Pork consumption, gastro-politics and social Islamisation in early al-Andalus (eighth to tenth centuries)

ABSTRACT This paper presents a novel perspective concerning the emergence of al-Andalus based on the study of food. This is a field of human behaviour that is highly informative in socio-cultural terms because of its links to ethno-religious identity. The aim is to demonstrate the usefulness of studying zooarchaeological evidence that provides information on the consumption of (or abstinence from) pork by different peninsular communities during the first centuries of al-Andalus. The results show the analytical potential of this line of research for examining the advance of the process of social Islamisation, understood as the fundamental dynamic of historical change in the shaping of Andalusi society, and one of the main mechanisms used by the Umayyad state to establish its power.


Introduction
The Islamic conquest of Iberia in 711 marked the beginning of a process that led to a profound transformation of the structures inherited from the late antique world. 1 From that point on, the Roman-Visigothic region previously known as Hispania was progressively transformed into an "Arab and Islamic polity" named al-Andalus. 2 The consequences of the peninsula becoming part of the social and cultural sphere of the medieval Arab-Islamic world were numerous, and they have been studied from different perspectives. However, the focus here will solely be on the relationship between foodways and two phenomena that were part of the basis for the formation of al-Andalus, namely the process of social Islamisation and the establishment of Islamic socio-political order. The starting point is to assume that foodways can sometimes be altered as a result of various factors that respond to the interests of dominant social groups. The motivations of those in power can have different goals, such as the establishment of measures aimed at discouraging the production of certain foods (i.e., the breeding of certain animals), or the promotion of market-based systems of food provisioning. However, the attitudes that are of greatest interest here are those related to the shaping of shared and homogenous socio-cultural or ethnic identities. In other words, food systems, 3 like any social product, are the result of historically determined negotiable and dynamic relationships that grew out of the merging of different interests. In certain historical contexts, this could have been manipulated by groups holding socio-economic, political, and/or religious power.
This contribution begins by assuming that the emergence of al-Andalus and the subsequent process of social Islamisation caused the diets which had been established in the late antique/Visigothic period to change. These changes took place in very different areas, including agricultural transformations resulting from the introduction of new plant species and new culinary habits that can be studied by using archaeobotanical 4 and ceramic 5 records, respectively. However, as will be shown here, the study of the consumption patterns of animal products-although these were a less substantial part of ancient diets than plant-based foods-enables the rhythms of the diffusion and implementation of the Islamic social order to be traced. By using an archaeological approach that focuses on regular dietary habits over extended periods, this study enables a "bottom-up" assessment to be carried out on the practical consequences of the process of social Islamisation in al-Andalus from an innovative (zoo)archaeological perspective.

Some thoughts on food and power
Diet and food consumption patterns are the result of the interaction of many different aspects, from purely economic and material ones to more profoundly cultural and symbolic ones. From a historical perspective, the multi-faceted nature of food makes it possible to offer insights into how past societies functioned at multiple levels. It is important to consider that the eating habits of socially organised groups and the ways in which food was produced, distributed, and consumed were not only the result of logical economic reasoning aimed at making the most of available resources. In addition to these aspects, the fundamental role played by other social, cultural, and political factors must also be considered when defining food systems. This paper emphasises the decisive importance of food and its consumption in the construction and reinforcement of social ties. It is the "glue" that bonds family and society while creating the individual. 6 Furthermore, it is consumed on a daily basis, following repetitive and recurrent patterns. 