UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Rural religion at rock sanctuaries in Roman Spain : an alternative model McCullough, Kimberly 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2018_september_mccullough_kimberly.pdf [ 4.67MB ]
JSON: 24-1.0369280.json
JSON-LD: 24-1.0369280-ld.json
RDF/XML (Pretty): 24-1.0369280-rdf.xml
RDF/JSON: 24-1.0369280-rdf.json
Turtle: 24-1.0369280-turtle.txt
N-Triples: 24-1.0369280-rdf-ntriples.txt
Original Record: 24-1.0369280-source.json
Full Text

Full Text

RURAL RELIGION AT ROCK SANCTUARIES IN ROMAN SPAIN:  AN ALTERNATIVE MODEL  by KIMBERLY MCCULLOUGH B.A., York University, 2008   A THESIS SUBMITTED IN PARTIAL FUFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF ARTS  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Classical and Near Eastern Archaeology)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2018  © Kimberly McCullough, 2018  ii  The following individuals certify that they have read, recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the thesis entitled:   Rural Religion at Rock Sanctuaries in Roman Spain: An Alternative Model                              _  Submitted by Kimberly McCullough___________ in partial fulfillment of the requirements for the degree of  Master of Arts______________________________________________________ in                   Classical and Near Eastern Archaeology__________________________________  Examining Committee:  Matthew McCarty, Classical, Near Eastern, and Religious Studies________________________ Supervisor  Alejandro Sinner, Greek and Roman Studies (Uvic)___________________________________ Supervisory Committee Member  Lisa Cooper, Classical, Near Eastern, and Religious Studies_____________________________ Supervisory Committee Member           iii  Abstract Rural religion in Roman Spain continues to be misunderstood due to problematic narratives in both ancient and modern sources. In the age of Augustus, the Greek geographer Strabo put forth misrepresentations of the religious beliefs and practices of inhabitants of Hispania as a result of two main problems, the polis-religion model view and idea of acculturation. Strabo and his sources’ shared a lack of familiarity with religious expressions in the rural sphere of Roman Spain due to their narrow view of religious rites. Moreover, writers under the Roman Empire like Strabo tend to emphasize cultural transformations to the “Roman mode of life” as positive and widespread experiences, even if in reality the process was much more gradual and varied. This Strabonian meta-narrative problematizes our understanding of religious change in the region of Hispania. What’s more, this meta-narrative has lived on in modern scholarship as scholars continue to focus their inquiries into religious change largely on the urban centers of society, conceptualize religious and cultural change in terms of acculturation models such as “Romanization,” and treat religious beliefs in isolation from practice by ignoring the spatial context of epigraphic evidence.  In response to such problematic frameworks, I propose an alternative model aimed at presenting a more complete picture of religion in Roman Spain. Throughout this framework, I privilege the study of the rural sphere, trace instances of inventing traditions in rural religion, and analyze the epigraphic evidence alongside its spatial context in order to look beyond the narrow range of material covered by past scholars.   In the first chapter I apply my alternative model to the sanctuary of Panóias and demonstrate the inability of past approaches to portraying the innovative agency taking place. In the second chapter I test the applicability of interpretations of Panóias to other rock sites in Spain iv  as done by past scholars. I conclude that Panóias is not necessarily applicable as a model to other sites, although interpretations made through the application of an alternative model does drive knowledge forward by helping us understanding individual agency and the invention of tradition in rural religion in Roman Spain.   v  Lay Summary  Rural religion in Roman Spain continues to be misunderstood due to problematic narratives in both ancient and modern sources. Ancient sources misrepresent the religious practices of local inhabitants due to their lack of understanding of rural religion and bias towards portraying an adoption of Roman culture. In turn, modern scholars focus too much on the urban sphere, models of acculturation, and textual evidence. In contrast, I propose an alternative model aimed at presenting a more complete picture of religion in Roman Spain by analyzing the rural sphere, tracing inventions of traditions and analyzing textual evidence alongside its spatial context. I use this alternative model to re-analyze the rock sanctuary of Panóias and then test the applicability of Panóias as a model for understanding other sites. Overall, I conclude that Panóias is not applicable as a model to other sites, although the application of an alternative model does drive knowledge forward.       vi  Preface  This thesis is the original, unpublished, independent work by the author Kimberly McCullough. All figures are the property of their original creators, who have been acknowledged in the image captions.           vii  Table of Contents Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface............................................................................................................................................ vi Table of Contents .......................................................................................................................... vii List of Figures ................................................................................................................................ ix Acknowledgements ........................................................................................................................ xi Dedication ..................................................................................................................................... xii 1. Introduction ................................................................................................................................. 1 1.1 – Problematic Narratives in Antiquity ................................................................................. 1 1.1.1 – The Polis-Religion Model ..................................................................................... 2 1.1.2 – Acculturation Ideology ......................................................................................... 3 1.2 – Problematic Narratives in Modernity ................................................................................ 4 1.2.1 – The Urban Takeover ............................................................................................. 4 1.2.2 – The Romanization Model ..................................................................................... 6 1.2.3 – Epigraphic Bias ..................................................................................................... 7 1.3 – An Alternative Model ....................................................................................................... 8 1.3.1 – The Rural Sphere .................................................................................................. 8 1.3.2 – The Invention of Tradition .................................................................................... 8 1.3.3 – Epigraphic and Spatial Context .......................................................................... 10 1.4 – Setting the Stage .............................................................................................................. 11 1.4.1 – Terminology ....................................................................................................... 11 1.4.2 – Nature of Evidence ............................................................................................. 12 1.4.3 – Historical Context ............................................................................................... 13 2. Sanctuary of Panóias ................................................................................................................. 15 2.1 – Past Interpretations .......................................................................................................... 16 2.2 – Historical Context ........................................................................................................... 18 2.3 – Understanding the Ritual Route ...................................................................................... 20 2.3.1 – First Rock............................................................................................................ 21 2.3.2 – Second Rock ....................................................................................................... 29 2.3.3 – Third Rock .......................................................................................................... 30 viii  2.4 – The Significance of the Ritual Route .............................................................................. 31 2.4.1 – Invention ............................................................................................................. 31 2.4.2 – Tradition ............................................................................................................. 35 2.4.3 – Agency ................................................................................................................ 38 2.4.4 – Performativity ..................................................................................................... 40 2.5 – Reconsiderations ............................................................................................................. 41 3. Applying the Panóias Model ..................................................................................................... 43 3.1 – Pena Escrita ..................................................................................................................... 44 3.2 – Cabeço das Fráguas ......................................................................................................... 49 3.3 – Pia dos Mouros ................................................................................................................ 57 3.4 – Peñalba de Villastar ......................................................................................................... 60 3.5 – Panóias as a Model .......................................................................................................... 68 4. Final Reconsiderations .............................................................................................................. 70 Figures........................................................................................................................................... 72 Bibliography ................................................................................................................................. 94   ix  List of Figures Figure 1. Map of Roman Hispania circa first century C.E. with distribution of discussed case studies. .......................................................................................................................... 72 Figure 2. General Plan of the archaeological site of Panoias (labels and proposed ritual progression by author) (Sousa and Ribeiro da Silva 2013: 68). ................................... 72 Figure 3. Topographical plan of rock outcrop with five numbered inscriptions (labelled by author) and a proposed ritual progression made by Alföldy (1997: 199). ................... 73 Figure 4. First rock monument in rock sanctuary group at Panoias (prehistoriadelsur). .............. 73 Figure 5. Rectangular and circular cavities cut out of the bed rock as well as post holes (prehistoriadelsur). ....................................................................................................... 74 Figure 6. CIL II 2395a (Argote 1732: VII: 568 ff.). ..................................................................... 75 Figure 7. CIL II 2395a according to Alföldy (1999: Abb. 3). ...................................................... 75 Figure 8. CIL II 2395a photo by Hispania Epigraphica Record No. 8213. .................................. 76 Figure 9. Traced drawing of CIL II 2395a following the application of MRM (Corriea Santos, Pires, & Sousa 2014: 208). ........................................................................................... 76 Figure 10. CIL II 2395b (Hispania Epigraphica Record No. 8214). ............................................ 77 Figure 11. CIL II 2395b according to Alföldy (1999: Abb. 4). .................................................... 77 Figure 12. Gastra located above CIL II 2395b (prehistoriadelsur). .............................................. 78 Figure 13. CIL II 2395c (Alföldy 1997: 185). .............................................................................. 78 Figure 14. CIL II 2395c (Hispania Epigraphica Record No. 8215).............................................. 79 Figure 15. CIL II 2395c Traced drawing made from polychromatic MRM (Corriea Santos, Pires, & Sousa 2014: 218). ..................................................................................................... 79 Figure 16. CIL II 2395d (Hispania Epigraphica Record No. 7696). ............................................ 80 Figure 17. Rectangular cavity above CIL II 2395d used to burn entrails (prehistoriadelsur). ..... 80 Figure 18. Inscription CIL II 2395d located to the left of small stairs leading to sacrifical area (prehistoriadelsur). ....................................................................................................... 81 Figure 19. CIL II 2395d, Alföldy (1999: Abb. 6). ........................................................................ 81 Figure 20. Traced drawing of CIL II 2395d made with polychromatic MRM (Corriea Santos, Pires, & Sousa 2014: 222). ........................................................................................... 82 Figure 21. Second rock at Panóias from afar (prehistoriadelsur). ................................................ 83 Figure 22. Second rock at Panoias (prehistoriadelsur). ................................................................ 83 x  Figure 23. Second Rock with missing inscription on left and stairs in the centre (prehistoriadelsur). ....................................................................................................... 84 Figure 24. Second rock at Panoias, Ariel view. ............................................................................ 84 Figure 25. Third rock at Panoias stairs leading up (prehistoriadelsur). ........................................ 85 Figure 26. Third rock at Panoias (prehistoriadelsur). ................................................................... 85 Figure 27. Ariel view of Pena Escrita (Correia Santos 2010a: 131). ............................................ 86 Figure 28. Pena Escrita Sanctuary in Vilar de Perdizes (Rodríquez Colmenero 1995: 191). ...... 86 Figure 29. Topographical drawing of Pena Escrita (Corriea Santos 2010a: 132). ....................... 87 Figure 30. Reconstruction of the sanctuary of Pena Escrita (Correia Santos 2010b: 186). .......... 87 Figure 31. Votive altar dedicated to Larcauco (Rodríquez Colmenero 1995: 127). .................... 88 Figure 32. Sanctuary of Cabeço das Fráguas granite outcrop with inscription painted over with chalk (Olivares ............................................................................................................. 88 Figure 33. Lusitanian Inscription at Cabeço das Fráguas (Untermann 2002: fig. 2). ................... 89 Figure 34. Possible second inscription at Cabeço das Fráguas (Rodríquez Colmenero 1995: 223). ...................................................................................................................................... 89 Figure 35. Pias dos Mouros (Correia Santos 2010b: 193). ........................................................... 90 Figure 36. Plan of Pias dos Mouros (Santos Júnior et al. 1989: 379). .......................................... 90 Figure 37. Two rectangular cavities with a detailed look at the inscriptions (Correia Santos 2010b: 194)................................................................................................................... 91 Figure 38.  General view of the sanctuary of Peñalba de Villastar (Alfayé Villa and Marco Simón 2008: 283)..................................................................................................................... 91 Figure 39. "Gran Inscripción" at Peñalba de Villastar (after J. Cabré 1909-1910) ...................... 92 Figure 40. Rock depressions and channels at Peñalba (Alfayé Villa and Marco Simón 2008: 286). .............................................................................................................................. 93      xi  Acknowledgements  I would like to express my profuse thanks and appreciation to the following individuals. First and foremost, I sincerely thank my committee members: Dr. Matthew McCarty, my supervisor not only throughout my thesis, but throughout my entire time at the University of British Columbia whose unrelenting encouragement, help, and genuine passion for teaching and spreading knowledge, has helped shape and challenge me as a scholar; Dr. Alejandro Sinner, for without whom I never would have been introduced to the world of archaeology and Roman Spain to begin with and who has served as a constant mentor of mine since embarking on my first dig under his guidance three years ago; and Dr. Lisa Cooper for asking such thought-provoking questions regarding my thesis findings and providing extensive feedback. I would also like to thank my colleagues in the CNERS graduate program for their continuous encouragement and support throughout my process researching and writing my thesis, as well as general great manner in and out of the reading room.         xii  Dedication  I would like to dedicate the final compilation of the endless hours spent researching and composing this thesis to two people who have served as my pillars throughout this process: my mother and best friend, Jill McCullough, and Faraj Edher. It is their constant support and belief in my ability as a scholar that have grounded me throughout my studies and helped me achieve my goals.   I would also like to thank my father Chris McCullough for his support and encouragement without which I would not be here, my sister Kristin McCullough-Sawh and my brother-in-law Mohit Sawh for always believing in me, as well as the rest of my family and friends for continuing to support me throughout my endeavors.     1  1. Introduction  1.1 – Problematic Narratives in Antiquity Some authors affirm that the Callaeci are atheists, but that the Celtiberians and their neighbours to the north offer sacrifices, on the nights of the full moon, to a nameless god, in front of their houses, spending the whole night long with their families dancing, singing in chorus and feasting.1  In the age of Augustus, the Greek geographer Strabo provides us with a brief and problematic description of the religious beliefs and practices of inhabitants of Hispania. Strabo references the Callaeci, also known as the Gallaeci, who were a tribal group inhabiting Gallaecia in the northwest corner of the Iberian Peninsula. He also mentions the Celtiberians who were a group of Celtic peoples that occupied the more central-eastern area of the Peninsula during the late centuries B.C.E. As for the religion of these peoples, it is largely accepted that Strabo’s classification of the Callaeci as “atheist” was rather misplaced.2 Contrary to Strabo’s writings, there is epigraphic evidence validating the existence of numerous local gods. In 1981 Alain Tranoy compiled a list of at least 35 deities worshipped by the Callaeci alone.3 It can be argued that as a result of two main problems, their polis-model view and idea of acculturation, Strabo’s sources were unable to detect the religious worship of such deities.4                                                   1 Strabo, Geographia, 3.4.16. 2 Alarcão 1988: 91. 3 Tranoy 1981: 266-86. 4 Strabo largely relied on observations from travellers that often embellished the savagery of the inhabitants of foreign lands (Le Roux 1995: 8-9). Strabo, like Livy and other ancient writers, was also known to have written with the motivation of pleasing an audience that resonated with the Augustan ideology of the empire (Keay and Terrenato 2001: 120). 2  1.1.1 – The Polis-Religion Model First, the discrepancies between textual sources and material culture can be explained by the lack of familiarity of Strabo’s sources with religious behavior in Hispania. Strabo’s unnamed sources were driven by a “polis-religion” model of urban cult which inhibited them from perceiving the practices of the Callaeci as forms of religious worship.5 The model of civic religion was rooted in the notion that religious practice was performed in a specified social context.6 It also privileges the communal performance of religious praxis and thus focuses too narrowly on predetermined religious organization in the urban cults.7 As a result, the polis-religion model does not take into account the religion of the individual, or the fact that religious behavior can be expressed in varying ways and is therefore not a universally consistent phenomenon.8 For instance, religion in the Iberian Peninsula did not manifest itself in ways familiar to the Romans, who included elements such as votive inscriptions, anthropomorphic representations of deities, and Graeco-Roman temple structures in their religious practices. As a result, expressions of religion, such the embodiment of a deity invoked alongside the sacredness of the rivers or the mountains, or the practice of ritual sacrifice at rock sanctuaries or open-air nemata, were often be misinterpreted. Yet such landscape features, with varying degrees of human intervention, seem to have played an important role in the religious life of the region.                                                   5 The polis was a classical city-state under which the city’s magistrates and priests were the primary religious authorities and oversaw public interactions with the gods under what Varro described as “civic theology” (City of God, 6.5; Rives 2007: 45 ff). In the present model, polis is used to refer to religious practice performed in the specific social context of a Graeco-Roman community that was bound by ties of citizenship and place. 6 James Rives also makes the distinction between civic authority’s concern for behaviour rather than belief as it was important in the polis-religion model for a certain level of communal participation in line with the customary performance of public rituals in the Graeco-Roman world (2007: 45). 7 For more on the weaknesses of the “polis-religion” model see Kindt 2009. 8 Scheid 2016: 1. 