieces. Within the corpus of intact objects are five gold discs with a central boss (Fig.  8.6), two basket-shaped hair ornaments, and nine penannular gilded rings often known as hair rings or ring money. Three fragments of at least two hollow beads or hollow pinheads decorated with filigree and granulation (and certainly of Mediterranean origin) have also been found (Warmenbol 1999, 2004). In addition were two ingots, one of exactly twice the weight (1.272 g) of the other (0.636 g). Also of Mediterranean origin were glass beads attributable to north Italian workshops (Cosyns et al. 2005). No less remarkable is the pottery, since the ‘tableware’ is closely related to that of the so-called ‘Groupe Rhin-Suisse-France Orientale’ and distributed predominantly in that part of Europe. Four hundred and fifty different sherds were identified, mostly belonging to deep bowls, which were internally decorated, and small goblets (which were also often decorated). Curiously, the clay used leaves no doubt about their local production (Leclercq 2012), though the forms on which these vessels were based are certainly not indigenous.

144

C. Delaere and E. Warmenbol

Han-sur-Lesse

A70-0

a

b

a

b

4706 a

b

a

b

4708

a

b

a

b

as 014 a

b a

b

a

b

as 015

a

b

as 008 a

b a

b

b a

a

as 016

b

as 013

as026

as025 a

b

a

b

5 cm

Fig. 8.5  The saw and socketed knives from the Lesse (provisional numbering) (Drawing and CAD by A. Stoll, CReA-Patrimoine, ULB). The wood in contexts 4706 and 4704 (bottom) produced radiocarbon dates of 2805 ± 30 BP (KIA-21782) and 2715 ± 25 BP

8  The Watery Way to the World of the Dead: Underwater Excavations …

145

Fig. 8.6  Five complete gold discs recovered at the bottom of the Galerie des Petites Fontaine (Photograph: M. Jasinski). Two of the discs were submitted for PIXE analysis, which indicated that they are made of alluvial gold (A 64-202 = Au, 81.2%; Ag, 13.0%; Cu, 5.8%. A 64-206 = Au, 79.6%; Ag, 13.7%; Cu, 6.7%)

The caves in Han-sur-Lesse offer a very peculiar context for deposition at this time, since there was no settlement or cemetery here in the Bronze Age. The bronze objects found at the bottom of the river do not all constitute hoards, but some may indeed have entered the dark world of the caves as parts of hoards. Neither does the nature of the finds, deposited over several centuries, allow for strict comparison with ‘classical’ Bronze Age river finds from other regions. Some of the objects found in the Lesse have, for example, been purposefully ‘destroyed’, which is never the case with objects from the Seine, the Thames, the Rhine, etc. As Muriel Mélin (2012) stressed, objects entrusted to water are usually given over intact, while those entrusted to the earth are often broken, especially in the Late Bronze Age. These hoards were until recently systematically described as ‘founders’ hoards’, which may well be true for some of them (those with ingots, for instance), but certainly not for all of them. An explanation of ritual behaviour is more usually favoured today (Mélin 2012; Warmenbol in press). As far as Han-sur-Lesse is concerned, with its similarities to descriptions of the Underworld in classical texts such as Homer’s Odyssey, we favour the idea that Bronze Age people saw in it the ‘Gates of the Realm of the Dead’ (Warmenbol 1996). The objects in the river would thus be deposits for the dead or offerings to them, which leaves us with one problem, i.e. explaining why some of them were destroyed or ‘killed’. Another major question also remains: were they placed in the river, or did they end up there accidentally? Could they have been cast into the water, as at the Swiss Iron Age sites of La Tène and Cornaux, where weapons and other military equipment were thrown from a bridge or platform? These kinds of

146

C. Delaere and E. Warmenbol

sites have Bronze Age antecedents. In Han-sur-Lesse, many wooden posts were observed by divers in the early excavations, particularly in front of the Galerie des Petites Fontaines, and they could indeed represent the base of a platform of the same kind as those described above. Regular floods have, however, destroyed these posts, and only one was radiocarbon dated. Bizarrely, it seems to belong to the Early Middle Ages (1380 ± 100 BP; Lv-689), and the latest excavations just outside of the cave in 2016 produced similar dendrochronological dates for other posts, putting the idea of a Bronze Age platform ‘on hold’.

8.2.4  The Galerie des Petites Fontaines The Galerie des Petites Fontaines opens onto the Lesse at the so-called Tournant du Jour, where full darkness sets in (Fig. 8.7). Some excavation took place here in the early twentieth century, but what we know today stems mainly from the excavations of M.-E. Mariën in 1963–1964. He established the stratigraphic sequence, noting the absence of Neolithic layers (which had probably been washed away by the River Lesse). The main sequence pertains to the Second Iron Age or La Tène period

Fig. 8.7  Daylight coming through the Trou de Han into the Galerie des Petites Fontaines (Photograph: G. Deflandre)

8  The Watery Way to the World of the Dead: Underwater Excavations …

147

(Mariën 1970, p. 256). One of the most extraordinary discoveries in the caves of Han-sur-Lesse was made in Level IV, where seven human mandibles were recovered, two or three of them with cut marks left by the sword or dagger used to behead their owner (Fig. 8.8). They are probably the surviving remains of heads placed on stakes, and it is certainly no accident that they were placed here, at the entrance to the ‘dark side of the world’. Radiocarbon dates (Warmenbol 2007a, b, p. 543–544) indicate that the group of mandibles is not homogeneous, with two being considerably older than the others and the younger specimens being of possible Roman date. While the older mandibles might represent curated bones, the younger examples indicate that the practice of collecting heads did not cease with the arrival of the Roman period; they in fact suggest quite the contrary.

8.2.5  Celts and Romans Material was again deposited in the Lesse or on its banks during the Second Iron Age. Relatively little material dates to the Early La Tène period but includes a magnificent phalera, a few fibulae and some decorated vessels with close parallels from the tumuli of the Groupe Ardennais (chariot burials,  probably of the Remi). The Middle La Tène period is also represented by a few fibulae and, again, by some decorated (and some plain) vessels, similar to those from the Trou de l’Ambre in Eprave (only 5  km away) but otherwise unknown in the Groupe Ardennais (Warmenbol 2013b). The Late La Tène period is the best represented in terms of artefacts, with numerous metal objects (mostly tools and ornaments), together with glass and amber, and also several hundred pots; the latter almost exclusively represent the Haltern 91-type, a medium-sized storage vessel (Fig. 8.9). Levels III and IV from the Galerie des Petites Fontaines must be contemporary, since Level IV contained the human mandibles, the latest examples of which have been radiocarbon-dated to the first century BC and/or AD.  This again demonstrates that one cannot understand the subaquatic site at the Trou de Han without knowledge of the finds from the Galerie de la Grande Fontaine and the Galerie des Petites Fontaines. There is no doubt that Roman material was also deliberately left in the caves, with the selective deposition of certain objects. Most of the Roman material belongs to the third century AD, and perhaps more specifically to the third quarter of the third century (Goffette 2013), when the ‘Gaulish’ emperors (Postumus and his successors) reigned. There are about 100 coins, 30 fibulae (of only a few different types) (Callewaert and Goffette 2011; Lallemand 1994), a brass bucket of the Hemmoor type, and a brass oval plate bearing the name of its last owner (Fig. 8.10) (Warmenbol and Raepsaet-Charlier 2014). The most significant object, however, is a metallic military diploma, dating from the reign of Trajan but probably deposited much later, and certainly deliberately considering the value of such an official item (Absil 1988).

148

C. Delaere and E. Warmenbol

1

3

2

4

Fig. 8.8  Four mandibles from the Galerie des Petites Fontaines, two of which (left) display clear cuts from a sword or large knife (Mariën 1970, Fig.  74). They produced radiocarbon dates of 1970 ± 30 BP (PF-1) and 2250 ± 25 BP (PF-3)

8  The Watery Way to the World of the Dead: Underwater Excavations …

149

HAN-GBG-85.142

HAN-GBG-86.116

HAN-GBG-86.95/105/125/173

HAN-GBG-85.58/59/163 HAN-GBG-86.129

HAN-GBG-86.20/133/167/173

HAN-GBG-86.83

0

10 cm

Fig. 8.9  A collection of Haltern 91-type pots recovered on the other side of the Massif de Boine, in the Galerie Belgo-Romaine (Drawing and CAD by A. Stoll, CReA-Patrimoine, ULB). GBG-­ 85.142 is a fragment of a coupe en parasol (a type of earthen lamp dated to the fifth or fourth century BC)

The pottery also demonstrates that we are not dealing with ‘domestic’ material. The assemblage comprises (almost exclusively) tableware and predominantly drinking vessels. This type of assemblage is encountered elsewhere, in the Trou des Nutons in Presles (Hainaut), for example, as well as in other caves.

150

C. Delaere and E. Warmenbol

Fig. 8.10  Brass oval plate of the third century AD. The base bears an inscription of the name of its last owner, Serenus; a man with a very Latin (as opposed to Celtic) name (Photograph: M. Timperman)

8.3  Present-Day Excavations 8.3.1  Challenges and Potential Since 2012, the Centre de Recherches Archéologiques Fluviales (CRAF) has been experimenting with new excavation and research strategies adapted to the underwater archaeological deposits of the caves of Han-sur-Lesse. The main area of ​​study corresponds to the stretch of water between the Trou de Han (at the current exit of the cave) and the resurgence of the Lesse (at the Passage du Diable), after its underground passage through the Massif de Boine (Fig. 8.1). Since 1963, when the first planned underwater excavations took place, thousands of objects (complete and fragmentary) have been removed from the riverbed inside the cave, mainly at the Tournant du Jour, 70 m upstream of the current cave exit (1963–1978), but also from the Trou de Han itself (1978–1983). In view of the history of excavation of the cave and its archaeological importance, objectives in 2012–2014 took on a dual focus: on the one hand, to carry out new underwater archaeological excavations in a limited and assumedly untouched area and, on the other hand, to combine data from the current excavations with that from past operations (as compiled in the excavation journals—the so-called LIBER (1963–1983)—but also from photographs, inventories, testimonies, etc.). The objective was not only to perform an archaeological survey of the site but also to document and study the contexts of objects from previous excavations through creation of the first stratigraphic cross-section through the river. Although the

8  The Watery Way to the World of the Dead: Underwater Excavations …

151

u­ nderwater archaeological site is, to date, mostly known for material from the Late Bronze Age and Iron Age, the first three years of excavation of the current campaign (2012–2014) have brought to light a particularly rich deposit dating from the sixteenth to twentieth centuries. The long-term objective is, however, to reach the socalled ‘ancient’ layers, in order to understand the context of the exceptional objects dating from the Late Neolithic and Metal Ages (i.e. those discovered primarily on the bed of the Lesse between 1963 and 1983). Since the middle of the nineteenth century, the Trou de Han has served as the exit for the part of the cave system visited by tourists. It is characterised by a body of water measuring around 2000 m2 (c. 100 m × 20 m), delimited by an ancient dock situated under the vault of the cave roof and a modern bridge located 30 m upstream of a modern dam. At the other end of the Massif de Boine, the entrance to the cave has probably been in use since its discovery in 1817 and certainly since the 1st May 1857 (Timperman 1989, p. 24). Before 1817–1857, however, the Trou de Han probably served as both the entrance and exit to the cave and formed part of a complex network of galleries (not all necessarily accessible in the past). There is no doubt that the natural morphology of the Trou de Han has been profoundly altered in recent times (i.e. between the sixteenth and nineteenth centuries), mainly by the building of the dam downstream of the caves in order to create the present-day artificial reservoir. The date of creation of this structure is difficult to determine, but the Relation d’un voyage fait à la grotte de Han au mois d’août 1822, written by two academics from Brussels, describes something very much like it (Kickx and Quetelet 1822, p. 7). Thus, the caves of Han-sur-Lesse appear to have been well-known by the beginning of the nineteenth century (though not for their archaeology). The presence of a wooden fence intended ‘to stop the fish’ in 1822 (ibid.) suggests that the reservoir was used for fishing, but construction of the dam primarily allowed easier access to the cave; the resulting increase in the level and volume of water flooded the entire gallery behind the Trou de Han, which then provided access to the caves. The topography of the site (now submerged) tells us that before the creation of the reservoir, it was probably possible to access the gallery by foot, along the left-hand bank. However, entering the cave in this way was difficult, and the bank was only accessible in the ‘dry’ season; the increase in water level allowed easy and permanent access by boat. The creation of the dam modified the river’s behaviour and changed its erosion mechanisms: new areas—originally on dry land (palaeoterrestrial)—were now eroded by the current, and sediment was deposited on the river bed; in this way, areas which had been previously eroded are now gradually silting up. These observations show that the spatial configuration of the cave has been very much altered. If there were terrestrial activities or structures in the past at the Trou de Han (in Roman times or earlier), they will now be difficult to identify, since they are most probably underwater. During the first terrestrial archaeological excavations carried out in 1902 on the left bank of the Lesse, inside and in front of the Galerie de la Grande Fontaine (de Pierpont 1904), the water reservoir was already in existence, and the height of the bank had probably also been raised by the Société des Grottes (the company owning the caves), who were concerned with accessibility for tourists.

152

C. Delaere and E. Warmenbol

8.3.2  Methodology The cave of Han-sur-Lesse is the ‘pioneer’ site for underwater archaeology in Belgium (Jasinski 1965, p. 51–54). However, 50 years of dives have changed the morphology of the riverbed, and it is now necessary to determine their impact on the preservation of the site and to define any intrusive elements in the new excavation area. In this regard, processing of data from previous excavations is vital. Previous subaquatic interventions have mainly taken place at the Tournant du Jour. Other areas, such as the Trou de Han, have, however, also been excavated for shorter periods, and preliminary data from this area has now also been processed. This area was excavated between 1978 and 1983, following the fortuitous discovery in 1978 of intact ceramics from the seventeenth to eighteenth centuries on the riverbed (LIBER 1978, p. 1). It focused on areas under rocky overhangs in the deepest part of the river. From the beginning of the interventions, underwater visual landmarks were left to demarcate the excavations, and were mapped in 1979 by Albert Henin and Michel Timperman (LIBER 1979, p. 38). We estimate that the old excavations at the Trou de Han covered an area of c​ . 216 m2 (or 10.8%) of the total area of t​he reservoir. The entire reservoir was also prospected (but not excavated) in 1970. In terms of volume, we can calculate from the excavation journals (the LIBER) that between 1978 and 1983, 815 h of dives were undertaken. Between 2012 and 2014, we explored c. 50  m3 of sediment in 503  h, which theoretically corresponds to a productivity of c. 1  m3 every 10  h. Applying this ratio retrospectively, we can estimate that the volume of material excavated between 1978 and 1983 was c. 82 m3. Although these figures are theoretical, they allow us to estimate the scale of past interventions and to suggest that the majority of the site is still relatively well preserved. Indeed, according to the LIBER 1978–1983, excavations were predominantly carried out in the deepest part of the river. As soon as the divers reached the zone of sediment between the riverbed and the left bank (the 2012–2014 excavation area)— representing two-thirds of the width of the river—they stopped because of the density of large blocks resulting from partial collapse of the vault (problems that we are currently tackling). The LIBER clearly describe that some areas of the riverbed were excavated to bedrock because they were yielding archaeological objects! In fact, the first dives in 1978 harvested objects in situ from the riverbed, though in the years that followed, the weakening of trenches caused adjacent sedimentary layers to collapse onto the riverbed each winter, and trenches had to be ‘re-excavated’.

8.4  Stratigraphy Between 2012 and 2014, the team from CRAF excavated a stratigraphic cross-­ section between the two river banks in an area apparently undisturbed by previous interventions. A 4  m wide trench was established between the left and the right

8  The Watery Way to the World of the Dead: Underwater Excavations …

153

banks and is demarcated underwater by two ropes made of orange synthetic material. The trench is itself subdivided into zones of 4 m × 4 m (A, B, C, D, etc.), which are themselves subdivided into areas of 2 m × 2 m (A1, A2, A3, A4, B1, B2, etc.) (Fig. 8.11). If underwater archaeological work requires the same rigour of excavation and data-recording as its terrestrial counterpart, it imposes specific constraints related to its working conditions (visibility, current, collapse of a section, etc.) and to the context of discovery (taphonomy, deposits of a fluvial or palaeoterrestrial nature, etc.). The theoretical and physical subdivisions of the trench into 2  m  ×  2  m squares allowed for the control and monitoring of the excavation and ensured that the divers stayed in their area of intervention for the purposes of accurate recording of data. The new operations, undertaken across three seasons (August 2012, 2013 and 2014), currently represent 35 days of excavation and 412 h of immersion, from a total of 341 archaeological diving operations. Technical dives, such as flood layer cleaning and trench reinforcement work, are not included in these figures. The 2012–2014 trench measures 23 m long by 4 m wide, with a total surface area of 92 m2. At the end of the 2014 operations, 42 m2 of this area had been excavated, equating to a volume of c. 58 m3 of material (2.5% of the 2000 m2 surface area of the reservoir). Over 900 objects or fragments of objects have been identified, including 349 pottery sherds. In the riverbed, we were confronted with a thick layer of stones between 30 cm and 130 cm in diameter, sealed by a matrix of clayey-silty sediment containing 40% gravel and pebbles. Though the excavated area resembles a single layer of stones sealed with sediment, we identified certain phenomena throughout the cross-­section. Firstly, we observed an alternation of mineral and organic layers over the whole elevation. Some deposits were also subdivided into a more subtle alternation of thin black layers (winter flood) and gravel. Two mineral layers in particular distinguished themselves by their size (≥40 cm) and their inclination. These (US2 and US4) corresponded with two important reconfigurations of the river banks. Finally, we observed intrusive elements which were at odds with the rest of the stratigraphy. These were mainly in the upper layers of squares C3–C4 and D1–D2 and probably result from the excavations of 1978–1983; the lowest level in this area does however appear intact (Fig. 8.12). On the whole, the excavated area corresponds to a transition zone between a palaeobank and deposits of a fluvial nature, indicated by an alternation of layers left by flooding and mixed materials in mineral layers. Some intrusive elements, such as stones having fallen from the roof, have reshaped some areas by creating distinct sedimentary ‘pockets’, though much of the morphology appears to have resulted from reorganisation of the banks. Indeed, between squares B1–B2 and B3–B4, at this rupture in sediments (i.e. between the palaeobank and the river deposits), one encounters the level of the ancient base of the river. This belongs to event E1 (reshaping of the river), as indicated by the inclination of collapsed blocks and their concentration at this level.

154

Rive gauche

C. Delaere and E. Warmenbol

A4 B2 B4 C2 A3 B1 C4 D2 B3 C1 C3 D1

Rive droite

Voute Mouillante

0

10m

Fig. 8.11  Location of the 2012–2014 excavations in the Trou de Han, which are recording a cross-­ section between the banks of the Lesse (Centre de Recherches Archéologiques Fluviales; CAD: C. Delaere)

8.4.1  Artefact Assemblages and Their Implications Between 2012 and 2014, 947 objects or fragments of objects were recovered and were divided into 5 categories for preventive conservation and stabilisation (Table 8.1).

1200-1500 AD

FAUNE (SANGLIER) MÂCHOIRES INFÉRIEURES ET TERRE CUITE GRISE

1692 AD (??)

US4 = EFFONDREMENT DE VOÛTES TREMBLEMENT DE TERRE DE

B1-B2

3

1

0

5cm

1650-1700 AD

2

B3-B4

5

4

0

ASSIETTE EN TERRE CUITE 17 FRAGMENTS (NMI = 1) DÉCOR À LA BARBOTINE REHAUSSÉ DE CUIVRE

A3-A4

Limnimètre 2012-2014 = 85 cm

PIEUX EN PIN

≥ 1870 AD 5cm

1600-1800 AD

PIPE HOLLANDAISE DÉCOR SPIRALÉ

C3-C4

0

5cm

± 1650-1750 AD (??)

PLASTRON EN FER

C1-C2

1600-1800 AD

PERLES EN BOIS CHAPELET 25 PERLES (NMI = 1)

D1-D2

0

3m

OBJETS DÉCOUVERTS DANS LE LIT DE LA RIVIÈRE DATÉS DE 3000 BC À 1950 AD

Voute mouillante

Relevés subaquatiques et coupe: C. Delaere (2014)

Topographie adaptée de M. Georges et et B. Huyghe (1998)

US

FOUILLES 1978-1983

- 400

- 300

- 200

- 100

NIVEAU D’EAU

Surplomb

Fig. 8.12  Upstream cross-section of the 2012–2014 trench, showing the distribution of seventeenth century ceramics (Centre de Recherches Archéologiques Fluviales; CAD: C. Delaere)

Rive Gauche

TROU DE HAN Han-sur-Lesse / Rochefort Coupe Amont - Tranchée 1 - T1 (fouilles 2012-2014)

8  The Watery Way to the World of the Dead: Underwater Excavations … 155

156

C. Delaere and E. Warmenbol

Table 8.1  Breakdown of artefactual assemblage from the 2012—2014 excavations at Han-sur-­ Lesse by material 2012 2013 2014 Total

Ceramic 28 186 135 349

Metal 5 58 106 169

Glass 0 14 67 81

Lithic 1 8 29 38

Organic 3 46 249 298 Total

Other 0 0 12 12 947

Ceramics represent 36.85% of the assemblage and include coarseware, glazed pottery, stoneware and faience. Metals account for 17.85% of the assemblage and consist mainly of nineteenth and twentieth century coins, together with ironwork from the sixteenth to twentieth centuries. Glass, which accounts for 8.55%, consists almost entirely of fragments of petroleum lamps used in the late nineteenth century, as well as non-industrial and modern bottle fragments. The ‘lithic’ objects (4%) and ‘others’ category (1.25%) are not representative, since they comprise unidentified fragments and contemporary material of all types. Finally, organics, which account for 31.5% of the assemblage, consist of charred wood elements, numerous macro-­ remains (fauna), and many fragments of leather shoes and wooden beads. From preliminary assessment by Sophie Challe (SPW), the chronological distribution of the 349 ceramic fragments is as follows: 7.15% dates from the Bronze Age or Iron Age (24 sherds), 0.85% from Late Antiquity (3 sherds), 0.55% from the fourteenth and fifteenth centuries (2 sherds), and 88.85% from the sixteenth to the twentieth century (310 sherds), while 2.6% remains chronologically unattributed (9 sherds). It must be stressed that, out of the 27 sherds dating to protohistory or antiquity, 18 were excavated during the last 4 days of digging in 2014, while only 2 were discovered in 2013. We thus believe that the stratigraphy is coherent and homogeneous, and that the alternation between multiple mineral and organic layers (floods) is testimony to five centuries of sedimentation and erosion of the Lesse. Among the 24 protohistoric sherds, Walter Leclercq (CReA-Patrimoine, ULB) has identified 6 from the Late Bronze Age and 3 from the later Iron Age. Some sherds may be intrusive, but the majority were recovered during excavation of the last homogeneous layer of scree in 2014 (US4 and E1). Therefore, future excavation will undoubtedly confirm whether in situ layers dating to these earlier periods exist, both in primary contexts (palaeoterrestrial occupation floor) and/or in secondary contexts (old layers later reshaped by the Lesse). In either case, these early deposits are sealed by US4 and have not moved since the seventeenth century. The distribution of the seventeenth century ceramic material was fairly coherent (Fig.  8.12), though significantly, for some specimens dating to this period, fragments of the same vessel were distributed throughout the cross-section. Other ceramic distributions provided information on the profile of the riverbed and also on the levels of the scree. For example, 17 sherds belonging to 1 or 2 ornamental glazed

8  The Watery Way to the World of the Dead: Underwater Excavations …

157

plates (decorated with slip) indicated the position of the base of the river; when this vessel was thrown into the water, it subsequently broke, with fragments scattered along the inclination of the riverbed. Ceramic sherds dating to the seventeenth to nineteenth centuries have sharp fractures which show no signs of erosion, indicating that their context of discovery was that of their initial ‘deposition’. However, the sixteenth century ceramics that are highly eroded and ‘rolled’ by the river may have originated slightly further upstream or from a closer but certainly modified area. Associated with the ceramics of the seventeenth century were other objects, such as the soles of leather shoes, an iron breastplate (Fig.  8.13), lead shot for an arquebus, a gilded bronze coin and 25 wooden beads, which were discovered during the last 5  days of excavation and which indicate a homogeneous context. An encouraging factor for future study of earlier layers is the recovery of between three and five fragments of the same Late Bronze Age goblet, with much better preserved decoration than identical objects found in the riverbed in the past. These layers could therefore be relatively well preserved and undisturbed. Nevertheless, the river was not always ‘channelled’, and before it was ‘frozen’ by creation of the water reservoir, its various meanders may have produced multiple distinct deposits and/or different morphologies of riverbed prior to the seventeenth century.