7 This all means that food is not only essential for the survival of communities, but also occupies a fundamental place among the practices that serve to inculcate habitus, that is, the set of dispositions or learned behaviours that structure the daily actions of individuals and that, unconsciously, generates identity and social reference. 8 This is why food and foodways have been historically used by human groups to establish and maintain the cultural boundaries that they hold with other communities. 9 As highlighted by Sidney M. Mintz and Christine M. Du Bois, "like all culturally defined material substances used in the creation and maintenance of social relationships, food serves both to solidify group membership and to set groups apart." 10 This function of food patterns being used for social organisation as a way of emphasising communal similarities or differences consequently transforms these patterns into instruments of identity demarcation, 11 that can potentially be subject to external political regulation aimed at societal control. In other words, food systems constitute a means by which power (see below) can convey its decisions and interests and transfer them (sometimes by force) to the rest of the population, in order to implement or strengthen-in both symbolic and practical terms-particular forms of socio-political organisation. As Yannis Hamilakis reminds us, food has historically played a prominent role in the dynamics of generating, maintaining, legitimising, and deconstructing authority and power. 12 Along with the concept of identity, power is arguably one of the most complex components of the conceptual repertoire which, particularly since the 1980s, has been used in historical and archaeological research. The notion of power that will be used here is partially based on Michel Foucault's idea that it is a force that constrains potential new situations and social phenomena while also making them possible. 13 Therefore, power is assumed not only to be the privilege of certain institutions or social groups whose legitimacy derives from the position they occupy within the structure of society, 14 but rather "the multiplicity of force relations" that exists in every social relationship and is not always monopolised by particular institutions or groups because it "is everywhere." 15 However, for our current purposes, Harold Barclay's loose notion of power as the capacity to exert influence-understood simply as the ability to make others do what you want them to do-will be adopted. 16 In his view, power can easily be understood as a continuum in which, at one extreme, there is domination and, at the other extreme, there is the exercise of influence without intent or pretence of domination. 17 The appeal of this definition of power lies in its simplicity; it is a useful analytical tool for the study of the topic addressed in this article.
The archaeological examination of eating patterns during the formative period of al-Andalus (eighth to tenth centuries) enables a series of observations to be made that emphasise the profoundly socio-political use of food in the context of historical change defined by the process of social Islamisation.
The social Islamisation of al-Andalus and the establishment of the Islamic social order The formation of al-Andalus was the result of a military conquest undertaken by an expanding empire with a fully structured value system ruled by an elite that was perfectly aware of its governing role. 18 If, as Sonia Gutiérrez Lloret notes, one of the most important results of the Arab-Berber conquest of 711 was the profound transformation of Hispanic society over the following centuries, how this process of social change took place needs to be explained. 19 The phenomenon can be tackled by using different approaches, but this paper focuses on the implementation process of the basic instruments of social and ideological control within the construction of a centralised state power which, at least from the establishment of the caliphate in the tenth century, did not allow its ideological definition to be questioned. 20 Indeed, the branch of the Umayyad dynasty founded in 756 by 'Abd al-Raḥ mān I, the first emir of al-Andalus, and which remained in power until the fall of Hishām III al-Mu'tadd in 1031, was mainly legitimised by the religious message it claimed to defend. 21 Historical testimonies confirm-no matter how many attempts there have been to distort the incontestable reality reflected in the abundant evidence available 22 -the Arab and Islamic identity of the conquerors and how the Andalusi state embraced these characteristics. As leaders and guides of the Muslim community, the Umayyad rulers identified completely with the principles of the strictest Islamic orthodoxy. 23 The fact is that, in al-Andalus, the triumph of the Umayyad state was also the triumph of Islam.