3  1.1.2 – Acculturation Ideology Second, Strabo also goes on to characterize the extent to which inhabitants of Roman Spain were changing over to the “Roman mode of life”:    The Turdetanians, however, and particularly those that live about the Baetis, have completely changed over to the Roman mode of life, not even remembering their own language any more. And most of them have become Latins, and they have received Romans as colonists, so that they are not far from being all Romans.9  In this regard, Strabo is largely referring to the inhabitants in southern Spain along the Guadalquivir valley, whom he claims have largely let go of their previous language and culture in exchange for Latin and Roman culture. To Strabo, it appears that the defining characteristic of “being Roman” is language, for what ultimately defines the Turdetanians as “not far from being all Romans” is not remembering their own language anymore and becoming “Latins.”10 While the adoption of Roman language can be indicative of cultural and religious change in the region, the manner in which he emphasizes the transition to the “Roman mode of life” demonstrates the bias of ancient sources in representing processes of cultural change. In other words, sources partial to Roman imperialism have a tendency to emphasize cultural transformations to Romanitas as positive and widespread experiences, regardless of whether or not that really was the case. While this strategy may have worked for ancient sources, relying solely on the textual record for representations of cultural and religious change in the Roman provinces in modern scholarship remains problematic. It becomes especially distorted when textual records such as Strabo’s have come to influence modern representations of religion in Spain. For example, scholars agree that Strabo may have embellished elements in his accounts, such as the nameless god worshipped by the Celtiberians at night during the full moon,11 to emphasize the presumed                                                  9 Strabo 3.2.15. 10 Ibid. 11 Strabo 3.4.16. 4  barbarity of the local inhabitants.12 But this Strabonian meta-narrative that only primitive people did not name their deities has lived on in modern scholarship.13 For instance, F.J. Wiseman in his description of inscriptions made to divinities refers to the worshipped gods and goddesses by the local inhabitants as “streams and springs, deified in the simple, direct way, of a primitive people.”14 The hint of British colonialism of the twentieth century aside, the generic labelling of local worship of deities as “simple” and “direct” not only follows ancient ideas of acculturation, but also erroneously limits the possibilities of the worship or ritual that these misunderstood religious ideologies may have entailed.   The acculturation model, as applied by modern commentators, directly relates to the development of polis religion. In other words, as “Roman” looking cities and institutions developed in a region; likewise, civic theology is thought to have followed. However, these meta-narratives are problematic as they impose limitations onto our interpretations of how individuals of the Iberian Peninsula experienced rituals under the Empire. What’s more, these problems are not confined to Strabo, but have been internalized by contemporary scholarship.  1.2 – Problematic Narratives in Modernity 1.2.1 – The Urban Takeover The polis-religion model prevalent in antiquity has been adopted by modern scholars through their focus on urban centres as a means to understand religious change in antiquity. Local inhabitants of the rural sphere have to an extent been discounted by scholarship in exchange for a                                                  12 Greek and Roman geographers were accustomed to juxtapose their own literate culture with what they perceived and portrayed as the barbarism of others (Curchin 2004: 17). 13 Marco Simón 1999: 36. 14 Wiseman 1956: 99.  5  focus on the civic monumental sphere of society where evidence largely speaks to the agency of administrators and the wealthy.   For instance, Leonard Curchin asserts that worship by local inhabitants was likely more widespread than the modest representation in Latin that our surviving epigraphy in Roman Spain suggests.15  He argues that it was due to the lack of means by locals to commission stone altars inscribed in Latin that our evidentiary support of local participation in religion is so limited.16 Yet there are more forms of evidence that can be analyzed in the rural sphere than solely Latin epigraphy.17 But like Curchin, scholars become rather resigned when it comes to examining the periphery due to its lack of preferred evidence and monumentality. Instead they turn their attention to the more appealing urban scope in Roman Spain. However, if we look elsewhere than solely at literary sources or Latin inscriptions for inspiration, we can see that there is a bounty of evidence of religious worship in the rural sphere as some scholars have managed to bring to light.18 Overall, Roman civilization is often associated with urbanization.19 However, scholars estimate that at least 90 percent of the population was rural.20 Therefore, a look at the religious beliefs and practices performed by individuals in the Iberian Peninsula that solely focuses on the urban centre fails to understand the religious praxis enacted by the vast majority of people.                                                   15 Curchin 2004: 176. 16 Ibid. 17 For instance, Tranoy not only recorded epigraphic evidence venerating local gods in Hispania as discussed earlier, but also the presence of altars whereby it becomes clear that the majority of religious worship conducted through votive inscriptions and altars, as well as rock sanctuaries, originated in the rural sphere (Alarcão 1988: 91; Tranoy 1981: 266-86). 18 Scholars include: F. Marco Simón, S. Alfayé Villa, T. Tortosa, F. Diaz de Velasco, I. Grau, as well as E. Richert who in her study of “Native Religion under Roman Domination” examines patterns of religious veneration in the rural sphere in addition to urban centres in the northwest of the Iberian Peninsula (2005: 15 ff.). 19 Curchin 2004: 96. 20 Hopkins 1978: 75-6; Woolf 1998: 138. 6  1.2.2 – The Romanization Model In addition to evidentiary issues, such as scholarly bias towards the urban centre influenced by past polis-religion models, there are also theoretical problems stemming from past conceptualizations of acculturation. The concept of acculturation, and by extension, Romanization, has been at the centre of scholarly debate for over a century now.21 While valuable contributions to the discourse have been made in recent years, even its basic mechanisms—the agency of Romans directing the adoption of their culture onto others, the trickle-down effect whereby elites of society are first to make such changes to “become Roman” and others follow, the standardization of the process across different areas and periods of time, the assumption that if people adopted Roman culture they were “becoming Roman” and if did not, they were actively resisting, and the presumed unilateral exchange whereby the Romans inflicted their culture on others without absorbing other cultures into their own—remain largely consistent.22 These elements which form the Romanization model continue to dichotomize Roman culture with other cultures to a disproportionate degree and presuppose that the process was much simpler than it really was. Moreover, scholars’ attempt to move away from such frameworks effectively complicates an already biased concept. As noted by Woolf in his “Romanization 2.0 and its alternatives,” scholars’ dissatisfaction with past approaches inciting their attempt to introduce new frameworks, whether it is “entanglement theory,” Mattingly’s “discrepant identity,” or Webster’s “creolization,” simply offer the substitution of new words for old ones by trading past prejudices with current ones, in an act of histoire inverse.23                                                   21 By “acculturation” I am referring to the process by which one culture adopts the beliefs and/or practices of another culture as a result of increased cultural interactions. And “Romanization” here is being used to specifically refer to the adoption of Roman culture by other cultures, such as those incorporated into the Roman Empire through conquest and acquisition.  22 For the various acculturation models see: Haverfield 1905; Mattingly 2013; Millett 1990; Webster 2001; and discussions in Van Oyen 2015; Versluys 2014. 23 Woolf 2014: 47. 7  1.2.3 – Epigraphic Bias Beyond the problematic approaches of privileging the urban sphere and models of acculturation such as Romanization, a third problem is worth noting: treating religious beliefs as system that operates in isolation from practice. This is made most apparent by the tendency of scholars to treat votive inscriptions as mutually exclusive from the sacred context in which they were created. Inscriptions, as part of the material culture left behind by ancient inhabitants, are at times decontextualized in their studies and thereby taken out of the practice and interaction they were conceptually part of. In other words, the material is separated from the culture it was created in. For example, Elizabeth Richert in her work, Native Religion under Roman Domination, focuses on what she regards as “our prime source of evidence, epigraphy,” whereby the context in which the investigated inscriptions were created and the religious significance of their natural surroundings are neglected.24 This example of Richert’s work not only highlights scholarly tendency to decontextualize material culture, but also to privilege the epigraphic corpus over other forms of evidence. As James Rives admits, “Without inscriptions, the study of religion in the Roman provinces would scarcely exist….”25 While an exaggeration, it nevertheless encompasses the epigraphic bias present in studies of religion in Roman provinces such as Hispania. Thus, there is a problematic inclination in scholarship to treat both inscriptions and religious beliefs as self-contained phenomena.                                                   24 Richert 2005: 2-3. 25 Rives 2015: 420. 8  1.3 – An Alternative Model Strabo’s narrative has raised two problems: the polis-model view and the idea of unidirectional acculturation. Strabonian narratives have further influenced modern scholarship through the enduring bias towards the urban centre, acculturation models such as Romanization, and treating religious beliefs as separate from practice and interaction.  In response to such problematic frameworks, I will present an alternative model aimed at presenting a more complete picture of religion in Roman Spain.  1.3.1 – The Rural Sphere To start, observing the religious realm of less central areas of the western provinces could enable a look at what unusual and new forms of worship are flourishing under less regulated agency. While these corners of the empire may offer less monumental or extensive evidence to facilitate such studies—at least in comparison to the urban centres—that does not mean that the periphery is incapable of further study. On the contrary, the remains of rock-cut sanctuaries and votive inscriptions in local languages can improve our understanding of religion in the provinces. It allows a look into what is going on in the far corners of the empire where the spread of Roman culture and simultaneous reformations of tradition are less understood. Thus, I will examine rural religion in rock-cut sanctuaries to get beyond the polis-religion model.    1.3.2 – The Invention of Tradition Strabo’s picture of gradual “Romanization” has also had a lasting impact on scholarship of religious change in Iberia, even on accounts that privilege material remains. Unidirectional acculturation models start from an external perspective and fail to recognize the variety of ways 9  that “traditions” might operate in lived religion. Moreover, modern descriptions of continuity and change in the material record often blur the lines with the ways ancient worshippers constructed connections or breaks with the past.26 These meta-narratives of continuity and change and their variants under the umbrella of “Romanization,” are modern interpretations of past processes that are often employed uncritically when the reality is that these processes were complex and less clear-cut than continuation or change.27 For instance, at sanctuary sites in Roman Spain where ancient religious rites are carried out alongside the adoption of Roman praxis, it is unfair to classify observed hybridizations of religious behavior as strict breaks with the past to institute new forms of worship. Perhaps instead we should look at such as evidence of inventing new traditions under the guise of conservatism in order to create links both with the past and present.28  As noted by Eric Hobsbawm, “Traditions which appear or claim to be old are often quite recent in origin and sometimes invented.”29 In other words, the “invention of tradition” can denote instances in which a set of ritual practices—aimed at instilling particular expressions of behavior through repetition—are intended to imply continuity with the past while in fact creating innovation in the present.30 Henceforth, in order to unravel the inventions of traditions in a society consumed by heightened cultural contacts and the development of new mechanisms of worship, we must be critical in our studies and aware of the shortcomings of past approaches in order to avoid what Woolf deemed the histoire inverse that shadows Romanization models.31                                                  26 McCarty 2013: 3. 27 Ibid. 28 Augé 2009; McCarty 2013: 4; Versluys 2014. 29 Hobsbawm 1992: 1. 30 Ibid. 31 Woolf 2014: 47. 10  Therefore, I will use the model of inventing tradition as a way to examine the various ways continuity and change can manifest beyond acculturation.  1.3.3 – Epigraphic and Spatial Context In order to move beyond the internal belief system whereby religious beliefs are analyzed devoid of their archaeological context and ritual practices that created them, I will privilege a focus on the sacred contexts of inscriptions and experience of space.   Words inscribed into rocks in the rural areas of Roman Spain do not only denote texts prescribing ritual enactments, but also engagement with the sacred landscape they have come to occupy. Rather than analyze the epigraphic corpus as textual evidence, I will incorporate the archaeological contexts of these inscriptions into our studies. Contextualizing epigraphic finds enables us to learn more about the inscriber’s religious behaviour, such as how the natural landscape influenced the performance of rituals. Furthermore, without taking into consideration the environment in which communication with deities was enacted, we risk disillusioning the representation and significance of ritual in religious practices. Thus, by analyzing the epigraphic culture and spatial organization of rock-cut sanctuaries, we can both look beyond the narrow range of material covered by literary sources and beyond the urban, “polis religion” model that still governs most accounts of Roman religion and sees worship in the landscape as “primitive” continuities or—like Strabo—as non-religion.  In general, I will apply an alternative model using the archaeological record from rock sanctuaries in the rural sphere, and the theory of inventing traditions, to present a more comprehensive understanding of religion in Roman Spain.  11  1.4 – Setting the Stage 1.4.1 – Terminology Given that terms such as “culture” and “religion” have become amorphous concepts in scholarship, I will clarify their respective definitions for the current study. To start, “culture” has taken on diverse meanings over time, from denoting the cultivation of one’s mind in antiquity,32 to its use to define differences between civilizations and cultural “superiority” in the twentieth century.33 However, the notion of “culture” for the present study will instead be consistent with Bourdieu’s definition of habitus as elements of culture consisting of a variety of practices, ideas, and material manifestations that over time express continuity and/or a transformation of the cosmological understanding of life.34 In other words, culture will be used to express the ideas and practices that shape an individual or collective group’s way of life.35  As for “religion,” this term will be used to describe the “system of ideas and values with which human communities define their place in the socio-cosmic universe.”36 How these ideas and values are materialized in the archaeological record will also be a focus of the present study. As noted by Claudia Moser and Jennifer Knust, “religion is not—and has never been—a disembodied, abstract, or purely conceptual category, despite a widespread modern Western assumption that it should be.”37 Thus, understanding culture and religion as praxis and as material indicates that these cannot be studied as isolated phenomena—names of gods or textual descriptions—but must be examined through their material presence.                                                   32 An early example of its etymon is in Cicero’s Tusculan Disputations from 45 B.C.E. where he wrote “cultura autem animi philosophia est,” translated as: “on the contrary, philosophy is the cultivation of the soul” (2.1.5).  33 For more on “culture” as a concept stemming from Western imperialist ideas colonialism see Grahame 1998: 157. 34 Bourdieu 1977. 35 James 2015: 53. 36 Derks 1998: 1.  37 Moser and Knust 2017: 1.  12  1.4.2 – Nature of Evidence Prior to engaging with the materiality of rock sanctuaries in Hispania, it is worth untangling how rock-cut sanctuaries are demarcated, as well as the limitations that analyzing rock-cut features entail. While I recognize that a sacred space can simply be demarcated through a matter of perception enacted by worshippers without any physical form of boundaries erected, the ritual spaces I will be examining in the present case studies are large granite rocks in the landscape that have been carved and/or inscribed on by man-made means as evidence of ritual performance.38  In regards to their evidentiary limitations, there is primarily the issue of chronology. Rock sanctuaries across the Iberian Peninsula are sacred spaces made-up of rock already present in the landscape. Seeing as there is no stratigraphy given the lack of soil overlying these features, it is almost impossible to date when exactly these rock-cut features came into use and when they were abandoned. The lack of stratigraphy and soil composition also means a lack of finds besides the cuttings themselves.   Fortunately, there are certain epigraphic elements—such as the use of particular grammar or word uses limited to a particular time period—which at times enable us to roughly conceptualize when these texts were inscribed. Therefore, I will use rock-cut sanctuary sites across the Peninsula dated from the late first to early third century C.E. to trace the invention of traditions and negotiation of religious identity spurred in part by Roman conquest (Figure 1).39 Focusing on such often neglected sites allows not only for a more complete picture of the religious life of the Iberian Peninsula, but a better understanding of the dynamics of religious change between the late Iron Age and the period of Roman domination.                                                    38 See Derks 1998: 132 ff. for more on the definition and demarcation of cult places such as temples and sanctuaries. 39 Alfayé Villa and Marco Simón 2008: 300. 13  1.4.3 – Historical Context Last of all, in order to understand religious change in Roman Spain from the late Iron Age to the imperial era, we must recognize its social and historical context, such as the extent of cultural interactions and how they varied from region to region.  Roman Contact Cultural interactions in Roman Spain were originally facilitated through commercial exchange as the first archaeological evidence of contact between inhabitants of northern Hispania and the rest of the Mediterranean dates to the late sixth to fifth century B.C.E.40 Afterwards, the first Roman soldiers arrived in Hispania in 218 B.C.E. at the start of the Second Punic War. Decades of warfare between the Romans and Carthaginians followed and led to a rather thorough entanglement of Roman troops in the Iberian Peninsula.41 After nearly two centuries of warfare the Iberian Peninsula was brought under subjugation by the Romans. What this meant for the inhabitants of the Peninsula in its simplest form was a centralized framework of government, exploitation of mineral and agricultural resources, and integration of Roman veterans into the landscape.42 Less simply put the intensified contact with Romans that culminated in the late Republic and early imperial era led to cultural and religious influences that can be traced in varying degrees across the sacred landscape of the Iberian Peninsula.                                                  40 Keay 1988: 14; Sinner 2015: 7. 41 Kulikowski 2013: 514. The presence of Roman troops in the Iberian Peninsula included the indigenous uprising in 197 B.C.E. subdued by the Roman senator Cato the Elder (Livy, History of Rome 34) where he also organized the eastern and southern parts of Hispania being organized into two provinces, Hispania Citerior and Ulterior. Fighting also reached another peak during the Celtiberian Wars of the 130s B.C.E. where campaigns moved further into the interior of the Peninsula. Battle also continued during the wars of Sertorius from 82-72 B.C.E., until finally culminating in final conquest by the Romans from 61-19 B.C.E. when Julius Caesar intervened to restore order after resurgence, followed by Augustus suppressing the rest of the turmoil upon coming to power in the late first century B.C.E.; For more on Roman conquest see Keay 1988: 25 ff; Wiseman 1956. 42 Keay 1988: 47.  14  Regional Discrepancies While the extent to which Roman contact brought about cultural nuances in Hispania is difficult to say, it is even more challenging to answer when one considers the range of intensities of interaction across different regions. Roman armies did not penetrate the Iberian Peninsula until the late third century whereby Roman conquest and provincialization of the area was a gradual process that varied across different regions. For instance, the archaeological record reveals that interactions were heightened in the south and east of Hispania seeing as urbanization, Punic and Hellenistic models of cult worship, and the use of votive inscriptions were already present in these regions prior to Roman conquest.43 In turn, worship of Roman gods is more prevalent in the south than in the north due to a more intensified contact not only with the Romans, but earlier contact with the Phoenicians, Greeks and Carthaginians.44 Thus, the process of socialization into a pan-Mediterranean pantheon and koine of practice began much sooner and thus ran deeper in these regions.  Religious praxis also varied due to the vast cultural differences between inhabitants of the coast and interior, river valleys, mountain chains, and other natural phenomena which altered the way in which they were adopted and admitted. Therefore, spatial distribution of cultural and/or religious praxis was by no means standardized across the natural and sacred landscapes of the Iberian Peninsula.45 A lack of standardization across the landscape becomes even more apparent when one looks at the rural sphere in these regions as their distance from centres of provincial administration arguably enabled more private innovation and agency to invent traditions.                                                   43 Kulikowski 2013: 515. More specifically, scholars such as Alarcão argue that the deeper penetration of Roman culture in the south is seen with the adoption of Latin nomenclature, mosaics, sculpture, villas and public building endeavors, while almost the opposite level of penetration is evident in the Northwest (1988: 98). 