Fig. 8.13  Seventeenth century iron breastplate (Photograph: A. Huybrechts and M.-J. Declerck)

158

C. Delaere and E. Warmenbol

8.5  Conclusion: An Enduring Ritual Landscape Tourism at the caves of Han-sur-Lesse is only rarely documented before the nineteenth century, yet ceramic material recovered from ‘recent levels’ between 2012 and 2014 indicates that visitors were already showing interest in the Trou de Han before this period. The presence of the reservoir is documented in 1822, but the taphonomy of the ceramics uncovered in 2012–2014 indicates that its first iterations could date from as early as the seventeenth century. Indeed, there is a change in the preservation of the underwater material around this time: ceramics of the sixteenth century are ‘rolled’ and suggest a fluvial context, while the ceramics of the seventeenth to eighteenth century are well preserved and sealed in situ by a spread of stones which may be associated with these initial constructions. Whether to facilitate access to the cave or to create an artificial reservoir or a recreational space, the development of the artificial lake could therefore be of considerable antiquity and has profoundly altered the morphology of the Trou de Han. Indeed, the mutation of this landscape has left ‘marks’ which are still visible in the sedimentary morphology of the river, as identified during the 2012–2014 excavations, and which demonstrate both palaeoterrestrial and more typical fluvial contexts. Almost 90% of the material discovered between 2012 and 2014 concerns recent periods (sixteenth to twentieth centuries) and constitutes a representative sample of tourism at the cave of Han-sur-Lesse, as well as various other constructions carried out between the seventeenth and twentieth centuries. This material, less well-­ documented in comparison to that of the earlier chronological periods usually studied at this site, demonstrates a major historical event for Han-sur-Lesse, whose socio-economic organisation is still largely dependent on tourism today. The caves of Han are known historically as one of the first tourist attractions in Belgium, predominantly from the nineteenth century onwards, though new evidence for ‘recreational’ visits as early as the seventeenth century makes them an exceptional and atypical case study. The river retains everything of this history of use: the first electric cables from 1903, fragments of petroleum lamps from the second half of the nineteenth century, bottles of OXO from the cave bistro, soles of the damaged shoes of the first visitors, coins deposited or lost in the river, and objects broken and then abandoned as ‘rubbish’. It seems quite clear, however, that use of the caves at Han from the end of the Neolithic to the Roman period was not of a ‘domestic nature’. The objects found by the divers were not casually lost, discarded or broken but were carefully selected and often transformed before their deliberate deposition within a theatre of ritual associated with the dead or with ‘Death’ itself (Fig. 8.14). The recent archaeological work documented above does, however, lead us to question our assumption that all (or even the majority) of these finds were originally cast into the river. The time has come for a holistic approach to the caves of Han-sur-Lesse: an approach which takes into account the many interconnected archaeological features (both wet and dry) that made up this ‘World of the Dead’, revered from the Neolithic to Roman times.

8  The Watery Way to the World of the Dead: Underwater Excavations …

159

Fig. 8.14  Artistic representation of a sword being offered to the River Lesse (Drawing and colouring by M. Rénier, 1995; collection of E. Warmenbol)

References Absil, M. (1988). Le ‘diplôme’ de Han-sur-Lesse: Un exemple de traitement informatisé des documents épigraphiques et d’application de la méthode PETRAE aux diplômes militaires romains. Annales de la Société archéologique de Namur, LXV, 353–371. André, P. (2013). Etude anthropologique des ossements de la grotte du Père Noël et de la Galerie de la Grande Fontaine (Rochefort, Province de Namur). Unpublished MA thesis, Université Libre de Bruxelles. Bonenfant, P.-P. (1982). Stratigraphie de Han-sur-Lesse. Annales d’Histoire de l’Art et d’Archéologie, IV, 115. Bonenfant, P.-P. (1984). Stratigraphie de Han-sur-Lesse. Annales d’Histoire de l’Art et d’Archéologie, VI, 106. Callewaert, M., & Goffette, Q. (2011). Analyse typologique et technologique des fibules romaines de Han-sur-Lesse. Actes de la Journée d’Archéologie Romaine, 2011, 21–30. Cosyns, P., Warmenbol, E., Bourgeois, J., & Degryse, P. (2005). Pre-Roman glass beads in Belgium. In M.-D.  Nenna (Ed.), Annales du 16ème Congrès de l’Association Internationale pour l’Histoire du Verre, London, 2003 (pp.  326–330). Nottingham: L’Association Internationale pour l’Histoire du Verre. de Pierpont, E. (1904). Fouilles et explorations archéologiques de la Grotte de Han (1902 à 1904). In E. de Pierpont (Ed.), Fédération Archéologique et Historique de Belgique. XVIIe session. Congrès de Dinant organisé par la Société Archéologique de Namur, 9–13 août 1903 (pp. 519– 521). Namur: Federation Archeologique et Historique de Belgique.

160

C. Delaere and E. Warmenbol

Fourny, M. (1995). Armatures de flèches et pointes de silex de la Grotte de Han (Han-sur-Lesse/ Rochefort, Nr.): Du Paléolithique supérieur à la fin du Néolithique… voire au-delà. Amphora, 77, 3–31. Goffette, Q. (2013). Le matériel d’époque romaine de la grotte de Han (Han-sur-Lesse, province de Namur). Annales d’Histoire de l’Art et d’Archéologie, XXXV, 107–117. Jasinski, M. (1965). Plongées sous la terre. Paris: Flammarion. Kickx, J., & Quetelet, A. (1822). Relation d’un voyage fait à la grotte de Han au mois d’août 1822. In Nouveaux mémoires de l’Académie Royale des Sciences et Belles Lettre de Bruxelles, volume 2 (pp. 315–362). Brussels: P. J. De Mat. Lallemand, J. (1994). Les monnaies antiques de la grotte de Han. Amphora, 75, 4–28. Larocca, F. (2010). Le Grotte dell’Angelo a Pertosa: Il sistema sotterraneo e il giacimento archaeologico. Pertosa: Fondazione Musei Integrati dell’Ambiente. Leclercq, W. (2012). L’âge du Bronze final dans les bassins de l’Escaut et de la Meuse moyenne: Culture matérielle et cadre socio-économique. Unpublished PhD thesis, Université Libre de Bruxelles. Mariën, M.-E. (1970). Le Trou de l’Ambre au Bois de Wérimont, Eprave. Monographies d’archéologie nationale 4. Brussels: Musées Royaux d’Art et d’Histoire. Mariën, M. (1981). Cuillères en os de type Han-sur-Lesse (Néolithique S.O.M.). Helinium, XXI, 3–20. Mariën, M. (1982). Fouilles à la Grande Fontaine à Han-sur-Lesse (XVIIe siècle et champs d’urnes). Archaeologia Belgica, 247, 31–35. Mélin, M. (2012). Existe-t-il un lien entre les pratiques de dépôts métalliques non funéraires et les variations climatiques? In M. Mélin & C. Mougne (Eds.), L’Homme, ses ressources et son environnement, dans le Nord-Ouest de la France à l’âge du Bronze: Actualités de la recherche. Actes du Séminaire Archéologique de l’Ouest, 22 mars 2012, Université de Rennes I. Mémoires Géosciences Rennes, hors série no. 8 (pp. 113–127). Rennes: Éditions de Géosciences. Pleuger, L. (2011). A propos du Néolithique récent/final de la Grotte de Han à Han-sur-Lesse… (Rochefort, B). Notae Praehistoricae, 31, 5–13. Sohn, M. (2009). La notion de dépôt ‘collectif’ dans les sites funéraires de la fin du Néolithique en Europe occidentale. In S.  Bonnardin, C.  Hamon, M.  Lauwers, & B.  Quilliec (Eds.), Du matériel au spirituel: Réalités archéologiques et historiques des ‘dépôts’ de la préhistoire à nos jours. XXIXe rencontres internationales d’Archéologie et d’Histoire d’Antibes (pp. 131–141). Antibes: Éditions APCDA. Timperman, M. (1989). La grotte de Han au fil des siècles. Gembloux: Imprimerie Duculot. Van Strydonck, M., & Warmenbol, E. (2012). Une séquence radiométrique du Néolithique final à La Tène finale: Le ‘pilier stratigraphique’ de Han-sur-Lesse (prov. de Namur, Belgique). Lunula: Archaeologia Protohistorica, XX, 3–9. Warmenbol, E. (1996). L’or, la mort et les Hyperboréens: La bouche des Enfers ou le Trou de Han à Han-sur-Lesse. In P. Schauer (Ed.), Archäologische forschungen zum kultgeschehen in der Jüngeren Bronzezeit und Frühen Eisenzeit Alteuropas. Ergebnisse eines kolloquiums in Regensburg, 4–7 Oktober 1993 (pp. 203–234). Bonn: Universitätsverlag Regensburg GMBH. Warmenbol, E. (1999). Le soleil des morts: Les ors protohistoriques de Han-sur-Lesse (Namur, Belgique). Germania, 77, 39–69. Warmenbol, E. (2004). Gold pickings and PIXE analysis: More about the Bronze Age gold found in the cave of Han-sur-Lesse (Namur, Belgium). Nuclear Instruments & Methods in Physics Research (Section B), 226, 208–221. Warmenbol, E. (2007a). Un dépôt de mandibules humaines dans la grotte de Han-sur-Lesse (Rochefort, Namur). In Actes du VIIe congrès de l’Association des Cercles Francophones d’Histoire et d’Archéologie de Belgique. Congrès d’Ottignies – Louvain-la-Neuve, 26, 27 et 28 Août 2004, volume 2 (pp. 641–650). Bruxelles: Éditions Safran. Warmenbol, E. (2007b). Le dépôt d’ossements humains en grotte aux âges des Métaux en Belgique: Nouvelles questions. In P. Barral, A. Daubigney, C. Dunning, G. Kaenel, & M.-J. Roulière-­ Lambert (Eds.), L’âge du Fer dans l’arc jurassien et ses marges. Dépôts, lieux sacrés et territorialité à l’âge du Fer. Actes du XXIXe colloque international de l’AFEAF, Bienne, 5–8 Mai 2005, volume II (pp. 537–547). Besançon: Presses Universitaires de Franche-Comté.

8  The Watery Way to the World of the Dead: Underwater Excavations …

161

Warmenbol, E. (2009). Natures mortes: Les dépôts subaquatiques de Han-sur-Lesse (Belgique). In S. Bonnardin, C. Hamon, M. Lauwers, & B. Quilliec (Eds.), Du matériel au spirituel: Réalités archéologiques et historiques des ‘dépôts’ de la préhistoire à nos jours. XXIXe rencontres internationales d’Archéologie et d’Histoire d’Antibes (pp. 143–154). Antibes: Éditions APCDA. Warmenbol, E. (2013a). Un nouvel exemplaire de cuillère en os de type ‘Han-sur-Lesse’, en provenance du site éponyme: Contexte et chronologie (B). Notae Praehistoricae, 33, 147–152. Warmenbol, E. (2013b). Le deuxième âge du Fer (fin Ve-début Ier s. avant notre ère) dans la grotte de Han (commune de Rochefort, province de Namur, Belgique). Revue du Nord, 95 (Hommages à Germaine Leman-Delerive), 91–112. Warmenbol, E. (2015). Vestiges archéologiques et occupations en grotte dans la Lesse Calestienne. In G.  Michel & G.  Thys (Eds.), Atlas du Karst Wallon. Bassin de la Lesse Calestienne: Inventaire cartographique et descriptif des sites karstiques et des circulations d’eau souterraine (pp. 100–109). Jambes: CWEPSS/Service Public de Wallonie. Warmenbol, E. (in press). Les dépôts d’objets métalliques fragmentés du Bronze final dans la grotte de Han (Rochefort, Belgique): ‘Les morceaux choisis, c’est toujours les morceaux choisis par un autre’. In B. Toune & E. Warmenbol (Eds.), Pezzi scelti. Distruzione e manipolazione di beni tra età del Bronzo e del Ferro: Dal riciclo al sacrificio. Atti del Convegno Internazionale, Roma, 16–18 Febbraio 2012. Rome: Institut Historique Belge de Rome. Warmenbol, E., & Raepsaet-Charlier, M.-T. (2014). Un plat ovale en alliage cuivreux du IIIe siècle provenant du Trou de Han (Nr.). Signa, 3, 177–182. Warmenbol, E., & Van Strydonck, M. (2011). Quelques dates radiométriques pour les haches à douille (Bronze final) du ‘Trou de Han’ à Han-sur-Lesse (province de Namur, Belgique). Lunula: Archaeologia Protohistorica, XIX, 55–59. Warmenbol, E., & Van Strydonck, M. (2012). Quelques dates radiométriques pour les couteaux à douille (Bronze final) du ‘Trou de Han’ à Han-sur-Lesse (prov. de Namur, Belgique). Lunula: Archaeologia Protohistorica, XX, 45–47.

Part II

Innovative Digital Capture Techniques in Cave Archaeology

Chapter 9

The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French Heritage Site Albane Burens, Laurent Carozza, Raphaelle Bourrillon, Stephane Petrognani, Pierre Grussenmeyer, Samuel Guillemin, François Lévêque, Vivien Mathé, Yves Billaud, Aurélie Brodard, Pierre Guibert, Stéphane Jaillet, Olivier Jest, and Mathieu Koehl

9.1  Introduction This chapter details the study of the exceptional and atypical Bronze Age French site of Les Fraux (Dordogne, France), which combines cave art and testimonies of domestic and symbolic activities (Fig.  9.1). The interdisciplinary research conducted on the site since 2007 attempts to untangle the interwoven uses of this cave for later prehistoric communities, during short but regular visitation. A. Burens (*) · L. Carozza Environmental Geography Laboratory, CNRS, UMR 5602 GEODE, Toulouse, France e-mail: [email protected]; [email protected] R. Bourrillon TRACES-UMR 5608, CREAP, Toulouse, France S. Petrognani UMR 7041 ArScAn, Ethnologie Préhistorique, MAE René Ginouvès, Nanterre, France P. Grussenmeyer · S. Guillemin · O. Jest · M. Koehl ICube Laboratory, CNRS, UMR 7357, Photogrammetry and Geomatics Group, INSA Strasbourg, Strasbourg, France e-mail: [email protected]; [email protected]; [email protected]; [email protected] F. Lévêque · V. Mathé UMR 7266 LIENSs, CNRS, Université de La Rochelle, La Rochelle, France e-mail: [email protected]; [email protected] Y. Billaud · S. Jaillet UMR 5204 EDYTEM, CNRS, Le Bourget du Lac, France e-mail: [email protected]; [email protected] A. Brodard · P. Guibert UMR 5060 IRAMAT CRPAA, CNRS, Pessac, France e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_9

165

166 Fig. 9.1  The Bronze Age cave of Les Fraux (Saint-Martin-de-­ Fressengeas, Dordogne, France): (a, b) location of Saint-Martin-de-­ Fressengeas, Aquitaine (© Google Maps; A. Burens), (c) location of the underground network on aerial photograph (orange) (© Google Maps; M. Koehl)

A. Burens et al.

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

167

The opportunity to study a cave which temporarily hosted small Bronze Age communities over nearly three centuries allows us to gain a detailed understanding of the nature of activity in relation to the long duration of Palaeolithic occupation that characterises most of the major French painted caves. Indeed, there is no evidence of human activity inside the cave of Les Fraux before the Middle Bronze Age, and none after the Late Bronze Age, when a roof collapse sealed the cave and preserved intact the archaeological remains until 1989: the date of the site’s rediscovery. Thus, Bronze Age activity in the cave corresponds to a single period around 1450–950 BC (the maximum duration of human activity inside the cave). The extraordinary conditions of preservation of the site (due to the collapse of its entrance), and the plurality of human activity of different kinds over a short period of time, force us to take advantage of new technologies in order to elucidate information unattainable in most French prehistoric painted caves. By combining chronology, interactions between people and their environment, and emerging technologies, our broader objective was to gain a better understanding of the complexity of perceptions of the cave of Les Fraux by local or exogenous prehistoric communities. One aim was the interpretation of archaeological data within a framework which combined space and time (in which all stages of the research are integrated), from data acquisition, implementation of site monitoring (temperature, CO2 levels, etc.), experimentation and simulation to reconstruction. We employed an integrative and interdisciplinary research approach, based on new technologies able to acquire full 3D documentation of the entire site (even in inaccessible and non-visible parts) and to process interdisciplinary data according to accurate 3D models of the cave that constituted the common framework for the various studies and partners involved. The 3D geographic information system (GIS) of the cave aggregates all types of data produced since 2008 within the same system, including overall 3D recording of structural elements, maps, pictures, analyses, databases, drawings and reports. It allows, for example, for spatial analyses between scientific data and archaeological and geomorphological contexts. By using non-invasive techniques such as laser scanning and photogrammetry, we aimed to preserve the archaeological remains and the integrity of the site, in which all of the material survives in situ, across 9 years of study.

9.2  T  he Contribution of the Cave of Les Fraux to Existing Palaeolithic Sites The cave of Les Fraux is located in south-west France, in the Périgord-Limousin Regional Natural Park, within the township of Saint-Martin-de-Fressengeas (Fig. 9.1). It consists of an extensive underground network (Fig. 9.2), difficult to access and comprising narrow, damp and sinuous galleries of more than 1  km, formed in Middle Jurassic geology. Its development is nearly horizontal, following

168

A. Burens et al.

Fig. 9.2  Map of the underground network (Y. Billaud and A. Burens) and view of gallery no. 13 (Photograph: A. Burens)

two preferential directions (Burens et al. 2014). The cave comprises two independent parts, simultaneously occupied during the Bronze Age. It is poorly ventilated and does not benefit from natural light: Les Fraux is an entirely dark-zone cave (Faulkner 1988). It might have been chosen for its discreet entrance, which is hardly

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

169

visible on the surface of the landscape. Geomorphological studies indicate that several entrances could have been used since the network’s formation, though our reconstruction suggests that the Bronze Age point of entry could have existed near the current one. The cave is characterised by a succession of phases in which it was alternately accessible and inaccessible. The presence of cave bears demonstrates that the cave was open for a time during the Pleistocene, before its first (undated) obstruction, probably at some point before the Holocene. After a long period of inaccessibility, the cave became accessible once again in the Middle to Late Bronze Age; radiocarbon dating and archaeological investigations suggest that the cave was used from 1450 to 950 BC. At the end of the Bronze Age, the collapse of the entrance sealed the cave and preserved intact the archaeological remains until 1989. In short, there is no evidence of human activity in the cave before the Middle Bronze Age, and none after the Late Bronze Age. As a result, the site displays very little evidence of multi-layered deposits (except in one gallery characterised by two different palaeosols). This peculiarity provides a considerable advantage in our ability to study the relatively short-term perception of the site by ancient communities, in comparison, for example, to Upper Palaeolithic sites, which were generally occupied for longer. For instance, the study of 259 radiocarbon dates from the exceptional Chauvet-Pont d’Arc Cave in Ardèche, France (mainly related to rock art and other anthropogenic material), recently highlighted two periods of human activity: one from 37,000 to 33,500 years ago and the other from 31,000 to 28,000 years ago (Quiles et al. 2016). This outstanding and unparalleled work has allowed for a global overview of the site’s history and argues in favour of creation of the parietal art before 28,000 years ago. But the temporal sequence remains too large to allow for a detailed understanding of the tempo of human activity at the cave, in contrast to Les Fraux where the short duration of activity helps us remove ambiguity in the nature of the various visitations. The site was discovered in 1989; it was immediately nicknamed the ‘Lascaux of the Bronze Age’ by French media, owing to its extraordinary archaeological and heritage significance, and was registered with the French Historical Monuments in 1995. According to the French Office of Geological and Mining Research (BRGM), the cave’s geology is highly unstable, and it may unfortunately collapse in the coming years. The cave has, therefore, been sealed off to the public, and access is now severely restricted. The cave of Les Fraux is thought to contain rare examples of Bronze Age cave art in Europe and is the only protohistorical cave site with evidence for domestic, symbolic and/or votive activities and cave art. When comparing the site with other French Bronze Age caves, there are several similarities. Mostly, however, there are distinct differences, such as the absence of cave art and/or the presence of funerary activity in association with the deposits (as is the case, for instance, in Charente, where multiple graves have been found in Bronze Age caves; Gomez 1980). The Bronze Age activity at Les Fraux, which occupies nearly 70% of the cave (Fig. 9.2), left very well-preserved archaeological deposits, domestic features (such as stairs

170

A. Burens et al.

dug into the clay), torches and postholes, together with no less than 66 hearths and a further 24 traces of hearths. The same galleries included 32 deposits of ceramics and metalwork. The cave also contains many perfectly-preserved carvings on the walls and ceiling. Most of the wall art comprises finger-strokes or is otherwise engraved in the clay and comprises linear, schematic and geometric designs (Carozza et  al. 2009). All available evidence (typology, radiocarbon dating, etc.) suggests that the cave art and some material deposits should be assigned to the Middle Bronze Age, whilst other deposits date to the Late Bronze Age. The cave of Les Fraux represents an extraordinary opportunity for the study (in all its complexity) of prehistoric cave use (practical versus symbolic) and the significance of this place for prehistoric communities. The short duration of human activity witnessed has also allowed us to create fine-grained chronologies of the various activities which took place.