Therefore, the process that began with the conquest of 711 led to the full integration of al-Andalus into the socio-cultural context of the Arab-Islamic world that was being formed at the same time in the east and west. However, it was not until the establishment of the caliphate in 929 by 'Abd al-Raḥ mān III-the leading force behind what Gabriel Martínez-Gros calls "Andalusi identity" 24 -that the establishment of the Islamic order was complete. This has been defined by Eduardo Manzano Moreno as "a set of practices, norms and expressions that consecrate a certain way of organising and disciplining human experience [and serves] to consolidate power." 25 Julián M. Ortega Ortega has recently been reluctant to accept the role of Islam as practised by the conquerors as an expression of shared identity. In his opinion, "Islam served in this context as a system of elite recruitment, an instrument to regulate access to power," 26 dismissing the idea that the Arabs had an interest in the acculturation of the local population, who were guaranteed "the possibility of persevering in their ancestral customs." 27 However, despite the existence of temporal and spatial differences, as will be discussed below, the central role that Islam played in the shaping of Andalusi culture seems undeniable. As Alejandro García Sanjuán mentions in his review of Ortega Ortega's book, Arab and Islamic components were present in the political and social structure of al-Andalus since it began, meaning that the idea of acculturation, derived from the "imposition of the conquerors' own markers," must be considered. 28 Therefore, the feature that best defines al-Andalus is the Islamised character of its society, which is equivalent to the meaning that Marshall G.S. Hodgson gives to the term Islamicate to designate "the social and cultural complex historically associated with Islam and Muslims, both among Muslims themselves and even when found among non-Muslims." 29 The semantic reference that will be used here to refer to this process of historical change linked to the establishment of the Islamic order in Iberia during the formative phase of al-Andalus is social Islamisation. Since a detailed definition of this concept has been the subject of a recent publication, 30 in the present article, social Islamisation is simply defined as the set of social and cultural practices and behaviours that made up the orthopraxy of medieval Islam being incorporated into the fabric of the new Andalusi society. 31 This process, which represents the behaviours of the social actors, has a material expression that can be understood archaeologically.
It is necessary to highlight that this process of historical change, which fits under the conceptual umbrella of social Islamisation, is not confined solely to the imposition of a "top down" form of social life led by political power and religious authority. Islamisation should rather be understood as the product of a fluid process of discursive negotiation. This process enabled individuals to assess the social possibilities and options that unfolded before them, allowing them to enter the Islamic social order and adopt, in practice, the forms of medieval Islamic socialisation and orthopraxy. This included, as shown below, various provisions that directly affected food and foodways.
However, the role of the ruling dynasty during the first centuries of al-Andalus was the guarantor and maximum defender of Islam, and therefore it is important also to view Islamisation as one of the operational resources used in the process of building and consolidating the symbolic and material power of the Umayyad state. The idea is driven by the fact that the construction of the structure of the Andalusi state both required and fostered the spread of the Islamic order and social Islamisation. Therefore, both processes must be considered to have been linked and mutually conditioned. 27 Ortega Ortega, La conquista islámica, 295: "la posibilidad de perseverar en sus ancestrales costumbres." 28 García Sanjuán, review of La conquista islámica, 470: "imposición de los marcadores inherentes a los conquistadores." 29 Hodgson, The Venture of Islam, 59. The same idea underlies the work of other authors. For example, Acién Almansa, Entre el feudalismo y el islam, viii, notes that the Coptic groups in Egypt "can be considered as a socially 'Islamicised' population, insofar as they participate fully in the features of what I have called 'Islamic social formation'." More recently, Gutiérrez Lloret, "Early al-Andalus," 59-60, defined al-Andalus as "a totally Islamized society […] that collectively assumed the new ideology regardless of the religion practiced or the language spoken, as happened with the Christian Mozarabs." Similarly, Carvajal López, "The Archaeology of Al-Andalus," 334, stated "we can understand al-Andalus as an Islamicised society (that is, a society where Islam influences everyone, not only Muslims)." 30 García García, "La dimensión alimentaria." 31 Orthopraxy can be defined as the set of behavioural markers (both subjective and social) that define the practical expression of religious affiliation. For more on this concept, see Denny, "Orthopraxy in Islam and Judaism;" Ruthven, Islam: A Very Short Introduction; Lindsay, Daily Life in the Medieval Islamic World.