44 Alarcão 1988: 98. 45 Derks 1998: 158. 15  2. Sanctuary of Panóias The sanctuary at Panóias consists of three granite rock outcrops with cavities carved out of them for ritual sacrifice and five inscriptions describing the dedication of the site by a private senator, the deities to whom the site is dedicated, and manner in which the dedication should be fulfilled (Figure 2).46 Given its rich epigraphic evidence and extensive rock-cut features manipulated to form an unusual sacred space in this peninsular corner of the Roman Empire, the sanctuary of Panóias offers an unparalleled site for analysis (Figure 3).47 As a result, the rock sanctuary has come to act as a model in scholarship for understanding the religious praxis of other rock sanctuaries across the Iberian Peninsula that present less epigraphic and contextual evidence than Panóias. However, applying Panóias as a model to other rock sanctuaries across the Peninsula is perhaps more problematic than beneficial as the agency and mechanisms of worship at each sanctuary warrant their own specific analysis and consideration.  Re-analysis of the site at Panóias can also demonstrates the inability of traditional Romanization paradigms to portray the innovative agency of individual worshippers, or at least those who share in enough wealth and power to dedicate sanctuaries and facilitate new technologies of worship. Therefore, I will examine the applicability of Panóias as a model for understanding agency and the invention of tradition at rock sanctuaries in Roman Spain. To start, I will explore past interpretations of Panóias, contextualize the site, and then trace its evidence of ritual progression and innovative traditions driven by a private senator.                                                  46 The use of cavities for ritual purposes was to an extent customary given that they are further attested in other areas of Galicia such as in Sanca Marina das Aguas Santas (Orense) with a group of three basins, the Pia da Santa, or the Pia dos Mouros that will be discussed in the subsequent section, located near the castro of Cameixa, as well as the Fonto do Largarto to name a few (Tranoy 1981: 338 ff).  47 Cortez 1947: 8. 16  2.1 – Past Interpretations The rock enclave at Panóias was first recognized in 1721 by Rodrigues de Aguiar, a parish priest of Vale de Nogueira.48 However, the earliest and most extensive record of the site was published by Jerome Contador de Argote over a decade later in 1732.49 Although data was compiled without Argote ever having visited the site, his work has gone on to influence later analyses.50 While at first glance his work can be commended for its unparalleled attention to detail, later records reveal that many liberties were taken with his presentation of data from Panóias, including inaccurate measurements, records of epigraphy, and drawings.51 The lack of accurate interpretations of eroded rock inscriptions during the transmission of epigraphic texts by Argote’s team has been rectified over time by other scholars who have visited the site.52 As for the drawings, Argote’s illustrations were made at a later time by a draftsman whose work clearly demonstrates the intent to simplify and idealize certain aspects of the site.53 Thus we are confronted at times with poor technique and artistic liberties taken at the expense of objectivity in some instances in the early records of Panóias.  The scarce preservation of the site has also problematized an ample understanding of the space. For instance, remnants of explosions at Panóias demonstrate that the nineteenth century saw to the destruction of parts of bedrock for the purpose of extracting building material.54 The site has seen further disruption due to work carried out to level the ground and build surrounding                                                  48 Corriea Santos, Pires, & Sousa 2014: 199. 49 Argote 1732; Alföldy 1997: 228; Corriea Santos, Pires, & Sousa 2014: 199-223; Tranoy 1981: 336. 50 Alföldy 1997; Corriea Santos, Pires, & Sousa 2014; Cortez 1947; Tranoy 1981. 51 See Alföldy 1997: 176 ff for a more comprehensive correction of erroneous records from the earliest accounts of Panóias; as well as Rodríquez Colmenero 1999. 52 For instance, engineer JH von Hafe visited the site in person in 1883 and correctly deciphered various words that were previously misunderstood by Argote—such as reading IMMOLANTAR instead of IMMANTVR, SANGUIS instead of SANTVS to name a few (Alföldy 1997: 186). Other scholars also speak to the challenges in interpreting the ritual formulas of worship decreed to the divinities of Panóias given the illegibility of the inscriptions (Corriea Santos, Pires, & Sousa 2014: 199; Cortez 1947: 60; García y Bellido 1967: 135). 53 Alföldy 1997: 184.  54 Alföldy 1997: 212; Corriea Santos, Pires, & Sousa 2014: 199. 17  residences, as well as causes from exposure to poor weather over the centuries.55 Thus, the preservation of the remaining three rock outcrops is increasingly at risk of further determent.56  While flawed documentation techniques and a lack of preservation of the rock features have made comprehensive analysis of the site challenging, a wealth of evidence remains to enable further interpretation. In light of the persisting challenges to fully understanding the use of space at Panóias, the work of Geza Alföldy remains one of the most thorough analyses of the sanctuary to date.57 Correia Santos’ attempts at reinterpreting the site as well as reconstructing the inscriptions using new methodologies also further enhances our understanding of such a complex cultic space.58  In general, the consensus in scholarship has been that Panóias represents a site of “indigenous” origin reclaimed in the Roman empire by a Roman senator.59 This interpretation remains problematic as it follows the traditional acculturation trajectory, as well as presupposes hasty dichotomizations between “indigenous” and “Roman” religious veneration and will thereby be challenged in the subsequent analysis.60                                                      55 Tranoy 1981: 336; Tranoy 2004: 91. 56 Vasconcellos 1905-13: 346. 57 The significance of Alföldy’s contribution is further recognized by Corriea Santos, Pires, & Sousa 2014 as Alföldy’s work (published over two decades ago) attempts to present a corrective model to past misinterpretations, whilst also supplying his own—and at times also rather imaginative—interpretations which remain largely accepted until now. 58 Corriea Santos, Pires, & Sousa 2014. 59 Tranoy 1981: 338 ff. 60 Examples of such works following the traditional Romanization model include Alvar 2002: 205 whereby the appearance of eastern deities at Panóias are seen as “vehicles of Romanity”; see also Sousa and Ribeiro da Silva’s interpretation of Panóias as a Serapeum and “harmonious combination between native cults and Roman cults and the cult of Serapis” (2013: 7). 18  2.2 – Historical Context The Sanctuary of Panóias is located in Vila Real in ancient Lusitania, modern Northern Portugal. It resided in the northwest of the Roman province of Hispania Citerior, within the Gallaecia region in the Conventus Bracaraugustanus, approximately 50 km from the Roman municipium, Aquae Flaviae (Chaves).61  The sacred space is also located in an area far from the coastal region and urban centres of Hispania Citerior. It was situated on the border of territory belonging to the Celtic tribes Astures and Galleci.62 To the east of the sanctuary where there is a rather characteristic and elevated topography, ceramic fragments and select testimonies have provided evidence of inhabitation.63 In addition, Fernando Russell Cortez in the mid twentieth century discussed in part the pre-Roman clans dwelling in the area surrounding Panóias.64 He relied on inscriptions to conclude that a pre-existing political and economic structure survived into imperial times. Local inhabitants in the outskirts of Roman Spain were able to also maintain their local autonomy through continued practice of social customs and ways of living. In other words, it has generally been agreed upon that the area maintained its rural character, with no major urban centres located nearby.65  Yet the sanctuary itself seems to suggest a different picture. The sanctuary as it survives was heavily renovated or built new by a Roman senator, Gaius Calpurnius Rufinus, named in all of the inscriptions that will be discussed below. Given his engagement here as a private citizen, and the lack of evidence for the engagement of any civic or tribal institutions, it seems most                                                  61 Alföldy 1997: 176; also see Tranoy on the neighbouring mining sector and its proximity to Panóias (2004: 90). 62 Castiglione 1970: 100. 63 Cortez 1947: 9-10. 64 See Cortez 1947: 9ff for more on prior inhabitation in the area nearby Panóias.  65 Alarcão 1988; Blázquez 1962; Cortez 1947; Rodríquez Colmenero 2000: 187. 19  probable that the sanctuary was on land that belonged to Rufinus.66 In other words, the territory presumably formed part of a senatorial estate; like much of the Iberian Peninsula, whatever land-owning and land-working arrangements had been in place in the Iron Age gave way to an elite land-grab by the High Empire.67 Russell Cortez’s evidence for continuities, then, simply suggests that prior residents of the land stayed, perhaps working as tenant farmers for Rufinus or the conductores who presumably leased parcels of his land.    Rufinus’ alteration to the territory could have been motivated by the fact that the site at Panóias occupied a prime location within this part of Hispania Citerior. The sanctuary’s relatively visible and accessible position within the region can primarily be exhibited through the height and monumentality of the three granite rocks.68 While the three surviving rocks feature visible accessibility to onlookers, it remains ambiguous how the rest of the site related to them as Argote’s eighteenth century records indicate that the religious complex originally consisted of eleven rocks used as cult space.69 However, neither a cohesive nor an extensive map exhibiting the topography of the holy district of Panóias was published by either Argote or Cortez. As a result, obstacles remain when it comes to trying to fully assess the spatial relation of the eroded granite rocks at Panóias.70                                                  66 An example of private investment in innovation is when another senator, Pliny the Younger, asks the decurions to assign him a plot of land in order to erect a temple at his own cost and for his own predilections (Pliny the Younger, Letters 10.8.); For more see Scheid 1996 and Revilla Calvo 2002: 216-8. 67 While in theory all conquered land belonged to Rome, in practice there were individuals and families who occupied the arable land as owners or tenants. But Roman law saw to the sale of land to others whereby plots were often purchased by foreign elite and therein large private estates spread throughout the region (Curchin 2004: 98). 68 Although I examine the three rocks in relation to one another, in line with popular belief in scholarship on the sacred precinct and its relation between the rocks, Lambrino has argued that this sacred space did not extend to the others but rather they served different ritual purposes for other cult followers (Lambrino 1954: 65). Sousa and Robeiro da Silva have also argued that past cults operated independently of Roman cults (2013). However, given the overwhelming similarities among the depressions in the three rocks, as well as the succeeding height of the third, I position my study on the belief that the rocks were related. Regardless, even if the sacred complex did only consist of the first rock, the sacrificial nature of this ritual and its dedications, as well as its followers, remain significant. 69 Alföldy 1997: 228; Argote 1732: VII; Blázquez 2001: 200; Corriea Santos, Pires, & Sousa 2014: 223; Tranoy 1981: 336. 70 Castiglione 1970: 100. 20  2.3 – Understanding the Ritual Route Past interpretations of Panóias can be problematic to our understanding of the ritual space. However, further analysis of the spatial layout of the site and the progression of ritual rites that it prescribes will demonstrate the significance of agency and the invention of tradition at rock sanctuaries in Roman Spain. For instance, tracing the progression of the ritual space, as opposed to simply reading the inscriptions as texts, demonstrates how each space had a different use, and the performance of rituals could set precedents for new traditions to arise.  In order to understand how the space came together to create such ritual performativity, it is necessary to consider the “social logic” of the sanctuary.71 While the topography of the site and relation of the various rocks remains difficult to understand, due to the destruction of the other eight remaining rock sites in the area, the sequence of the three remaining rocks, as well as their inscriptions, appear to signal a predetermined route given their close proximity and sequential ascension up the summit of the hill (Figure 2).72 The route begins with an approximately 1.5 m wide ancient path, partly carved into the rock, which leads up to the summit of the hill where an onlooker would have approached the first large boulder.73 A pool of water is also visible and in close proximity to the beginning of the road.74 While we know that water played an important role in religion, the purpose of this source of water, whether ritual for its use in lustration rites, or practical for cleansing purposes following sacrificial rites, remains debatable.75 Perhaps given the presence of lustration rites in                                                  71 Fisher 2009: 440. 72 Correia Santos, Pires, & Sousa 2014: 223. 73 Alföldy 1997: 178, 199; Sousa & Ribeiro da Silva 2013: 67. 74 The closest natural water source from the cult district is the Rio Corgo which is within walking distance, though not within immediately proximity at approximately seven kilometers away. However, given the practicality of human needs, alongside the ritual requirements, readily accessible water was a necessity in a cultic site. 75 Lustration rites have been regarded as a rather universal aspect of religious rites, seeing as the desire for appropriate purification prior to engaging in spiritual activity was, and remains, common in many cultures (Richert 2005; Wild 1984: 129). At Panóias, lustration rites are believed to have served purification purposes for those 21  Hispania Citerior, as well as the relatively close proximity to a source of natural water, it is more likely to assume that the water may have served ritual purposes. If that was the case, a prospective cult follower would fulfil an act of self-cleansing via the “purification pool” before continuing on the carefully carved out path towards the rest of the sanctuary complex.  2.3.1 – First Rock CIL II 2395e After completing potential purification rites, the individual would then continue north on the path towards the first granite rock of the series (Figure 4). As one approached the rock the individual would immediately come across the first inscription located on its eastern face, CIL II 2395e. This inscription was inscribed in Latin and located just south of four corresponding cavities carved out of the rock (Figure 3). While the text was destroyed in the nineteenth century, it lives on in Argote’s records.76 Although given the aforementioned liberties taken with Argote’s records, the validity of the text will be treated with caution.  According to Argote, the lost inscription read: Diis(!) [deabusque templi] / huius. Hostiae quae ca/dunt hic immolantur / exta intra quadrata / contra cremantur / sanguis laciculis iuxta / superfunditur / [G(aius!) C(- - -) Calp(urnius) Rufinus v(ir) c(larissimus)]77                                                                                                                                                               intending to be initiated into the cult (For more discussion on self-cleansing as a precursor to initiation rites see Alföldy 1997: 200). We also see a similar practice at Greek precints where basins of water were positioned prior to entering a sacred area so that individuals wishing to perform religious rites were purified prior to participation (Wild 1984: 130). However, we must also consider that there are many practical reasons as to why participants might have been in need of water, or frankly, sanitation, given that inscriptionary evidence points towards sacrificial rites taking place at Panóias. Obviously the slaughtering of animals for ritual purposes would have been a rather messy affair and water nearby to help clean up would be greatly appreciated by ritual participants (Blázquez 2009: 225-6). 76 Contador de Argote 1732: VII. 77 CIL II 2395e. 22  This can be translated as: “To the gods and goddesses of this sacred precinct. The victims which fall here, they are sacrificed. The viscera are burned in the square units opposite. The blood is poured nearby in the small cavities. [Built by] by Gaius C. Calpurnius Rufinus, member of the senatorial order.”78 While there is a lot to unpack here, it is worth understanding how an onlooker would have experienced, rather than simply read, this text.  At first glance, the text stipulates not only that sacrifices were made at the sanctuary, but also that they had a prescribed order whereby animals were to be killed, burned, and drained of blood. More specifically, “hostiae quae cadent” indicates the victims are killed, “hic immolantur…” designates that they were sacrificed on the spot, “exta intra quadrata contra cremantur” specifies that the entrails were burned in the front square cavities, and “sanguis laciculis iuxta superfunditur” directs the blow flow into the smaller cavities nearby (Figure 4).79 After reading the final instructions outlined in the inscription, a ritual participant would turn their attention to the square units in front where the described sacrifices would be carried out prior to continuing through the circuit of rock inscriptions.80 More specifically, the individual would execute the sacrificial victims in the larger rectangular basins as instructed and then burn the entrails in the adjacent square basins. The correspondence of the texts to the carved out spaces at Panóias indicate that these inscriptions were meant to be experienced and acted out, not simply read and forgotten. The use of the preposition contra also provides a concrete, almost deictic, spatial reference, creating a strong relationship between the text and the rock depressions in close proximity to them. This                                                  78 While Alföldy’s translation of CIL II 2395e remains most accepted in scholarship, translation remains debatable. I will not delve into the debate here but for more information regarding the varying translations made by scholars ranging from Vasconcelos to Lambrino see Corriea Santos, Pires, & Sousa 2014: 204-5. 79 Tranoy 2004: 95. 80 CIL II 2395e. 23  correlation is further reinforced by iuxta. Thus, even the language gives the rites a strong spatial character that cannot be comprehended when the inscription is excerpted from its location. In addition to directing the onlooker to the performance of sacrificial rites, the first inscription also reveals the consecrator of the site, Gaius C. Calpurnius Rufinus, member of the senatorial order, as well as demonstrates his intent to instruct the worship of varying deities at the site. Thus, the introductory inscription to the site not only points towards the significance of ritual sacrifice carried out in the cavities nearby, such as referencing the hostiae and managing where the victims will be immolantur and cadunt, or sacrificed and killed, but it also informs onlookers of the deities worshipped there, and the agency of the consecrator.81 After reading the first inscription and presumably carrying out said rites, an individual would then come across the series of texts inscribed at four different points along the sides of the first rock with one on the north face, and the subsequent three inscriptions on the western rock face (Figure 3 and 5).  CIL II 2395a The second inscription a ritual participant would come across is located just south of where the individual would have performed sacrificial rites (Figure 3). This inscription measures 50 cm high and 110-120 cm across and was inscribed directly below a rectangular cavity (Figure 6). Argote’s record of the text reads: Diis Severis locatis in hoc / Templo :::::: / Gneus Caius Calpurnius Rufinus.82 This inscription has been translated as “GC Calpurnius Rufinus dedicated this (temple) to the Severe gods, who dwell in this temple.”83 However, Alföldy challenges the validity of this recording due to the aforementioned liberties taken with Argote’s records and so he offers an                                                  81 CIL II 2395e. 82 CIL II 2395a; Contador de Argote 1732: VII.568.  83 Contador de Argote 1732: VII.568. 24  alternative “Diis(!) Seve[r]is in hoc / templo lo[ca]t[i]s” which can be translated as “For the severe gods in this temple” (Figure 7).84 Though as noted by Vasconcellos in the early process of interpreting these inscriptions, it is evidently difficult to provide a translation that satisfies all, especially given the heavy erosion of the text (Figure 8).85  Yet, Maria Correia Santos and her team have recently provided a more convincing transcription of the texts given their application of the Morphological Residual Model (MRM).86 This latest interpretation reads: “Diis Serapidi Isidi / Diis Deabus Omni/bus lacum et [hanc?]/aedem G(aius?) [C(?) C]alp(urnius) Ru(finus) v(ir) c(larissimus)” (“To the gods Isis [and] Serapis / To all the gods [and] goddesses. Gaius C Calpurnius Rufinus, senator, [dedicated] the lake and this temple”) (Figure 9).87 The new reading of CIL II 2395a presents a more comprehensive translation of the text by transcribing a line previously unknown to us where deities of Eastern origin, Serapis and Isis, as well as the various gods and goddesses in the area, are venerated by the patron Calpurnius Rufinus.88 Through this methodology Correia Santos was also able to more precisely distinguish the theonyms in the first line of the text which read “Diis Serapidi Isidi,” instead of “Diis Sever[r]is in hoc” as put forth by Alföldy.89  Not only does this inscription reveal a formerly unidentified dedication to Serapis, Isis, and other deities, but also that the dedication was accompanied by the offering of a lacum. The                                                  84 This reinterpretation of the inscription comes from later observations made of the letters inscribed on the stone. For further discussion on the proposed reading of the second inscription by Geza Alföldy see 1997: 189-92; 231-3. 85 Vasconcellos 1905-13: 468. 86 The MRM method enables the detection of subtle differences not observable to the human eye due to erosion or other circumstances as it uses a process of segmentation of the stone’s morphology at varying scales. Like the majority of methodologies, this one is not without its flaws. The process of recording lines through drawing may maintain a level of subjectivity, but according to Correia Santos the readings presented in her article were based on records of residue and not interpretive measures (Corriea Santos, Pires, & Sousa 2014: 197, 202-3); For more on the theoretical framework of the MRM and its application at Panóias see Corriea Santos, Pires, & Sousa 2014. 