9.3  B  enefits of New Technologies in Understanding the Complexity of Sites and Their Interwoven Uses It was decided that, in pursuit of understanding the complex use of decorated cave sites, archaeological deposits at the site of Les Fraux would not be destroyed by excavation; their integrity would be maintained for potential future studies. In the course of the study, almost no archaeological material was handled or removed from the site; it was, instead, studied, registered, measured and drawn in situ, without physical contact, or was studied virtually through the use of 3D models (Fig. 9.3). This methodology had a significant impact on how the site was studied and on the conditions under which data were obtained. Working with non-invasive techniques, such as contact-free measurement of ceramic deposits and wall art, was challenging. The analyses were performed using equipment designed to record the maximum amount of information possible, even in inaccessible or non-visible parts of the cave (narrow galleries, cracks, areas underneath roof collapse, etc.). This 3D protocol corresponded to an attempt by the team to georeference each archaeological and geological dataset from the site and to enforce large-scale spatial analysis, allowing for the full integration of interdisciplinary and multi-scalar data. Furthermore, we aimed to produce 3D models in order to test our archaeological hypotheses and generate accurate visualisation documentation. We integrated different techniques of data recording based on terrestrial laser scanning, digital photogrammetry and spatial imaging systems (Fig. 9.4), in order to generate geometric and photorealistic 3D models of all structural elements of the cave and of the main archaeological artefacts from a combination of point clouds and photogrammetric images. Furthermore, our approach integrated different datasets generated at different scales using the various types of 3D recording, including the overall volume of the cave (Fig. 9.5a, b), wall art representations (Fig. 9.5d) and archaeological deposits (Fig.  9.5c), demonstrating that it was in fact possible to combine, in the same depiction system, several different sources of information.

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

171

Fig. 9.3  3D recording of a Bronze Age ceramic deposit inside gallery no. 13 (using FARO Focus 3D): top view of the deposit and of the 3D recording system, middle and bottom visualisation of the deposit as a point cloud (A. Burens, S. Guillemin and P. Grussenmeyer)

172

A. Burens et al.

Fig. 9.4  Recording of the total network of galleries by surveying techniques and terrestrial laser scanning (TLS). The cave is linked to the terrain above thanks to surveying techniques (total station underground and GPS measurements in the field): (a) GPS measurements, (b) surveying techniques using total station, (c, d) TLS, (e) TLS combined with photogrammetry, (f) photogrammetry (Photographs: A. Burens)

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

173

Fig. 9.5  3D recording of archaeological and geological information from the site based on terrestrial laser scanning, digital photogrammetry and spatial imaging system: (a) image of gallery no. 13 of the cave (left) and its 3D model in PDF 3D format for visualisation in Adobe Reader (right), (b) image of gallery no. 18 (left) and its 3D model in PDF 3D format (right), (c) 3D recording of a Bronze Age vessel from the cave, coated with the charred remains of animal fat, (d) 3D recording of the main decorated panel from gallery no. 13 (S. Guillemin, P. Grussenmeyer and A. Burens)

174

A. Burens et al.

These data not only included drawings (such as sections and elevations) but also photorealistic perspectives from different viewpoints and visual 3D models. We also implemented various other interdisciplinary studies, including different modelling approaches, magnetic field mapping, palaeoenvironmental reconstructions and a simulation of the cave’s thermal dynamics. In addition, we tested innovative and experimental approaches which combined photogrammetry and terrestrial laser scanning, including photorealistic rendering of the cave art. The documentation and recording work contributed to accurate indexing and georeferencing of the entire set of surveys, images, structural models and relief drawings acquired in different parts of the cave. The following section provides a technical overview of some of the basic methods used in the study, in order to reconstruct and model the cave, and to help us understand the character of the various human activities which took place within it.

9.3.1  Overall 3D Recording of Structural Elements Overall recording of the cave by terrestrial laser scanning (TLS) was our first goal, since it allowed for georeferencing of any type of object, image or measurement and accurate indexing of the data between the various researchers involved in the project. The study began with consolidation of the topography and the creation of a polygonal network inside the cave with a total station. This network was connected to the landscape outside the cave by several geodetic points, measured using differential global navigation satellite system (GNSS) techniques. The first TLS campaigns were undertaken using a FARO Photon 120 Scanner, which was well-­adapted for the central underground passage. Since 2012, data acquisition has been undertaken using a FARO Focus 3D Scanner. The expected accuracy of both scanners is about 2–5 mm at 25 m. The scanner was moved along the galleries, and an overlapping area between two stations was routinely provided. More than 600 m of galleries have been recorded so far (Fig.  9.6), requiring more than 130 stations and 30 days in the cave. Handling data from multiple stations required resampling of the individual point clouds, which were then imported and merged into a single point cloud. The need to facilitate the exchange and dissemination of data within the team led us to export the global 3D model of the cave, resampled at 1 point/cm, in 3D PDF format. This format allowed for the retention of measuring functions and toolkits (distances, sections, etc.), as well as visualisation options (such as wireframe, shaded, solid, vertices). We also recorded ceramic and metalwork deposits within the galleries (Figs. 9.3 and 9.5c). When the 3D model was transformed into 3D PDF, it was then possible to draw, measure and define the section and typology of every ceramic artefact, without the need for any contact with the deposit (though this approach was not possible for completely inaccessible deposits, such as pottery almost entirely hidden within and between rocks).

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

175

Fig. 9.6  Global 3D models of the underground network of the cave; data recorded using the French Geodetic Reference System (Lambert 2) (P. Grussenmeyer, S. Guillemin and A. Burens)

9.3.2  Photorealistic Rendering In archaeological caves, studies of wall art require the processing of high-definition images with greater accuracy than is usually necessary for the total volume of the underground network. A common solution is the correlation of high-resolution digital photographs of specific panels with 3D volumetric models of the cave (Grussenmeyer et  al. 2012). The aim of the experimental work developed at Les Fraux since 2008 was to find an interface between the 3D imaging methods and the

176

A. Burens et al.

Fig. 9.7  3D recording of Bronze Age finger-strokes in the clay wall of the main panel from gallery no. 13 of the cave: (a) recording test with the FARO ScanArm, (b) view of the high-density point cloud recorded using the FARO ScanArm, (c) digital surface of a selected piece of the panel, (d) textured 3D model, (e) projection of drawn finger-strokes onto the 3D model, (f) detail of the digital drawing from the orthophoto extracted from the 3D model (S. Guillemin, P. Grussenmeyer and A. Burens)

photographic and manual illustrations of cave art specialists; an interface which reconciled the subtleties of traditional drawing with the rigorous precision and swiftness of 3D survey (Fig. 9.7). We managed to negotiate, for the first time in an archaeological cave, access to the latest generation FARO ScanArm V3 (19,200 points/s and 0.035  mm accuracy). The exceptional quality of the resulting point cloud is due to the accuracy of the fusion arm and ensuring that the laser V3 head remained only a few centimetres from the object during the time-consuming recording process. This new equipment, providing submillimetre resolution, has enabled the team to succeed in mapping high-definition images onto a very accurate 3D model, in order to obtain photorealistic rendering of the decorated panels by combining photogrammetry and TLS. Orthophotos from the 3D models were used (after digital image enhancement processing) to support vectorisation of the decorated panels and produce an alternative to traditional and time-consuming drawings by creating an accurate, georeferenced image. This technique provided better recording of superimposed archaeological depictions, in order to gain a greater understanding of their relative chronology. The recording of cave art using orthophotos, based on 3D models, will prove very useful for archaeologists. We were also able to carry out experimental work on identifying the various techniques and tools used to create the decorated panels, by comparing 3D models of several in situ panels with their experimental counterparts (Fig. 9.8). The latter were made in soft clay materials placed in vertical frames. Finger-strokes were

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

177

Fig. 9.8  3D recording of a decorated panel from gallery no. 10 and its experimental duplicate with the FARO ScanArm: (a) view of the decorated panel, (b) 3D recording of the decorated panel by TLS with the FARO ScanArm, (c) preparation of experimental duplicates with different tools, (d) scanning of the experimental duplicate used as reference, (e) view of section of the finest carvings from the decorated panel (A. Burens, S. Guillemin and P. Grussenmeyer)

made by hand, whilst engravings were made using a wide range of tools on the surface of the clay: flint, bone, bronze and copper implements, dry wood, freshly cut branches, etc. Drawings were created in conditions close to those observed on site (e.g. moisture content of the  clay, position of the  drawing inside the cave, etc.). The aim was to attempt to differentiate between the various techniques used to create the motifs, such as finger-strokes, engraving using tools or etching. The flexibility of the head of the ScanArm used for this work allowed for recording of the base of even the thinnest engravings, including oblique ones, both on the in situ panels and their experimental counterparts. Indeed, the laser can be oriented effortlessly in all directions. Afterwards, we compared sections through the 3D archaeological motifs with the 3D experimental duplicates in order to assess the morphology and types of tools used. By using this approach, we were able to propose the probable use of wooden and metal tools for the execution of engravings.

178

A. Burens et al.

9.3.3  The Les Fraux 3D Geographic Information System The Les Fraux 3D GIS was designed to store, manipulate, manage, present and share the huge amount of 2D and 3D spatial and geographical data produced and georeferenced by the total station and laser scanning (Fig. 9.9). This system allowed us to create interactive queries, interrogate spatial information and edit data in maps, whilst also taking into account the location and nature of archaeological artefacts and geomorphological contexts. A 3D approach is crucial in the context of underground networks. Thus, we chose a system which could manage 3D point clouds (ArcMap 10.2) and temporalities. Unlike traditional relational databases, this GIS processes the spatial dimension of multi-scalar and heterogeneous data of various different disciplines. The conceptual data model thus expresses relationships between multidisciplinary information. Very large datasets, including images, maps, analyses, reports in JPEG and PDF formats, and associated objects and features, have been integrated within this GIS. According to this system, we are able to produce 2.5D thematic maps—in which the coordinates of a point [x, y] are associated with only one altitude [z]—to simulate 3D movement within the galleries, to manage archives (images, analysis, etc.), and to create automated dynamic mapping of atmospheric measurements. Some tools deal with the spatial analysis of artefacts, hearths, deposits and decorated panels in relation to other entities (e.g. the close association between the location of decorated panels and symbolic deposits). Overall 3D recording of the cave's structural elements constitutes the basic element of the GIS. 3D data were resampled before being integrated in the GIS. The georeferenced point clouds allowed for 3D visualisation of the cave and the semantic data linked to artefacts and geological objects recorded on site. Data were either quantitative or qualitative. Semantic data were located according to the point clouds (i.e. the structural elements of the cave).

Fig. 9.9  Left screenshot from the cave’s 3D geographic information system (gallery no. 28), right handling of 3D point clouds in  the ArcMap 10.2 viewing window and the creation of sections (O. Jest, M. Koehl and A. Burens)

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

179

They were processed in a table before being integrated into the GIS. Vector data (2D maps) were exported in DWG format for georeferencing and integration into the GIS.  Raster data were used as the background of the GIS and also required coordinates. The GIS included, for example, all of the information related to the hearths (morphology, location, spatial context, presumed function, analysis, radiocarbon date, etc.). Preliminary analyses suggest that spatial correlations exist between hearths and other elements which structured space, such as votive deposits, cave art or domestic debris. Thus, the GIS constitutes an important tool for analysing the location of the hearths within their immediate environment. Now that the georeferenced data-gathering period is complete, our goal is to create a reconstruction and simulation of the various past activities which took place in the cave, using the 3D models of the cave and the GIS. The integration of temporality in 3D numerical simulation is an important challenge for our understanding of the development of the underground network, from the modern day to various hypothetical past states, particularly in terms of palaeoclimate. Furthermore, this tool already provides a way for us to compare 3D models of different archaeological cave systems using spatial analysis, in order to deepen our knowledge of human interaction with these environments. In this regard, archaeologists and surveyors from the team have committed themselves to a new project based on archaeological and ethnographic analogy: studying a historical underground network, temporarily occupied by a small population and documented by direct testimony (living witnesses, pictures, written accounts, historical archives, etc.), akin to Cyril Marcigny’s study of the Norman quarries of Caen (Carpentier et al. 2016). Our collective goal is to study, under the same conditions as those implemented at Les Fraux, the quarry of Saingt, Fleury, Normandy. This quarry was occupied by the local population in June and July 1944 (during allied troop bombing raids which accompanied the so-called Normandy Landings). It has been digitised in 3D according to the same techniques as those implemented at Les Fraux, in order to produce 3D models of the site and a 3D GIS. Preliminary results allow us to observe a wide range of similarities in the spatial distribution of artefacts (the remains of meals) and living spaces between these two underground networks. The comparison of characteristics between one site and the other encourages us to reconsider some of our interpretations of Les Fraux, notably that some activity in the cave is more domestic in character than might otherwise have been expected. It is important to note, however, that ethnographic and ethnohistorical evidence cannot always be used as a direct parallel for archaeological interpretations. As Tolan-­Smith and Bonsall (1997, p. 217–218) warn us, ‘the dimension of human behaviour… is notoriously difficult to appreciate in periods before the emergence of documented history and recorded ethnographic observation’, and ‘the distinction between economic and ritual behaviour may be partly false as it arises from the application of a twentieth century rationalist perspective which may not be entirely appropriate in other contexts’.

180

A. Burens et al.

9.4  The Visible and the Unseen: The Issue of Hearths Hearths, which are widely considered as structuring elements, especially in caves, provide important information about the nature of space occupation. Sixty-six domestic hearths (Figs. 9.2 and 9.10) have been recorded in the cave of Les Fraux, together with a large amount of combustion products, across the entirety of the area which saw activity during the Bronze Age (that is to say, nearly 70% of the network). Some structures are partially washed away or buried by sand from the decay of sandstone blocks or flood events. We have recorded 41 fire-settings built directly on the ground surface, 10 hearths not attributable to a specific type because of their poor state of preservation, 7 fire pits (both in natural and anthropogenic features), 5 instances of the indirect evidence of hearths (such as traces of reddening on the cave ceiling or walls and isolated heat-effected sandstone blocks) and 3 hearths made from clay extracted from inside the cave. One of these latter features comprises a semi-elliptical shape, a ridge of clay and a pattern made with a thumb; the only protohistorical parallel for this is in Ouessant, Finistère (Le Bihan and Villard 2010). The exceptional preservation of most of these structures has allowed us to collect quantitative and qualitative data relating to their function. We have established a classification based on morphology, location in relation to their immediate surroundings, dimension, typology (see Beeching and Gasco 1989; Gasco 2003) and the materials used in their construction. Most of the hearths are characterised by their small dimensions (300 °C), are located in narrow galleries (with low ceilings) and more open spaces (rooms). Some of them are close to material deposits and domestic debris; their uses seem more variable. Flat hearths might have been used for baking. All of the information related to fires has been registered in the 3D database (including morphology, typology, location, archaeozoological remains, photographs, description of geological alterations and combustion remains, taphonomy, radiocarbon date, charcoal analysis, fire intensity, etc.). A novel approach to understanding the intensity of these fires has already been published, based on a combination of archaeological observation and analyses of hearth sediments (i.e. thermoluminescent properties of the sedimentary quartz; see Brodard et al. 2012) and relevant proxies obtained from experimental structures (Brodard et al. 2016). This study is thought to allow for evaluation of the fire intensity of hearths (temperature/time/energy) and to contribute to our understanding of their role and their impact on the ground, walls and atmosphere of the cave. In addition, we wanted to find a way to map potential hidden, displaced or ruined structures. This was another innovative approach, managed by two of the team (F.  L. and V.  M.), and was particularly interesting in terms of 3D modelling, in establishing significant progress in our methodological framework. On outdoor

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

181

Fig. 9.10  Overview of some of the hearths in the cave: (a) gallery no. 13, (b) gallery no. 18, (c, d) gallery no. 29, (e) Bronze Age palaeosol from gallery no. 13 and its five hearths (F3, F5, F7, F9 and F17) (Photographs: A. Burens)

182

A. Burens et al.

archaeological sites, cartography of local distortions of the earth’s magnetic field (e.g. generated by the thermal impact on soil or sediment) allows researchers to locate ancient hearths (visible or not) or other structures. For the first time, François Lévêque and colleagues successfully adapted this technique of magnetic survey to the underground network at Les Fraux. The complex topographic configuration of the underground galleries required full 3D data acquisition (instead of the usual 2D acquisition), together with higher-resolution data capture. Rocks interact to a greater or lesser extent with the earth’s magnetic field, depending on their nature and their thermal history. The interpretation of local variation in the magnetic field required us to clearly identify the source of an anomaly (i.e. the highest and lowest values of intensity of the local magnetic field). In order to distinguish topographic effects from the impacts of fire, both magnetic and topographic information were combined in the same 3D model of the cave. The 3D cartography of local distortions of the earth’s magnetic field in relation to Les Fraux was useful in helping to complete the inventory of ancient hearths, both visible and invisible. Measurements of the magnetic field were recorded in the cave using a dual-sensor magnetometer coupled to a 360° prism reflector (Fig.  9.11). Geolocation of measurements was ensured by tracking the 360° prism with a total station. This magnetic field mapping project succeeded in combining, in the same depiction system, several sources of magnetic, topographic and archaeological information (Fig. 9.12). Due to this innovative 3D approach, we were able to map unseen hearths within the cave and to differentiate in situ structures from displaced hearths (Burens et al. 2014). This new method is now applied to other French painted caves, for example, the caves of Cussac, Dordogne and Chauvet Vallon-­Pont-­d’Arc, Ardèche.

Fig. 9.11  The 3D magnetic prospection device inside gallery no. 40 (Photograph: A.  Burens). A device comprising a telescopic boom-pole fixed on a tripod allows for good control of the space covered by the superimposed sensors (white cylinders). Geolocation of the measurements are obtained by a tracking method thanks to a total station

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

183

Fig. 9.12  Gallery no. 40: 3D mapping of the magnetic field (georeferenced in Lambert 2). Thanks to a second magnetometer, the diurnal variation of the earth’s local magnetic field is corrected. (a) point clouds of both the intensity of the magnetic field and the wall surface (the top of the cave is truncated), (b) vertical section 20 cm thick (F. Lévêque)

184

A. Burens et al.

9.5  Types of Space Use and the Meaning of Cave Art As mentioned above, nearly 70% of the cave of Les Fraux displays significant evidence for Bronze Age human activity (Fig.  9.2). That said, we have also found traces of lesser activity throughout the whole cave (isolated charcoal, imprints on walls, etc.), which probably indicates a willingness to recognise and utilise the entire underground network. Our results show that a whole range of evidence for different types of activities (such as hearths, fire pits, faunal remains and wood, ceramic sherds, postholes, votive deposits, stairs dug into the clay, toolmarks on the walls, torches, etc.) was present in a wide range of galleries, even those which were narrow, had low ceilings or were located a long way from the entrance. It must be understood that no remains have been recovered from around the cave entrance, owing to the burial of archaeological deposits by sediment that has flooded in from outside (Fig. 9.13). Nevertheless, it appears that people visited all parts of the cave, from the entrance to remote parts of the interior, with the exception of inaccessible or inhospitable galleries (which were too damp, too poorly ventilated, rich in carbon dioxide, difficult to access, temporarily flooded or affected by subsidence, etc.). Multiple lines of evidence indicate that at least some of the archaeological deposits are associated with domestic activities (Carozza et al. 2009); conceivably, the cave may have been used as a temporary shelter during a period of climatic change (e.g. the 3.2ky BP event). But regarding the spatial location of the cave art, spread over 75 cumulative linear metres (the network extends over more than 1 km), our understanding of the cave’s use over time needs to be reconsidered. The cave contains a vast amount of cave art (Figs. 9.13, 9.14, 9.15 and 9.16), featuring a wide range of schematic and geometric drawings widely dominated by parallel traces and finger-strokes (>80%) (Petrognani et al. 2014). This preponderance gives the impression of a low variability of patterns, made in a repetitive way, using predominantly one technique (fingers: >75%). The drawings have preserved their integrity (Figs. 9.14, 9.15 and 9.16), whilst blocking of the cave and the effective sealing of this archaeological context suggest a high degree of authenticity. The spatial distribution of art does not represent continuous development, even if some small sections present an almost uninterrupted continuous sequence of patterns; it is random in its composition (Fig.  9.14). Drawings have been recorded in 13 different galleries; only 4 display hundreds or thousands of marks. Around 1500 graphical entities and 73 panels contain non-­ figurative motifs. The right network of passages (looking in from the entrance) has delivered a remarkable concentration of 39 decorated panels and 90 groups of motifs. The left network is characterised by a less dense concentration of drawings, with 26 decorated panels and 54 groups of motifs, but it contains the largest and most elaborate panel of cave art (in gallery no. 13). Motifs have been formed using four different techniques (Fig. 9.17). By far the most common is finger-strokes, with more than 70% rendered using this technique. Two, three or four fingers were deliberately pressed several times onto the moist clay of the wall and were moved ­vertically, horizontally or obliquely, forming sequential unexplained patterns. The

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

185

Fig. 9.13  Map of the underground network displaying the location of cave art and other material (Y. Billaud and A. Burens)

186

A. Burens et al.

Fig. 9.14  A select range of decorated panels from gallery no. 13 (Photographs: A. Burens and R. Bourrillon)

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

187

Fig. 9.15  A select range of decorated panels from the cave: (a, c) gallery no. 13, (b) gallery no. 28 (Photographs: A. Burens and R. Bourrillon)

188

A. Burens et al.

Fig. 9.16  A select range of decorated panels from the cave: (a) gallery no. 6, (b) gallery no. 10, (c) gallery no. 9 (Photographs: A. Burens and R. Bourrillon)

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French… Fig. 9.17  Les Fraux cave art: (a) technical characteristics, (b) presentation of the main patterns (A. Burens, R. Bourrillon and S. Petrognani)

189

picking 0.06% % painting 3.24

engraving 21.5%

finger-strokes 75.2%

a

rs 7.4

%

gs

3.3

-za

.7% ts 2

poin

de

an

me zig

grid pattern undet. motifs chevrons - scalariform circles - U shapes - crosses (1.1-0.1%)