Food and Islamic orthopraxy
The focus of this study is the dissemination of the set of social norms and behaviours that structured the orthopraxy of medieval Islam, which is just one of the numerous consequences of the process of social Islamisation of al-Andalus. Although this is a complex phenomenon that concerns various spheres of social experience, this study is only concerned with food. Islam cannot be considered a totalising entity that was immutable and fossilised from the time of its emergence and first expansion. Therefore, it is important to keep this in mind when defining what the "correct" or "authentic" practices of medieval Islam were. 32 However, it is equally important to remember that, as part of the process of the emergence of the ideological and symbolic system of classical Islam, a type of religious discourse with strong legal connotations was generated. Not only did this discourse contribute to blurring the boundaries between the religious and secular spheres, 33 it was also decisive in shaping what would become the identity traits of the Muslim community. Therefore, considering the scriptural nature of Islam, the basic text that must be referred to is the Qur'an, which, from the birth of this new religion, was the reference point for Islamic identity. 34 The centrality of the Qur'anic text derives from the preeminent role it occupies in terms of both the orthodoxy-as the "source and perfect matrix of faith" 35 -and orthopraxy-given that it defines itself as the "guide for the pious" 36of Islam. The sharia (šarī'a), which is usually called "Islamic law," comes from the Qur'an and the Sunna. Rather than constituting a body of legislation in the modern sense, it should be understood as a code of behaviour comprising duties and obligations that are incumbent on all members of society and influence some of the most important aspects of their daily lives. 37 As was previously mentioned, it is important to consider that the formation of Islam took place simultaneously in the east and in the west. 38 For this reason, we must regard the definition of the rules that make up sharia as the product of a design shaped over centuries which was thus subject to historical fluctuations and changes that were required as a result of adapting to new circumstances that arose from the spread of Islam outside the Arabian arena. 39 However, it should also be noted that some of the behaviours regulated by sharia were already well established as part of the basic doctrine of Islam that had existed since the formation of the community around the Prophet Muhammad in the first half of the seventh century, including certain provisions concerning food. 40 In any case, we need to be aware that Islamic rules on food did not crystallise at any single point in time. In fact, many of these issues (which, in the specific case of food of animal origin, 32 Edwards, "The Archaeology of Religion," 128. 33 Marín Niño, "Espacios religiosos," 96. 34 On the centrality of the Qur'anic text to Islamic identity, see Rippin, Muslims; Robinson, "The Rise of Islam;" Silverstein, Islamic History. 35 Maíllo Salgado, Vocabulario de historia árabe e islámica, 193. 36 García Sanjuán, "El consumo de alimentos de los d  immíes," 109. 37  ranged from products that were permitted and others that were prohibited to how livestock should be slaughtered), caused legal and juridical controversy during the centuries when classical Islamic societies were being formed. In the specific case of al-Andalus, the Malikī jurists created and established different rules based on both the Qur'ān and the Sunna that believers had to follow in their diet. 41 Nonetheless, the present study is specifically concerned with one of the prescriptions on food which did not cause controversy in terms of the Qur'anic text or of the Prophetic tradition. 42 Pork ( janzīr) was one of the few foods considered impure (naŷāsa) by all the juridical doctrines of Islam-including the malikī, which was considered by Maribel Fierro as "the 'backbone' of the identity of al-Andalus" 43 -and its consumption was absolutely prohibited (ḥ arām). Mohammed Benkheira states that neither the Qur'an nor any of the later medieval interpretations make any distinction between domestic pigs and wild boar, 44 and therefore, the prohibition includes both swine species. 45 The origin of the Islamic taboo surrounding the consumption of pork (influenced, as is well known, by the Semitic tradition) has been the subject of enormously diverse discussion, so this study will not review the cultural, 46 hygienic-sanitary, 47 ecological-functional, 48 or socio-political 49 arguments put forward, as there are already works of synthesis, criticism, and elaboration on this issue. 50 This paper focuses on, whatever the reasons for its origin, the prohibition of the consumption of these animals in the Islamic world, which is clearly stated in the earliest known versions of the Qur'anic text. 51 The importance, at many different levels, of the proscription on pork derives primarily from the central role that the concept of purity (t ahāra) plays in Islamic religiosity. 52 Since, as Manuela Marín points out, Islamic social life is strongly imbued with normative contours of belief, the concept of t ahāra has profound implications for the everyday life of all members of the Muslim community as a whole, not only the most pious believers, but also those who share their residence with Muslims in urban or rural areas. 