87 The end of line three has proven impossible to distinguish even with the application of MRM but has been restored as “hanc” after taking into account the text that follows. Lines four and five of the inscription have also not been reinterpreted as much as the previous readings of this text has been confirmed as accurate (Corriea Santos, Pires, & Sousa 2014: 205-207). 88 Corriea Santos, Pires, & Sousa 2014: 209, 221. 89 Corriea Santos, Pires, & Sousa 2014: 221. 25  lacum is the basin located directly above the inscription, thought to have contained blood from the sacrifice, or possibly water for purification or washing purposes after killing the victims.90 CIL II 2395b After sacrificing the victims and pouring their blood into basins, the ritual participant would turn the corner towards the western face of the rock outcrop and come across the third inscription, CIL II 2395b (Figure 3).91 As seen in Figure 19, the cavities located above the inscriptions are in relatively close proximity to one another. This inscription was also carved below the rectangular cavity and is almost perpendicular to the cavity of similar dimensions featured above CIL II 2395a. Although the state of the inscription remains highly eroded (Figure 10), the morphological residue model performed by Correia Santos and her team has substantiated past interpretations of this inscription (Figure 11).92 The inscription reads: Diis(!) deabusque aeternum lacum omni/busque numinibus et Lapitearum cum / hoc templo sacravit / G(aius!) C(ornelius?) Calp(urnius) / Rufinus v(ir) c(larissimus) / in quo hostiae voto / cremantur93  It can be translated as: "For the Gods and Goddesses, as well as for all the deities of the Lapiteae, senator Gaius C. Calpurnius Rufinus consecrated a lake with this sacred precinct for all times, in which the victims are burned by vow." Once more, we have the familiar diis deabusque initially regarding the various gods and goddesses the space is dedicated to, followed by a reference to the deities of the “Lapiteae,” direction of the ritual rites, and signed by the consecrator and vir clarissimus, Gaius C. Calpurnius Rufinus. The inscription also specifies “in quo hostiae voto / cremantur,”94 which directs the burning of sacrificial victims presumably in the corresponding                                                  90 Garcia y Bellido 1967: 135. 91 The epigraphical field is 66 x 150 inches with the height of letters at 7-5 by 7 inches; Hispania Epigraphica Record No. 8214. 92 Corriea Santos, Pires, & Sousa 2014: 211. 93 CIL II 2395b. 94 Ibid. 26  cavity carved out above the inscription. This cavity is known as the gastra (Figure 12). It consists of two small holes on each side of the upper edge, interpreted as inserts for grates to burn the animal, a hypothesis supported by the presence of interior burn marks.95 The rock surface of the such cavities also maintain traces of moulding, as well as post holes as seen above the inscription, which are thought to have facilitated the use of wooden concealments or other semi-fixed structures to cover the cavities. The need for concealing cavities could also point towards the use of some of these spaces to store ritual instruments such as tools or statues used in the rituals.96 What is more, the relatively smooth surface of this bedrock, absence of any confirmed system in place to drain rainwater, and reference to an “aedem” in CIL II 2395a, has prompted scholars to believe that there may have been a roof covering or possible temple structure built there.97 Although what form these structures may have taken remains a mystery. In addition to coverings, the sacred complex would have needed other organizational spaces, such as: somewhere to host a ritual feast that presumably took place after the offering of the animal sacrifice was made to the gods, a place to store the animals prior to their execution, somewhere to purify oneself following the sacrificial rites, and spaces where the ritual participants could gather to witness such sacrifices—if the rites were in fact performed collectively or with an audience. However, these elements of the spatial organization of Panóias remain a mystery as the evidence does not corroborate where they may have taken place.                                                  95 Alföldy 1997: 207; Castiglione 1970: 101; Contador de Argote 1732: VII.569; Garcia y Bellido 1967: 135; Tranoy 2004: 95; while scholars maintain the argument that the burn marks indicate their use to incinerate the sacrificial victims, burning post-antique use cannot be ruled out. 96 Alföldy 1997: 211. 97 Alföldy 1997: 210; Castiglione 1970: 101; though largely overlooked by later sources, Vasconcellos from 1905-13 argues the possibility of a system to drain rainwater as he references a visit made by engineer JH von Hafe in 1883 whereby he says there was a groove by the edges of the cavities intended to prevent rainwater from entering the vats (472) though wear over the years has prevented further confirmation of such a system originally being in place. 27  CIL II 2395c Continuing along the west face of the first rock outcrop a ritual participant would come to view the fourth inscription, CIL II 2395c. It is located directly in the middle of the previous inscription CIL II 2395b and just before the fifth and final inscription CIL 2395d (Figure 3). Visually, this inscription stands out from the others for two main reasons. First, the surrounding border carved out of the rock is circular instead of rectangular like the others (Figure 13).98 This may have something to do with the corresponding cavity above the inscription. While all of the aforementioned inscriptions correspond to rectangular rock depressions above them, the shape of the cavity above CIL II 2395c is distinct, as like its text, it is circular (Figure 5). The other distinguishable feature of this inscription is its script. Unlike the other inscriptions which are transcribed in Latin, CIL II 2395c fuses Latin with Greek (Figure 14): ᾽Υψσ(τω) τῶ Σε/ρἀπιδι σὺν ἄν Κὀ/ρᾳ καὶ μυστα/ρἰοις G(aius!) C(- - -) Calp(urnius) / Rufinus v(ir) c(larissimus)99  This can be translated as “To the highest Serapis, together with Destiny and the Mysteries, consecrated by GC Calpurnius Rufinus, most renowned.”100 Correia Santos’ reinterpretation (Figure 15) largely validates the readings done by past scholars, such as Alföldy who observed the section in situ.101  The bilingual nature of the text has also been a fascination of scholars as CIL II 2395c is composed in Greek while only the dedicator’s name is in Latin. Tranoy argues that Rufinus                                                  98 Tranoy 1981: 337; While it has not been explored why the outline of this text is formed differently than the other seemingly standard form exhibited on the same rock, perhaps it was intentionally meant to designate this text as distinct from the others or simply the handiwork of another inscriber at an earlier or later time. The significance for the form of this text in relation to the others is hard to assess given there is no other evidence that the script style or form is any different from the others. 99 CIL II 2395c. 100 García y Bellido 1967: 133. 101 Although Correia Santos does point out some slight peculiarities in the text, such as the use of the Hypsistos epithet which was a formula attested in an inscription from Leptis Manga that also became frequent in the east following the start of the imperial era (Corriea Santos, Pires, & Sousa 2014: 213-4). 28  likely ensured he left his name in Latin so that inhabitants could understand who dedicated the text, given that Greek was not a common language in these regions.102 As for the reasoning behind using a language so uncommon in this region, scholars have argued a range of motives.103 While hypotheses range from using Greek as a status symbol to honouring a Graeco-Egyptian deity in its mother tongue, the reasoning behind using Greek in this inscription, as well as the function of the circular vat it corresponds to, remain unclear. CIL II 2395d And finally, a participant would continue southwest to the last inscription of the group, CIL II 2395d, featured under a comparatively larger rectangular cavity where a sacrificial mixture is presumed to have taken place (Figure 16 and 17). This engraving was also inscribed next to a small staircase that provided access to ascend the rock (Figure 18).104 The inscription reads:  Diis cum aede / et lacum qui / voto misce/tur / G(aius) C(- - -) Calp(urnius) / Rufinus v(ir) c(larissimus)105  This dedication can be understood as “To the gods GC Calpurnius Rufinus [dedicated] a pool mixed out of his vow along with a temple.” The mentions of lacum and miscetur in the text have been taken to refer to the rectangular cavity in close proximity to the inscription where part of                                                  102 Tranoy 2004: 95. 103 The partial employment of Greek could demonstrate the distinct culture of the individual making the dedication, as Rufinus effectively demonstrates his knowledge of the Greek language by employing it. Besides acting as a status symbol, Rufinus could be paying due respect to a deity of partially Greek origin by invoking the language of the Greeks (Lambrino 1954: 65). It is also arguable that the Greek text was meant to be read by the gods rather than the worshippers of the gods. Sanctuaries could act as spaces that embodied the deities they honoured, similar to the Graeco-Roman belief of gods living in their temples, and so Rufinus further honours Serapis by using a script understood to be one of the deities’ mother tongues. A clue from the inscription is the use of the epithet “highest” in reference to Serapis. This designation is frequently found among inscriptions to the Greek god Zeus, as Serapis was known to have been identified with the father of the gods and deity of the heavens. The association with a Greek god could further explain why Greek script was exercised at this stage, although the exact reasoning behind such a choice remains unclear (Cortez 1947: 58; Vasconcellos 1905-13: 346). 104 Tranoy 1981: 337. 105 CIL II 2395d; With an epigraphical field of 73 by 75 cm and the letters as high as 6-8 cm, this inscription was found in situ located to the right of two other inscriptions, on the north side of a boulder that consists of four other inscriptions (Alföldy 1997). 29  the blood ritual may have taken place (Figure 19).106 Otherwise, Correia Santos’ morphological methodology was only able to recognize minor corrections, such as confirming the presence of “aede” at the end of the first line, and the previously debated presence of a “t” in “voto” in line 3 (Figure 20).107 Overall, this inscription consists of an ambiguous mention of “diis,” the indication of a temple structure with “aede,” and the direction of rituals rites to be carried out in the nearby cavities cut out of the bedrock. After reading the final instructions provided in CIL II 2395d, a ritual participant would face a small set of stairs leading up to the top of the first monument of the group. From there, it becomes even clearer how the cavities carved out of the rock outcrops correspond to their aforementioned texts, such as the mention of hostiae, informing us of the victims awaiting sacrifice at the sanctuary, the viserca and their cremantur in different cavities, as well as the act of miscetur in the various lacum.  2.3.2 – Second Rock After reading the texts and progressing through the ritual rites, an individual would have continued to ascend the hill towards the second large granite rock located about 40 m southwest (Figure 2 and 21).108 Like the first, this rock also shows signs of previous semi-fixed structures in place, namely architectural vestiges such as columns and/or walls, as well as a roof to shelter it (Figure 22).109 Also similar to the first rock, the second shows remnants of an inscription flanking steps to the top of the monument (Figure 23). Thus, the complexity of the ritual                                                  106 García y Bellido 1967: 135; Cortez also speaks to the possible mixture of blood and wine as a symbolic and ritual drink of initiates (Cortez 1947: 53). 107 Corriea Santos, Pires, & Sousa 2014: 221. 108 Alföldy 1997: 213. 109 Tranoy has argued that it is not necessary to assume a full structure was in place to shelter the rock outcrop as perhaps a roof was not necessary, but rather there was a simple wall of protection closed by a door seeing as the site served as a open-aired sanctuary with rocks acting effectively as altar spaces (1981: 337). 30  performances here could have been even more intricate given the appearance of another inscription too eroded to have ever been deciphered by scholars. But the similarities to the first stone space do not end there as there are also various rectangular depressions in the rock, as well as square cavities and a small round opening which likely functioned as an anchor for a post (Figure 24).110 The cavities varying in size and shapes are almost identitical to those of the first rock monument and so presumably performed similar if not the same functions: ritual ablutions, burnt offerings, and possibly even the preparation of ritualistic meals.111 Alföldy hypothesizes that given their dimensions and lack of depth, they also served as shafts for the insertion of altars or possibly even statues.112 There was also a large shallow square basin nearby next to a square shaft providing outflow for the water that was presumably drained from there (Figure 24).113 While the absence of epigraphic explanation like we saw at the first rock makes the functions of these cavities difficult to discern, the similarities with the first monument indicate a continuation of ritual sacrifices.  2.3.3 – Third Rock Finally, leaving the second rock one comes face to face with the third and final rock of the group (Figure 25). The last rock is primarily notable for its considerable size in comparison to the preceding two rocks. The third rock features a height of approximately 3.5 m, a width of 7 m from east to west and length of 13 m from north to south, consisting of nine steps to reach the peak.114  This rock is at the summit of the hill to the northeast of the precinct and dominates                                                  110 Contador de Argote 1732: VII.548. 111 Alföldy 1997: 215. 112 Alföldy 1997: 213; Garcia y Bellido 1967: 134. 113 Alföldy 1997: 215; It is also possible that these basins served as purification ponds. 114 Tranoy 1981: 38; Although the lack of preservation of the base step has led to some sources citing eight steps, the majority and breadth of scholarship supports nine steps ascending to the top of the third granite rock. 31  much of the natural space given its stature. The upper surface of the third rock also exemplifies similar traces of the foundations of a temple to the other rocks (Figure 26). Overall, given the dimensions for the larger rectangular cavities115 and the inner impressions of all five basins thought to have served as seals, Alföldy convincingly argues that the main act of the ritual took place here.116  2.4 – The Significance of the Ritual Route Now that we have walked through the sacred space, it is worth turning our attention to the significance of this rock sanctuary to our understanding of rural religion in Roman Spain and how traditions could be invented under a senatorial agent such as Gaius C. Calpurnius Rufinus. The significance of ritual performance at the site of Panóias is enriched by the existence of dedicatory inscriptions, alongside corresponding space to enact ritual rites. Without the consideration of how the space relates to the ritual instruction of the text, representations of religion, such as the performativity piece taking place at the anepigraphic third rock space, would go unnoticed. After all, the ritual rites at Panóias were not simple, or rudimentary, but rather, they required specific and sequential instruction, and were above all, new.  2.4.1 – Invention Chronology While some scholars argue that the rock outcrops served a religious function prior to Calpurnius Rufinus’ intervention in the second to third century C.E., this hypothesis remains conjectural.                                                  115 The dimensions are between 2.10 and 2.70 m in length, 0.60 to 0.90 m in width, and 0.40 to 0.60 m in depth. 116 Tranoy 1981: 338; Alföldy further hypothesizes that symbolic grave goods may have been placed in vessels with incense in order to further the symbolism of death and rebirth though evidence of this or similar interpretations of the data following this line of thinking are nonexistent in the sources (1997: 225). 32  Certain indications, such as the lettering, grammatical forms, and word usages at the site, help us in dating the inscriptions and alteration of the site by Rufinus. However, given the nature of evidence, the geology of the site prevents the detection of past phases of use. Thus, all we know for certain from the site is Rufinus’ intervention and any allegations of “pre-Roman” worship taking place at Panóias should be met with caution.117 Just because the site at Panóias does not follow traditional models of worship known in other regions of the empire, does not mean that it was consecrated by “primitive” peoples prior to Roman hegemony.118 With no earlier evidence of use at the site, it is perhaps safest to assume that Rufinus was innovative in his establishment of a sacred precinct at Panóias.  Imported Deities Not only did Rufinus consecrate the site at Panóias, but he also chose to import deities such as Serapis and his consort Isis to this corner of the empire. Worship of Serapis was introduced shortly after the foundation of Alexandria in 332 B.C.E., under the orders of Ptolemy I of Egypt in order to help unify the Greeks and Egyptian inhabitants. Serapis can be thereby seen as a syncretic deity, embodying both attributes of deities such as the Greek god Zeus and Egyptian god Osiris to name a couple.119 The cult of Serapis then spread throughout the Graeco-Roman world and was officially recognized under Nero’s reign, when ancient temples were restored and Egyptian venerations celebrated publicly.120 Serapis was also a deity immensely favoured under the reign of Severus which has further been used to reinforce the widely accepted belief that Rufinus’ occupation at Panóias began from the late second to early third century C.E.121                                                   117 Salinas de Frías 2001: 155-6. 118 This relates back to Wiseman’s previously discussed portrayal of the deifications of springs by local inhabitants in Roman Spain as ‘primitive’ simply because they were not in line with Roman forms of worship (1956: 99).  119 Cortez 1947 28 ff. 120 Vasconcellos 1905-13: 341. 121 Tranoy 2004: 90. 33  While cult worship of Serapis spread throughout the Roman Empire, dedications to both Serapis and Isis were relatively rare in the Iberian Peninsula.122 Only fourteen dedications to Isis are currently known, three to Serapis, and five to Serapis and Isis together, with CIL II 2395a as the latest addition.123 Thus, Correia Santos’ latest transcription of CIL II 2395a provides us with the first of two direct dedications to the Graeco-Egyptian god Serapis at the site of Panóias.  The presence of not one, but two inscriptions directly venerating Serapis has also made the notion of the sacred space as a Serapeum even more convincing. Prior to the recent interpretation by Correia Santos and her team, scholars were already convinced that the sanctuary at Panóias was in fact a Serapeum. For instance, Cortez and Rodríquez Colmenero have regarded the site as a Serapeum, while other scholars such as Alföldy have disputed this interpretation. Nonetheless, I maintain that while Serapis may have a privileged veneration at this site, the incorporation of worship with other deities, as well as lack of criteria—such as architectural characteristics—a serapeum usually posesses, does not signify that the site at Panóias was a Serapeum in the normal sense of the word.124 Instead what we are dealing with is not a sanctuary dedicated to one god, or gods and goddesses of similar origin and cultural significance, but rather dedications to the worship of different deities under varying belief systems that come together under the guise of renewed tradition. Therefore, imposing a categorical label such as a space solely worshipping the god Serapis would only limit our understanding of the multifaceted function of this space.                                                  122 The designation of “Serapidi” in this new reading also demonstrates the growing frequency with which we see this form in the third century C.E. when Serapis becomes increasingly associated with Jupiter and Sol Invictus, thus providing another chronological indicator for the inscription (Corriea Santos, Pires, & Sousa 2014: 210). 123 Corriea Santos, Pires, & Sousa 2014: 209 and for a specific contextualized list of the inscriptions see notes 45-46. 124 See Wild 1984 and Sousa & Ribeiro da Silva 2013 for more on the typical characteristics of a Serapeum in the Graeco-Roman world and how the space at Panóias does not conform to the traditional model of a Serapeum. 34  But the question remains, why was an eastern deity venerated in this corner of the Peninsula? In particular, what connection did the consecrator of the site, Rufinus, have to Serapis? Rufinus’ devotion to the Graeco-Egyptian god could be due to his speculated eastern origins. For instance, Alföldy uncovered that only one family is known with the nomenclature Calpurnius Rufinus, a small Asian family of Calpurnii Rufi who flourished in the second century B.C.E.125 Apparently the family had two lines in ancient Anatolia, modern Turkey, with one family residing in Antioch ad Pisidiam and the other in the Pamphylic city of Attaleia.126 If the Rufinus in the inscriptions at Panóias was related to this family, the abundance of coin iconography of Serapis in Attaleia and neighbouring Perge would demonstrate where Rufinus first became acquainted with the cult of Serapis.127 However, the genitilicial name is rather common, and the cognomen is not necessarily indicative of a family, so perhaps Alföldy’s hypothesis is farfetched. Moreover, there is a chronological gap between when the family Alföldy references is known in the sources during the second century B.C.E. and when Rufinus is actively engaged in cult practices during the second century C.E. In contrast, Russell Cortez has put forth a more plausible hypothesis that Rufinus likely gained exposure to Serapis through military experience as the expansion of eastern religions to the northwest corners of the empire was often a result of Roman soldiers serving overseas and then settling in the Roman provinces.128 However, the motivation behind Rufinus’ interest in importing an eastern cult to this part of the Iberian Peninsula remains uncertain. What can be said for certain is that the presence of a cult space devoted largely to an eastern deity such as Serapis in this region was innovative to say the least.                                                   125 Alföldy 1997: 240. 126 Ibid. 127 Ibid. 128 Cortez 1947: 29. 