%

unorganised finger-strokes 12.3% s ke ro -r st ge fin d se ni ga or .1% 13

parallel lines 57.4%

b

190

A. Burens et al.

second most common technique is engraving (with more than 20% of motifs rendered in this way): patterns were engraved by pressing metal and wooden tools into the soft clay surface of the rock (Burens-Carozza et al. 2011). Lastly, are painting and pecking techniques. Only two techniques have been discovered in the left part of the cave network, which contains the densest concentration of motifs rendered by finger-strokes, together with some engravings. In contrast, all four techniques have been used in the right part of the cave network. Two are exclusively represented in this area: the painting and picking technique. Engraving is also predominant in this part of the cave, but finger-strokes are less common. Motifs are organised into 13 different patterns, reflecting a rich symbolic system (Fig. 9.17). They comprise a relative wide repertoire, including parallel lines (which predominate), organised and unorganised finger-strokes, zig-zags, meanders, points, grid patterns, undetermined motifs, chevrons, scalariforms, circles, U shapes, crosses and so on. For Marcel Otte, the Mediterranean Bronze Age is characterised by an extreme degree of symbolism, but often excluding human representations (except in the south-east of France and in the mountains; Otte et al. 2002, p. 193). Variations can of course be explained by various cultural influences, such as those from the Atlantic associated with the Tumulus Culture in Aquitaine and the north of France, or the emergence of the Duffaits Group in Middle Bronze Age in Perigord (Carozza et al. 2009; Gomez de Soto 1995; Otte et al. 2002). In karstic regions, however, caves are rarely decorated; the cave of Les Fraux constitutes an exception. Even though direct dating of the cave art of Les Fraux remains somewhat speculative and is based on the maximum duration of the human activity on the site (from 1450 to 950 BC according to relative chronology and radiocarbon dates), it seems that we should assign at least some of the wall art to the Middle Bronze Age. Indeed, radiocarbon dating has been undertaken on a charcoal sample from inside one of the motifs on a large decorated panel in gallery no. 18 and returned dates of 1410– 1190 BC (at 98%) and 1380–1210 BC (at 68%) (Lyon SacA 15,773: 3035 ± 35 BP). The cave art only features in a small part of the underground cave network. Its spatial distribution, however, exhibits good correlation between art, hearths, the remains of domestic activities and votive deposits. In seven galleries, cave art is closely associated with votive deposits, whilst in the same areas decorated panels have been found near to pot sherds and other archaeological remains (animal bone, metalwork, etc.). In the four most decorated galleries (nos. 8, 13, 18, 28), panels are closely associated with contemporaneous hearths, perhaps in order to provide light. On the contrary, however, in six other galleries (nos. 4, 5, 6, 9, 10, 14), no visible hearths have been found in close proximity to cave art. Here we observe a clear disconnect that cannot be explained purely by problems of taphonomy. Domestic and ritual activity is rarely witnessed simultaneously in the same cave. Work in progress, especially in gaining additional radiocarbon dates and refining our relative chronology, could eventually create a more detailed understanding of these patterns of activity and whether they were indeed contemporary or sequential. From a different perspective, research on reverberation sounds in caves suggests links between prehistoric paintings and the acoustic properties of the places in

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

191

which they were created (e.g. Bahn 1997, p. 37; Clottes 2003, p. 21; Lewis-Williams 2002, p.  225; Reznikoff and Dauvois 1988; Sieveking 1997, p.  29). In France, ­studies conducted in Palaeolithic caves (e.g. Grotte du Portel, Niaux, Fontanet) have shown—by measuring the intensity and duration of sounds, and changing resonances in relation to the point of emission—a correlation between the location of paintings and the acoustic properties of that peculiar location within the cave. Reznikoff and colleagues (1988) have found that major points of resonance within caves tend to correspond with the location of paintings. Furthermore, the location of some paintings seems to be exclusively explained by the acoustic properties of that specific place. It might be suggested that prehistoric communities developed close associations between the ritual uses of the voice (or other sound sources, such as musical instruments) and cave paintings. In the cave of Les Fraux, work in progress has revealed that the main decorated panel of the cave is located in the most important room within the network (gallery no. 13), in that it offers the best resonance of any part of the cave. Furthermore, some unusual resonances have been observed between two different galleries, not in direct contact with one another, and in two different branches of the cave system. At these peculiar locations, deposits of pottery and metalwork have been found, in some cases associated with engraved motifs and finger-strokes. Of course, more concrete evidence is needed, and this preliminary work is based on qualitative observations, but it seems that, in the cave of Les Fraux, the acoustic properties of the cave might also have been important in the Bronze Age for artistic or ritual expression. It is worth at this stage remembering that we have no idea of the meaning of the cave depictions. The significance of geometric patterns remains unknown. In Spain, some figures (e.g. zig-zags, meanders) which are associated with figures, such as deer, could depict water or thunderstorms (Hameau 2009). The grid patterns and scalariforms are, however, dissimilar from Alpine rock art, such as those discovered in the Val Camonica (Anati 2009). At the cave of Les Fraux, interpretations remain largely speculative and range from utilitarian functions, artistic expression and space appropriation to a counting or measuring system. Alternatively, do the signs represent the property or the presence of a group, or were they made consciously or under altered states of consciousness, induced by sensory deprivation linked to silent isolation and a totally dark environment (Barnatt and Edmonds 2002, p. 125; Lewis-Williams 2002, p. 214)? We have explored different and alternative avenues of research, including modern fetishist practices, such as those performed in Burkina Faso and Mali, which involve heaps of pot sherds and finger-strokes in the ground surface, in order to ward off evil (Pataux 2010): evidence very similar to that observed at Les Fraux. The motifs might also be associated with story-telling, in a similar way to the mnemonic Rongorongo tablets used by people who recite oral traditions on Easter Island. The difficulty arises when the culture and its traditions are of great antiquity. The originality of the iconography of the Les Fraux cave art sets it apart from traditional European protohistorical rock art. There are few parallels for this non-figurative art, and often the archaeological contexts in which it is found are undated. Locally, Rouffignac Cave displays some finger-rendered motifs that could possibly be assigned to the Bronze Age (Barrière 1974, 1975).

192

A. Burens et al.

Some parallels can also be found in Spain, but the activity represented there is much older (e.g. Ojo Guareña Cave, Tune de la Varaime; Gómez-Barrera 1999) or the parallels are less obvious (e.g. Cueva Mayor). Furthermore, Bronze Age rock art, megalithic art and cave art share a wide range of geometric designs with other protohistorical periods, but the indiscriminate comparison of artefacts across time and space leads to inappropriate analogies (Wylie 1985).

9.6  From Domestic Use to Votive Practice Through the course of this chapter, we have discussed the rich archaeological character of the cave of Les Fraux. Some of the activities represented seem to relate to the domestic sphere, whilst others relate to ritual practices. In relation to Neolithic caves in Ireland, Marion Dowd (2008, p. 305) notes that ‘Understanding the use of caves… has to begin with a careful evaluation of the material available and its context’. For William Walker (1995), objects used during rituals are made sacred and must be disposed of in respectful ways. ‘Many of us report finding broken objects in caves that may be the result of ritual activities occurring at the site, so broken votive offerings may be an imperative of ritual practice’ (Moyes 2012, p. 8). If this is the case, how do we set ritual remains apart from domestic objects? Despite renewed interest in this issue, it remains difficult to objectively define the notion of ‘deposit’. Inside the cave of Les Fraux, the galleries include 32 deposits, the majority of which comprise ceramics and, more rarely, metalwork (3), faunal remains (5) or speleothems (1) (not to mention pebbles deliberately deposited on raised areas of calcite) (Figs. 9.18 and 9.19). Unlike other protohistorical caves in France characterised by rich deposits—especially those excavated in Charente-Maritime by José Gomez de Soto (1980)—the cave of Les Fraux has not yielded any human remains, and thus no association with the funerary sphere can be objectively confirmed. Half of the deposits are located in eight galleries which also include cave art, where they are located under or close to decorated panels. But in seven galleries, deposits are located some distance away from wall art panels. Some of the deposits are organised in a highly visible way (at the junction of galleries, on the top of stone blocks, etc.); others are intentionally hidden (under boulders, inside crevices, etc.). Meanwhile, metal artefacts (necklaces, small bronze rings, pins, etc.) and faunal deposits are often spatially distinct from ceramic deposits (which include a large range of storage, cooking and fine ceramic vessels). Deposits sometimes comprise complete vessels, but most often they are broken; most bear traces of use, with some coated with the charred remains of animal fat (ongoing work by Arnaud Mazuy and Martine Regert, UMR7264). The requirement to leave all archaeological material in situ, and working with non-invasive techniques, forced the team to study the ceramic deposits using contact-free measurements only. As a result, we were unable to undertake any petrographic study or technological analysis which might provide information about the origin of the pots.

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

193

Fig. 9.18  A select range of deposits in the cave: (a) gallery no. 6, (b, d) gallery no. 18, (c) gallery no. 12, (e) gallery no. 8 (Photographs: A. Burens)

194

A. Burens et al.

Fig. 9.19  A select range of deposits in the cave: (a, b) gallery no. 13, (c) gallery no. 27, (d, e) gallery no. 18 (Photographs: A. Burens)

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

195

However, typological analysis suggests that some of the vessels belong to the Duffaits Group, dating from the Middle to the beginning of the Late Bronze Age. This corresponds with the outlying Tumulus Culture and the Atlantic Complex (Gomez de Soto 1995, p. 201). Both radiocarbon dating and typological study of the artefacts suggest that votive practices were performed within the cave, together with and disconnected from cave art and domestic activity. In each phase of activity, however, previous deposits and decorated panels were respected and left intact.

9.7  C  onclusion. Residential Activity Versus Cave Art: Towards Overlapping Uses? According to Tolan-Smith and Bonsall (1997, p. 217–218): ‘Most researchers concerned with the human use of caves find it convenient to summarize their data in terms of economic or ritual behaviour… [and]…When people decide to adopt a sedentary mode of existence they usually come out of their caves and the use of these natural shelters is given over to ritual’. Furthermore, ‘the use of caves as theatres for ritual… [is]… usually evidenced by cave art and/or the presence of votive deposits and the use of caves as burial vaults’.

The atypical decorated Bronze Age cave of Les Fraux combines cave art with evidence for domestic and symbolic activities, simultaneously but temporarily within two distinct branches of one large underground network, over nearly four centuries. Even if both residential and ritual uses are rarely evidenced simultaneously in the same cave, the diversity of archaeological remains here suggests a plurality of uses, perhaps shifting in time between the Middle and Late Bronze Age. The interpretation of the site is closely linked to the control of time within the cave and is the reason why we developed a contextual approach, taking into account the economic, cultural, topographic and ecological factors of this environment. Difficulties have included a lack of stratigraphy, and the absence of direct links between deposits, archaeological sediments and decorated wall panels; at this stage, we have only 14 radiocarbon dates from 5 galleries of the cave. Nevertheless, according to archaeological investigation, the cave appears to have witnessed activity from 1450 to 950 BC. Work in progress suggests three periods of activity within this broader date range: a period between the Middle and the Late Bronze Age (from 1450 to 1250 BC), based on dates from hearth nos. 3, 12 and 14, archaeological deposits and ‘torch holes’; a period at the beginning of the Late Bronze Age (around 1250 to 1125 BC), associated with hearth nos. 8 and 18 and one decorated panel from gallery no. 18; and finally, a period between about 1200 and 950 BC, based on dates from hearth nos. 4 and 19. It comes as no surprise that activity within the cave is more complex than previously thought. One must bear in mind that the cave of Les Fraux is unique in Europe. When comparing the site with other French Bronze Age caves, there are several similarities but mainly distinct differences (such as the absence of cave art and the almost systematic presence of funerary activity).

196

A. Burens et al.

Whilst not conclusive, several possibilities emerge. Evidence for interruptions in the archaeological record between the Middle and Late Bronze Age, combined with the low density of features and other remains within the cave over such a long period of use, and the minimal disturbance of older deposits, suggests that activity at the site took the form of numerous short but regular visits over this broader time period. It is worthwhile at this stage to consider that the nature of the various activities may have been very different. The cave may first have been utilised for symbolic activities (cave art, votive deposits, etc.), becoming increasingly used for more domestic activities, with occasional continuation of some symbolic practices, but with respect for the ancient wall art panels. In order to take these interpretations further, more radiocarbon dates from a wider range of galleries is required to complement the chronological framework based on ceramic typology. The difficulty lies in the definition and meaning of ritual or symbolic gesture, as practiced in a cave context. Our contribution to the study of Norman quarries (Cyril Marcigny pers. comm.) will allow us to compare these two sites, recorded according to the same 3D technology, and will improve our understanding of underground networks by combining the results of archaeological investigations (such  as 3D simulation) with the benefits of 2.5D spatial analysis. Acknowledgements  The study of Les Fraux was a planned multi-year archaeological excavation funded by the regional Aquitaine office of the French Ministry of Culture and Communication. The French National Research Council is also supporting this project, under the umbrella project Global Ecology (SEEG). Thanks also to E. Moisan, V. Léglise and B. Cazalet for their contribution as a part of their training at INSA. Our warmest thanks to Edmond (†) and Marcelle Goineaud, owners of the cave of Les Fraux, and their daughter, who allowed us to publish images taken inside the cave. We would also like to express our sincerest gratitude to Delphine Cross.

References Anati, E. (2009). L’art rupestre du Valcamonica: Évolution et signification. Une vision panoramique d’après l’état actuel de la recherche. L’Anthropologie, 113, 930–968. Bahn, P.  G. (1997). Dancing in the dark: Probing the phenomenon of Pleistocene cave art. In C. Bonsall & C. Tolan-Smith (Eds.), The human use of caves. British archaeological reports (international series) 667 (pp. 35–37). Oxford: Archaeopress. Barnatt, J., & Edmonds, M. (2002). Places apart? Caves and monuments in Neolithic and earlier Bronze Age Britain. Cambridge Archaeological Journal, 12(1), 113–129. Barrière, C. (1974). Rouffignac: L’archéologie. Travaux de l’Institut d’Art Préhistorique de Toulouse, 16, 3–210. Barrière, C. (1975). Rouffignac: L’archéologie. Travaux de l’Institut d’Art Préhistorique de Toulouse, 17, 3–83. Beeching, A., & Gasco, J. (1989). Les foyers de la préhistoire récente du sud de la France (descriptions, analyses, et essais d’interprétation). In M. Olive & Y. Taborin (Eds.), Nature et fonction des foyers préhistoriques. Actes du colloque international de Nemours (12–13–14 ai 1987). Mémoire du Musée de Préhistoire d’Île de France (pp. 275–292). Nemours: APRAIF. Brodard, A., Guibert, P., Lévêque, F., Mathé, V., Carozza, L., & Burens, A. (2012). Thermal characterization of ancient hearths from the cave of Les Fraux (Dordogne, France) by thermoluminescence and magnetic susceptibility measurements. Quaternary Geochronology, 10, 353–358.

9  The Bronze Age Decorated Cave of Les Fraux: Ritual Uses of an Atypical French…

197

Brodard, A., Lacanette-Puyo, D., Guibert, P., Lévêque, F., Burens, A., & Carozza, L. (2016). A new process of reconstructing archaeological fires from their impact on sediment: A coupled experimental and numerical approach based on the case study of hearths from the cave of Les Fraux (Dordogne, France). Archaeological and Anthropological Sciences, 8(4), 673–687. Burens, A., Grussenmeyer, P., Carozza, L., Lévêque, F., Guillemin, S., & Mathé, V. (2014). Benefits of an accurate 3D documentation in understanding the status of the Bronze Age heritage cave “Les Fraux” (France). International Journal of Heritage in the Digital Era, 3(1), 179–195. Burens-Carozza, A., Grussenmeyer, P., Guillemin, S., Carozza, L., Bourrillon, R., & Petrognani, S. (2011). Numérisation 3D de la grotte ornée des Fraux, Saint-Martin-de-Fressengeas, Dordogne, France: Approche multiscalaire. Images et modèles 3D en milieux naturels. Collection EDYTEM, 12, 183–189. Carozza, L., Burens, A., Billaud, Y., Ferrulo, O., Bourrillon, R., Petrognani, S., Fritz, C., Tosello, G., Goineaud, E., & Goineaud, M. (2009). L’horizontal et le vertical: L’âge du Bronze de la grotte des Fraux (Saint-Martin-de-Fressengeas, Dordogne). In D. Fabre (Ed.), De Méditerranée et d’ailleurs… mélanges offerts à Jean Guilaine (pp. 159–172). Tolouse: Archives d’Écologie Préhistorique. Carpentier, V., Dujardin, L., Marcigny, C., avec Burens, A., Carozza, L., Grussenmeyer, P., Guillemin, S., Mazet, S., & Vipard, L. (2016). ‘Archéologie du refuge’ ou de ‘l’enfermement-­ volontaire’: La carrière-refuge de la brasserie Saingt à Fleury-sur-Orne (Calvados). Les Nouvelles de l’Archéologie, 143, 59–63. Clottes, J. (2003). Caves as landscapes. In K. Sognnes (Ed.), Rock art in landscapes: Landscapes in rock art (volume 4) (pp. 11–30). Trondheim: Tapir Akademisk. Dowd, M.  A. (2008). The use of caves for funerary and ritual practices in Neolithic Ireland. Antiquity, 82(316), 305–317. Faulkner, C. H. (1988). Painters of the “dark zone”. Archaeology, 41(2), 30–38. Gasco, J. (2003). Contribution pour une proposition de vocabulaire des structures de combustion. In M.-C. Frère-Sautot (Ed.), Le feu domestique et ses structures au Néolithique et aux Âges des métaux. Actes du colloque de Bourg-en-Bresse et Beaune (7–8 Octobre 2000) (pp. 109–112). Montagnac: Éditions Monique Mergoil. Gomez de Soto, J. (Ed.). (1980). Les cultures de l’âge du Bronze dans le basin de la Charente. Périgueux: Fanlac. Gomez de Soto, J. (1995). Le Bronze Moyen en occident: La culture des duffaits et la civilisation des tumulus. L’âge du Bronze en France tome 5. Paris: Picard. Gómez-Barrera, J. A. (1999). Approche de l’étude des gravures rupestres post-paléolithiques de la péninsule Iberique. Préhistoire Ariégeoise, LIV, 265–292. Grussenmeyer, P., Alby, E., Landes, T., Koehl, M., Guillemin, S., Hullo, J.-F., Assali, P., & Smigiel, E. (2012). Recording approach of heritage sites based on merging point clouds from high resolution photogrammetry and terrestrial laser scanning. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XXXIX-B5, 553–558. Hameau, P. (2009). Site, support et signe: Une cohérence de sens. L’expression graphique picturale au Néolithique. L’Anthropologie, 113, 861–881. Le Bihan, J.-P., & Villard, J.-F. (2010). Archéologie d’une île à la pointe de l’Europe: Ouessant. Tome 2: L’habitat de Mez-Notariou des origines à L’âge du Bronze. Saint-Thonan: Centre de Recherche Archéologique du Finistère. Lewis-Williams, D. (2002). The mind in the cave. London: Thames & Hudson. Moyes, H. (2012). Introduction. In H. Moyes (Ed.), Sacred darkness: A global perspective on the ritual uses of caves (pp. 1–14). Boulder: University Press of Colorado. Otte, M., David-Elbiali, M., Eluère, C., & Mohen, J.-P. (2002). La protohistoire. Bruxelles: De Boeck Université. Pataux, A. (2010). Cœur blanc, ventre blanc: Fétiches et féticheurs. Montreuil: Gourcuff Gradenigo. Petrognani, S., Bourrillon, R., Burens, A., & Carozza, L. (2014). À la poursuite du temps court: Les expressions pariétales de l’âge du Bronze de la grotte des Fraux (Dordogne). PALEO, numéro special. Les arts de la Préhistoire: micro-analyses, mises en contextes et conservation. Actes du colloque ‘micro-analyses et datations de l’art préhistorique dans son contexte archéologique’, MADAPCA, Paris, 16–18 Novembre 2011, 163–169.

198

A. Burens et al.

Quiles, A., Valladas, H., Bocherens, H., Delqué-Količ, E., Kaltnecker, E., van der Plichtf, J., Delannoy, J.-J., Feruglio, V., Fritz, C., Monney, J., Philippe, M., Tosello, G., Clottes, J., & Geneste, J.-M. (2016). A high-precision chronological model for the decorated Upper Paleolithic cave of Chauvet-Pont d’Arc, Ardèche, France. Proceedings of the National Academy of Sciences, 113(17), 4670–4675. Reznikoff, I., & Dauvois, M. (1988). La dimension sonore des grottes ornées. Bulletin de la Société Préhistorique Française, 85(8), 238–246. Sieveking, A. (1997). Cave as context in Palaeolithic art. In C. Bonsall & C. Tolan-Smith (Eds.), The human use of caves. British archaeological reports (international series) 667 (pp. 25–34). Oxford: Archaeopress. Tolan-Smith, C., & Bonsall, C. (1997). The human use of caves. In C. Bonsall & C. Tolan-Smith (Eds.), The human use of caves. British archaeological reports (international series) 667 (pp. 217–218). Oxford: Archaeopress. Walker, W. H. (1995). Ceremonial trash? In J. M. Skibo, W. H. Walker, & A. E. Nielsen (Eds.), Expanding archaeology (pp. 67–79). Salt Lake City, UT: University of Utah Press. Wylie, A. (1985). The reaction against analogy. Advances in Archaeological Method and Theory, 8, 63–111.

Chapter 10

From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave, North-East Scotland Lindsey Büster, Ian Armit, Adrian Evans, Tom Sparrow, Rachael Kershaw, and Andrew S. Wilson

10.1  Introduction The Sculptor’s Cave lies on the south coast of the Moray Firth in north-east Scotland (Fig. 10.1). Excavations in the 1920s and 1970s revealed that the site was used for mortuary rituals during the Late Bronze Age (c. 1100–800 BC) and Roman Iron Age (late first–fourth centuries AD), whilst a series of Pictish symbols carved into its entrance walls confirm the cave’s continued importance into the Early Medieval period. In 2014, the archive from the 1979 excavations came to the University of Bradford for post-excavation analysis and publication. Though a 30-year time lag between excavation and publication presented challenges, it also afforded opportunities to bring new questions and new methods of analysis to this old dataset. This chapter charts the use of a range of digital capture technologies (terrestrial laser scanning, structured light scanning, photogrammetry) in the analysis, presentation and management of the Sculptor’s Cave and its Pictish symbols. From the

Electronic supplementary material: The online version of this chapter (https://doi.org/10.1007/9783-319-99022-4_10) contains supplementary material, which is available to authorized users. L. Büster (*) School of History, Classics and Archaeology, University of Edinburgh, Edinburgh, UK e-mail: [email protected] I. Armit School of Archaeology and Ancient History, University of Leicester, Leicester, UK e-mail: [email protected] A. Evans · T. Sparrow · R. Kershaw · A. S. Wilson School of Archaeological and Forensic Sciences, University of Bradford, Bradford, UK e-mail: [email protected]; [email protected]; [email protected]; [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_10

199

200

L. Büster et al.

Fig. 10.1  Location map of the Sculptor’s Cave

elucidation of additional carved details and the monitoring of surface degradation of the motifs, to the dissemination of this difficult-to-access site to the wider public via online platforms, we demonstrate the central role that digital capture technologies can play in the study of these enigmatic spaces.