53 Given that purity is defined negatively-that is, it is lost through contact with contaminating agents-food is one of the most obvious channels of contamination, meaning that "the ingestion of food and drink has religiously marked limits that turn an everyday, natural act into a creedal manifestation." 54 It is for this reason that the avoidance of 41 See Arcas Campoy, "Los alimentos en el derecho māliki;" García Sanjuán, "El consumo de alimentos de los d  immíes;" Fernández Félix, Cuestiones legales del islam temprano; Safran, "Rules of Purity and Confessional Boundaries." The question of the legality of Muslim consumption of the food of the dhimmis (protected Christians and Jews) also has an important role in these studies. 42 Benkheira, Islâm et interdits alimentaires; Benkheira, "Tabou du porc et identité en islam." 43 Fierro Bello, "La política religiosa," 137. 44 Benkheira, "Tabou du porc et identité en Islam." 45 Benkheira, "La nourriture carnée," 85. 46 Lods, Israël; Douglas, Purity and Danger, and De la souillure. 47 Smith, "Animal Domestication and Animal Cults." 48  pork consumption constitutes not only "the most basic symbol of belonging to Islam," 55 but also "one of the ways to update it." 56 The surviving written documentation makes it possible to argue strongly that the observance of this fundamental precept of Islamic dietary regulation was already fully established at the time of the first Arab conquests in the seventh and eighth centuries. A few illustrative cases of this are the documentary references that prove the interest and concern of the various parties which, after the Prophet's disappearance, took over part of his religious leadership and political legacy. These parties put an end to the presence of pigs in several of the conquered regions where these animals were reared and consumed by the local populations. This is reflected in the document known as Šurūt  'Umar, the Stipulations or Treaty of Umar, a text traditionally attributed to Caliph 'Umar b. al-Khattab (r. 634-644), which, in its present form, dates from the late eighth and early ninth centuries. It contains the rules and restrictions to which non-Muslims had to submit in the conquered lands, including prohibiting the keeping of pigs and selling of wine which concerned dhimmis. 57 However, as Milka Levy-Rubin also mentions, several of the later versions of the Šurūt  only include the reference to wine which, in her opinion, could indicate that, for pork, "the danger had gradually faded away, or had even become irrelevant, in the sense that the dhimmis observed this prohibition strictly, as the penalty was well known and understood." 58 Similarly, it is documented that the caliph 'Abd al-Malik (d. 705) ordered the slaughter of all pigs in Syria and northern Mesopotamia, 59 and the same pattern was repeated centuries later during the Arab conquest of the Nubian region of Ibrim (modern Egypt) in 1173, when more than 700 pigs raised by local Christians were slaughtered. 60 There are abundant data for al-Andalus that all indicate the same thing, such as the information derived from the 'Utbiyya studied by Ana Fernández Félix, 61 or the opinions of various jurists expressed in the treatises of fiqh (Islamic jurisprudence) from the eighth to the sixteenth centuries examined by María Arcas Campoy, in which "the absolute and unanimous prohibition of pork" is clear. 62 It is also worth mentioning the alleged dialogue between a Cordoban bishop named John and a monk from Lorraine sent by Emperor Otto I (r. 962-973) to Qurt uba/Córdoba in the mid tenth century. According to the translation by Antonio Paz y Meliá, the monk reproached his co-religionist for the abstinence from the consumption of "certain foods that God created for the faithful to partake of with thanksgiving" by the Christian community of Córdoba. This was an attitude that the Cordoban bishop justified by stating that the social environment in which they lived made such behaviour necessary, adding that "we have received it from our elders through a long tradition." 63 García Sanjuán assumes that the foods referred to are pork, which he interpreted as one more manifestation, among others, of how Christians 55 Benkheira, "Lier et séparer," 94. 56 Benkheira, "Alimentation, altérité et socialité," 275. 57 Levy-Rubin, "The Pact of ʿUmar," 562. 58 Levy-Rubin, Non-Muslims in the Early Islamic Empire, 81. 59 Benkheira, "Tabou du porc et identité en Islam," 46. 60 Lobban, "Pigs and Their Prohibition," 64; Simoons, "Traditional Use and Avoidance of Foods," 179. 61 The 'Utbiyya is a compilation of legal issues initiated by al-'Utbī (d. 869). It was studied by Fernández Félix, Cuestiones legales del islam temprano. 62 Arcas Campoy, "El criterio de los juristas malikíes," 96. 63 Paz y Meliá, "La embajada del emperador de Alemania," 268. outside the Iberian Peninsula felt detached from and rejected those from al-Andalus, because of the acculturation of this latter group. 64 Consequently, it can be reliably inferred that the express prohibition of the consumption of pork was clearly established from the very emergence of Islam as it constituted part of the core of the system of ritual purity. 65 For this reason, when al-Andalus was founded in the early eighth century, this provision was already part of Islamic orthopraxy and was an integral part of Arab-Islamic identity. This association between food and Islamic orthopraxy enables an analytical link to be established between the study of the process of social Islamisation using the approach proposed here: analysing certain food behaviours that can be archaeologically identified.