35  Ritual Recipe The argument that the sanctuary of Panóias under Rufinus saw to innovation is further reinforced by the presence of such meticulous instruction provided at the site. The effort put in place to prescribe such a detailed recipe implies the need for specified instruction and lack of previous knowledge regarding how to operate the ritual space. For instance, a ritual participant at Panóias would not only be guided by written directions, but also by the close proximity of each inscription to corresponding cavities where the aforementioned rituals were meant to be carried out in sequence.129 This endeavor to ensure onlookers understood the ritual route of the site indicates that ritual worship imposed here was innovative and less familiar to users of the site.   2.4.2 – Tradition Traditionalizing Language While the importation of eastern deities and innovative systems of ritual worship at the site of Panóias point towards invention, there are elements underlying such invention that are traditionalizing. For instance, the use of aeternum in CIL II 2395b effectively casts the rituals as tradition instead of something new as it reads “Aeternum cum hoc templo.”130 Rufinus carefully chose his words in this consecration to denote his intention for an innovative ritual progression to become eternal and thus a tradition in the sacred landscape of Panóias.  How the consecrator phrases the performance of ritual in the inscriptions also denotes his intention to traditionalize said rituals. For instance, Rufinus detaches himself from the enactment of rituals at Panóias by enforcing words that do not restrict agency to him alone. We can see this                                                  129 Vasconcellos revisits examples of cavities similar in shape, size, and function in temples in Egypt during excavations he observed in 1909 affirming that the processes accompanying the sacrifices of animals in rituals were not entirely unparalleled (1905-13: 470). 130 Alföldy 1997: 241. 36  in inscription CIL II 2395e, the first text that a ritual participant would encounter at the site, as the present indicative tense is applied—rather than an imperative command or the past tense that figures predominantly in Roman votive dedications—in order to imply to anyone encountering the site that the ritual recipe is carried out in his present time. Similarly, instead of using active verbs, the ritual description uses a passive voice, depersonalizing agency and allowing the rites to be enacted by any individual. Thus, Rufinus effectively traditionalizes the rituals in an area previously unfamiliar with them by clearly indicating through his choice of words that anyone can enact such rituals, and that they will continue to be performed for generations to come.  Gods and Goddesses To further traditionalize the ritual worship at Panóias, Rufinus included any and all deities in his dedication. More specifically, the first inscription at the site demonstrates that the prescribed sacrifices are offered to the diis deabusque, or “gods and goddesses.”131 The homogeneity of this formula remains consistent with the majority of inscriptions at Panóias. García-Bellido argues that perhaps the deities not specified in the inscriptions relate to divinities of the surroundings region incorporated in order to make the sanctuary more accessible to all.132 Arguably, Rufinus endorses the worship of any gods and goddesses in order to ease the adoption of ritual rites at Panóias as a new tradition.133                                                   131 CIL II 2395e; The formula diis deabusque also aids in our chronological understanding of the space seeing as theonyms preceded by the plural “diis” is a form of invocation that has been dated to the middle of the second century C.E., when the space is believed to have been consecrated (Corriea Santos, Pires, & Sousa 2014: 210; Tranoy 2004: 90). 132 García-Bellido 1997: 29-96. 133 CIL II 2395a; CIL II 2395b; CIL II 2395d; CIL II 2395e; Corriea Santos, Pires, & Sousa 2014: 219. 37  Local Deities In addition to the generic formula targeting the worship of various gods and goddesses in the region, Rufinus also includes a dedication to the local deities of the Lapiteae in CIL II 2395b.134 The beginning of the inscription designates: “Diis(!) deabusque aeternum lacum omni/busque numinibus et Lapitearum” (“For the Gods and Goddesses, as well as for all the deities of the Lapiteae”). Rufinus clearly indicates his intention at Panóias to honour both new deities of the empire, such as imported deities Serapis and Isis,135 and old deities of the Lapiteae.  Although, there is a statal differentiation in the naming of the gods worth noting as deus and dea normally denote higher status than numen. Perhaps the distinction between how the local gods are regarded compared to other deities can be interpreted as an indication of their status as spirits as opposed to gods. Moreover, the numen are referenced in regards to a people and not a place, which stands in contrast to most toponymically specific deities. The connection between the people and the divinities could signal the importance of maintaining traces of their local identity as opposed to the more common connections to the sacred landscape. Moreover, it is not the local inhabitants themselves ensuring they maintain ties to these divinities, but rather, Rufinus who is overseeing the recognition of the numen of the Lapiteae. By balancing the dedication to his own imported deities with that of the local inhabitants—and in effect, traditionalizing the present ritual worship with elements of past reverence—Rufinus is more likely to have ensured the implementation of innovative ritual rites at Panóias.                                                   134 The consensus among scholars is that the Lapiteae were a local indigenous community, commonly mentioned in northwestern Spain and likely the inhabitants of the settlement located to the east of the sanctuary mentioned earlier (Alföldy 1997: 193). Argote attributes the Lapiteae as sacred to the peoples of Thephalia and discusses its importance to the Greeks in antiquity (Contador de Argote 1732: VII.573); Lapites is also mentioned in book XII of Ovid’s Metamorphoses; for more regarding the Lapiteae see Blázquez 2001: 200-202; Keay 1988: 162; Vasconcellos 1905-13. 135 CIL II 2395a and CIL II 2395c. 38  2.4.3 – Agency Gaius C. Calpurnius Rufinus What is even more remarkable about the prescription of innovative ritual rites at Panóias is that they were driven by a private senator, Gaius C. Calpurnius Rufinus. Rufinus is not only mentioned in all of the inscriptions, but his name is accompanied by the abbreviation “C · V,” which stood for clarissimus vir, an epithet that came to recognize senatorial rank in the first century C.E.136 While it is clear that Rufinus was a senator and agent behind the consecration of innovative rituals at Panóias, his designation as a senator does not necessitate assumptions that he was an administrator in Hispania. For instance, Tranoy has theorized that Rufinus was possibly a relative of the African senator Fronto Aemilianus Calpurnius Rufilianus.137 Scarlat Lambrino has also hypothesized that Rufinus was the provincial administrator of the Tarraconensis.138 Some scholars such as Géza Alföldy use this formula to regard Rufinus as not only an official of the provincial government at the time, but also a “pilgrim founder of religion.”139 While I would argue that “inventor of traditions” is perhaps a more appropriate designation for the senator, given his active prescription of new rituals feigning conservatism, the classification of senator does not necessitate provincial administrator. After all, these assumptions are rooted in the inherent Romanization narrative that only administrators were capable of bringing religion to foreign territories, which the present model is trying to move beyond.                                                   136 Sandys 1919: 193; The epithet can literally be translated as a “highly renowned man,” but it also defined the status of the consecrator as not only a Roman citizen, but a senator. Unless he acquired special favor from the emperor, he likely met all the conditions of eligibility. This means that he was a Roman citizen who was free born, maintained an exceptional reputation in society, and owned property worth at least 1,000,000 sesterces (Abbott 1963: 381). Rufinus was then also someone who successfully secured election to the quaestorship prior to becoming a senator. 137 Tranoy 1981: 393. 138 Lambrino 1954: 61. 139 Alföldy quoted in Rodríquez Colmenero 2000: 187. 39  A closer look at the agency at Panóias also demonstrates that applying the Romanization model in the traditional manner of assuming Roman imperialism drove cultural change directly is problematic. After all, it was not necessarily Roman administrative powers that enabled Rufinus to invent traditions at Panóias, but rather the ability to own private land and establish sacred space accordingly. Instead of presupposing a trickle-down effect of Romanitas from Roman administrators down to the lower strata of society, studies should recognize the other agents who could drive change, or more appropriately, invent traditions to coalesce the new alongside the old. Imperial Religion The agency at Panóias also demonstrates that just because an individual bears the title “senator” does not mean that they were obliged to act as a promoter of imperial religion. Quite the contrary, Rufinus’ interest in religious phenomena extends far beyond the expected social norms of Roman senators.  Zsuzsanna Várhelyi focuses on the ability of Roman senators to promote imperial religion through their authoritative roles in the empire. She uses evidence of their investment in the establishment and application of innovative religious ideas to prove the active agency of senators in the religious sector of society.140 While Rufinus’ agency may have been active, it was arguably his power as land owner and not senatorial authority which enabled him to establish innovative religious praxis at Panóias. After all, the inscriptions at the sanctuary of Panóias do not bear the usual “locus datus decreto decurionum” designating civic control over land rights. Thus, the land at Panóias was not overseen by civic or territorial administration such as decuriones, but rather was private land owned by Rufinus. Without having to answer to a                                                  140 Várhelyi 2010: 186. 40  governing body, Rufinus was able to establish whatever cult he wanted, like Pliny the Younger had.  Rufinus may have invested in the “establishment of such innovative religious ideas” Várhelyi refers to, but under the pretense of senatorial authority. And if we move away from normative ideas of senator’s role in religious innovation, we can see that they were not just agents of the imperial cause, but agents of their own individualized motives, at times disparate from those of the emperor.  2.4.4 – Performativity Rufinus’ role may not have been as a senatorial administrator, but he did go to great lengths to establish his authority in cult practice at Panóias. For instance, the inscriptions and spatial organization demonstrate the extent to which Rufinus ensured individuals had to follow his prescribed ritual route in order to participate in the ritual rites at Panóias. According to Sousa and Robeiro da Silva’s work, ritual rites such as initiation into the cult of Serapis did not typically require a sacred course as we see at Panóias and so this prescription of a ritual route is either a continuation of past practices at the site (following traditional—and problematic—assumptions made about “Pre-Roman” use of the site), or Rufinus’ establishment of such a ritual route was entirely innovative.141 In addition, Rufinus saw organized the space so that each rock outcrop built up to a climatic ritual spectacle commencing at the third and final rock. After all, if the ritual killing was meant to be visibly accessible to all, then it likely took place on the third rock. Moreover, the third rock shows no trace of burning in the carved out cavities which encompass more space, and are positioned at a higher elevation than the other two rocks in the precint.142 Thus, the third rock                                                  141 Sousa and Rebeiro da Silva 2013: 106-7. 142 Tranoy 1981: 338. 41  arguably acted as a performativity piece and spectacular stage for the officient. The height and visibility of the third rock not only puts emphasis on the performer of the sacrificial rite, but also entrenches him in a social hierarchy where the person on display is elevated above the audience. Therefore, Rufinus’ authority as consecrator of the site is not only explictly indicated in the repetition of his name in the inscriptions, but also impliclty inferred through the spatial denouement at Panóias.   2.5 – Reconsiderations Overall, the sacred space at Panóias is significant for its unique set of inscriptions which prescribe the ritual route of the space, as well as the intricacies of the order and application of ritual rites mandated by consecrator Calpurnius Rufinus. This site offers us the ability to apply new methodologies whereby we can hone in on the agency of the invention of tradition taking place in the sacred landscape of Roman provinces such as Hispania.  For instance, rather than cling to Romanization model stereotypes of Roman administrators being the only agents of religious change in the provinces, the agency at Panóias demonstrates that new ideas could be mixed with the old by land owners. The religious praxis of local inhabitants in this area of the Peninsula were not replaced by Roman models instituted by a Roman administrator, but rather they experienced a level of hybridization of new and old religious praxis under the guidance of a senator. Thus, the nuanced traditions taking place at Panóias were not “pre-Roman” or “Roman,” but a product of the empire. While Rufinus may never have been in Hispania in the first place without his ties to the empire—or at a position of wealth and status enabling him to own the land the sanctuary of Panóias is consecrated on—that does not mean that the agency under which the space was defined should be classified as 42  “Roman.” Instead what we have is an elite landholder, probably not local to the area, who initiated this change, but in a way that looks both local and universal. Therefore, the situation at Panóias signifies a hybrid of religions driven by a single agent who created tradition through claims of eternalness, the prescription of ritual recipes, and structure of piety.  This alternative approach to the study of Panóias enables us to move away from predetermined ideas of what constitutes Roman and non-Roman systems of religious worship in the Roman Empire. Rather than abide by traditional models which have defined the components of what is to be accepted as a “primitive” form of worship—in other words applying an umbrella term to any abnormal patterns of worship or ritual space to primitive ideas—the sanctuary of Panóias shows us that Roman senators could initiate the weird and unusual. In fact, at Panóias, they were the lead experimenters as Rufinus’ land ownership granted him the ability to prescribe new traditions to others. Therefore, traditional dichotomizations of continuity and change, or Roman and non-Roman, no longer work. Instead, the model that will continue to be applied going forward in the analysis of rural sanctuaries in Roman Spain is rooted in the conception that religious change was often negotiated under pretenses of new practices alongside the old, as well as under agency both varied and unique.      43  3. Applying the Panóias Model The sanctuary of Panóias with its rich material culture and epigraphic evidence has come to act as a model for the interpretation of many others.143 In 1947, Russell Cortez, in his exploration of the various levels of worship present at Panóias and its cultic practices, perceived the site as an unparalleled monument and a true archaeological jewel.144 While I agree with the latter statement, the former deserves elaboration. Although the sacred space of Panóias is remarkable in its physical manifestation of inventing traditions and such detailed levels of instruction overseeing the union of worshipping old deities alongside new ones, there are other sites worth further consideration. As noted by Tranoy, the site of Panóias is not an isolated case and is actually part of a series of rock sanctuaries which testify to the integration of the sacred landscape crucial to religion in Roman Spain.145 In order to ensure these other case studies are not subjected to traditional ideas of agency and acculturation prescribed by Romanization models, the interpretation of Panóias as a space that facilitated the invention of tradition under unique agency will be applied as an archetype for understanding other spaces. Arguably, by using Panóias as a model of interpretation for other rural sanctuaries it can enable us to move beyond traditional models applied to studies of religious life in Central and Western Hispania under Roman hegemony. It is therefore essential to discuss the presence or absence of new religious praxis at other sites, such as Pena Escrita, Cabeço das Fráguas, Peñalba de Villastar, and Pias dos Mouros, as well as assess how well they correspond to the canonical insights into religious praxis already discovered at Panóias. Thus, further analysis will evaluate the extent to which Panóias can be used as an applicable model for understanding other rural sanctuaries in Roman Spain.                                                   143 Richert 2005: 15. 144 Cortez 1947: 8. 145 Alfayé Villa and Marco Simón: 2008; Tranoy 1981: 340; Tranoy 2004: 88. 44  3.1 – Pena Escrita The Pena Escrita Sanctuary, located in Vilar de Perdizes, Montalegre, is another sacrificial site under the Roman Empire that features a coalescence of religious identities (Figure 27).146 This sacred site was first recorded in 1978 by Father Lourenço Fontes.147 Pena Escrita has been analyzed primarily by Rodríguez Colmenero in the 1980s, Tranoy, Blázquez, and then with more thorough detail and analysis by Rodríguez Colmenero in the 1990s, followed by Correia Santos revisiting the site and consolidating the data and new interpretations in the past decade.148 With the exception of some of these more focused accounts, extensive analysis of the site and its spatial significance has been scarce. Like Panóias, the peculiar nature of this site and its inability to fit certain predetermined molds of sanctuary structures has, to an extent, limited interest and further analysis of its significance. The epigraphic bias present in scholarship could also have influenced the lack of interest in regards to this largely anepigraphic sanctuary. Moreover, the focus in scholarship has been on the discovery of two votive altars in proximity to the site, resulting in little attention paid to the sacred space itself and its significance to the religious identity of its occupants.149 Spatial Organization The site itself is located in a well irrigated plain encircled by agricultural fields and lying at the bottom of the Larouco Mountain.150 It consists of a granite outcrop slightly elevated in an open field with three narrow steps which lead to a rectangular cavity above and a smaller circular                                                  146 Correia Santos 2010a: 133. 147 Rodríquez Colmenero 1995: 190. 148 Blázquez 1983; Rodríguez Colmenero 1979; Rodríguez Colmenero 1993; Tranoy 1981. 149 Correia Santos 2010a: 129; Correia Santos 2010c: 7. 150 Correia Santos 2010c: 7. 45  cavity complete with a narrow drainage channel (Figure 28).151 Given the narrowness of the steps, Correia Santos presumes that they must have served more of a symbolic than a functional need.152 The carved rock also follows the configuration of a votive altar with a central focal point, the rectangular cavity (Figure 29).153 This cavity resembles the cavities we see at Panóias, especially the larger rectangular depressions on top of the third highest granite outcrop at Panóias. While there is no epigraphic evidence directly pointing to the function of this cavity, the similarity to Panóias has led scholars to interpret similar ritual or sacrificial use. Relation to Votive Altars Following the last step to the centre of the rock outcrop there are also two square carvings parallel in dimensions and placed in an intentionally symmetrical axis in respect to the rectangular cavity, much like what we see with the spatial relations at Panóias.154 Correia Santos posits that these two square carvings were placeholders for the two votive altars discovered in close proximity to the sanctuary as they are of comparable dimensions (Figure 30).155 Scholars such as Richert have also privileged the proximity of the altars in an aim to understand the deities worshipped at Pena Escrita.156 With one altar dedicated to a deity of place, Larauco157 (Figure 31), and the other to Jupiter, Rodríguez Colmenero further suggests that the sanctuary at                                                  151 Correia Santos 2010a: 130; Tranoy 2004: 89; The measurement of the steps range from 8-12 cm in depth and 12-18 cm in height while the rectangular cavity measures 50 x 20-25 x 26 cm in size. 152 Correia Santos 2010a: 130. 153 Correia Santos 2010a: 130. 154 Correia Santos 2010c: 7. 155 Correia Santos 2010a: 133; Correia Santos 2010c: 7; The proximity of the altars to remains questionable though as the altars were not found in immediate relation to Pena Escrita but rather were discovered at the foot of the mountain nearby Pena Escrita (Tranoy 2004: 89) and are thus regarded by Rodríguez Colmenero as related to Pena Escrita given they were found within the “vicinity” (1995: 192). Nonetheless, the similar proportions of the rectangular depressions at Pena Escrita and the bases of the altars found make it plausible that they were related.  Near the serra de Larauco, in the region of Montalegre, village of Vilar de Peridizes (Pena Escrita) 156 Richert 2005: 15. 157 Tranoy refers to Larauco or Laraucus as a deity whose worship manifests itself through multiple testimonies and who operates as a mysterious power governing the lives of the inhabitants around the mountain and following Roman conquest become associated with the Roman god Jupiter (1981: 281). Larauco was also associated with the natural landscape of the mountain nearby and so the presence of a mountain cult has been privileged by scholars. 46  Pena Escrita was a cult place of the divinity of the nearby Larouco Mountain, thus positing the altars as extensions of Pena Escrita’s cult practices.158 He even goes as far as to state that the ritual function is “proven” by the two altars found in the vicinity.159 Tranoy advances this interpretation as he, too, reasons that the discovery of the dedicatory altars nearby Pena Escrita “confirm” the religious use of the space.160 However, I remain cautious of viewing the proximity to aforementioned altars as confirmation of ritual function at Pena Escrita given that the presumed “altar bases” could have served other purposes. Alternative Interpretations While scholars have interpreted the space as serving a ritual function given its similar appearance to the rock outcrops at Panóias, the space could have served a funerary function instead. For instance, the bases could have been used as insertion points for stelae—common funerary commemorations in the region—rather than altars. And the drainage channel could have functioned as a pour channel for lead to affix these stelae, given its orientation towards the back of the space, rather than for the blood from ritual sacrifices. While applying the Panóias model can illuminate unique interpretations, we should also keep in mind that there is still a lot we do not know in regards to these sites and only further analysis can elucidate their function.  