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

201

10.2  A Dynamic Space The Sculptor’s Cave has never been an easy place to reach. Though relative sea level appears to have altered little in the immediate environs of the cave, the terrestrial environment has changed markedly since prehistory. Indeed, the stretch of coast on which the cave is located would have been substantially cut off from the mainland by a large sea loch—the Loch of Spynie—and its surrounding marshlands, making it a virtual island for much of the prehistoric and early historic period (Stratigos in press). Though the sea loch was largely drained in the nineteenth century, as part of a programme of agricultural improvement, access to the cave by land remains extremely difficult and requires negotiating high sandstone cliffs (Fig. 10.2a) or walking for around a kilometre along a boulder-strewn beach (Armit and Büster in press). The bay in which the Sculptor’s Cave is located is itself cut off from both of these access routes for several hours either side of high tide. Access in prehistory may also have been by sea, but this too would have involved a tortuous approach over dangerous rocks that could only be negotiated with small, light boats. The remoteness and inaccessibility of the cave is reflected in the words of its first excavator, Sylvia Benton (1931, p. 178), who noted that ‘the sun never touches it’. Whilst not strictly true (sunlight does enter the cave at certain times of the day), her words remind us of what marginal places caves were—both physically and cosmologically—and their enduring perception as sites outside the realm of the everyday world. The Sculptor’s Cave derives its name from a group of enigmatic Pictish symbols (probably dating to around the sixth century AD; Büster and Armit 2018) carved on the walls of its twin entrance passages (Fig. 10.2b, c). Such carvings are uncommon in caves and attest to the enduring significance of the Sculptor’s Cave. Nonetheless, it is the later prehistoric activity for which the site is best known. Excavations by Sylvia Benton in 1928–30 (Benton 1931), and Ian and Alexandra Shepherd in 1979, unearthed an assemblage of c. 1750 disarticulated fragments of human bone. The 1979 excavation campaign was never fully published, being disseminated only through a series of interim reports (Shepherd 1995, 2007; Shepherd and Shepherd 1979), but the arrival of the archive at the University of Bradford in 2014 allowed for its full reanalysis, integration and publication nearly 30 years on. Though the vast majority of the assemblage (which came from Benton’s excavations) is now lost, surviving handwritten bone lists and more recent stratigraphic and radiocarbon analysis indicate that the assemblage is the product of a long association of the cave with complex, changing mortuary practices. At least 33 individuals are represented, including 9 adults, 13 young individuals, 9 children and 2 infants (Büster et al. in press), but the actual figure is likely to be significantly higher. Broadly contemporary with the deposition of human remains in the Late Bronze Age (c. eleventh–ninth centuries BC), alignments of stakeholes suggest the presence either of a hurdle barrier blocking off the interior of the cave or a platform potentially associated with the laying out and/or display of the dead. Valuable items

202

L. Büster et al.

Fig. 10.2  Photographs showing (a) the Sculptor’s Cave, (b) the Sculptor’s Cave twin entrance passages as viewed from the cave exterior, (c) the entrance passages as viewed from the cave interior

of personal adornment, including gold-covered hair rings and copper alloy bracelets, accompanied the human remains, or may have been votively deposited as part of the funerary rites. The fact that they survived for excavation in the present day suggests a likely taboo surrounding the removal of objects that now belonged to the world of the dead.

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

203

Despite this potential taboo, a large assemblage of Late Bronze Age pottery and the presence of hearths suggest that people spent considerable periods of time in the cave: cooking, eating and drinking in close proximity to the dead. Each trip would have required significant effort, and may have necessitated spiritual as well as practical preparations. The logistics would have been especially demanding on those occasions when additional human remains had to be brought to the cave; they could have been lowered down the cliff in some way but would have been difficult to manoeuvre (as experienced first-hand during fieldwork when lowering scanning equipment weighing around half that of the average human body; see below). As well as bones and objects, the complex funerary rituals carried out at the Sculptor’s Cave would also undoubtedly have included less tangible elements such as song, dance, stories and music, which do not survive as material remains. The experiential qualities of the cave would have changed over time. As hurdle barriers were built, repaired and degraded/were dismantled, the interior would have been illuminated (at least at certain times of the day) or daylight would have been blocked out entirely, altering views from inside and out. The accumulation of deposits in the entrance passages, where human activity appears to have been concentrated, would also have gradually altered people’s experience, not least in terms of the ingress of natural light. Modifications, such as the addition of fittings and furnishings, together with the presence or absence of an internal pool of stagnant water, would likewise have altered the acoustic properties of the cave (Rupert Till pers. comm.). In addition to the laying out of bodies as part of normative funeral rites during the Roman Iron Age (late first–fourth centuries AD), a group of at least six individuals (four adults and two children or adolescents) were decapitated inside the cave (most probably around the third century AD), as evidenced by the presence of cut-marked cervical vertebrae (Armit et al. 2011, p. 258, 272–274). Bayesian modelling of the AMS dates obtained on these bones suggest that the decapitations may represent a single event. The angle of the cut marks, with blows struck from behind whilst the chin was flexed onto the chest, indiactes a controlled process of execution, potentially involving a group of assailants. The presence of cut-marked vertebrae which would have remained with the torso after decapitation, as well as those that would have remained with the detached head, suggests that these individuals were beheaded within the cave itself, rather than, for example, arriving there as trophy heads. Although this violent episode is of an entirely different character to the apparently reverential funerary rites also taking place in the cave at this time, it is likely that the decapitations themselves formed part of some ritualised practice, no doubt influenced by long-standing veneration of the site. Clearly, the power and significance of the Sculptor’s Cave outweighed the practical difficulties of coercing a group of living (and presumably unwilling) victims down the cliffs and across the rocks or in small boats along a rocky coast. Later, during the fourth century AD, the deposition of a large assemblage of Roman Iron Age coins marked the final archaeologically visible act within the cave interior (Moorhead in press). Representing one of the latest, and most northerly, of these types of deposit (cf. Hunter 2007, p. 35, Fig. 15), it again marks out the Sculptor’s Cave as a place of enduring significance. It was sometime later, most likely in the sixth century AD (Büster and Armit 2018), that a series of Pictish symbols—normally found on free-standing stones—was

204

L. Büster et al.

carved on the walls of the entrance passages. Though various hypotheses have been proposed, such as the possibility that they represent personal names (e.g. Samson 1992; Forsyth 1997; Lee et  al. 2010), the meaning of Pictish symbols remains unknown. The fact that they are confined to the entrance passages, and are absent from the interior, suggests that they may represent some kind of ritualised closure of the cave, especially since there seems to be no sign of contemporary human activity inside (Büster and Armit 2018). Indeed, the context of reuse of many Pictish symbol stones belies a preoccupation with thresholds and subterranean structures, suggesting that they may have been perceived as having liminal and transformative qualities (cf. Gondek 2015). With the exception of one crescent symbol at Clashach Cove approximately 1.5 km to the west (Fraser 2008, p. 106), the Sculptor’s Cave is the only known cave along this stretch of coastline to contain carvings of this period, and is the only cave to have attracted substantial quantities of later graffiti. Pictish carvings are extremely rare in caves, though not entirely absent (Büster and Armit 2018; see also Hambly et al. Chap. 11, this volume). Moreover, they tend not to conform to the common corpus found on free-standing symbol stones, leading to speculation that they might be relatively early examples (cf. Alcock 1996; though see Forsyth 1997, p.  93 regarding likely differences between formal versus informal renderings). Activity post-dating the Pictish period appears to have been limited to sporadic visits and is evidenced only by later carvings; these include a cross attributed to the  later Medieval period on stylistic grounds (Armit and Büster in press) and a seventeenth-century curse. The walls of the cave interior are also covered in graffiti dating from the nineteenth century to the present, whilst a make-shift altar (at the far rear of the cave) is adorned with all manner of objects from seashells, driftwood and seabird carcases to bundles of twigs and plastic combs.

10.3  Digital Documentation The long and complex biography of the Sculptor’s Cave dictates that any detailed understanding of the changing rites and rituals which took place there (and the landscape context of these practices) must take account of subtleties that are not always easy to present in conventional recording formats. The experiential nature of the site and its location—sounds, sights, smells, etc.—was presumably integral to ritual activities (Armit and Büster in press) and is thus as important in the telling of its stories as any archaeological plan, section, artefact or bone. Though the 1928–1930 excavations at the Sculptor’s Cave were published in a timely fashion (Benton 1931), they are, necessarily, a product of their time. The 30-year time lapse between excavation and publication of the 1979 archive is however providing an opportunity to augment this existing data (and that from Benton’s excavations) with innovative recording methods, primarily digital approaches developed with Visualising Heritage and Fragmented Heritage at the University of Bradford.

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

205

10.3.1  Augmenting an Archive Recording the Cave in 3D With the passing of time and with a multitude of different activities taking place, deposits began to fill the Sculptor’s Cave; particularly its entrance passages, which became heavily clogged with material. Not only, then, does the cave appear quite different to us today than it would have in prehistory, reconstructing even a basic working plan of the site is fraught with difficulties. For example, overlaying the Shepherd’s 1979 plans with that from the 1920s has been challenging, since the extent of the cave floor is different in each (and alters even within the confines of the 1979 campaign, as deposits were excavated and different sections of the undulating walls exposed). A basic problem, then, was deciding which plan to use as the basic footprint for the Sculptor’s Cave; a footprint over which to lay other data, such as the extent of archaeological deposits, stakehole alignments, and artefact and human bone distribution plots. With such a large and irregular space, conventional recording methods (e.g. with only a total station and prism) would be inadequate. Even electronic distance measurers (‘distos’), such as those used for recording  buildings, rely on straight and regular lines, not the contours of a natural space. Terrestrial laser scanning (Fig. 10.3), however, creates a fully three-dimensional model of the entire cave interior that can be manipulated and ‘sliced’ in any plane and at any angle, enabling the creation of bespoke plans for post-excavation analysis (Figs. 10.4 and 10.5).

Fig. 10.3  Terrestrial laser scanner in operation at the Sculptor’s Cave (Photograph: L. Demay)

206

L. Büster et al.

Fig. 10.4  Comparison of plans taken from (a) Benton’s 1928–30 archive, (b) from the terrestrial laser scan data captured in 2014

Terrestrial laser scanning of the Sculptor’s Cave was undertaken using a FARO (Lake Mary, FL) Focus3D S 120 laser scanner running FARO Scene software, on a 16 min 15 s scan setting (parameters of half resolution and 3 × quality). The scanner measured 24 cm × 20 cm × 10 cm and weighed 5 kg; as such, it fit neatly into a hard case which could be carried ‘rucksack style’ down the cliff and across the beach to the Sculptor’s Cave. Each laser scan comprised an average of 174.8 megapoints, with an average point distance at 10  m of 3.068  mm and an average file size of 568 MB. In total, 28 scans captured the entire cave interior, in around 8 h, over a 2 day period. Scan positions were located using a Trimble S6 robotic total station. Off-site, the individual scans were combined initially into a 3D digital model in Trimble (Sunnyvale, CA) Realworks, but for the creation of more sophisticated models, point clouds were converted into meshes in Autodesk (San Rafael, CA) ReCap 360 and then subsequently cleaned (removal of background ‘noise’) using

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

207

Fig. 10.5  Sections through (a) East Passage, (b) West Passage, generated from terrestrial laser scan models of the Sculptor’s Cave

open-source Cloud Compare software. This process took approximately 48 h: 16 for aligning and combining the scans, 8 h for converting the point clouds into a mesh and a further 24 h of data cleaning. The resultant model is nearly 2  GB in size and can be opened in open-source Cloud Compare or MeshLab software, both of which are freely available online. Finally, walk-through animations were created using a combination of FEI (Hillsboro, OR) Avizo and Adobe (San Jose, CA) After Effects software; these were outputted as AVI files and can be played in open-source multimedia VLC software (Fig. 10.6). Capturing Features of Interest High-resolution structured light scanning, photogrammetry and reflectance transformation imaging (RTI) were used for those portions of the cave walls containing Pictish symbols and significant carvings of later periods. Three pairs of Pictish

208

L. Büster et al.

Fig. 10.6  Still images taken from the walk-through animation generated from the terrestrial scan models, showing (a) the cave exterior, (b) the twin entrance passages from the cave interior, (c) the modern ‘altar’ at the rear of the cave

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

209

Fig. 10.7  Structured light scanning of the salmon and V-rod in the East Passage

carvings were subject to structured light scanning in situ: a salmon and V-rod on the east wall of the East Passage (perhaps the best-executed of the corpus of Pictish symbols in the cave) (Fig. 10.7); a flower and triple oval on the canopy between the two entrance passages which, due to their height and angle, are difficult to see in detail with the naked eye; and a double rectangle and mirror case on the west wall of the West Passage (Figs. 10.8, 10.9 and 10.10). Scanning was undertaken using a Mechinnovation Ltd (Leamington Spa) Mechscan: a specially commissioned macro structured light scanner produced for the Fragmented Heritage project at the University of Bradford. This scanner has a field of view of 30–110 mm and approximate point to point distance of 0.01–0.05 mm, and captures up to 2.6 million points per scan. The carvings were recorded using 140, 39 and 98 scans, respectively, taking 4–12 h to record each group. In order to acquire the highest possible resolution, scans were captured in greyscale and were later overlain with colour images produced using photogrammetry (see below). Scan data were captured and processed using LMI Technologies (Vancouver) FlexScan 3D software. The scanner, in its protective case, measured 79.5 cm × 51.8 cm × 39.4 cm and weighed around 30 kg, so coaxing it down the cliff-face with ropes, and manoeuvering it across the boulder-strewn beach and back up the cliffs on the backs of team members, was difficult and arduous. Furthermore, the scanner produces optimum results only in the darkest of conditions, such as might be encountered in a conventional karst cave with a narrow entrance. In a sea cave like the Sculptor’s Cave, where the panels of interest are located in the entrance passages, levels of natural light were too high to enable the scanner to function effectively, so a screen was rigged-up across the cave entrance to artificially darken the space.

210

L. Büster et al.

Fig. 10.8  Salmon and V-rod symbols in the Sculptor’s Cave (a) photograph, (b) decimated structured light scan data processed using FlexScan 3D software, (c) synthetic RTI model

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

211

Fig. 10.9  Flower and triple oval symbols in the Sculptor’s Cave (a) photograph, (b) decimated structured light scan data processed using FlexScan 3D software, (c) synthetic RTI model

Colour photographs of the carvings were taken to complement the greyscale scan data (Figs. 10.8, 10.9 and 10.10). The photographs were built into a 3D model using structure from motion photogrammetry with Agisoft (St Petersburg) Photoscan software. The models produced by both digital capture methods were studied and enhanced using synthetic reflectance transformation imaging (RTI), which uses light sources at various angles to illuminate details not visible under static lighting conditions (Earl et al. 2011). RTI can be undertaken in the field but for the Sculptor’s Cave was created digitally using a combination of photographs and structured light scan data. Synthetic RTI of the structured light scans was achieved by creating 3D models from the orthoimages and artificially lighting them using open-source Blender software (as per English Heritage guidelines; Duffy et al. 2013). Images were then captured using one light source at a time from the lighting array. Finally, RTI files were created using open-source RTI Builder software and open-source PTM (Polynomial Texture Mapping) Builder algorithms (ibid.), which allowed for interactive light source control and export of the images using RTI Viewer software. Some new details of the carvings were observed using combined structured light scan data and virtual RTI. The scan of the salmon revealed, for example, additional fins on the upper and lower sides of its body (Fig. 10.8). Furthermore, internal stratigraphy within the salmon and V-rod group was evidenced by an external fin on the lower body of the salmon truncating the adjacent crescent. This provides a level of time-depth to the carvings which was not previously apparent. Though no specific information could be identified regarding the types of tools used to carve the symbols, it was possible on the scan of the double rectangle and mirror case to see individual ‘pick’ marks (Fig. 10.10); this could indicate a different mode of carving

212

L. Büster et al.

Fig. 10.10  Double rectangle and mirror case symbols in the Sculptor’s Cave (a) photograph, (b) decimated structured light scan data processed using FlexScan 3D software, (c) synthetic RTI model

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

213

to the other symbols and suggests that not all symbols were necessarily carved contemporaneously (see Clarke 2007 for discussion on the possible cumulative nature of Pictish symbols on Class I stones). Alternatively, it could indicate that these two symbols were singled out for recarving sometime after their initial creation; the restating of a pertinent social message perhaps (cf. Clarke 2007, p. 36).

10.3.2  Managing Heritage Although the Sculptor’s Cave is a Scheduled Monument, frequent site visits for heritage management purposes are not always possible, particularly given the inaccessible nature of the site. The position of the Pictish symbols in the cave’s entrance passages makes them particularly vulnerable to weathering and erosion (Fig. 10.2b); the surface of the coarse-grained sandstone is particularly susceptible to delaminating in sheets large enough to destroy entire symbols in a single event. Furthermore, a large rockfall at the entrance to an adjacent cave (Covesea Cave 2), located just 100 m west of the Sculptor’s Cave, attests to the inherently unstable nature of these sites (Fig. 10.11). This problem is exacerbated by the cave’s location under a flight path for the nearby Royal Air Force base at Lossiemouth; the vibrations of passing aircraft could be felt in nearby Covesea Cave 2 during recent fieldwork. The Sculptor’s Cave has also been subject to a range of more recent graffiti, some of which has already encroached upon the Pictish carvings (see below; Fig.  10.7). Despite its inaccessibility, the site continues to attract visitors, and (irrespective of its legal protection) it is only matter of time before further carvings are damaged or lost. Indeed, the terrestrial scan model has already been used by the national heritage body, Historic Environment Scotland, to assess the extent of damage to the Sculptor's Cave caused by recent illicit excavations. Digital documentation of the Sculptor’s Cave has captured the site as it currently exists and has thus preserved it (in virtual form at least) for future generations.

Fig. 10.11  Large rockfall at the mouth of Covesea Cave 2, 100 m west of the Sculptor’s Cave

214

L. Büster et al.

Fig. 10.12  Scans of a later Medieval cross from the East Passage of the Sculptor’s Cave (a) structured light scan of 1979 fibreglass cast with colour, (b) structured light scan of 1979 fibreglass cast with colour ‘stripped’, (c) lower-resolution terrestrial laser scan image of the same carving in situ

The 3D scans will also help with the assessment of heritage management strategies, through comparison, for example, between fibreglass casts taken of many of the symbols—both Pictish and later—during the 1979 excavations and the current state of preservation of the carvings in situ, as represented by the structured light scans taken on site in 2014. The casts, held at the National Museum of Scotland, were themselves scanned at the University of Bradford using a structured light scanner with a field of view of 15–50 cm and a resolution of 100 microns (and supporting FlexScan 3D software) (Fig.  10.12), built in-house for Visualising Heritage. As such, it is now possible to compare the fibreglass casts of 1979 with the carvings as they survive today (Fig. 10.13). Another benefit of scanning the casts is that they are

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

215

Fig. 10.13  Structured light scans of the salmon and V-rod symbol (a) 1979 fibreglass cast with colour, (b) 1979 fibreglass cast with colour ‘stripped’, (c) scan image of carving in situ as it survives today

216

L. Büster et al.

artefacts of their time; they are large, brittle and unwieldy—the largest measuring 155 cm × 84 cm and taking 119 scans to capture—and have themselves been subject to a certain degree of physical and chemical deterioration during their 35 years in storage. This last point has implications beyond the Sculptor’s Cave, since carvings at other sites have been similarly recorded. Some of the most iconic of these are perhaps the Late Neolithic and Early Bronze Age cup-and-ring marks on the summit of Traprain Law, East Lothian (Armit and McCartney 2005) which, having been destroyed during quarrying of the site in the 1930s, now survive only as photographs and casts. Recording the Sculptor’s Cave symbols by structured light scanning has also allowed for identification and assessment of the impact of past conservation interventions. For example, infilling by conservators in the 1990s of certain letters in the name ‘Xavier’, which had been carved across the lower portion of the Pictish salmon on the east wall of the East Passage, was visible on the scan images. With this in mind, a further advantage of digital documentation is that it does not physically disturb the carvings. This is in contrast, for example, to the fibreglass casts of 1979; those symbols subject to casting are now characterised by a greener hue than the surrounding cave wall (Fig. 10.14), suggesting that the materials used have made the rock more susceptible to moss and lichen growth.

10.3.3  Experiencing Caves One of the biggest problems associated with the study, management and presentation of the Sculptor’s Cave is its inaccessibility: a feature that was no doubt fundamental to its use for prehistoric mortuary rituals, in a liminal place halfway between land and sea. The cave is internationally significant for its apparent use as a site for primary excarnation; a funerary rite which has generally been postulated for later prehistoric Britain on the absence of evidence for other visible normative rites (Carr and Knüsel 1997). It also plays a significant role in understanding the later prehistoric landscape of the region, together with the broadly contemporary sites of Birnie and Clarkly Hill (Hunter 2009, 2012). In addition to disseminating the archaeology of the Sculptor’s Cave to the academic community, digital documentation of this inaccessible site allows its integration into local narratives and, fundamentally, tourism. The local museum in Elgin is staffed by volunteers and is relatively small; digital capture technologies offer the prospect of showcasing the archaeology of the Sculptor’s Cave through space-­ saving media such as moveable and temporary projections, screens, laptops and VR headsets. Immersive media such as these also engage a wide range of audiences, particularly children, thus increasing the reach and inclusivity of displays. The high-­ resolution scans also lend themselves to 3D printing as a method for off-site (yet still tactile) interaction with the Pictish carvings (Fig. 10.15). With this is in mind, the team has created a walk-through animation of the cave, blending terrestrial laser scan data of the entire cave interior with high-resolution

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

217

Fig. 10.14  Seventeenth-century curse (a) in situ, immediately after the cast was taken in 1979, (b) scan of the fibreglass cast taken in 1979, (c) the curse in situ today, showing enhanced lichen growth in relation to the surrounding cave wall

218

L. Büster et al.

Fig. 10.15  3D-printed replica of the salmon and V-rod symbol (at 3:4 scale, with z-values amplified to 1.5 × their original), produced from digital scan data and augmented with an applied textured surface

structured light scan data of both the carvings scanned in situ and the fibreglass casts (Figs. 10.6, 10.8, 10.9, 10.10 and 10.12). As such, three different types of scan data (at different resolutions) have been combined, synthesising data captured both in situ and off-site into one integrated model and allowing the viewer to experience the Sculptor’s Cave simultaneously at a number of scales. The animation allows the viewer to ‘walk’ through the entire cave interior and to ‘zoom-in’ to particular carvings in greater detail and at higher resolution. From a single model, different animations can be created following virtually any route around the cave, with pauses at different points of interest, and future applications could include bespoke, user-led simulations. The walk-through animation also has the potential for augmentation with further sets of experiential data, such as archaeoacoustic models and recordings.

10.4  Conclusion The Sculptor’s Cave is a dynamic but inaccessible site with a long and complex biography. Digital capture technologies have allowed for augmentation of the existing excavation archives through the creation of 3D models, allowing for the integration of multiple datasets at different scales. These range from site-wide animations to high-resolution structured light scans of individual carvings, and serve not only to enhance archaeological interpretations of this enigmatic site but also to convey the less tangible, experiential qualities of the cave to a diverse range of audiences.