The contribution of zooarchaeology to the study of social Islamisation
The Islamisation of al-Andalus-understood as the process of socio-cultural change linked to the spread of the Islamic order and the dietary regulations of the orthopraxy of medieval Islam-could mean that pig farming was gradually abandoned by a large section of the indigenous population that had previously been able to freely exploit this animal. The economic implications of this phenomenon were enormously important, given that they directly affected the subsistence of the majority of the Andalusi population by proscribing an economic practice that yielded great benefits. 66 However, the discussion here focuses exclusively on the identification of the possible ethno-religious identity of human groups by analysing eating habits.
For this purpose, the information derived from the zooarchaeological record, a line of research whose main, but not only, area of study is the analysis of material remains derived from the consumption of food products of animal origin, is extremely useful. On the Iberian Peninsula, the recognition of the analytical usefulness of bioarchaeological samples (botanical and zoological) from medieval contexts was granted later than similar recognition in the rest of Europe. 67 However, in recent years, the importance of this discipline, together with geoarchaeological analysis as an operational and conceptual resource, has become apparent. 68 Among the many questions related to the forms of interaction between humans and animals that can be explored through zooarchaeological study, the focus here is on the habits of consuming food of animal origin. In particular, the informative potential of this discipline is illustrated by examining how frequently pig remains have been found in a set of peninsular contexts dated to the first millennium CE. As such, it can be contended that the presence of archaeological food residues that provide information on 64 García Sanjuán, "El fin de las comunidades cristianas," 270-77. 65 Benkheira, Islâm et interdits alimentaires, 51; Safran, "Rules of Purity and Confessional Boundaries," 204. 66 The consequences of this phenomenon on the livestock management systems have been recently addressed by García García, "La dimensión alimentaria." 67 The current state of Iberian zooarchaeology from the medieval period has been discussed by García García,  pork consumption contribute to the recognition of the process of social Islamisation of al-Andalus. Figure 1 shows the relative taxonomic frequencies of the main mammal species exploited for food (caprines, cattle, and swine) in a selection of peninsular zooarchaeological assemblages recovered from sites that were different in terms of chronology, cultural context, and provenance, and in which the total number of identified remains of these three taxonomic groups was over 100. The quantification, which is not exhaustive and does not include all the case studies analysed to date, is intended to help in broadly assessing whether the zooarchaeological record reflects meaningful changes in the habits regarding the consumption of animal products (mainly pork) in early al-Andalus (eighth to tenth centuries).