In addition to function, there are also alternative ideas concerning which deities were worshipped at Pena Escrita. For instance, scholars such as Richert are quick to associate the religious veneration at this site with the nearby mountain thus resulting in an interpretation of the sanctuary as a manifestation of a predominant mountain cult in Lusitania.161 While this argument is convincing given the significance of the natural landscape to sacred beliefs in the region, there                                                  158 Rodríguez Colmenero 1993: 75-6; Tranoy 1981: 281. 159 Rodríquez Colmenero 1995: 192. 160 Tranoy 2004: 89. 161 Richert 2005: 4. 47  are other variables to consider.162 Kulikowski refers to associating the names of deities in inscriptions with nearby natural features, such as rivers or mountains, as a rather “tenuous” method of interpreting gods seeing as it can result in hasty characterizations and misrepresentations.163 Nonetheless, the prominence of the mountain cult in this area of Lusitania does lead one to believe that the sacred space could have been related to the altar dedicated to Larauco.  Ornamentation Prevalent interpretations of the site at Pena Escrita also rely on the ritual interpretation of  what Rodríguez describes as crescent moon or horseshoes oriented in several directions carved into the rock at a range of positions that effectively decorate the site in what he refers to as a pre-Roman style of ornamentation.164 Yet again, the distinction of pre-Roman in regards to a form of decoration deemed abnormal is problematic. What’s more, these motifs are frequently found across the Mediterranean and as we saw at Panóias, unusual forms of worship can be influenced in part by the Romans.165 Thus, while the presence of the motifs at Pena Escrita are likely ritual, their exact function, be it religious, funerary, or protective, remains to be proven, as does agent(s) behind their orientation.                                                  162 See Richert 2005 on cultic veneration of the natural landscape and Horden & Purcell 2000 on the geography of religion. 163 Kulikowski 2013: 518-9. 164 Rodríguez Colmenero 1993: 75-6; The ornamentations are lined up directed to the north and inscribed in different positions on the northwest face of the rock. While Rodríguez Colmenero maintains that the rock ornamentations could be also represent attempted inscriptions of letters such as “o” or “c,” it is only through further analysis, such as the application of the Morphological Residual Model, that their significance the can be uncovered (1995: 191-2). 165 For more on the morphology and orthogonality of the carvings see Correia Santos 2010a: 133 as she discusses the typological parallels to comparative studies in sites across the Roman Empire. 48  Military Influence While the function of the site remains somewhat ambiguous, the chronology and level of interaction with Romans is more readily understood given remains of an inscription on the south side of the rock. It has been interpreted by Rodríguez Colmenero as “le(gionis) VII G(eminae) P(iae) (followed by an inverted C)” and attests to the use of the rock during the Roman period.166 The Legio VII Gemina was a legion from the Imperial Roman army which was founded under Nero’s reign in Hispania and remained there until the fourth, possibly fifth century C.E.167 The presence of legionary camps nearby (Figure 1) demonstrate the facilitation of interaction between soldiers and inhabitants near Pena Escrita.168 The inscription thereby aids in our chronological understanding of the space. The structure itself also shows no signs of previous phases of use, unlike other sites where there is a clear manifestation of the adaptation of space over time.169 Thus, the assumption that such sanctuaries are evidence of “pre-Roman” practice must remain tentative, as the only certain evidence for the site’s use is dated to the Roman era. Applicability of Panóias Model As expected, the level of evidence available to us at Panóias is unmatched at other rural sanctuaries in the region. Still, Pena Escrita does share similarities to Panóias, such as the geological makeup of the outcrop, the rectangular cavity carved out of the centre of the rock, the stairs leading up to it, and the presence of Latin recognized in the presumed related altars. However, there is also a dearth of similarities to Panóias as there is no prescription of ritual worship dictated through inscriptions, or a variation in the size and shape of the cavities carved                                                  166 Richert 2005: 4; the letters on the rock are vastly eroded though and further analysis, such as application of the MRM could help us understand the translation more soundly in the future (Rodríquez Colmenero 1995: 190-1). 167 The military presence in Hispania was relatively minor compared to other western provinces but the peninsula still accommodated various legions under Augustus and then only one legion remaining following the succession of the Flavian dynasty (Kulikowski 2013: 526). 168 Kulikowski 2013: 526. 169 A comparable site is Mogueira in Resende, Viseu, as analyzed in Correia Santos 2010a: 133 ff. 49  out of the rock for varying stages of ritual sacrifice as we have come to known in the previous case study.170 Thus, while these similarities could point to a uniform physical manifestation of space to serve ritual worship, they also remain largely vague and indirect relatives of the intricate mechanisms of religion taking place at Panóias. Contribution Although much remains a mystery at Pena Escrita, what we do know is that the granite outcrop was used under the Roman Empire, that it features physical manifestations of possible ritual use such as altars and a cavity carved out of the rock similar to what we see at Panóias, and that the site occupied a highly visible and accessible area within its natural landscape. What we can take away from the evidence at Pena Escrita is that while the site may not share enough similarities with Panóias to characterize it as parallel in form and function, the appearance of Latin inscribed on the rock, as well as on the votive altars located nearby, do speak to a level of epigraphic habit developing. The use of votive inscriptions in itself is a new mechanism of religious worship that was previously absent from the sacred landscape in the Iberian Peninsula. It points not necessarily to change in its purest form as opposing continuity, but rather to a new tradition of commemorating ritual worship taking form in Roman Spain.  3.2 – Cabeço das Fráguas Cabeço das Fráguas features a rock outcrop similar to what we see at Panóias. It is situated about fifteen km away from an urban centre in Central Portugal to the south of the river Duero in Pousafoules do Bispo, Sabugal. The site was also positioned within an Iron Age hillfort located                                                  170 Despite a lack of similarities, scholars continue to classify Pena Escrita as a rock sanctuary similar to Panóias. For example, one of the more recent publications regarding Pena Escrita by Inês Vaz states that the site has cavities for libations in proximity to traces of rock engravings even though in reality there is only one cavity and no trace letters where he regards them (2002: 41). For more on challenging such assumptions see Corriea Santos 2010a: 130. 50  at the top of a mountain 1015 m in the air and occupied until the first century C.E.171 Most scholars argue that the occupation of the hillfort was post-abandonment.172 If this was the case, the deliberate choice to reoccupy an Iron Age hillfort and use it as a cultic venue exhibits an intent by the creators to create links with the past through what has been referred to as “symbolic recycling.”173 It also points to innovation as the repurpose of space for ritual use effectively constructs new narratives about the past.174 Past Scholarship The rock was recognized in the eighteenth century by General Joa de Almeida.175 Although it was not properly analyzed until 1956 when visited by Adriano Vasco Rodrigues and Dr. Arsénio Rodrigues da Silva whereby they located a rock inscription and proceeded to outsource it for proper interpretation. They then set up a camp at Cabeço das Fráguas in the following year to further carry out analysis enabling them to publish their findings in Rodrigues’ early report of 1959.176 Afterwards, Utermann and Russell Cortez made further transcriptions, followed by additional analysis by Tovar and lastly Patricio Curado in the 1990s.177 Chronology As far as dating goes, the lack of archaeological context prevents proper dating of the site. On the one hand, most scholars date the inscriptions at Cabeço das Fráguas to between the second and third century C.E., at the same moment in time as Panóias.178 On the other hand, scholars                                                  171 Correia Santos 2010c: 8; Rodrigues 1959: 71. 172 Scholars such as Alarçao and Curado argue that the hillfort was already abandoned when the inscription at the site was carved and it was undergoing ritual use (Alarçao 2001: 315-316; Curado 2002,:71, 76); For more on the “cultural biography” of the location of Cabeço das Fráguas see Alfayé Villa and Marco Simón: 2008: 293. 173 Woolf 1996: 31. 174 Alfayé Villa and Marco Simón: 2008: 293. 175 Rodríquez Colmenero 1995: 221. 176 Correia Santos 2010c: 8-9; Marco Simón 1999:153. 177 Rodríquez Colmenero 1995: 221. 178 Alfayé Villa and Marco Simón: 2008: 291. 51  such as Correia Santos use the discovery of a first century C.E. votive deposit of clay figurines found at the foot of a hill nearby and other findings from comparable sites to aid in dating Cabeço das Fráguas.179 While such comparisons may prove useful, they also remain unsubstantiated.180 Epigraphic Evidence  Despite chronological challenges, Cabeço das Fráguas remains prevalent in the study of religious practices in the Iberian Peninsula, not only because of its similar environment to Panóias, but also due to the discovery of a linguistically rare inscription written on the horizontal surface of the rock outcrop (Figure 32).181 Given that no evidence of rock structures serving ritual means survive on the site, if there was a sanctuary at the top of the hill similar to the rock outcrops at Panóias, then the surviving inscription is currently our only indicator of such a space.182 The inscription reads:  OLIAM. TREBOPALA INDI. PORCOM. LABBO. COMAIAM. ICCONA. LOIM INNA. OLIAM. VSSEAM. TREBARVNE. INDI. TAVROM IFADEM REVE. *RE…183  The last two lines of the inscription are too damaged to make out (Figure 33). As for what can be made out, the etymology and interpretation of words as well as their arrangement remains controversial given that the text may be written in the Latin alphabet, but is composed in Lusitanian, a language that remains only moderately understood by scholars.184 Nevertheless,                                                  179 Alfayé Villa and Marco Simón: 2008: 293. 180 Encarnaçao 1995, 269 in Alfayé Villa and Marco Simón 2008: 293; Rodríguez Colmenero 1995. 181 Alfayé Villa and Marco Simón: 2008: 290. 182 Ibid: 294. 183 As described in Prósper 1999: 153 following Untermann 1997: 756-57. 184 Alfayé Villa and Marco Simón: 2008: 290. 52  Tovar has translated the text as, “A lamb for Trebopala and a pig for Labbo, a calf (?) for Iccona Loiminna, a sheep of the year for Trebarune and a stallion bull for Reve Tre .. .?”185 Another rock inscription has also been detected by Rodríguez Colmenero next to this one given its appearance of punctuation marks and remnants of non-Latin writing, but has yet to be deciphered (Figure 34).186  Scholars continue to debate the meaning of words from the inscription, such as whether they denotes places rather than deities, the meaning of nominative or dative case endings in regards to the naming of the deities, and so on.187 Regardless, what we are clearly looking at is a series of rites directed towards a set of local deities.  Suovetaurilia More specifically, the combination of deities mentioned in the inscription at Cabeço das Fráguas speak to a sacrificial rite with particular references to the offering of several animals to multiple distinctive indigenous deities.188 Scholars have compared the sequential list of sacrifices made to differing deities to the Roman suovetaurilia, a ritual that entailed a threefold sacrifice of a pig, sheep, and bull respectively.189 However, the suovetaurilia was a rather specific sacrifice conducted at particular lustral moments on behalf of the Roman state and featured different—not                                                  185 Alfayé Villa and Marco Simón: 2008: 294; Correia Santos 2007; Curado 1984; for an in depth discussion of other alternative translations see Correia Santos 2008: 261-2 and Tovar 1985 6 ff; and for more on the Lusitanian language see Bruun and Edmondson 2015; Prósper 1999. 186 Alfayé Villa and Marco Simón: 2008: 290. 187 For instance, Labbo has been identified by some scholars as Laepus due to the presence of a god under that name which was venerated on various granite altars found near the site in Sabugal, Portugal. Trebaruna is likened to Varuna and Ouranos as a sovereign deity and Trebopala as a protective chthonic deity. While little is known of Iccona Loiminna, Reve is a relatively well-known deity in the Iberian Peninsula likened as “a native equivalent of Jupiter” prevalent in mountain cults. For interpretations of the deities mentioned in the inscription see: Alfayé Villa and Marco Simón: 2008: 294; Curado 1984; García 1991; Prósper 1999; Marco Simón 1999: 155; Richert 2005: 4. 188 Curado 1984; Prósper 1999; Rodríguez Colmenero 2000: 157; Schattner and Correia Santos 2010: 109. 189 For more information on the practice of suovetaurilia in Lusitania see: Alarcão, J. d. 1988; Alfayé Villa and Marco Simón: 2008: 294; Correia Santos 2008: 253 ff; Correia Santos 2010c: 8-9. Also see Prósper 1999: 153 for comparisons to similar practices in Rome, India, Iran, Greece, Umbria, and early German cultures. 53  to mention only three—animals than those listed in the inscription.190 Instead, what we see at Cabeço das Fráguas is the performance of sacrificing particular animals to different deities indicating that there was a significant link between the victim and deity.191 Therefore, the suovetaurilia is not entirely comparable to what is taking place at Cabeço das Fráguas.  The relationship between the texts and their fulfillment by worshippers is also worth further consideration as the question remains whether the inscriptions were performative texts, commemorative, or prescriptive, as is the case at Panóias.192  Regardless, the inscription speaks to the prevalence of practicing animal sacrifice to appease the gods. What’s more, both situations attempt to maintain local tradition, in terms of the deities named, alongside the integration of new traditions, such as the epigraphic habit. Epigraphic Habit While analyzing the inscription on a microscale reveals the intricacies of sacrifices made to varying deities, if we step back and evaluate the significance of these texts on a macroscale, we can see the vast implication of cultural exchange marked by the discovery of this inscription at Cabeço das Fráguas. As stated, veneration in Hispania was not performed through epigraphic means prior to increased cultural interaction with the Romans and other cultures.193 As argued by Marco Simón, “the adoption of epigraphy was itself a consequence of the process of cultural contact between the native populations and Rome.”194 While I reject the hasty dichotomization of natives and Romans in this context, as the specific cultural identity of inhabitants inscribing                                                  190 Bendlin 2013; Huet 2008; Vermaseren 1957. 191 As noted by John Scheid, worship remained inextricably linked to the exercise of power and any modifications to such worship that occurred were a result of the frameworks in place. The significance of these relationships, whether they reflected the deity’s individual status, or their statal relation to others, remains unclear. (Scheid 1990: 751). 192 Alfayé Villa and Marco Simón: 2008: 291. 193 Widespread epigraphy in the Iberian Peninsula did not appear until from the second century B.C.E. onwards whereby it underwent a process of intensification. For instance, around 1,750 of the surviving inscriptions date from after Roman conquest and onwards into the following centuries (Beltrán Lloris 1999: 138). 194 Alfayé Villa and Marco Simón: 2008: 281. 54  rocks at this time remains unclear, the adoption of epigraphy does speak to a new tradition taking place in the northwest of Hispania. For instance, of all five of the Lusitanian inscriptions found in Lusitania, the use of vernacular language has only been found in the ritual sphere accompanying a sacrificial connotation.195 While this could be seen as evidence of traditionalism and a persistence of local language amidst hegemonic influence, the appearance of the epigraphic habit suggests that rather what is taking place is hybridization of different religious practices. Role of Language There are various ways one can interpret the appearance of Lusitanian language in sacred contexts. For one thing, traditional models of acculturation could treat this presence of the epigraphic habit taking form as a direct instance of cultural change and local inhabitants adopting the superior practices of the Romans. Or, opposing models could view the persistence of Lusitanian language despite the spread of Latin as active cultural resistance to hegemonic powers such as the Romans.196  Yet, conceding to either of these meta-narratives is too simplified.  Instead, at Cabeço das Fráguas what we have is a new practice coupled with a local language, meant to commemorate a complex rite in a new way. Negotiation of Identities The discovery of a Lusitanian inscription at Cabeço das Fráguas thereby highlights the role of language in the renegotiation of individual and collective identities in the religious sphere.197 With language as one of many indicators of identity, and the hybridization of using features of                                                  195 Alfayé Villa and Marco Simón: 2008: 281. 196 See Alfayé Villa and Marco Simón 2008: 282 for their application of Elsner’s 2001: 271-2 definition of cultural resistance on the situations in the Iberian Peninsula. Although it is worth considering that the use of the indigenous language of the Lusitanians could also be because the language was critical to the performance of the religious rite and thus could not be rendered in Latin. Unfortunately, not enough about the requirements of the ritual rites is known to ascertain the particular motive behind the use of language in this context, and as such, we should be mindful of such conjecture. 197 Alfayé Villa and Marco Simón: 2008: 281. 55  both Lusitanian and Latin language present at the sacred text inscribed at Cabeço das Fráguas, aspects of differing cultural identities are coalescing. What would further aid in our understanding of this process of inventing traditions through the renegotiation of identities is if we knew the agency under which the sacred text at Cabeço das Fráguas was performed. What drove individuals or groups to negotiate the terms of their identities in the religious sphere? Although we cannot rely on evidence of agency to help us understand this process of hybridization at Cabeço das Fráguas, we can compare it to other instances in sacred contexts whereby the epigraphic habit is espoused. Applicability of Panóias Model Like Panóias, the inscriptions at Cabeço das Fráguas also demonstrate the significant role of language in sacred contexts. The majority of inscriptions at Panóias are in Latin, with one dedication in Greek, either composed in Greek to be read by the Graeco-Egyptian deity it honours or to demonstrate the knowledge of the Greek language by the dedicator Rufinus. Either way, the employment of Latin and Greek at the site reveal the language spoken by the dedicator, as well as designates the languages as understood by those using the space seeing as their understanding of the language was necessary to carry out the prescribed ritual instructions. In contrast, the use of Lusitanian language and only the Latin alphabet at Cabeço das Fráguas, points towards a different social milieu using the site altogether. The agent(s) of the site at Cabeço das Fráguas could be attempting to communicate their performance of the sacrificial offerings to the prescribed deities through the deity’s preferred method of communication. The appearance of both languages could also simply denote an attempt to prescribe to a language the agent(s) have yet to fully comprehend, thus explaining why only the Latin alphabet is applied and not its syntax. Nonetheless, what the use of Lusitanian and Latin together in a sacred text 56  inscribed on a rock outcrop at Cabeço das Fráguas does denote is evidence of traditionalizing religious worship via language, a practice we see not only at Panóias, but other case studies to be discussed shortly.  The materiality of the site and its visible accessibility can also be compared to that of Panóias. For instance, at 1015 m in the air a sanctuary site featuring the performance of hierarchical ritual sacrifice would have attained quite the viewership and is rather comparable to what we can expect at the third rock at Panóias.   Overall, the Panóias model can only be applied to Cabeço das Fráguas insofar as the significance of agency and performance of ritual performance are privileged. While both spaces consist of inscriptions denoting the performance of ritual sacrifices for their respective deities, it does not mean that the manner in which they were enacted is the same. For instance, detailed instructions prescribing the intricate ritual recipe we see at Panóias is missing from the archaeological record at Cabeço das Fráguas. Instead, all we can say for certain about Cabeço das Fráguas is that there was a significant link between deity and victim as prescribed in the inscription. Moreover, we still do not know for certain if the inscription itself on the rock outcrop was intended to be prescriptive, as we see at Panóias, or commemorative of a onetime act. Therefore, while other scholars are quick to compare Cabeço das Fráguas to Panóias solely based on the observation of a rock outcrop with inscribed text, the social milieu to which Cabeço das Fráguas speaks to, as well as the motivation behind the inscription, remain distinct from Panóias. Contribution What Cabeço das Fráguas does signify in the grand scheme of rural religion in the Iberian Peninsula is the negotiation of identities through a hybridization of religious praxis such as adopting the epigraphic habit and inscribing a prescription of ritual sacrifices to various deities 57  sacred to the region.  