10  From Macro to Micro: Multi-Scalar Digital Approaches at the Sculptor’s Cave…

219

Accurate three-dimensional recording also aids in the monitoring and protection of this vulnerable site and allows for its digital preservation for future generations. The potential applications of digital capture technologies in cave archaeology are wide-­ranging and look increasingly to lie at the heart of how we understand, protect and promote these enigmatic places. Acknowledgements  The authors would like to thank Historic Environment Scotland for funding the scanning work and our collaborators Visualising Heritage and Fragmented Heritage at the University of Bradford, funded by HEIF (via the University of Bradford) and the Arts and Humanities Research Council (AH/L00688X/1), respectively.

References Alcock, L. (1996). Ur-symbols in the pictograph-system of the Picts. Pictish Arts Society Journal, 9, 2–5. Armit, I., & Büster, L. (in press). Darkness visible: The Sculptor’s Cave, Covesea, from the Bronze Age to the Picts. Edinburgh: Society of Antiquaries of Scotland. Armit, I., & McCartney, M. (2005). The new rock art discoveries at Traprain Law. PAST: Newsletter of the Prehistoric Society, 49, 4–5. Armit, I., Schulting, R., Knüsel, C. J., & Shepherd, I. A. G. (2011). Death, decapitation and display? The Bronze and Iron Age human remains from the Sculptor’s Cave, Covesea, north-east Scotland. Proceedings of the Prehistoric Society, 77, 251–278. Benton, S. (1931). The excavation of the Sculptor’s Cave, Covesea, Morayshire. Proceedings of the Society of Antiquaries of Scotland, 65, 177–216. Büster, L., & Armit, I. (2018). Signs from the Pictish underground: Early Medieval cave ritual in Scotland. In K. A. Bersgvik & M. Dowd (Eds.), Caves & ritual in Medieval Europe, AD 500–1500 (pp. 85–96). Oxford: Oxbow Books. Büster, L., Armit, I., Schulting, R., & Knüsel, C. (in press). Human remains. In: I.  Armit & L. Büster, Darkness visible: The Sculptor’s Cave, Covesea, from the Bronze Age to the Picts. Edinburgh: Society of Antiquaries of Scotland. Carr, G., & Knüsel, C. (1997). The ritual framework of excarnation by exposure as the mortuary practice of the early and middle Iron Ages of central southern Britain. In A.  Gwilt & C. Haselgrove (Eds.), Reconstructing Iron Age societies (pp. 167–173). Oxford: Oxbow Books. Clarke, D. V. (2007). Reading the multiple lives of Pictish symbol stones. Medieval Archaeology, 51(1), 19–39. Duffy, S., Bryan, P., Earl, G., Beale, G., Pagi, H., & Kotouala, E. (2013). Multi-light imaging for heritage applications. Swindon: English Heritage. Earl, G., Basford, P.  J., Bischoff, A.  S., Bowman, A., Crowther, C., Dahl, J., Hodgson, M., Martinez, K., Isaksen, L., Pagi, H., Piquette, K. E., & Kotoula, E. (2011). Reflectance transformation imaging systems for ancient documentary artefacts. In J. P. Bowen, S. Dunn, & K. Ng (Eds.), EVA London 2011: Electronic visualisation and the arts (pp. 147–154). London: British Computer Society. Forsyth, K. (1997). Some thoughts on Pictish symbols as a formal writing system. In I. Henderson & D. Henry (Eds.), The worm, the germ and the thorn: Pictish and related studies presented to Isabel Henderson (pp. 85–98). Brechin: Pinkfoot Press. Fraser, I. (Ed.). (2008). The Pictish symbol stones of Scotland. Edinburgh: Royal Commission on the Ancient and Historical Monuments of Scotland. Gondek, M. (2015). Building blocks: Structural contexts and carved stones in Early Medieval northern Britain. In H. Williams, J. Kirkton, & M. Gondek (Eds.), Early Medieval stone monuments: Materiality, biography, landscape (pp. 87–113). Woodbridge: Boydell Press.

220

L. Büster et al.

Hunter, F. (2007). Beyond the edge of empire: Caledonians, Picts and Romans. Rosemarkie: Groam House Museum. Hunter, F. (2009). Excavations at Birnie, Moray, 2008. Unpublished Data Structure Report, National Museum of Scotland. Hunter, F. (2012). Excavations at Clarkly Hill, Roseisle, Moray 2011: Interim report. Unpublished Data Structure Report, National Museum of Scotland. Lee, R., Jonathan, P., & Ziman, P. (2010). Pictish symbols revealed as a written language through application of Shannon entropy. Proceedings of the Royal Society A: Mathematical, Physical & Engineering Sciences, 466(2121), 2545–2560. Moorhead, S. (in press). Roman coins. In: I. Armit & L. Büster, Darkness visible: The Sculptor’s Cave, Covesea, from the Bronze Age to the Picts. Edinburgh: Society of Antiquaries of Scotland. Samson, R. (1992). The reinterpretation of the Pictish symbols. Journal of the British Archaeological Association, 145(1), 29–65. Shepherd, I. A. G. (1995). The Sculptor’s Cave, Covesea, Moray: From Bronze Age ossuary to Pictish shrine? Proceedings of the Society of Antiquaries of Scotland, 125, 1194–1195. Shepherd, I. A. G. (2007). “An awesome place”: The Late Bronze Age use of the Sculptor’s Cave, Covesea, Moray. In C. Burgess, P. Topping, & F. Lynch (Eds.), Beyond Stonehenge: Essays on the Bronze Age in honour of Colin Burgess (pp. 194–203). Oxford: Oxbow Books. Shepherd, I. A. G., & Shepherd, A. M. (1979). Sculptor’s Cave, Covesea (Drainie): Occupational/ ritual site. Discovery and Excavation in Scotland, 14, 88. Stratigos, M. (in press). Landscape reconstruction. In: I. Armit & L. Büster, Darkness visible: The Sculptor’s Cave, Covesea, from the Bronze Age to the Picts. Edinburgh: Society of Antiquaries of Scotland.

Chapter 11

How a Community Digital Heritage Project Has Helped to Imagine the Circumstances of Pictish Symbols in the Wemyss Caves, Scotland Joanna Hambly, Marcus Abbott, and Mike Arrowsmith

11.1  Introduction Wemyss—from the Gaelic word for cave, uaimh—is a coastal area of Fife, around 30 miles north of Edinburgh in Scotland (Fig. 11.1). On a visit to East Wemyss in 1865, the medical doctor and antiquarian James Young Simpson first documented and recognised around 50 pictograms incised upon the walls of five of the numerous former sea caves along this coastline as Pictish in origin. They closely resemble abstract symbols and animal representations much more commonly portrayed on Pictish standing stones: the defining monument of a Late Iron Age and Early Historic people whose territories eventually stretched across northern and eastern Scotland between the sixth and ninth centuries AD. But what will be left of the Wemyss Caves and their inscribed histories in 50 years’ time? This is the question that concerns the present-day community of East Wemyss. The Save the Wemyss Ancient Caves Society (SWACS) was originally formed in response to an episode of vandalism when a car set alight in one of  the caves destroyed two Pictish beast symbols and a historic swan carving. Since 1987, SWACS has maintained a twin campaign of local education and wider awareness raising of the value and plight of the caves.

J. Hambly (*) School of History, University of St Andrews, Fife, UK e-mail: [email protected] M. Abbott York Archaeological Trust, York, UK e-mail: [email protected] M. Arrowsmith Centre for Archaeology, Technology and Cultural Heritage, University of St Andrews, Fife, UK e-mail: [email protected] © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_11

221

222

Fig. 11.1  Location map of the main group of the Wemyss Caves

J. Hambly et al.

11  How a Community Digital Heritage Project Has Helped to Imagine…

223

The Wemyss Caves pose the ultimate problem in heritage conservation and management. Located in a former coal mining area with attendant legacies of environmental degradation and economic deprivation, the setting of the caves and carvings has been damaged by coastal erosion, decaying infrastructure, general neglect, antisocial use of the caves and vandalism. The caves themselves are in various states of structural instability, access to some of them is difficult, and—tucked away out of sight between the populous villages of East Wemyss and Buckhaven— it is all but impossible to keep an eye on them. The understanding, condition and setting of a monument has a profound effect on its perceived value. There has been a negative but persistent perception that the integrity of Pictish motifs in the Wemyss Caves is compromised and that the caves that house them are unsafe, which has been a major contributing factor in curtailing sustained effort to conserve and protect them. We also argue that the damaged condition of the carvings and the erosion of their setting have influenced later twentieth and twenty-first century judgements as to the quality of design of the symbols and have detracted from their potential contribution to research and debate. It is extraordinary that a collection of nearly 50 Pictish symbols, of which 26 survive in their original context—however rough and damaged they may be—has not contributed more to discussion about the purpose and meaning of Pictish symbols. Although for some people today perception of the condition of the caves is negative, it wasn’t always so. James Young Simpson (1867, p. 166) describes the West Doo Cave as ‘one of the most magnificent … In some lights the cryptograms on its high walls and dome-like ceiling show masses of beautiful and changing colours’. Christian MacLagan (1876, p.  108) writes, in the Glassworks Cave, about ‘sunbeams, lighting up a singular scene of pastoral beauty’. For John Patrick (1905a, p.  82), the Court Cave is ‘truly grand and striking’ in a natural setting of ‘tidal-­ washed Skerries edged with yellow sand’. Even excusing the romantic writing style of the period, early researchers up to and including J.  Romilly Allen and Joseph Anderson (1903) made no distinction between the Pictish motifs in the Wemyss Caves and the higher art of similar designs inscribed on the standing stones. They were interested in the meaning of the symbols in their cave setting. The caves also have a history extending beyond that of the Pictish and early Christian carvings documented by visiting antiquarians. Ghosts, malevolent pipers and Covenanters all feature in local stories about the caves. Tradition associates the Court Cave with a visit by James IV who ‘cavorted with gypsies’ there. The same cave is still remembered in the village as the place for the gambling game of Toss, played by miners in the mouth of the cave into the 1960s and watched over by children posted as lookouts in case of attention from the local police. The Well Cave was the focus of Handsel Monday New Year celebrations for young people in East Wemyss, which continued up to the late nineteenth century (Deas 1948). The challenge of enabling the current generation to see beyond the condition of the caves and understand their significance and history has provided the reason and purpose of the Wemyss Caves digital documentation and visualisation project. The close involvement of the local community has opened up local archives of information and memory relating to the caves and has directed the focus of the project to encompass the wider history of the caves. In collaboration with the East Wemyss

224

J. Hambly et al.

community, coordinated through SWACS, we have applied digital technologies to document and communicate the rich history and heritage of the Wemyss Caves. In doing so, we aim to restore the reputation of the Pictish heritage of the Wemyss Caves as worthy of protection and study, and play our part in keeping the wider history, traditions and stories alive for the current generation. In this chapter, we set out the heritage of the Wemyss Caves, describe the methods of data capture and creation of the digital resource and, in the light of our improved understanding of the monument, consider possible circumstances of why Pictish symbols are found in the Wemyss Caves.

11.2  The Inscribed Heritage of the Wemyss Caves The incised symbols in the Wemyss Caves share the uniformity of repertoire found on standing stones, and occasional survivals of portable stones and other personal artefacts across Pictland. These were comprehensively described and catalogued in 1903 by J. Romilly Allen and Joseph Anderson in their book The Early Christian Monuments of Scotland. Since then a number of new symbol stones and artefacts have come to light, but they generally fall within the repertoire as described by Allen and Anderson. Allen describes around 40 different types of incised symbol, of which 16 are found in the Wemyss Caves. These include animal representations of a horse or lion; a fish; what is thought to be a wolf (Allen described it as a man driving a beast); the stylised Pictish beast; and abstract symbols of the double disc, crescent with V-rod, flower, arch and rectangle (Fig. 11.2). The use of identical symbols in caves, on standing stones and on portable artefacts from the Forth to Shetland reminds us of the incredible conformity of the symbols—the meaning of which must have been commonly understood—but also invites us to think about what this meaning imbued to such diverse contexts. As an example, one of the Wemyss motifs, a double disc and Z-rod paired with an animal

Fig. 11.2 (a) Abstract double disc symbol in the Sliding Cave, (b) representation of a possible horse or lion in Jonathan’s Cave (see Fig. 11.7 for historic illustrations of horse/lion) (© SCAPE)

11  How a Community Digital Heritage Project Has Helped to Imagine…

225

Fig. 11.3  Double disc and Z-rod with animal head in collapsed Doo Cave (Photographed by John Patrick around 1900)

head (Fig. 11.3), has almost exact parallels to a symbol on a silver plaque recovered from a nearby cairn at Norrie’s Law in 1819 (Allen and Anderson 1903, Fig. 387); on a fragment of a standing stone incorporated into the foundations of the old church at Rhynie, Aberdeenshire (and recovered when the church was demolished in 1878) (ibid., Fig. 198); and on a bronze crescent-shaped plaque (now lost) found at the possible broch or fort of Laws, Monifieth, Angus, in the eighteenth century (ibid., Fig. 298). Recent research has identified the silver plaque, recovered from a sixth-­ century hacksilver hoard 8  miles from the Wemyss Caves at Norrie’s Law, as a probable component of a composite helmet similar to the spangenhelmet originating in the Late Roman period and common in sixth-century continental Europe. This is the only known Scottish example of a Pictish helmet (Blackwell 2016). Of the 49 Pictish symbols documented in the Wemyss Caves in the nineteenth and early twentieth centuries (Allen and Anderson 1903; Simpson 1867; Stuart 1867), 26 survive in three caves: the Court Cave, Jonathan’s Cave and the Sliding Cave (Table  11.1). Seventeen of the lost symbols were located in the West Doo Cave, which is thought to have collapsed under the weight of a heavy gun emplacement in 1914 (Fig. 11.4). Three were lost when the Glassworks Cave, undermined by coal mining, collapsed in 1901. Three more were lost as the result of vandalism in Jonathan’s Cave in 1986. The double disc and the Pictish beast (along with the crescent and mirror) are the most frequently occurring of all Pictish symbols (Allen and Anderson 1903). This is reflected at Wemyss, where the most common symbol is the double disc (of which there are 15 examples), followed by the Pictish beast (six examples) and the arch symbol (six examples). Seven bird-like figures, which do not all conform to standard

226

J. Hambly et al.

Table 11.1  Symbols documented in the Wemyss Caves by J.  Romilly Allen in 1890 (*British Library Add_37542 and Add_37573) and by this project in 2014/2015 Recognised Pictish symbol (P) or other antique symbol (O) P O

Recorded by Allen in field notebooks, mostly in 1890* 3 6

Recorded 2014/2015 0

Notes Cave undermined by mining and collapsed in 1901

Cave Glassworks

Symbol Arch Cross

Court

Pictish beast Double disc (1 with floriated rod) Crescent and V-rod Arch Double crescent Man holding a club Cup-mark cross Deer-like figure Triangle Other shapes Pictish beast

P P

3 4

3 4

P

1

1

P P

1 1

1 1

In passage

O

1

1

In passage

O

1

1

In passage

O

1

Not found

In passage

O O P

3 Not known 1

5 8 0

Double disc Double disc with Z-rod and beast’s head Arch Flower Bird Disc and rectangle (mirror) S-shaped figure Rectangle Serpent Crosses Other shapes

P P

5 1

0 0

P P P P

2 1 2 2

0 0 0 0

P

1

0

P P O O

1 1 2 c. 12

0 0 0 0

West Doo

1 floriated rod added post-1890

All post-1890 West Doo Cave collapsed in 1914 beneath weight of heavy gun emplacement above cave

(continued)

11  How a Community Digital Heritage Project Has Helped to Imagine…

227

Table 11.1 (continued)

Cave Jonathan’s Cave

Symbol Pictish beast

Double disc Fleur de Lys Bird Rectangle Lion (horse/ bull) Fish Wolf/dog Boat

Crosses Other shapes Sliding Cave Double disc Rectangle with double disc inside Rectangle Total Pictish carvings

Recognised Pictish symbol (P) or other antique symbol (O) P

Recorded by Allen in field notebooks, mostly in 1890* 2

Recorded 2014/2015 0

P P P P P

4 1 5 1 1

4 1 4 1 1

P P O

1 1 1

2 1 1

O O P P

10 >6 1 1

19 c. 15 1 1

P

1 49

1 26

Notes Destroyed by vandalism in 1987

Swan destroyed

One modern Described by Simpson in Doo Cave Photograph in Allen’s notebooks could be a copy of John Patrick’s

Fig. 11.4  Aerial view of the coastline and main group of Wemyss Caves (Photograph: A. Liggat and S. Russell, UK Civil Air Patrol)

228

J. Hambly et al.

Pictish forms, are probably ancient carvings. The rest of the symbols occur no more than four times. Animal representations occur usually once, and most are found in Jonathan’s Cave. Crosses were documented by the antiquarian researchers, but are not systematically presented in the nineteenth-century records available to us. We recorded 19 crosses alone in Jonathan’s Cave, compared to the 10 in J. Romilly Allen’s notebooks. These are in most cases very damaged and much altered; however, it is still possible to recognise forms such as the Greek, pedestalled and trident motifs similar to those ascribed to Early Medieval forms by Fisher (2001) in cave contexts in western Scotland. Of the numerous other historic shapes inscribed into the Wemyss Caves walls, three are noteworthy. A pecked motif that appears to form a six-oared boat survives on the north wall of Jonathan’s Cave. Although its authenticity has been questioned (Le Bon 1992), there is no reason to doubt its antiquity or that it could be Pictish (Ritchie and Stevenson 1993). In the passageway adjacent to the Court Cave, a figure brandishing a club next to an abstract shape—locally referred to as Thor and his hammer with a goat (Fig. 11.5a)—is also carved with a pecked technique. On the opposite wall of the passage, nine midget cup-marks form a cross shape (Fig. 11.5b). These are the only carvings of this type in the surviving Wemyss Caves, although a cup-and-ring mark was recorded in the Michael Cave prior to its infilling as a result of mining activity in 1929 (Edwards 1933). Midget cups are found in prehistoric cup-and-ring art; however, another parallel might be the occurrence of groups of small cups on cross slabs of the tenth or eleventh centuries AD from Craignarget and Sinniness, west of Whithorn in Dumfries and Galloway (Collingwood 1925). Finally, although devoid of Pictish symbols, the Well Cave contains inscribed evidence upon its walls relating to later historical use of the cave. Until the late nineteenth century, young people from East Wemyss visited the eponymous cave on Handsel Monday (the first Monday after January 12th) to drink from the holy well (St Margaret’s Well) that bubbled up at the back of the cave. Many hundreds of nineteenth century monograms, and occasional crosses and drawings (some finely carved) adorn the cave walls (Fig. 11.6).

Fig. 11.5 (a) ‘Thor and his goat’, (b) cup-marked cross shape, both in passageway next to the Court Cave (© SCAPE)

11  How a Community Digital Heritage Project Has Helped to Imagine…

229

Fig. 11.6  Nineteenth-century graffiti in the Well Cave (© SCAPE)

11.3  The Archaeology of the Wemyss Caves The scale of archaeological evidence lost through subsequent use of the caves (e.g. stabling for animals, workshops, etc.), antiquarian ‘turning over’ (Simpson 1867) of sediments within the caves and coastal erosion of deposits in front of the caves is significant. However, coastal section recording and a number of small-scale modern excavations within and outside the caves (summarised by Gibson and Stevens 2007; Guttmann 2002) have identified fragmentary stratified contexts associated with Iron Age and Early Medieval use of the caves and their immediate surroundings. There is good evidence for an extended Iron Age presence and farming of the then coastal plain in front of the caves. Five radiocarbon dates from anthropogenic soils developed over raised beach deposits all fall within 800–170 BC (see Guttmann 2002 for details), and, in 2005, cut marks were recorded outside the Well Cave, sealed by a securely stratified deposit dated to 770–400 BC (Gibson and Stevens 2007). The most tantalising evidence—and the only dated context associated with an undisturbed cave interior where there are surviving Pictish carvings—is in the Sliding Cave. Here, a date of AD 240–400 was obtained from barley in a secure context interpreted as a possible floor surface of trampled clay (Gibson and Stevens 2007). Ten radiocarbon dates obtained from recent excavations at Dunnicaer promontory fort in Aberdeenshire—a site associated with Pictish symbols that bear many similarities with those in the Wemyss Caves—date the occupation of the fort to AD 200–400 (Noble 2016). This is significant, because it provides supporting evidence for what was thought to be an anomalously early date for the carving of Pictish symbols in the Wemyss Caves.

230

J. Hambly et al.

Later Iron Age and Medieval activity was detected in deposits outside Jonathan’s Cave, where soils contained archaeological material dated to AD 60–420, AD 420– 640 and AD 960–1250. A diagnostic ninth- to tenth-century Norse pin was also recovered from these deposits (MacKie 1986). Two Christian burials—a male in his 30s or 40s and a female in her 20s—revealed by coastal erosion outside Jonathan’s Cave in 1988 and 1992 have been radiocarbon dated to AD 890–1220 and AD 1020–1190, respectively.

11.4  A  Brief Retrospect of Documentation of the Wemyss Caves The inscriptions in the Wemyss Caves are very well recorded. Since James Young Simpson’s first hurried visit in 1865, every documentation technique of the time has been applied to them. Simpson recruited the service of his friend and fellow antiquarian, James Drummond, a well-known and accomplished Edinburgh artist who drew the carvings ‘with great care and scrupulous exactitude’ (Simpson 1867, p.  164) (Fig.  11.7a). In the same year, John Stuart published the Wemyss Caves material in volume 2 of the Sculptured Stones of Scotland, illustrated with Aberdeen engraver Andrew Gibb’s drawings (Stuart 1867). Gibb’s technique beautifully brings out the three-dimensional character and economic lines of the symbols (Fig. 11.7b). Records of the carvings were brought up to date in 1983/1984 by a Royal Commission on the Ancient and Historical Monuments of Scotland survey (Ritchie and Stevenson 1993). This considerable body of illustration is a reliable record of the lost carvings and of the damage to those that survive. Unfortunately, despite extensive research by volunteers during this project in Scotland and London, Christian Maclagan’s (1876) rubbings made in the Wemyss Caves have not been located. Rubbings of some of the symbols by George Seton in 1866 (RCAHMS catalogue nos. 544/545) and J. Romilly Allen (British Library

Fig. 11.7  Drawing of horse/lion by (a)  James Drummond (Simpson 1867), (b) Andrew Gibb (Stuart 1867) (see Fig. 11.2 for photograph of current condition)

11  How a Community Digital Heritage Project Has Helped to Imagine…

231

catalogue nos. Add_37542 and Add_37543) provide valuable direct documentation of the inscriptions. The most important surviving archive is that of Allen. His collated sketches, rubbings and photographs, held in the British Library, mostly date from his main period of fieldwork in May 1890, with occasional sketches dated September 1879. Allen’s notebooks contain rubbings and photographs of every probable Pictish and/or historic engraving present in the Wemyss Cave in the late nineteenth century. They form an objective record of the appearance of the symbols prior to modern damage and alteration. These are accompanied by detailed measured sketches of the caves and the location of the symbols within them, which form the basis of our digital recreation of the collapsed West Doo Cave with its 17 Pictish symbols. Allen undertook his own photography but also appears to have used material taken by local professional photographer John Patrick. Patrick published a number of his photographs in a series of articles for The Reliquary and Illustrated Archaeologist (1905a, b, 1906), in which he drew attention to the Pictish carvings, and local traditions and folklore associated with the caves. In the second half of the twentieth century, coal mining began to have a seriously adverse impact upon this stretch of coastline; coastal erosion and vandalism also became more pressing issues. Concerns over the structural integrity and vulnerability of the caves and carvings prompted the then Ministry of Works to take casts of the most significant carvings in 1951 (held by the National Museum of Scotland). In 1991, the National Museum of Scotland took moulds and casts of all surviving carvings in the Court Cave and Jonathan’s Cave (now in Kirkcaldy Galleries). In 2005, an early application of terrestrial laser scanning was applied to the Wemyss Caves by Archaeoptrix Ltd (Wessex Archaeology 2005). Driven by the same concerns to record, raise awareness, research and ultimately protect the caves and carvings, our project is the latest chapter in the history of their documentation and a new chapter in applying digital technologies to communicate their rich and multifaceted heritage.