As can be seen in Figure 1, in the Roman, late antique, and medieval Christian assemblages, pigs are generally abundant, with the exception of a few cases in which little of Figure 1. Relative abundance of swine (domestic pigs/wild boar), caprines (sheep/goat), and cattle in Roman, late antique/Visigothic, medieval Christian, and early Andalusi sites in Iberia, with an indication of the chronology and the province where the site is located. Only those cases that numbered more than 100 identified specimens (NISP) of these three taxonomic groups were considered. Raw data and references are provided as electronic supplementary material.
these remains are to be found. It should be noted that there is a significant contradiction between the written documentation and the zooarchaeological information regarding the relative importance of pigs during Late Antiquity. On the basis of different types of written references (legal documents, ecclesiastical texts, tax records, and literary references), pigs have commonly been understood to play a leading role in the food system of the late antique period. Thus, for example, the mention of the abundance of woodland and areas that favoured extensive pig rearing in legal texts has led authors to suggest that pork was the main meat in the Visigothic diet. 69 However, despite the fact that these animals are present in many of the postclassical cases considered in Figure 1, it seems that their role was far less important than has traditionally been attributed to them in studies of animal husbandry during Late Antiquity that depend solely on documentary sources. 70 This secondary quantitative importance of pigs compared to other livestock during this period contrasts with their relevance in symbolic terms. In fact, in the words of Jamie Kreiner, "it was probably first in the Visigothic kingdom of Iberia that the idea was floated that pigs were essential to Christian identity." 71 This seems to be confirmed by documentary sources, both ecclesiastical and legal, highlighting, for example, the importance given to the consumption of these animals as definitive proof of the conversion to Christianity by a group of "Hebrew" citizens of Toledo, 72 reflecting the symbolic role of the pig as representative of "Christian Europe in its entirety." 73 By contrast, in the assemblages from early Andalusi contexts (see Figure 1), the partial or total absence of pig remains is a dominant feature, given that in twenty-five of the thirty cases considered (83%) these animals represent less than 5% of the total amount of remains identified, and in twenty cases (67%) they were not present at all or represented by less than 1%.
Pig remains were identified in four of the ten rural sites considered (Jerigas, Humanejos, La Gavia III and Los Hitos), all of them located in the central region of the peninsula and studied by the same authors. 74 The information available in these cases is problematic, given that there are significant discrepancies between several of the reports on these sites (both published and unpublished) regarding the chronologies of the samples studied, the taxonomic frequencies (particularly with regards to the occurrence of pigs), as well as a lack of a clear description of the archaeological context. 75 Therefore, although it would not be unreasonable to assume that these rural populations maintained their traditional subsistence practices, which included pig farming, after the Islamic conquest, 76 it is necessary to be cautious about the results taken from these sites. 69 Arce Martínez, Esperando a los árabes, 176; García Moreno,Historia de la España visigoda,204. 70 As demonstrated, among others, by Grau Sologestoa, "El registro faunístico," and The Zooarchaeology of Medieval Alava; Lewit, "Pigs, Presses and Pastoralism;" Morales Muñiz, "Pig Husbandry in Visigoth Iberia." It is therefore surprising that the recent zooarchaeological review by Estaca et al., "Economic Implications of Livestock Management Strategies," 1301, states-without adding any references to justify the assertion-that "pork meat was among the main elements of the Hispano-Visigothic diet." 71 Kreiner, Legions of Pigs, 12. 72 García Moreno, Los judíos de la España Antigua, 79-80. 73 Montanari, "Introduction," 91. 74 Estaca, "El yacimiento de Humanejos;" Estaca and Yravedra, "Informe arqueozoológico;" Estaca et al., "Economic Implications of Livestock Management Strategies." 75 This makes it difficult for the reader to understand the archaeological arguments for dating the deposits from which the archaeofaunal samples were drawn. This is of particular importance when dealing with multi-period sites, as is the case in many of these contexts which contain phases of occupation that both predate and postdate the eighth century. 76 This has been proved for Islamic Sicily, for which it has been suggested that the countryside was Islamicised to a lesser degree than urban areas. See the zooarchaeological studies including Aniceti and Albarella, "Who's Eating Pork?," and, from the organic residue analysis of cooking pots, Lundy et al., "New Insights into Early Medieval Islamic Cuisine." On the other hand, pig remains are only abundant in two of the twenty urban contexts considered. These two sites are Conimbriga 77 (now known as the city of Coimbra in Portugal) and the Emirate phase of Cercadilla in Qurt uba (modern-day Córdoba), the capital of al-Andalus during the Umayyad period. In both cases, as will be argued below, the data available might indicate that we are dealing with food residues generated by Christian Andalusi populations.