Though the practice of ritual sacrifice is similar to what we see happen at Panóias, the manner in which it is conducted, who’s using the space based on their language and dedication, as well as the monumentalization of the space, differs vastly from the Panóias model. Cabeço das Fráguas uses technologies of worship, such as inscribing in rock and using the Latin alphabet, that is Latinate and interested in hierarchies and relationships of divinities, but made to be local with Lusitanian language. Thus, instead of invention driven by an elite landowner, we see invention taken on by inhabitants of Lusitania trying to negotiate their identities under the religious complexities present in Roman Spain.  3.3 – Pia dos Mouros At an enclave nearly unnoticed within the general landscape at Pias dos Mouros in Argeriz, Valpacos, we find a site comparable in chronology and findings to the sacred environment of Panóias (Figure 35).198 Though like Pena Escrita, this sacred space was first published relatively recently in 1989 by Santos Júnior, as more and more attention has been paid to rock sanctuaries. Spatial Organization The physical makeup of the space consists of an orthogonal disposition with two symmetrical stairs composed of nine and eleven steps, respectively approaching the west and east side of the largely rectangular rock outcrop (Figure 36).199 Leading upwards towards the top of the rock structure one is confronted with two vast rectangular depressions at the west end, much like what we see at the top of the third rock at Panóias where performative rituals were hypothesized to have taken place (Figure 37).200 Pia dos Mouros also features depressions surrounded by                                                  198 Correia Santos 2010c: 8. 199 Correia Santos 2010b: 194; Correia Santos 2010c: 8; Freitas 2001. 200 Correia Santos 2010b: 194; the rock depressions measure 47 x 0.60 x 0.35 m and 2.02 x 0.60 x 0.40 m. 58  orthogonal carvings which indicate a foundation of a three sided precinct,201 similar once more to what is believed to have existed at the tallest rock structure at Panóias. Inscriptions Epigraphic evidence is also visible at Pias dos Mouros as Rodríguez Colmenero in 1995 recorded the existence of Latin lettering to the left side of the upper cavity which reads APADAV.202 In addition, there are also two inscriptions to the left side of the smaller depression. While the first is rather worn and appears to have been intentionally damaged, the other positioned below it is appears to have reproduced the former as both inscriptions feature similar lettering in the same positions.203 For instance, the first inscription though challenging to read denotes ‘APRO’ which emerges in the second inscription as APAO, arguably replacing the first after reading it became too problematic to make out.204  Ritual Function One of the primary arguments for interpreting this space as ritual is the similarity of the size of the cavities to Panóias.205 For an example, Tranoy argues that the presence of two basins and a double access staircase indicate organized ritual similar to Panóias.206 However, the similar shape of the cavities does not necessarily equate a similar function, as the spaces could have performed other functions, such as burial.207 Thus, explanations for the function of these                                                  201 Correia Santos 2010c: 8. 202 Rodríguez Colmenero 1995: 77-8. 203 Correia Santos 2010b: 194; Correia Santos 2010c: 8. 204 Correia Santos 2010b: 194; Correia Santos 2010c: 8. 205 See Rodríquez Colmenero 1995: 194 where it is posited that excavation would validate interpretations that the site follows the Panóias model. 206 Tranoy 2004: 88. 207 Cremation was customary in the late Republic and early imperial period, but funerary clubs were also known throughout Hispania whereby tombstones and communal mausoleums have been discovered. Wealthier individuals are also known to have commemorated their death with private burials and the erection of tombstones. Although this particular form of burial rites, in carved out cavities of rock outcrops, remains Thus, it is hard to discern whether or not spaces such as the rock outcrop at Pias dos Mouros would have facilitated burial means seeing as scholars have 59  inscriptions range from possible burial use after the site’s use for ritual means was abandoned, thereby denoting the name of the deceased, to an invocation to a particular local deity previously unknown to us.208 In absence of real epigraphic formulae, as is available to us for Panóias, further interpretations for other sites either lacking existence of rock structures, or more indicative inscriptions, as is the case at Pias dos Mouro, are difficult to obtain. Applicability of the Panóias Model Panóias has become our best model for understanding ritual spaces such as these in the north western corner of the Iberian Peninsula. The similar size of the rock outcrop, as well as dimensions of the cavities, has led interpretations to rationalize strong ritual functions akin to those at Panóias. However, Pias dos Mouros is largely lacking the archaeological context and detailed inscriptions corresponding to specific cavities used for ritual sacrifice that warranted such levels of certainty for the interpretation of the sanctuary of Panóias’ ritual function. Perhaps instead, the cavities similar in size to the two cavities on the top of the third rock of Panóias previously noted to suitably fit a body lying supine comfortably, were used for human burials. The inscription would then have been a name denoting the occupant of the burial site. Thus, it is somewhat unfair to treat other sanctuaries that appear to have one or two similar features to Panóias as identical in function.  Contribution Pias dos Mouros uses the natural landscape to create a physical manifestation of ritual—whether religious or funerary—with the carved outcrop and man-made depressions. While the function may not be analogous to that of Panóias, evidence at Pias dos Mouros remains consistent with                                                                                                                                                              yet to compare these sites in depth to the burial practices of inhabitants of Roman Spain. (Alarcão 1988; Keay 1988 89-93).  208 Correia Santos 2010b: 195. 60  the discovery of the epigraphic habit becoming engrained in Hispania. Moreover, Pias dos Mouros contributes to our understanding of the significance of sites largely similar to Panóias. These sites see to the manipulation of the sacred landscape for ritual means conducted by inhabitants in the Roman Empire where epigraphy is being traditionalized and sacrificial rites prescribed, as inhabitants continue to embark on a negotiation of their provincial identities.  3.4 – Peñalba de Villastar In addition to exploring the applicability of the Panóias model to other expressions of rural religion in the northwestern corner of the Iberian Peninsula, it is worth briefly expanding our search to the other side of the Peninsula where ritual spaces continue to be compared to Panóias (Figure 1). A look at the case of sacred inscriptions and space at Peñalba de Villastar is a suitable extension of the already discussed themes of inventing traditions through the epigraphic habit, as well as the problematic layer of assumptions that at times presuppose interpretations of spaces sharing similarities with Panóias. For instance, scholars such as Francisco Marco Simón when describing examples of “religious Romanization” at Peñalba makes assumptions like: “It seems evident that the groups of worshippers who frequented the sanctuary were traditionally Celtiberians, although we do not have any firm evidence of their origin” which prove problematic when agency at the site remains unclear and there is no concrete evidence supporting interpretations derived from the Romanization model.209 Instead, we should focus on the evidence at hand and what indications of new traditions taking form that it exemplifies, as well as how it responds to what we already know concerning religious praxis from the sanctuary of Panóias.                                                   209 Beltrán Lloris and Marco Simón 2007: 178. 61  Epigraphic and Spatial Context The sanctuary of Peñalba de Villastar is located in Teruel in northeastern Spain on a remarkably high cliff measuring nearly 2 km in length and 1000 m above ground level (Figure 38).210 Discovered in 1910 by J. Cabré, it was originally interpreted as a sacred mountain that facilitated local pilgrimages211 and served as a meeting place for varying ethnic and religious groups.212 Peñalba de Villastar was also located in a rural area far removed from any neighbouring cities.213 What is more, the lack of proximity to urban centres implies a less watchful eye over the religious praxis taking place. Epigraphic evidence and comparative analysis have exemplified the levels of religious worship taking place at Peñalba de Villastar, ranging from public rituals to individual practices.214 The cliff thereby served as a canvas for votive inscriptions, while a rock outcrop at the site served sacrificial purposes. And while many votive Paleohispanic as well as Latin inscriptions etched on the walls indicate the space was used for ritual purposes, the specificity of those rituals remain a mystery due to an absence of other material culture or monumental structures found in past excavations.215 Nevertheless, perhaps further analysis of the significance of such inscriptions and presumed sacrificial space will elucidate a better understanding of not                                                  210 Chronology for the site remains uncertain as the discovery of etched verses from Virgil’s Aeneid have acted as a terminus post quem roughly dating the inscriptions to the end of the first century B.C.E. and the beginning of the following century. However, there is no further archaeological evidence to corroborate this. The date was instead likely accepted by scholars because it coincides with a period of transition in Hispania. For instance, Alfayé Villa and Marco Simón have asserted that the majority of graffiti found on the side of the cliff at Peñalba served as dedications to various Celtic gods, with votive inscriptions appearing at the end of the first century B.C.E. (Alfayé Villa and Marco Simón: 2008: 283-4; Marco Simón 1999: 153). 211 The term ‘pilgrimage’ here is used to refer to travel to venerate a place god, that is a god who is limited to the natural landscape or place they embody and so individuals or collective groups travelled to fulfill pious acts at said places. For more on the distinction between religious pilgrimage and simply travelling to a sanctuary see Turner 1978. 212 Alfayé Villa and Marco Simón: 2008: 283. 213 Marco Simón 1999: 153. 214 Alfayé Villa and Marco Simón: 2008: 284. 215 Alfayé Villa and Marco Simón: 2008: 283-6; While no structures have yet been discovered at the site, holes in the walls of the cliff have led some to believe sacred structures could have been present during occupation. 62  only how—or if—Peñalba corresponds with the Panóias model, but how it contributes overall to our study of rural sanctuaries in Hispania. Votive Inscriptions Peñalba de Villastar features a wide array of votive inscriptions demonstrating the adoption of the epigraphic habit and other cultural nuances.216 Fortunately, despite precarious care and retrieval—and even ruining archaeological context at times—many of the inscriptions have survived to us.217 While the majority of existing inscriptions appear in the traditional language of the Celtiberians inhabiting the region, some of the inscriptions feature the use of the Latin alphabet, a custom also implemented at Cabeço das Fráguas.218 The use of Latin in some of these inscriptions implies that the language of power and prestige was only deliberately applied some of the time, given that most were consecrated in the Celtiberian language.219 As we saw with Pena Escrita, whether or not these inscriptions in varying languages can be taken to signify defiance or compliance to a new order of languages under Roman hegemony remains conjectural. What we do know is that these rock inscriptions denote the adoption of epigraphy as a new ritual form, similar in significance to what we have seen happening at all of the previously discussed case studies.220                                                  216 While the present study will only touch upon the inscriptions briefly, more extensive interpretations of further inscriptions at the sanctuary of Peñalba can be found in Beltrán Lloris and Marco Simón 2007. 217 Twenty-four of the inscriptions, ranging from inscribed names to the rare mention of ritual activity, have been properly documented thus far. However, the first excavation did see to the removal of thirty of the inscriptions in an attempt to protect them which only led to extraction holes and a lacking in our understanding of the epigraphy in its spatial setting (Alfayé Villa and Marco Simón: 2008: 283). 218 Beltrán Lloris and Marco Simón also interpret the cave inscriptions of Peñalba as an indication of the moment after which the Celtiberian language began to fade and give way to an increasing prominence of Latin (2007: 172).  219 Marco Simón, 2017: 206. 220 Geertz 1966: 28; Marco Simón 2017: 208. 63  Latin Names The majority of what epigraphists have made out are described as “masculine indigenous personal names,” such as Turos, Calaitos, Guandos or Aios,221 which exhibit Latin style names repeatedly carved at different points of the cliff. Also graffiti devoted to local deities such as the veneration to Cordonus by the dedicant Caius or Aius Atilius, exemplifying individuals with adopted Latin names worshipping local deities.222 While the quality of the preservation of these inscriptions inhibits further quantitative analysis, the presence of both Latin style and Latin names in consecrations to local deities honoured in the region show that individuals were communicating with their deities through Latinate names and at times even script. How individuals endeavoured to communicate with their deities is significant as it reflects their own renegotiation of cultural and religious identities. The “Gran Inscripción” A significant piece of evidence to our understanding of the sacred space at Peñalba de Villastar is the so-called “Gran Inscripción.”223 The Celtiberian text inscribed in the Latin alphabet honours the Cultic god Lugus, and it confirms the undertaking of pilgrimage at the cliff site (Figure 39).224 While scholarship has yet to produce an accepted translation, given that the text is composed in the Celtiberian language but transcribed in the Latin alphabet, Alfayé Villa and                                                  221 Alfayé Villa and Marco Simón: 2008: 284; Untermann 1997: 7. 222 Text 3 from the “Gran Inscripción” as described in Table 1 in Alfayé Villa and Marco Simón: 2008: 287. 223 The panel itself includes four epigraphs with three appearing above one another and the fourth situated on the right of the top one with one panel denoting an alphabet, one invoking deities, another recording the day of the year, and the final one indicating a sort of ritual. In any case, the readings of these epigraphs remain problematic due to difficulty of reading the texts and deciphering the language. For more see Beltrán Lloris and Marco Simón 2007. 224 There are other dedications to specific indigenous gods at Peñalba such as Cordonus though due to the majority and variety of gods I will not go into additional detail here, for more information see Alfayé Villa and Marco Simón: 2008; Marco Simón 1999: 153; Marco Simón 2017: 207. 64  Marco Simón in their recent work have described this text as the longest known inscription to the Celtic god Lugus as it reads:225   ENIOROSEI  VTA. TIGINO. TIATVMEI  ERECAIAS. TO. LVGVEI  ARAIANOM. COMEIMV  ENIOROSEI. EQVEISVIQVE  OGRIS. OLOCAS. TOGIAS. SISTAT. LVGVEI. TIASO.  TOGIAS.  Scholars interpret this text as an oath that would have been taken under the protection of deities mentioned in the inscription, such as Eniorosei, Tiatumei and Equaisos. However, the only agreed upon interpretation is that the text describes a “collective religious meeting” as denoted by “coneimu,”226 and that it involved the erection of “religious construction(s)” for the deities as mentioned with “togias,” despite the lack of any structural monuments found in past excavations.227 Thus, what we have is evidence that there were collective groups meeting at the site to fulfil pious promises that entailed the dedication of religious monuments or structures of some kind. While the particulars of such prescriptions remain ambiguous, it does demonstrate the mechanisms by which inhabitants were honouring their deities, such as through collective worship, dedication, and epigraphic means.  Language While the text exhibits the language of Celtiberians, the attempt to transliterate the text into Latin script points towards different systems of language training in place and the limitation of scribal education in this area. Much like what we saw at Cabeço das Fráguas, though individuals are                                                  225 Alfayé Villa and Marco Simón: 2008: 285. 226 Scholars have also taken the verb form comeimu as evidence of a performance of a pilgrimage at the sanctuary (Beltrán Lloris and Marco Simón 2007: 176). 227 Alfayé Villa and Marco Simón: 2008: 285-6. 65  trying to adopt the Latin language, the limited manner in which they are able to do so helps us understand how extensive their interactions with the Latin language were at the time.  In contrast, it can also demonstrate how transactional relationships with the gods materialized in the epigraphic habit, through Latin script or not.228 Perhaps Alfayé Villa and Marco Simón characterize this phenomenon best when they asserted: These rock inscriptions force us to rethink the role played by language and writing in the construction of provincial identities, and the importance of sanctuaries as public arenas where the individuals and the community could negotiate, display and strengthen their identities and status in the supernatural and in the human spheres through new ceremonial forms.229   The display and negotiation of identities through epigraphic means is definitely a phenomenon we have already seen at Panóias—not to mention Cabeço das Fráguas—though in a different context and with differing levels of meaning present.230 Measuring Time A further indication of new phenomena taking form in the sacred context of inscriptions at Peñalba de Villastar is the use of Roman calendar dates.231 For example, the inscription by Caius or Aius Atilius used the Roman calendar to appropriately date his visit on the day prior to the kalendas Maias, possibly corresponding to the Palilia which celebrated the founding of Rome.232 Another text in the “Gran Inscripción” also corresponds to the Roman religious calendar as it dates to the Larentalia when Marcus Carbo, inscribed “X kalendas Ianuarias.”233 Curchin                                                  228 Beltrán Lloris and Marco Simón compare the relationship between language and veneration to deities with examples at other sites in Roman Spain (such as Castejón and Navarra) where the appearance of not only Latin script, but a Latin alphabet inscribed nearby was used in a votive or magical context. 229 Alfayé Villa and Marco Simón: 2008: 300. 230 Beltrán Lloris 1999. 231 Alfayé Villa and Marco Simón: 2008: 287. 232 Alfayé Villa and Marco Simón: 2008: 286. 233 Ibid; the Larentalia were funeral rites held in honour of Acca Larentia by the pontiffs and the Flamen Quirinalis and has been postulated by Beltrán Lloris and Marco Simón to be evidence of pilgrims engaging in “Romanitas worship” (2007: 177). While a convincing argument on the surface, we still know little for sure of the agency at this 66  interprets the use of Roman calendar dates as proof that the cosmological understanding of the year was now in Roman terms.234 However, is it fair to presuppose the transition to Roman conceptualizations of time from a limited sample of inscriptions when the agents inscribing them use Latin names? Just because a small sample of the inscriptions denote Latin names and Roman calendar dates, does not mean that we should assume they signify the adoption of Roman culture as traditional acculturation models would have us suppose. Instead, an agent inscribing their Latinate name and recording time according to the Roman calendar could already be familiar with Roman culture to begin with, be it through prior military experience or former settlement. Given the scarcity of evidence, perhaps all we can—and should—concretely argue from the use of Latin names, script, and Roman calendar dates, is that they denote the presence of individuals familiar with mechanisms of expression previously unfamiliar to the region. Rock Depressions and Channels In addition to epigraphic evidence, the sacred landscape at Peñalba also consists of depressions and associated draining channels that are both similar and different from the types of sacrificial rites taking place at other sanctuaries like Panóias (Figure 40). The cavities at Peñalba de Villastar are believed by scholars to have served a similar function to Panóias.235 Conversely, it is worth keeping in mind that cavities carved out of rocks such as these, especially differing in shape and dimensions from those at the granite rocks at Panóias, could warrant different functions in antiquity, such as the collection of rain water, use as oil or wine presses, and other                                                                                                                                                              site and so it is hard to say what said participation in Roman form of worship would say about the inhabitants besides the adoption of new traditions taking place in a sacred context.  234 Curchin 2004: 240. 235 Alfayé Villa and Marco Simón: 2008: 286; Burillo 1997: 234; Marco Simón 1986: 746-747. 67  mechanical structures.236 Therefore, the presence of drainage channels carved out of rocks does not necessarily necessitate sacrifice, but could have facilitated a number of more practical means.  Applicability of Panóias Model Overall, the difference in size, shape and general form of the carvings at the rock outcrop at Peñalba to confirmed cavities facilitating sacrificial rites at Panóias augur that while sacrificial rituals may have taken place there, we lack the necessary evidence to corroborate that theory. Moreover, the technologies of worship and motivation behind adopting the epigraphic habit at this space are quite different than what we see at Panóias. Instead of a senatorial landowner consecrating a space for the worship of his own preferred deity alongside those of the local inhabitants, we have evidence of a range of ritual worship taking place. Furthermore, at Panóias we have a single agent attempting to create a catalyst for eternal worship at his site through sacrificial means, while at Peñalba, the majority of our evidence points instead towards isolated moments in time in which an individual inscribes his name at the site to commemorate the fulfillment of a pious promise, not necessarily expressing intent to return or continue use there. Perhaps the religious expressions at Peñalba were more fleeting and commemorative in nature than the intended eternal and prescriptive character we see at Panóias. Thus, the site at Peñalba de Villastar facilitates engagement with deities on a whole other scale given and can therefore be looked at as a different model than Panóias in various ways.  