11.5  Methodology Our aim was to create a comprehensive and accurate record of the surviving Pictish heritage and present it as an interactive digital resource that would allow users to explore the caves in their coastal setting, closely examine the carvings, and access historic documentation about them. Integrating local, traditional and intangible heritage within their cave settings to tell the wider story of the caves was also an important aim. Furthermore, we wanted to experiment with digitally restoring lost caves and symbols, such as the collapsed West Doo Cave, and recreate cave spaces and the coastline in different historical periods. Local volunteers, coordinated through SWACS, were involved in every aspect of the research, fieldwork and creation of interpretive material for the website, which was built by one of the authors of this paper (and subsequent chair of SWACS), Mike Arrowsmith.

232

J. Hambly et al.

11.6  Collation and Digitisation of Archives Quite apart from the inscribed and archaeological heritage of the Wemyss Caves, there are significant archives of documentation, research, local history and folklore related to the caves that are dispersed and, in some cases, difficult to access. Research by volunteers tracked down archives about the caves held by Kirkcaldy Galleries, Historic Environment Scotland and the British Library. SWACS also themselves hold extensive local history archives, including historic photographs of the Wemyss Caves and surrounding coastline. Wherever possible, a digital copy of material most relevant to the caves was obtained or created. Additionally, we put out a call in the local community for information, stories, memories and documents relating to any aspect of the caves. This elicited new photographs and first-hand accounts of stories and (mostly) childhood memories relating to the caves, which gave us a privileged insight into the personal and community value of the caves in East Wemyss. We used video and audio recording to document these memories and stories through interviews. These inspired, and provided content and cast members, for short films developed for the website to tell some of the local stories of the caves.

11.7  Digital Data Capture and Processing 11.7.1  Laser Scan Survey Digital data capture was conducted using multiple techniques and technologies. Eight hundred metres of coastline and eight caves were scanned using a longrange Faro Focus X330 laser scanner. The Focus X330 was the choice of instrument for this project because of the speed and range necessary for accurately capturing cliff-­faces and coastal features that were too distant for some other scanners to record, especially when working in the dynamic tidal environment of the foreshore. A total of 376 individual scans (resulting in 400 GB of data) were captured to cover the coastline, caves and MacDuff’s Castle. The castle is located on the coastal cliff above the caves and is an integral part of the historic coastline. Once registered to the common Ordnance Survey grid, each scan was exported into E57 format for archive and distribution. A plan and corresponding table allows the end-user to select and load the data required for an individual cave or portion of coastline. A subsample point cloud of the entire survey is also available so that the end-user is able to load complete coverage of the caves and coast at a lower resolution than the original scans. Thus, the data is transformed into a management resource. The scan data will be archived with Historic Environment Scotland and Fife Council (since both organisations have a management interest in the caves), so that long-term access to the data is ensured.

11  How a Community Digital Heritage Project Has Helped to Imagine…

233

11.7.2  Laser Scan Survey In addition to the laser scan survey, a convergence photogrammetric or structure from motion survey was conducted. The photogrammetric survey consisted of three distinct levels of recording. Firstly, we created high-resolution photogrammetricderived meshes of the individual carvings at submillimetre level. During a pilot phase, we trialled this technique alongside structured light scanning and found that photogrammetry gave results on par with the structured light scanner, so photogrammetry was utilised for all subsequent high-resolution meshes of the carvings. Secondly, we created photogrammetric-derived meshes for each cave. Finally, a drone was employed to capture a photogrammetric survey of the wider landscape. In total, over 24,000 images were taken and processed into digital models, each registered to the same grid as the laser scan survey.

11.7.3  Reflectance Transformation Imaging Individual carvings, and panels of carvings, were also recorded using reflectance transformation imaging (RTI). The use of raking light to study features has a long pedigree in archaeology, notably for rock art, epigraphy and aerial landscape photography taken under advantageous light conditions. It is this attribute that is recreated in RTI: a technique based on image-based texture mapping, named polynomial texture mapping (PTM) by its developers (for explanation and application of PTM see Malzbender et al. 2001). The technique was subsequently developed by the non-­ profit organisation Cultural Heritage Imaging, which also distributes software, guidance and equipment for use with RTI (Cultural Heritage Imaging 2016). In this technique, multiple photographs of the subject are taken from a fixed camera position, with light directed from different positions at a constant distance. A reflective sphere next to the subject enables the software to determine the direction of light in each photograph, which it uses to create the RTI file. This file can then be viewed in a browser that provides a moveable virtual light source. RTI proved ideal for detailed recording and examination of the carvings in the Wemyss Caves because it was possible for volunteers to undertake the fieldwork and create the interactive output. Around half of more than 50 RTIs for the Wemyss Caves carvings were captured and processed by members of SWACS and other volunteers, who continue to document the hundreds of examples of nineteenth century graffiti using this technique. Viewers can shine a virtual torch over a carving, exploring and discovering features for themselves. It also enables the carvings to be examined under lighting conditions which may not be possible in situ (Fig. 11.8). In order to allow the RTI images to be viewed on the website, a lower-resolution version of the PTM files has been uploaded. Although these do not permit the application of the extra lighting filters, the overall results do simulate the effect of shining a torch across the carvings.

234

J. Hambly et al.

Fig. 11.8  Image of the boat carving in Jonathan’s Cave generated through RTI rendered with specular enhancement

11.8  The Digital Resource 11.8.1  C  reation of the 3D Virtual Caves and Coastal Environment Our challenge was to take the huge dataset and simplify it to produce a realistic virtual ‘Wemyss Caves’ that forms the basis of the online interactive experience. The cave models were created from a workflow created for this project. The E57 laser scan files were cleaned in CloudCompare to remove unwanted ‘noise’. The cleaned data was subsampled to reduce the number of points. The surface normals for the point set were computed and the results exported to MeshLab to generate the mesh model for the cave. This was then exported to Cinema 4D, where the final texturing and digital set was assembled. Once in Cinema 4D, we filmed within the models to produce our interactive panoramic ready to be included in the website. One of the aims of the project was to experiment with digitally restoring the collapsed West Doo Cave and lost carvings using J.  Romilly Allen’s 1890 survey. Allen’s notebooks provided sketches of the cave with a set of detailed measurements. The laser scan survey captured the external collapsed entrance, part of the infilled passageway and the large, now tree-filled, depression in the ground surface caused when the main chamber of the cave collapsed. It therefore framed the void where the cave should be and allowed us to accurately locate and position the lost cave. By using Allen’s measurements to fill in the void in our data, we were able to fairly accurately recreate the interior cave space. Allen also noted the position of each carving on the cave walls and provided measurements of the distance between each carving and measurements of the relative position of each carving to the cave

11  How a Community Digital Heritage Project Has Helped to Imagine…

235

Fig. 11.9  Screenshot of carvings sketched by J. Romilly Allen arranged on a grid and projected onto the reconstructed West Doo Cave walls to accurately locate the lost carvings

Fig. 11.10  Screenshot of coast model with time slider

floor and the cave roof. Using this information, we were able to plot each carving onto a grid and then projected this grid onto the virtual cave to reinstate the carvings in their correct locations within our digital model (Fig. 11.9). The dynamic nature of the landscape and the ever-changing coastline is also represented in our digital models by the introduction of a fourth dimension in the form of a time slider. The laser scan and drone survey provided us with a baseline model which we have adjusted to represent different time periods (Fig. 11.10). Within the digital set, we were able to experiment with different levels of coastal erosion and reinstate parts of the eroded coastline virtually. We also reconstructed the collapsed cave mouths, as well as the Medieval Macduff’s Castle and the nineteenth-century gasworks, and placed them within their virtual landscape.

236

J. Hambly et al.

It is from the digital coastal model that website visitors travel back in time to access content in the caves, but the model itself also provides the context for the rich timeline of historical events that have created this coastline.

11.8.2  T  he 4D Wemyss Website and Integration of Content (http://www.4dwemysscaves.org) The historic study of the caves has produced a wide range of content which, until now, has not been collated or presented as a coherent collection. A key aim of the website was that it should bring together all the accurate and reliable information about the caves within a realistic digital cave environment generated from the 3D digital survey. Our approach to content delivery from the main interface of the website was that information and media could be called up from within an immersive 3D environment which remained in the background. From each cave, context-­ relevant material can be accessed through overlay panels and by clicking on the carvings highlighted on the virtual cave walls (Fig. 11.11). Throughout, we sought to create a balance between the interactive interface designed to give some simulation of the experience of being in the caves with a practical and intuitive way of accessing the rich and detailed content, whilst also presenting a narrative about the historic use and study of the caves. Each content type (image gallery, video, RTI scan, etc.) is delivered via a pop-up viewer. These viewers are self-contained modules which integrate into the main code, allowing for more content types to be seamlessly added to the site should they become available. It was a project requirement that content for the RTI viewer should be responsive on all modern platforms—mobile as well as desktop—irrespective of screen resolu-

Fig. 11.11  Screenshot of user interface from within Jonathan’s Cave

11  How a Community Digital Heritage Project Has Helped to Imagine…

237

tion or interface (mouse, touchpad, etc.). This entailed amending, in JavaScript for HTML5, the WebRTI viewer developed by the Visual Computing Lab of the Italian National Research Council. For researchers and others who want to view the entirety of the comprehensive resources without having to explore each cave, a catalogue incorporated into the website allows content to be viewed through a single click. Content is also indexed and organised so that it can be filtered by cave and classification. The content is stored and managed in a back-end database, which allows us easily to add new or rediscovered content relating to the caves at any time in the future.

11.9  Some Achievements of the Digital Resource The project has digitised the entire coastal landscape, together with the caves and the carvings within them. Not only has this provided the 3D structure and architecture for communicating the heritage and telling the stories of the Wemyss Caves, in a worst-case scenario (e.g. cave collapse, coastal slope failure or destructive vandalism) we have the best possible record of them that is currently available. As a management resource, this highly accurate dataset forms the essential baseline against which to measure change. For example, we can compare our data to historic casts of the carvings  and (deliberate damage aside) it should be possible to predict the expected condition of the carvings at defined time intervals in the future. This level of information is needed to assess options for preventing deterioration and for improving the condition, safety and setting of the monument (e.g. remedial structural engineering works, vegetation management, access improvements). A Wemyss Caves Management Group, with membership drawn from project partners, other local stakeholders, the landowner, Historic Environment Scotland and Fife Council, is already taking forward some of this work and using the digital resource for description and monitoring purposes. In addition, for the first time since antiquarian records, anyone interested in Pictish symbols in the Wemyss Caves can easily access authoritative current and historical information about them. This has allowed us to disentangle carvings originally recorded as Pictish from subsequent alterations and later carvings, and allows us to review what has been damaged or lost. All data relating to each carving recorded in this project is available either in situ in a virtual cave context or in a catalogue (which includes destroyed carvings). Each carving is linked to an RTI viewer, so that the present condition of the carving may be examined, as well as to historic illustrations and photographs where available. With access to a comprehensive catalogue about the original location and form of each carving, it is now much easier to compare the Wemyss Caves material with the way in which these symbols were used at other sites and in other situations. Additionally, integration in the digital resource of local stories and traditions relating to historic use of the caves (via embedded video content and text) invites connections to be made between uses of the caves in all periods.

238

J. Hambly et al.

11.10  Discussion ‘Why they should be found alike on sepulchral stones and on the walls of caves is a problem not apparently admitting of any easy solution’ (Simpson 1867, p. 176).

11.10.1  T  he Setting and Placement of the Symbols in the Wemyss Caves There are only nine recorded caves—five of which are in East Wemyss—and two living rock-faces which bear probable Pictish symbols, and a further cave and rock outcrop with possible Pictish symbols (Fig. 11.12), compared to the c. 200 monumental Pictish Class I symbol stones with which they are probably contemporary (see Table 11.2 for a summary of recorded surviving and destroyed Pictish symbols in cave and rock outcrop contexts). There is no reason for the standing stones to have survived better than symbols in other contexts—prehistoric rock art, for example, is much more common than Early Medieval rock art—so it is likely that cave and rock outcrop sites were always rare. However, the presence of Pictish symbols in these contexts, and also occasionally on objects such as metal artefacts, gaming pieces, portable stone objects and stone plaques, shows that they were involved in a wider set of meanings than just those relating to the standing stones, which have dominated thinking about the symbols (Clarke 2007). The traditional imagined setting of Pictish symbol stones as monumental markers in the landscape has been challenged by recent work centred on the Craw Stane at Rhynie (Noble et al. 2013). Here, excavations have revealed that the Craw Stane originally stood at an entranceway into a group of buildings within a doubleditched enclosure surrounded by a 4–5 m high wooden palisade. As more contextual evidence for the use of Pictish symbols comes to light through archaeological excavation, cave and rock-face carvings may cease to be perceived as being so different or unrelated to their use on standing stones (e.g. for this traditional view see Jackson 1984, p. 225). And unlike the majority of standing stones, caves retain the original unambiguous context of where the act of carving took place, and so offer a unique opportunity to examine closely the relationship between a symbol and its immediate setting. Although most of the project team are very familiar with the Wemyss Caves, the recording process entailed unusually extended periods of being in and around the caves at all times of the day, in all weathers and at all states of the tide. A consequence of this was a much fuller awareness and appreciation of the cave spaces themselves and of their immediate setting within the great panoramic arena of the Outer Forth. It became clear that the question of placement of Pictish symbols in the caves was one of the areas of research we could and should explore. Is there a pattern of placement in the caves? Could the digital outputs help our understanding and interpretation? How does the placement of the symbols in the Wemyss Caves compare with other Pictish symbol-bearing caves?

11  How a Community Digital Heritage Project Has Helped to Imagine…

239

Fig. 11.12  Map showing location of all recorded cave and rock outcrops with Pictish symbols

Location Wemyss Caves, East Wemyss, Fife

c. 8

c. 10

17

West Doo Cave

Other antique symbols 1

10

Probable Pictish symbols 3

Court Cave

Cave/rock surface Glassworks Cave

2

0

1 known in the Well Cave

1 known in the Well Cave

Holy Crosses wells 8 1 known in the Well Cave

Burials/human remains

Location in cave or aspect on rock-face High up on north-east wall Assume would be lit by light from west-south-west and south-facing cave openings opposite. Simpson (1867) and MacLagan (1876) describe the cave as being full of light Mid-way into the cave on the Pictish symbols: 3 Pictish beasts; 4 double discs (2 with north-east wall. The carvings are floriated rod); 1 crescent and located on a panel from approximately half way up the wall V-rod; 1 arch; 1 double to the top as it meets the cave roof. crescent. Other antique Lit by light from east-facing cave carvings mainly geometric shapes. Figure holding a club mouth and cup-marks occur in passage to south of the Court Cave On back north wall and north part of Cave now destroyed west wall of the cave Pictish symbols: 1 Pictish Located ‘a few feet above the beast; 5 double discs; 1 ground... some very low down double disc with Z-rod and requiring the earth to be brushed beast’s head; 2 arches; 1 away... one group requires looking up flower; 2 birds; 2 disc and rectangle; 1 S-shaped figure; to’ (as described by Simpson 1867) Opposite south-facing cave mouth. 1 rectangle; 1 serpent Assume would be well lit. Simpson describes them as being lit by the sun Notes on inscriptions Cave now destroyed Pictish symbols: 3 arches

Table 11.2  All recorded Pictish symbols (surviving and destroyed) in cave and rock outcrop contexts with relevant placement information where available and summary of shared features

Caiplie Caves, Kilrenny, Fife

Location

3

?1

1

Mortuary Cave

Chapel Cave

Probable Pictish symbols 16

Sliding Cave

Cave/rock surface Jonathan’s Cave

0

Other antique symbols c. 13

9

>50

0

Burials/human remains Outside cave. 1 male, 30–40 yrs; 1 female 20s. E–W orientation. AD 890–1220 and AD 1020–1190 (Guttmann 2002)

Hermit’s Outside cave. 4 Well un-coffined E–W burials 1 'head first' in a hole near the entrance of the cave (Stuart 1867) Hermit’s Well

1 known in the Well Cave

Holy Crosses wells 10 1 known in the Well Cave

1 arch

1 Z-rod. Very uncertain, could be natural. Not identifiable now. There are no other examples of singly occurring Z rods

Pictish symbols: 1 double disc; 2 rectangles (one with double disc inside)

Notes on inscriptions Pictish symbols: 2 Pictish beasts; 4 double discs; 1 Fleur de Lys; 5 birds; 1 rectangle; 1 lion?; 1 fish; 1 wolf. Other antique symbols include an oared boat and a range of animal shapes, including 4 deer-like shapes and geometric shapes

(continued)

High up on ledge on north wall. Lit by light from south-facing cave mouth, opposite. Crosses on west wall near entrance. Well lit

Crosses throughout cave. Some well lit, some in darkness at rear of cave

Location in cave or aspect on rock-face From just beyond the entrance to approximately 2/3 along the length of the west wall. Spread out individually and in small groups from ground level to top of wall where it curves to meet the roof. Lit by light from south-east-facing cave mouth 3 crosses towards rear of cave on well-lit west wall. Majority of crosses on dark back wall above wave-cut ledge. Lit by candles? Double disc at entrance on west wall low down. Rectangles about 1/3 of the way into the cave on east wall half way up the wall

Covesea Caves, Drainie, Moray

Location Fife Ness, Crail, Fife

The Sculptor’s Cave

8

Probable Pictish Cave/rock symbols surface Constantine’s Cave

Table 11.2 (continued)

2

Other antique symbols 2

?1

Holy Crosses wells >35 Notes on inscriptions At least 4 deer-like animal figures. Two previously recorded (Stuart 1867), at least 2 further examples identified in 2017: https:// canmore.org.uk/site/35369/ constantines-cave-fife-ness Human remains Pictish symbols: 1 fish; 2 crescent and V-rods; 1 (Late Bronze rectangle; 2 disc and Age and Late Iron Age) inside rectangle; 1 triple oval; 1 step cave (Armit et al. symbol 2 pentangles, not a usual 2011) Pictish form. Pentangle is one of the very few ritual protection marks for which there is evidence to support its supposed meaning and function (Champion 2015) 1 flower symbol on passageway roof not recorded until the late twentieth century but could have been overlooked because of its unusual location (John Borland pers. comm.)

Burials/human remains Stone coffins have been found in immediate vicinity of cave (RCAHMS 1933)

One group at entrance of east passage on east and west walls. Those on east wall close to the floor, those on west wall, mid-way up. One group at entrance of west passage on east and west walls. Half way up the walls. Both groups in best lit part of the cave at entrance

Location in cave or aspect on rock-face High up on all 3 walls. Whole cave well lit by light from north-facing cave mouth

Location Clashach Cove, Duffus, Moray Trusty’s Hill, Anwoth, Kirkcudbright, Dumfries and Galloway Rock surface Dunadd, Kilmichael, Glassery, Argyll and Bute Rock surface Eggerness, Garlieston, Sorbie, Dumfries and Galloway

Cave/rock surface Unnamed cave Rock surface

Other antique symbols 0

?1

2

?5

Probable Pictish symbols 1

2

1

?1

Rockcut basin

Rockcut basin

Holy Crosses wells 0

Burials/human remains

Well-defined pecked stag, scrolled joints in Pictish style but could be influenced by Irish manuscript tradition via Iona. 2 fainter deer carvings on the same stone. 3 backward-looking incised horses on a separate nearby stone, interpreted as Iron Age (Morris and van Hoek 1987)

Pictish symbols: 1 double disc with Z-rod; 1 hippocampus. On rock surface at entrance to vitrified fort Boar in Pictish style with footprints and ogham script

Notes on inscriptions Faint crescent and V-rod

Located at entrance to summit enclosure of Dunadd fort opposite rock-cut basin

North face of southerly of 2 projecting rocks which form entrance to summit enclosure. Opposite rock-cut basin

Location in cave or aspect on rock-face East wall at south (landward) end

244

J. Hambly et al.

Fig. 11.13  Plan generated with point cloud data showing shape of caves behind coast edge; all caves are tunnel-shaped except the Well Cave

The Wemyss Caves are relict sea caves. They display the typical architecture of wave-cut caves carved into the soft sandstone carboniferous geology of the East Fife coast: a wide cave mouth, tendency for a tunnel shape that narrows into the coastal cliff, and wave-smoothed cave walls and ledges inside (Fig.  11.13). This creates open, lofty light-filled entrances which gradually narrow and darken towards the rear of the caves. Looking out from the Wemyss Caves are panoramic views to the coastline on the other side of the wide expanse of the Outer Forth. The geography of this stretch of the Fife coast highlights a sense of boundary at what would have been the southern limits of Pictish territory. Slope failure and structural failure—both natural processes that may have been exacerbated by the effects of coal mining (Saiu and McManus 1998)—and deliberate blocking of cave entrances has resulted in most of the Wemyss Caves being less open now than probably at any time in their history. However, despite these obstructions, parts of the caves are still illuminated during daylight hours most of the time. Without exception, the Pictish symbols are located on the sunlit walls. The flexibility of the digital dataset allowed us to experiment with lighting effects in our virtual caves. By removing material at the mouths of the caves and digitally restoring the full spans of the cave mouths, the relationship between where daylight strikes the cave walls and the location of the Pictish carvings is even more evident. This is illustrated on the website in the Court Cave, where the user is invited to go back in time to experience the cave space as the Picts would have experienced it. A simulated morning sun outside the cave illuminates the Pictish symbol-bearing panel inside (Fig. 11.14). This relationship is also observed and presented in the digitally recreated West Doo Cave (collapsed in 1914), which unlike the tunnel forms of the  Court Cave, Jonathan’s Cave and the Sliding Cave, is shallower and more circular in plan. The digital reconstruction demonstrates daylight entering the very wide cave mouth and illuminating the carved symbols as it traverses across the cave walls throughout the day.

11  How a Community Digital Heritage Project Has Helped to Imagine…

245

Fig. 11.14  Screenshot of reconstructed the Court Cave entrance to show illumination by daylight of wall bearing Pictish symbols (http://www.4dwemysscaves.org/cave/index.php?ccode=cc)

Looking to the three other known Scottish caves with surviving or recorded Pictish symbols, the pattern in the placement of carvings observed in the Wemyss Caves is repeated. In the Chapel Cave, Caiplie, light streaming in from the entrance hits the crescent symbol located on the back wall; in Constantine’s Cave, Fife Ness, light from the cave mouth illuminates the site of the inscribed animal figures and the inscribed and pecked crosses covering every wall of this small cave; and in the Sculptor’s Cave, Covesea, the Pictish symbols occur within the double entrance passages where daylight partly illuminates the large, otherwise dark space. Although any acoustic relationship with the placement of the carvings has not been studied in this project, we experienced the important contribution of sound to the cave environment during data capture. Wave and seabird noises are amplified within the wide entrances of the seaward-facing caves. Gondek (2015) has recently discussed the juxtaposition of Pictish symbols at the entrance (but not the interior) of the Sculptor’s Cave as a mnemonic device between the living and the dead, linked to Late Iron Age ritual killings that may have happened there (Armit et al. 2011). This echoes the relationship she finds between the placement of smaller, portable symbol stones found in structural contexts with thresholds, between inside/outside, aboveground/underground and, by extension, living world/underworld (see Gondek 2015, Table 4.1, Figs. 4.4 and 4.5), an important concept in pre-Christian Pictish cosmology. However, in the six or seven Fife caves that have recorded Pictish symbols, although some carvings are inscribed near to cave entrances, it is the relationship with light that is the determining factor in their placement. If the back wall of a cave is well lit, it is likely to bear symbols. This is not to say that there is no association between the use of pre-Christian symbols and transitions into otherworldly liminal places, but if we include the Wemyss Caves and other Fife caves in the analysis, it is visibility and light that are the universal factors in the choice of placement of the

246

J. Hambly et al.

symbols. They are not found in the darker recesses of the caves where access to them could be more easily mediated or restricted but are located where they can be seen by people. It has been observed that symbols on standing stones nearly always occur in pairs (Allen and Anderson 1903). We examined the arrangement of symbols in the Wemyss Caves with the aid of digital reconstructions of all surviving and lost carvings in their 3D cave spaces, but did not recognise patterns in their distribution in relation to each other. Beyond placement on cave walls illuminated in daylight (as discussed above), the arrangement of the carvings seems informal. In every cave that contains Pictish carvings, a similar range of carving types can be found; the exception is Jonathan’s Cave, which also bears the only examples of clearly Pictish animal symbols: the fish, horse and dog or wolf. The meaning denoted by the pairing of symbols on standing stones does not appear to be relevant in cave contexts.

11.10.2  Why Are There Pictish Symbols in the Wemyss Caves? By bringing the archaeological, historical and intangible evidence together in one place, we can compare and evaluate the evidence found at Wemyss with other sites, helping to gain insight into the possible circumstances of the Pictish symbols. The Wemyss Caves share some of the defining features of other rare examples of caves bearing Pictish symbols and sites where Pictish symbols retain or closely retain their original context. The Wemyss Caves and the Caiplie Caves share a historical association with a holy well: St Margaret’s Well in the Well Cave and the Hermit’s Well next to Mortuary Cave. Although a very different type of site, it is suggested that rock-cut basins found in close association with rock-faces bearing Pictish symbols at the sixth- to seventh-century AD royal strongholds of Trusty’s Hill in Dumfries and Galloway (Toolis and Bowles 2013) and Dunadd in Argyll and Bute (Campbell and Lane 2000) held holy water for religious or ceremonial uses (Gondek 2015). The remarkable rock-cut well in the fourth- to seventh-century AD Pictish fort at Burghead in Moray contained one of the original 30 stone plaques with incised carvings of bulls found in and around the fort, and reputedly a Celtic stone head which may hint at an Iron Age origin for the site (Foster 1996). Whether a fort or cave, where Pictish symbols exist, there seems to be a link between the carvings and ritual practice that involves water and wells. Are we seeing a continuity of practice in the East Wemyss Handsel Monday celebrations (still strong in community memory)—where young people drank the waters of St Margaret’s Well in the Well Cave—with a Late Iron Age tradition centred on a sacred spring? In common with the Caiplie Caves and Constantine’s Cave in Fife, the most obvious shared feature of the Pictish symbol-bearing caves is the numerous crosses also found within them. For Joseph Anderson (Allen and Anderson 1903, p. cxi), the juxtaposition of Pictish symbols with crosses provided proof of the Christian character of, and intentions in, caves: ‘which were practically churches’. In Fife, many caves still retain their association with an early Christian saint, e.g. St Adrian with

11  How a Community Digital Heritage Project Has Helped to Imagine…

247

the Caiplie Caves, St Constantine’s Cave at Fife Ness, St Serf’s Cave at Dysart, St Regulus with a cave in St Andrews (now destroyed) and St Fillan’s Cave in Pittenweem (St Fillan’s Cave is still a church and also has a holy well). Interestingly, St Gernadius is associated with a cave in the historical parish of Kinnedar in Moray, where the Sculptor’s Cave is located (ibid.). There is no surviving tradition of a historical saint associated with the Wemyss Caves (other than St Margaret with the well), but they do have important defining features of crosses and a holy well, and there is unambiguous evidence of Christian practice in the two tenth- to thirteenth-­ century Christian burials in front of Jonathan’s Cave, which may have been the last remnants of a burial ground lost to the significant coastal erosion that has taken place here. The presence of features of Christian religious practice in the caves does not imply that the Pictish symbols found here are Christian in origin or meaning, although their meaning may have been transformed to become so. It is far more likely that the Wemyss Caves became a site of early Christianity because it was already a sacred place. The relationship between Class I symbol stones and early Christian sites has been demonstrated by Alcock (1988) and may be explained by the appropriation of existing holy places as a deliberate policy of early missions: a practice attested by Bede in the early seventh century (Wallace-Hadrill 1988). The symbols in the Wemyss Caves, even in their original undamaged condition, do not attain the artistic virtuosity and do not appear to be as structurally-organised as those on the standing stones. In the light of the recent third- to fifth-century AD chronology obtained at Dunnicaer, the similar date obtained in the Sliding Cave raises the prospect that the Wemyss Caves contain some of the earliest expressions and applications of the Pictish symbol. The symbols carved into the cave walls, that may have identified the site as a holy place, would eventually be incorporated into Christian practice at the same site. Does the high number of Pictish symbols and crosses in the Wemyss Caves suggest that this was a particularly significant place for pagan and early Christian communities in this area of Fife?

11.11  Conclusion The great value of caves that contain Pictish symbols is that they are different from other Pictish symbol-bearing contexts. There is no surviving evidence at Wemyss for high-status activity, royal affiliation or defensive intent, although Roman artefacts (presumably denoting some status) are known from similar cave contexts, such as Constantine’s Cave and the Sculptor’s Cave. Along with the growing number of symbols found on artefacts and smaller portable stone plaques in structural contexts, the manner of the carvings suggests informality rather than commissioned artistry, and the circumstances within which the symbols are found may be described as personal, domestic and sacred, or combinations of all of these; this is in contrast to the monumentality expressed by the Pictish symbol stones, whether or not they were originally positioned to be seen in the landscape. Caves offer an additional line

248

J. Hambly et al.

of evidence through which to explore the meaning that symbols imbued on the places where they are found. We have explored them through the digital journey we have taken with the East Wemyss community to document and share the Pictish and wider heritage and history of the Wemyss Caves. It has given us, and we hope others, the ability to see beyond the damaged monument of today to (on the balance of evidence) an ancient holy place and a site of early Christianity in eastern Scotland. The symbols bear the memorials, prayers, hopes and fears of the Pictish people who carved them, in a place that is still special to many in the present community. Acknowledgements  We are indebted to SWACS who made this project happen. Members of SWACS and volunteers from the local community generously contributed their time, knowledge, memories, stories, photographs, RTI skills and acting talent. Many of the archives used in the research and website, and referred to in this paper, were tracked down by Pam Cranston. We are extremely grateful to Fife Council, Historic Environment Scotland and the Heritage Lottery Fund for their funding and support.

References Alcock, E. A. (1988). Pictish stones Class I: Where and how? Glasgow Archaeological Journal, 15, 1–21. Allen, J. R., & Anderson, J. (1903). The early Christian monuments of Scotland. Reprinted 1993. Balgavies: Pinkfoot Press. Armit, I., Schulting, R., Knüsel, C. J., & Shepherd, I. A. G. (2011). Death, decapitation and display? The Bronze and Iron Age human remains from the Sculptor’s Cave, Covesea, north-east Scotland. Proceedings of the Prehistoric Society, 77, 251–278. Blackwell, A. (2016). Silver and symbols: Glenmorangie Project research on the Norrie’s Law Hoard. https://www.youtube.com/watch?v=bwHnGutK1ao&index=5&list=PL2O74qVmda4 M35xTrl2Mh5d9cS-EMheXE. Accessed 16 May 2017. Campbell, E., & Lane, A. (2000). Dunadd: An early Dalriadic capital. Oxford: Oxbow Books. Champion, M. (2015). Medieval graffiti: The lost voices of England’s churches. London: Ebury Press. Clarke, D. V. (2007). Reading the multiple lives of Pictish symbol stones. Medieval Archaeology, 51(1), 19–39. Collingwood, W. G. (1925). The early crosses of Galloway. Transactions of the Dumfriesshire and Galloway Natural History and Antiquarian Society (3rd series), 10, 205–231. Cultural Heritage Imaging. (2016). http://culturalheritageimaging.org/. Accessed 31 July 2016. Deas, G. B. (1948). The sculpturings on the caves of Wemyss. Offprint of the Rothmill Quarterly Magazine. Markinch: Tullis, Russell & Co. Ltd. Edwards, A. J. H. (1933). Short cists in Roxburgh and Sutherland, and rock sculpturings in a cave at Wemyss, Fife. Proceedings of the Society of Antiquaries of Scotland, 67, 171–176. Fisher, I. (2001). Early Medieval sculpture in the West Highlands and Islands. Edinburgh: Royal Commission on the Ancient and Historical Monuments of Scotland/Society of Antiquaries of Scotland. Foster, S. M. (1996). Picts, Gaels and Scots. London: Batsford. Gibson, C., & Stevens, C. (2007). Iron Age and Pictish activity at Wemyss Caves, Fife. Tayside and Fife Archaeological Journal, 13, 91–99. Gondek, M. (2015). Building blocks: Structural contexts and carved stones in Early Medieval northern Britain. In H. Williams, J. Kirton, & M. Gondek (Eds.), Early Medieval stone monuments: Materiality, biography, landscape. Woodbridge: The Boydell Press.

11  How a Community Digital Heritage Project Has Helped to Imagine…

249

Guttmann, E. B. (2002). Time and tide at East Wemyss: Excavations of the foreshore 1980–1995. Tayside and Fife Archaeological Journal, 8, 110–124. Jackson, A. (1984). The symbol stones of Scotland: A social anthropological resolution of the problem of the Picts. Kirkwall: The Orkney Press. Le Bon, E.  A. (1992). The Jonathan’s cave boat carving: A question of authenticity? The International Journal of Nautical Archaeology, 21, 337–342. MacKie, E. (1986). Iron Age and Early Historic occupation of Jonathan’s Cave, East Wemyss. Glasgow Archaeological Journal, 13, 74–77. MacLagan, C. (1876). Notes of the sculptured caves near Dysart in Fife. Proceedings of the Society of Antiquaries of Scotland, 11, 107–120. Malzbender, T., Dan Gelb, D., & Wolters, H. (2001). Polynomial texture maps. In SIGGRAPH ‘01, Proceedings of the 28th annual conference on computer graphics and interactive techniques (pp. 519–528). New York, NY: ACM Press. Morris, R. W. B., & van Hoek, M. A. M. (1987). Rock carvings in the Garlieston area, Wigtown district. Transactions of the Dumfriesshire and Galloway Natural History and Antiquarian Society, 62, 32–39. Noble, G., Gondek, M., Campbell, E., & Cook, M. (2013). Between prehistory and history: The archaeological detection of social change among the Picts. Antiquity, 87(338), 1136–1150. Noble, G. (2016). Discovering the northern Picts: Latest results of the University of Aberdeen Northern Picts Project. https://www.youtube.com/watch?v=Ttsp9Ql7fb0&index=7&list=PL2 O74qVmda4M35xTrl2Mh5d9cS-EMheXE. Accessed 16 May 2017. Patrick, J. (1905a). The sculptured caves of East Wemyss: Part 1. The Reliquary and Illustrated Archaeologist, 11, 73–84. Patrick, J. (1905b). The sculptured caves of East Wemyss: Part 2. The Reliquary and Illustrated Archaeologist, 11, 249–263. Patrick, J. (1906). The sculptured caves of East Wemyss: Part 3. The Reliquary and Illustrated Archaeologist, 12, 37–47. Ritchie, J.  N. G., & Stevenson, J.  N. (1993). Pictish cave art at East Wemyss, Fife. In R.  M. Spearman & J. Huggett (Eds.), The age of migrating ideas (pp. 196–208). Edinburgh: National Museum of Scotland. Royal Commission on the Ancient and Historical Monuments of Scotland. (1933). Eleventh report with inventory of monuments and constructions in the counties of Fife, Kinross, and Clackmannan. Edinburgh: Royal Commission on the Ancient and Historical Monuments of Scotland. Saiu, E. M., & McManus, J. (1998). Impacts of coal mining on coastal stability in Fife. In J. Hooke (Ed.), Coastal defence and earth science conservation (pp.  58–66). London: Geological Society. Simpson, J.  Y. (1867). Archaic sculpturings of cups and circles etc. upon stones and rocks in Scotland, England and other countries. Edinburgh: Edmonston & Douglas. Stuart, J. (1867). Sculptured stones of Scotland. Volume second. Edinburgh: The Spalding Club. Toolis, R., & Bowles, C. (2013). The Galloway Picts Project: Excavation and survey of Trusty’s Hill, Gatehouse of Fleet. Unpublished Data Structure Report 3309, GUARD Archaeology. http://archaeologydataservice.ac.uk/archiveDS/archiveDownload?t=arch-1168-1/dissemination/pdf/guardarc1-134833_1.pdf. Accessed 10 July 2016. Wallace-Hadrill, J.  M. (1988). Bede’s ecclesiastical history of the English people. Oxford: Clarendon Press. Wessex Archaeology. (2005). Wemyss Caves, Fife, Scotland: Archaeological evaluation and assessment of results. Unpublished Data Structure Report Ref. 55754.01, Wessex Archaeology.

Chapter 12

Hear Here: Prehistoric Artists Preferentially Selected Reverberant Spaces and Choice of Subject Matter Underscores Ritualistic Use of Sound Steven J. Waller

12.1  Introduction Regarding the use of caves as ritual spaces, a theoretical question that can be posed is: what makes caves unique? When I asked myself that question while visiting cave paintings in an Ice Age cave in France, I literally received an answer: the cave spoke back in the form of echoes. I remembered having learned from mythic literature (e.g. Bulfinch 1855  [2000]) that, prior to recent scientific understanding of the wave nature of sound, echoes were personified. Sound reflection was explained as echo spirits living in the rocks and calling back. I realised that this dynamic nature of caves—their ability to respond sonically in a way that was perceived as supernatural—could help explain the obsession with exploring these environments and artists’ responses in the form of magnificent cave art. A literature search showed that a connection between music and cave art had been made (Reznikoff and Dauvois 1988; see also Reznikoff 1995, 2008) prior to my independent discovery regarding sound reflection and rock art. However, the theory that I proposed (Waller 1993a, b) also made a connection with ancient mythology: since sound reflection gave the illusion of noises emanating from spirits in the rocks, the cave art could represent images evoked by the sounds. A testable hypothesis is that cave art placement would correlate to acoustic properties of the cave. A corollary is that the subject matter might also be expected to relate in some way to sound. This theory of acoustics serves as a framework for understanding and testing cave art and the possible role that the sonic context may have played in the ritualistic use of caves, as well as canyons and cliffs. When understood from the worldview of ancient cultures, in which myths have captured the spiritual agency of sound reflection, the soundscape of caves can thus be recognised as a vital element in rituals to communicate with the supernatural. Some examples of myths regarding caves and echo spirits from around the world include: S. J. Waller (*) Rock Art Acoustics, 1952 Sonoma Lane, Lemon Grove, CA 91945, USA © Springer Nature Switzerland AG 2019 L. Büster et al. (eds.), Between Worlds, https://doi.org/10.1007/978-3-319-99022-4_12

251

252

S. J. Waller

Aztec  The earth and cave god called Tepeyollotl (‘Heart of the Mountain’) or Tepeolotlec (‘Lord of Beasts’) was thought to cause echoes (Spence 1913). Tsimshian  Chief Echo speaks but is invisible (Thompson 1929). North-West Coast  The Saˆa (or ‘Echo’) is a human-like being that has the ability to imitate the sound or voice of any creature. She is found in a cave near Blunden Harbour. You will know that Echo is coming because she will imitate the sounds she hears (U’mista Cultural Society 2003). California  A site called Wikwip contains rock art for which there exists ethnographic information that the paintings were made by men preparing for ceremonial dances. The site name means ‘Echo Rock’ and is derived from the sound-focusing acoustical characteristics of the cave (Hedges 1993). Cherokee  A traditional story describes how a priest would go to the ‘Cave of Sacrifice, to gather the will of the Great Spirit from the hollow voice [echo] within it’ (Jones 1830, p. 226; the interpretation of the hollow voice as an echo is included as a note in the source). Greek  Echo was a beautiful nymph. Juno passed sentence upon Echo in these words: ‘“You shall forfeit the use of that tongue with which you have cheated me, except for that one purpose you are so fond of… reply. You shall still have the last word, but no power to speak first”… Narcissus… left her, and she went to hide her blushes in the recesses of the woods. From that time forth she lived in caves and among mountain cliffs. Her form faded with grief, ‘til at last all her flesh shrank away. Her bones were changed into rocks and there was nothing left of her but her voice. With that she is still ready to reply to anyone who calls to her, and keeps up her old habit of having the last word’ (Bulfinch 1855 [2000], p. 80–81). Hawaii  The echo god lived in the hollow grey rocks (Westervelt 1915). India  Echoes have religious significance to members of an indigenous tribe of India called the Korku. This tribe continues to produce rock art today, using echoes as selection criteria for choosing which caves to paint (Somnath Chakraverty pers. comm.). Malaysia  ‘There are echo-spirits of the mountains, like men and women in shape... a female echo-spirit lived in a cave in the face of a limestone bluff, a beautiful woman called the Princess of the Rice-fields by the Hot Spring’ (Winstedt 1925, p. 25). Zimbabwe  ‘It is generally agreed that Zimbabwe’s finest readily accessible rock paintings are found at Ngomakurira or ‘Mountain of Drums’... The name is derived from an acoustic effect that is believed to represent the beating drums of festive spirits. (This effect is most in evidence in several sacred caves that are off limits to visitors)’ (Chinula and Talbot 2002, p. 161).

12  Hear Here: Prehistoric Artists Preferentially Selected Reverberant Spaces…

253

Algonquian  ‘Hand prints were left by the mischievous, mysterious Maymaygwayshiuk [Memegwashio], showing where they touched the rock as they “closed the door” before disappearing into their dwellings in the cliffs’… ‘Maymaygwayshiuk [Memegwashio] supposedly seal their magic “entrances” to the rock with hand prints’ (Furtman 2000, p. 61, 113). Echo spirits called Memegwashio dwell within the rocks and call out in reply. The EiGf-2 rock art site in Quebec, which is noted for its acoustic properties and is known as a place inhabited by these Memegwashio beings, can thus be described as an echoing dwelling place for echo spirits who painted the rocks when entering and exiting the rock through a supernatural cave portal (Waller and Arsenault 2008). Ojibwa ‘Maymaygwashiuk [Memegwashio] fishing in front of rocks with markings; could hear them speaking… Suddenly the wall with paintings opened up in front of the fairies… he reached the dark opening just as the rock began to close… songs echoing… Low sound, like a crowd of voices, could be heard around the group of paintings… The Indian glided out of the cave. As soon as he turned around to look at his hosts, he saw only a wall of rock, covered with some paintings, now silent’ (Conway 1993, p. 155–157)

12.2  Recording To test the hypothesis that cave art placement correlates to the acoustic properties of caves and canyons, the sounds in the context of the art must be carefully studied. It is somewhat ironic that modern people tend to ignore and trivialise sound reflection as unwanted noise; a distinct echo in a concert hall is considered an unforgivable flaw. Yet a researcher must try to understand ancient peoples’ spiritual perspective of echoes while simultaneously needing to understand the physics of sound reflection and the psychoacoustics of echo perception. It is interesting that sound waves reflecting from a surface are indistinguishable from sound waves emanating from a virtual sound source on the other side of the sound-reflecting surface (just like Alice perceived another world through the looking glass; i.e. no amount of testing could disprove that the image was not ‘really’ a scene observed through a window)! One must keep in mind that a distinct echo can be discerned by the brain only at an appropriate angle and position such that the sound will reflect back to the listener, and can be distinguished from the original only if there is a sufficient delay due to the speed of sound. To avoid the ‘precedence effect’, in which a loud sound masks the perception of subsequent sounds, the test location should usually be at least

254

S. J. Waller

17  m from the sound-reflecting surface, to allow for  about a 0.1  s echo delay. Reverberation occurs when there are so many echoes that they blur together into a prolonged continuation of sound that decays like rolling thunder. Initially, I conducted an informal study to test the acoustics at a large number of rock art sites in Europe, North America and Australia, to determine if echoing was a rare or a common trait of rock art sites. Progressing from perception to first analogue and then digital equipment, I made recordings of the echoes to document that the strong sound reflections I could hear at nearly every rock art site tested were not just artefacts of biased thinking. The results indicated that the theory of acoustic motivation for rock art was not only theoretically plausible but is supported by objective evidence, making studies worth pursuing further.

12.3  Digital Capture Technology and Systematic Analysis To test for a correlation of rock art placement and sound reflection, some rock art sites were tested systematically and quantitatively for comparison of sound levels at decorated versus undecorated locations. The first site tested systematically was Horseshoe Canyon in Utah (Waller 2000). It was tested and analysed with analogue equipment, but the clear-cut results (statistical significance of p