Contribution While Peñalba de Villastar is by no means a direct replica of the Panóias model, the various levels of engagement with the gods and differing religious praxis speak to the development of the epigraphic habit and spread of Latin language and mechanisms to the region of Hispania. The                                                  236 See Pérez Jordá 2000 for non-ritual uses of rock structures. 68  heterogeneous array of inscriptions denoting practices and levels of devotion offer us a valuable look into the different stages of interaction individuals or groups could engage in with their deities. Moreover, the appearance of new forms of worship demonstrate how traditional customs in religious belief systems could adapt to fit present conceptualizations of religious expression, such as continuing to venerate the god of a particular place but through the inscription of one’s name and dedication via rock carvings.  While the context under which such rituals are taking place at Peñalba may not fit the Panóias model scholars have often become too keen to inflict on it, it presents its own valuable contribution to our understanding of religious expressions and traditions in Roman Spain.  3.5 – Panóias as a Model While the Panóias model helps us reconsider spaces with a lack of archaeological context or presence of descriptive inscriptions, it is not necessarily applicable to all sacred sites in the Iberian Peninsula that remain marginalized in scholarship or share slight similarities to features indicating ritual function at Panóias. Perhaps the influence that our modern understanding of the sanctuary of Panóias has had on other sites has become excessive.  For example, Prósper argued that Cabeço das Fráguas had to have been a natural sanctuary because its rock inscription was similar to Panóias.237 However, in order to emphasize the site’s connections to Panóias, its dissimilarities to Panóias, such as a lack of understanding who consecrated the site or participated in the rituals or an absence of cavities relating to the ritual performance of animal sacrifice, are largely ignored.  Rather than focus on the presence of other rock sanctuaries in the Iberian Peninsula as analogous with the beliefs or practices exhibited at Panóias, we should look at the overarching                                                  237 Prósper 1999: 153 no. 3. 69  ideologies that drive such instances of creating sacred spaces to perform ritual worship. An example of an ideology worth exploring in the archaeological record of the rural sphere as demonstrated through the aforementioned case studies is the adoption of the epigraphic habit in order to continue the veneration of old deities under new mechanisms of worship. Religious practices—and consequently beliefs—in Roman Spain were not only changing, but they were adopting new forms under the guise of remaining ancient. The physical manifestations of such cultural adoption and adaptations in the region through rock sites and inscriptions not only signify changing belief systems, but the manner in which they were mediated and identities’ negotiated.  Therefore, while the Panóias model does not necessarily “summarize all the others,”238 the other sites still provide their own contributions to our understanding of the various beliefs and practices of inhabitants of the Iberian Peninsula. What’s more, the consideration of other rock sites in the region demonstrates that Panóias was not alone in its adoption of the epigraphic habit and intent to invent traditions as a method of mediating past religious praxis with new ones.                                                          238 Rodríguez Colmenero 2000: 164. 70  4. Final Reconsiderations Perhaps it goes without saying that rock sanctuaries provide invaluable insight, not only into expressions of identity and religious beliefs and practices, but of inhabitants’ conception of the landscape in which they gathered to experience such religiosity. It becomes especially significant to study rock sanctuaries in Roman Spain when we consider that the institution and maintenance of religious practices arise out of choices made by individual or collective agents and can thereby be demonstrative of the negotiation of identities of a particular time and place.239   One such choice that emerged in Roman Spain is the introduction of inscription as a form of religious worship. Individuals and religious communities across the Iberian Peninsula came to understand their communication with deities through not only language in general, as demonstrated with the proliferation of votive inscriptions under the Empire, but also at times through the language of the Roman Empire, Latin. In any case, language and writing played a vast role in the construction of provincial identities and are crucial to our understanding of religious praxis.240  But the epigraphic record is not the only manner of evidence vital to our understanding of the invention of tradition in rural religion. Expressions of religious practices also survive to us through the physical manifestation of sacred spaces such as rock sanctuaries and the presence of cutout cavities. The use of carved rock outcrops despite the spread of Graeco-Roman temples in urban centres like Tarraco, Emporion, or Emerita Augusta, could exhibit the continuity of traditional practices alongside new manners of worship and conceptualizations of space. However, as we continue to move beyond Strabonian models of interpretation, a lot remains to be learned from these sites from an interpretative and analytical standpoint. The                                                  239 McCarty 2017: 395. 240 Woolf 1996. 71  endeavour to analyze and further understand these spaces remains scarce as out of almost 150 sites referred to as ‘rock sanctuaries,’ few have become the subject of in-depth studies.241 We can trace some elements of religious praxis, such as the negotiation of religious identities, the invention of tradition, the epigraphic habit, the use of rock structures as sacred space to perform of sacrificial rituals, and occasionally even agency. But seeing as unidirectional “Romanization” and acculturation models fail to capture the range of creative dynamics at play, it is only through the application of an alternative model and further excavation that we can better grasp the complexity of religious behavior in Roman Spain.                                                               241 Recently, scholars such as Correia Santos have called for more exhaustive analyses of these rock sanctuaries (2010a: 125). Sites such as Panóias and Pias dos Mouros are already undergoing new excavations and can hopefully provide us with more comprehensive understanding of the religious praxis of these cult spaces from “interdisciplinary and contextualized perspectives” in the future (Correia Santos 2010a: 151). 72  Figures  Figure 1. Map of Roman Hispania circa first century C.E. with distribution of discussed case studies.   Figure 2. General Plan of the archaeological site of Panoias (labels and proposed ritual progression by author) (Sousa and Ribeiro da Silva 2013: 68). 73   Figure 3. Topographical plan of rock outcrop with five numbered inscriptions (labelled by author) and a proposed ritual progression made by Alföldy (1997: 199).    Figure 4. First rock monument in rock sanctuary group at Panoias (prehistoriadelsur). CIL II 2395a CIL II 2395e CIL II 2395c CIL II 2395d CIL II 2395b 74   Figure 5. Rectangular and circular cavities cut out of the bed rock as well as post holes (prehistoriadelsur). 75   Figure 6. CIL II 2395a (Argote 1732: VII: 568 ff.).   Figure 7. CIL II 2395a according to Alföldy (1999: Abb. 3). 76   Figure 8. CIL II 2395a photo by Hispania Epigraphica Record No. 8213.  Figure 9. Traced drawing of CIL II 2395a following the application of MRM (Corriea Santos, Pires, & Sousa 2014: 208). 77   Figure 10. CIL II 2395b (Hispania Epigraphica Record No. 8214).  Figure 11. CIL II 2395b according to Alföldy (1999: Abb. 4).      78   Figure 12. Gastra located above CIL II 2395b (prehistoriadelsur).   Figure 13. CIL II 2395c (Alföldy 1997: 185). 79   Figure 14. CIL II 2395c (Hispania Epigraphica Record No. 8215).  Figure 15. CIL II 2395c Traced drawing made from polychromatic MRM (Corriea Santos, Pires, & Sousa 2014: 218). 80   Figure 16. CIL II 2395d (Hispania Epigraphica Record No. 7696).   Figure 17. Rectangular cavity above CIL II 2395d used to burn entrails (prehistoriadelsur). 81   Figure 18. Inscription CIL II 2395d located to the left of small stairs leading to sacrifical area (prehistoriadelsur).  Figure 19. CIL II 2395d, Alföldy (1999: Abb. 6).  CIL II 2395d 82   Figure 20. Traced drawing of CIL II 2395d made with polychromatic MRM (Corriea Santos, Pires, & Sousa 2014: 222).   83     Figure 21. Second rock at Panóias from afar (prehistoriadelsur).  Figure 22. Second rock at Panoias (prehistoriadelsur). 84   Figure 23. Second Rock with missing inscription on left and stairs in the centre (prehistoriadelsur).  Figure 24. Second rock at Panoias, Ariel view. Inscription 85   Figure 25. Third rock at Panoias stairs leading up (prehistoriadelsur).  Figure 26. Third rock at Panoias (prehistoriadelsur). 86   Figure 27. Ariel view of Pena Escrita (Correia Santos 2010a: 131).  Figure 28. Pena Escrita Sanctuary in Vilar de Perdizes (Rodríquez Colmenero 1995: 191). 87   Figure 29. Topographical drawing of Pena Escrita (Corriea Santos 2010a: 132).  Figure 30. Reconstruction of the sanctuary of Pena Escrita (Correia Santos 2010b: 186). 88   Figure 31. Votive altar dedicated to Larcauco (Rodríquez Colmenero 1995: 127).  Figure 32. Sanctuary of Cabeço das Fráguas granite outcrop with inscription painted over with chalk (Olivares  Pedreño 2002). 89   Figure 33. Lusitanian Inscription at Cabeço das Fráguas (Untermann 2002: fig. 2).  Figure 34. Possible second inscription at Cabeço das Fráguas (Rodríquez Colmenero 1995: 223).  90   Figure 35. Pias dos Mouros (Correia Santos 2010b: 193).  Figure 36. Plan of Pias dos Mouros (Santos Júnior et al. 1989: 379). 91   Figure 37. Two rectangular cavities with a detailed look at the inscriptions (Correia Santos 2010b: 194).  Figure 38.  General view of the sanctuary of Peñalba de Villastar (Alfayé Villa and Marco Simón 2008: 283). 92   Figure 39. "Gran Inscripción" at Peñalba de Villastar (after J. Cabré 1909-1910) 93   Figure 40. Rock depressions and channels at Peñalba (Alfayé Villa and Marco Simón 2008: 286).    94  Bibliography Abbott, F. F. (1911). A History and description of roman political institutions. (3 ed.). New York; Boston;: Ginn & company. Alarcão, J. d. (1988). Roman Portugal (Vol. 1). Warminister, Wiltshire: Aris & Phillips. Alarcão, J. d. (2001). Novas Perspectivas Sobre Os Lusitanos (e Outros Mundos). Revista Portuguesa De Arqueologia, 4 (2), 293-350. Alfayé, S., & Marco Simón, F. (2008). Religion, Language and Identity in Hispania: Celtiberian and Lusitanian Rock-Inscriptions. In R. Häussler, Romanisation et epigraphie. Etudes interdisciplinaires sur l'acculturation et l'identité dans l'Empire Romain. Montagnac. Alföldy, G. (1997). The Mysteries of Panóias (Vila-Real, Portugal). Madrid Communications, 38, 176-246. Alvar, J. (2002). Cultos Orientais e Mistéricos na Província da Lusitânia. In J. Cardim Ribeiro. Lisboa: Loquuntur Saxa: Religiões da Lusitânia. Argote, J. C. (1732). Da cicade de Panóias e das antiguidades e vestígios que actualmente existem dela. In Memórias para História Eclesiástica do Arcebispado de Braga (pp. 325-347). Portugal: VII, II. Augé, M. (2009). Pour une anthropologie de la mobilité. Paris. Beltrán Lloris, F. (1999). Writing, Language and Society: Iberians, Celts and Romans in Northeastern Spain in the 2nd & 1st Centuries BC. Bulletin of the Institute of Classical Studies, 43, 131-151. Beltrán Lloris, F., & Marco Simón, F. (2007). New Inscriptions in the Sanctuary of Peñalba de Villastar (Teruel). In A. C. King, & P. Andrews, Continuity and Innovation in Religion in the Roman West (pp. 168-182). Portsmouth, R.I.: Journal of Roman Archaeology. Bendlin, A. (2013). Suovetaurilia. In R. S. Bagnall, K. Brodersen, C. B. Champion, A. Erskine, & S. R. Huebner (Eds.), The Encyclopedia of Ancient History (pp. 6456-7). Blackwell Publishing. Blázquez, J. M. (1962). Religiones primitivas de hispania. Consejo Superior de Investigaciones Cientificas, Delegación de Roma, 14. Madrid. Blázquez, J. M. (1983). Primitivas Religiones Ibericas: Tomo II, Religiones Preromanas. Madrid. 95  Blázquez, J. M. (2001). Religiones, ritos y creencias funerarias de la hispania prerromana. Madrid: Biblioteca Nueva. Blázquez, J. M. (2009). Romanising oriental gods: Myth salvation and ethics in the cults of cybeles, isis and mithras. Gerión, 27 (2), 221-223. Bourdieu, P. (1977). Outline of a Theory of Practice (Vol. 16). Cambridge; New York;: Cambridge University Press. Bruun, C., & Edmondson, J. C. (2015). The Oxford Handbook of Roman Epigraphy. New York; Oxford;: Oxford University Press. Burillo, F. (1997). Espacios cultuales y relaciones étnicas: contribución a su estudio en el ámbito turolense en época ibérica. Cuadernos de Prehistoria y Arqueología Castellonense, 18, 229-238. Castiglione, L. (1970). Vestigia. Acta Archaeologica(22), 95-132. Correia Santos, M. J. (2010a). Inscrições rupestres do Norte de Portugal: novos dados e problemática. Sylloge Epigraphica Barcinonensis, VIII, 123-152. Correia Santos, M. J. (2010b). Inscripciones rupestres y espacios sagrados del norte de Portugal: nuevos datos y contextualización. Los casos de Pena Escrita, Mogueira y Pias dos Mouros. In A. Arenas-Estéban (Ed.), Celtic Religion Across Space and Time (pp. 180-199). Castilla-La Mancha. Correia Santos, M. J. (2010c). Rock Sanctuaries:the construction of a sacred landscape in Western Iberian Peninsula. Unpublished Working Paper. Retrieved from https://www.scribd.com/document/26051325/Rock-Sanctuaries# Correia Santos, M., Pires, H., & Sousa, O. (2014). Nuevas lecturas de las inscripciones del santuario de Panóias (Vila Real, Portugal). Sylloge Epigraphica Barcinonensis, 197-224. Cortez, F. R. (1947). Panoias Cividade Dos Lapiteas: Supsidios Para o Estudo dos Cultos Orientais e da Vida Provincial Romana na Regiao do Douro. Instituto do Vinho do Porto. Curado, F. P. (2002). A ‘ideologia tripartite dos indeuropeus’ e as religiões de tradição paleohispânica no Ocidente peninsular. In L. Raposo (Ed.), Religiôes da Lusitania. Loquuntur saxa (pp. 71-77). Lisboa. Curchin, L. A. (2004). The Romanization of Central Spain: Complexity, Diversity and Change in a Provincial Hinterland. London; New York: Routledge. 96  Derks, T. (1998). Gods, Temples, and Ritual Practices: The transformation of religious ideas and values in Roman Gaul (Vol. 2). Amsterdam: Amsterdam University Press. Fisher, K. (2009). Placing social interaction: an integrative approach to analyzing past built environments. Journal of Anthropological Archaeology(28), 439-57. García y Bellido, A. (1967). Les Religions Orientales dans L'Espagne Romaine. Leiden: E. J. Brill. García, J. M. (1991). Religiões Antigas de Portugal. Lisboa: Imprenta Nacional. Geertz, C. (1966). Religion as a Cultural System. In In Anthropological Approaches to the Study of Religion (pp. 1-46). London: Tavistock. Grahame, M. (1998). Material Culture and Roman Identity. In R. Laurence, & J. Berry (Eds.), Cultural Identity in the Roman Empire (pp. 156-178). London; New York;: Routledge. Haverfield, F. (1905). The Romanization of Roman Britain. ProcBritAc. Hobsbawm, E. (1992). Introduction: Inventing Traditions. In E. Hobsbawm, & T. Ranger, The Invention of Tradition (pp. 1-14). New York; Cambridge: Cambridge University Press. Hopkins, K. (1978). Economic Growth and Towns in Classical Antiquity. In P. Abrams, & E. Wrigley (Eds.), Towns in Societies (pp. 35-77). Cambridge: Cambridge University Press. Horden, P., & Purcell, N. (2000). The corrupting sea: A study of mediterranean history. Malden, Mass: Blackwell. Huet, V. (2008). Le relief dit des suovétauriles de Beaujeu : une image sacrificielle hors de l'Italie. In S. Estienne, D. Jaillard, N. Lubtschansky, & C. Pouzadoux (Eds.), Image et religion dans l'Antiquité gréco-romaine (pp. 399-408). Naples: Centre Jean Bérard. James, P. (2015). Urban Sustainability in Theory and Practice: Circles of Sustainability. London; New York: Routledge. Keay, S. J. (1988). Roman Spain. London: British Museum Publications. Keay, S. J., & Terrenato, N. (2001). Italy and the West: Comparative Issues in Romanization. Oxford: Oxbow Books. Kindt, J. (2009). Polis Religion - A Critical Appreciation. Kernos, 22, 9-34. 97  Kulikowski, M. (2013). Religions of Roman Spain. In M. A. Sweeney, & M. R. Salzman (Eds.), The cambridge history of religions in the ancient world (Vol. 2, pp. 510-532). Cambridge;New York;: Cambridge University Press. Lambrino, S. (1954). Les Divinites Orientales en Lusitanie et le Sanctuaire de Panoias. Bulletin des Etudes Portuagaises et de L'Institut Francais au Portugal, 18. Le Roux, P. (1995). Romains d'Espagne. Paris: Armand Colin. Marco Simón, F. (1986). El dios céltico Lug y el santuario de Peñalba de Villastar. VVAA., Estudios en homenaje del Dr. A. Beltrán, 731-759. Marco Simón, F. (1999). Divinidades indígenas en la Hispania indoeuropea. Veleia, 16, 33-49. Mattingly, D. J. (2013). Imperialism, Power, and Identity: Experiencing the Roman Empire. Princeton: Princeton University Press. McCarty, M. M. (2013). Continuities and Contexts: The Tophets of Roman Imperial-Period Africa. Studi Epigrafici e Linguistici, 29-30, 1-26. Millett, M. (1990). The Romanization of Britain. Cambridge: Cambridge University Press. Moser, C., & Knust, J. (2017). Ritual Matters: Material remains and Ancient Religion. Michigan: University of Michigan Press. Olivares Pedreño, J. C. (2002). Los Dioses De La Hispania Céltica (Vol. 15). Alicante; Madrid: Real Academia de la Historia. Pérez Jordá, G. (2000). La conservación y la transformación de los productos agrícolas en el mundo ibérico. Saguntum Extra, 3, 47-68. Prósper, B. M. (1999). The Inscription of Cabeço Das Fráguas Revisited. Lusitanian and "Alteuropäisch" Populations in the West of the Iberian Peninsula. Transactions of the Philological Society, 97(2). Revilla Calvo, V. (2002). Santuarios, élites y comunidades cívicas: consideraciones sobre la religión rural en el conventus Tarraconensis. In F. Marco, F. Pina, & J. Remesal (Eds.), Religión y propaganda política en el mundo romano (pp. 189-226). Barcelona. Richert, E. A. (2005). Native Religion under Roman Domination: Deities, Springs, and Mountains in the North-West of the Iberian Peninsula. Oxford, England: Archaeopress. Rives, J. B. (2007). Religion in the Roman Empire. Malden, MA: Blackwell Publishing. 98  Rives, J. B. (2015). Religion in the Roman Provinces. In C. Bruun, & J. C. Edmondson, The Oxford Handbook of Roman Epigraphy (pp. 420-444). Oxford; New York: Oxford University Press. Rodrigues, A. V. (1959). O castro do Cabeço das Fráguas e a romanização das suas imediações. Separata de Beira Alta, XVIII (1-2), 111-128. Rodríguez Colmenero, A. (1993). Corpus-Catalogo de Inscripciones Rupestres de Epoca Romana del Cuadrante Noroccidental de la Peninsula Iberica. La Coruna. Rodríquez Colmenero, A. (1979). Augusto e Hispania: Conquista y organización del norte peninsular. Cuadernos de Arqueologia de Duesto Seminario de Arqueologia, 7. Rodríquez Colmenero, A. (1995). Corpus de inscripciones rupestres de época romana del cuadrante noroeste de la Península Ibérica. In A. Rodríguez Colmenero, & L. Gasperini (Eds.), Saxa Scripta (Inscripciones en Roca), Actas del Simposio Internacional Ibero-Itálico sobre Epigrafia Rupestre (Santiago de Compostela y Norte de Portugal, 1992), Anejos de Larouco 2, Sada (pp. 117-259). Rodríquez Colmenero, A. (1999). O santuário rupestre galaico-romano de Panóias (Vila Real, Portugal): Novas achegas para a sua reinterpretação global. Lisboa: Ministério da Cultura. Rodríquez Colmenero, A. (2000). Deorum Temene. Espacio sagrado y santuarios rupestres en la Gallaecia romana. Un intento de clasificación. (V. O. Jorge, Ed.) Arqueología da antiguidade na Península Ibérica. Actas do III Congreso de Arqueología Peninsular, VI, 153-196. Salinas de Frías, M. (2001). La religiosidad de las poblaciones antiguas de Salamanca y el Norte de Cáceres. Paleohispanica, 1, 151-172. Sandys, J. E., & Campbell, S. G. (1919). Latin Epigraphy: An Introduction to the Study of Latin Inscriptions.  Schattner, T., & Correia Santos, M. (Eds.). (2010). Porcom, Oilam, Taurom: Cabeço das Fráguas, o santuário no seu contexto. Guarda. Scheid, J. (1990). Romulus et ses frères: Le collège des Frères arvales, modèle du culte public dans la rome des empereurs (Vol. 275). Rome: Ecole française de Rome. Scheid, J. (1996). Pline le Jeune et les sanctuaires d’Italie. Observations sur les Lettres IV, 1, VIII, 8 et IX, 39. In A. Chastagnol, S. Demougin, & C. Lepelley, Splendidissima civitas. Études d’histoire romaine en hommage à François Jacques. Paris. 99  Scheid, J., & Ando, C. (2016). The Gods, the State, and the Individual: Reflections on Civic Religion in Rome. Philadelphia: University of Pennsylvania Press. Sinner, A. G. (2015). Cultural contacts and identity construction: A colonial context in NE spain (2nd - early 1st c. B.C.). Journal of Roman Archaeology, 28, 7-37. Sousa, R., & Ribeiro da Silva, J. (2013, February). Serápis Nos Confins Do Império: O Complexo Sagrado De Panóias. (R. Â. Araújo, Ed.) Tovar, A. (1985). La inscripción del Cabeço das Fráguas y la lengua de los Lusitanos. Actas del III Coloquio sobre Lenguas y Culturas Paleohispanicas, 227-254. Tranoy, A. (1981). La Galice Romaine. Paris: Recherches sur le Nord-Ouest de la Péninsule Ibériquedans l’Antiquité. Tranoy, A. (2004). Panóias ou les rochers des dieux. Conimbriga, 43, 85-97. Turner, V., & Turner, E. (1978). Image and Pilgrimage in Christian Culture. Columbia University Press. Untermann, J. (1997). Monumenta Linguarum Hispanicarum. Wiesbaden: Band IV. Die tartessichen, keltiberischen und lusitanischen Inschriften. Van Oyen, A. (2015). Deconstructing and Reassembling the Romanization Debate Through the Lens of Postcolonial Theory: From Global to Local and Back? Terra Incognita, 6, 205-226. Várhelyi, Z. (2010). The Religion of Roman Senators in the Roman Empire: Power and the Beyond. Cambridge; New York;: Cambridge University Press. Vasconcellos, J. L. (1905-1913). Religiones Da Lusitania (Vols. II-III). Vaz, J. L. (2002). Tipologia dos santuários rupestres de tradição paleohispânica em território português. In Loquuntur Saxa, Religiões da Lusitânia. Lisboa. Vermaseren, M. J. (1957). The suovetaurilia in Roman art. Bulletin Antieke Beschaving, 32, 1-12. Versluys, M. J. (2014). Understanding objects in motion. An archaeological dialogue on Romanization. Archaeological Dialogues, 21 (1), 1-20. Webster, J. (2001). Creolizing the Roman Provinces. American Journal of Archaeology, 105 (2), 209-225. 100  Wild, R. A. (1984). Water in the Cultic Worship of Isis and Serapis. Études Préliminaires aux Religions Orientales dans l'Empire Romain, 1831-1832. Wiseman, F. (1956). Roman Spain: An Introduction to the Roman Antiquities of Spain and Portugal. London: Bell. Woolf, G. (1996). Monumental Writing and the Expansion of Roman Society in the Early Empire. Journal of Roman Studies, 86, 22-39. Woolf, G. (1998). Becoming Roman. Cambridge/New York: Cambridge University Press. Woolf, G. (2014). Romanization 2.0 and its alternatives. Archaeological Dialogues, 21 (1), 45-50.                


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items