In short, the quantification presented clearly shows a remarkable difference in the frequency of pig remains in the zooarchaeological record before and after the eighth century. However, it is obvious that the changes following the conquest of 711 were not immediate and that, as the zooarchaeological information suggests, there are exceptions to the general pattern defined by the abandonment of pig breeding and pork consumption after al-Andalus came into existence. Qurt uba is one of the most illustrative cases in which the abundant archaeofaunal record has proved useful in linking changes in dietary patterns and the spread of the process of social Islamisation from both a spatial and temporal perspective. 78 A fairly representative volume of animal remains recovered from rubbish dumps in the capital of al-Andalus has been studied in recent years, providing insight into the food consumption habits of various population groups between the eighth and early eleventh centuries. Specifically, the suburb of Šaqunda and the Cercadilla Archaeological Zone are of interest (Figure 2). The first case corresponds to an extensive, complex urban area frequented between the mid eighth century and the first decades of the ninth century by a population group that, according to all the information available, can reliably be identified as Muslim. 79 The second case, Cercadilla, is located to the northwest of the walled enclosure of the madīna (city) and there is evidence of several phases of occupation, two of which are of interest here. The first phase dates to the emirate period (eighth to mid tenth centuries) and is characterised by an intense domestic occupation in which the presence of a Christian cult centre with an associated necropolis is noteworthy. The second phase dates to the caliphate (mid tenth to early eleventh centuries), when a large, fully urban suburb was developed that razed the previous structures and, as a result, led to a significant reduction in the space occupied by the Christian necropolis which continued to be in use until the suburb was abandoned at the beginning of the eleventh century. 80 The frequency of the main mammals (i.e., caprines, cattle, and swine) in these case studies appears in Figures 3 and 4. For comparison, the species composition of the archaeofaunal sample recovered from another urban area occupied during the caliphate period and studied by Laura Aparicio Sánchez and José A. Riquelme Cantal is also included in Figure 4. 81 Firstly, Figure 3 shows a significant difference in the taxonomic composition of the samples from Šaqunda and Cercadilla during the emirate period, mainly due to the absence of pig remains in the Šaqunda rubbish dumps, reflecting the fact that these 77 Detry, Cardoso, and Correia, "What Did the Romans and Moslems Eat in Conimbriga (Portugal)?" 78 The following synthesis of the zooarchaeological evidence from the early Andalusi period in Córdoba was originally elaborated in the author's doctoral thesis: García García, "Explotación y consumo de los animales." 79 For a detailed study of the suburb of Šaqunda, see Casal García, "El arrabal de Šaqunda." 80 Fuertes Santos and Hidalgo Prieto, "La transformación del paisaje del área noroccidental cordobesa," and Guía arqueológica de Cercadilla; Hidalgo Prieto, "El complejo monumental de Cercadilla." 81 Aparicio Sánchez and Riquelme Cantal, "Localización de uno de los arrabales noroccidentales." animals were not part of the diet of its inhabitants. In contrast, at Cercadilla, not only are pig remains abundant but also, as suggested by other zooarchaeological parameters, 82 they could have come from domestic pigs that had been reared and consumed locally at the same location. Therefore, the data available enable the existence of different socio-alimentary food consumption contexts to be shown in Qurt uba during the emirate. These contexts can be defined as Islamised in the case of Šaqunda and as non-Islamised in the case of Cercadilla. Our data on the latter area confirmed other lines of archaeological evidence-specifically, the aforementioned Christian cult centre with its associated necropolis in use between the sixth and eleventh centuries-which suggest the presence of a Christian dhimmi community in Cercadilla during the emirate.
Secondly, there is another contrast-in this case of a temporal rather than a spatial nature-related to the taxonomic composition of the samples from the emirate and caliphate phases of settlement at Cercadilla (Figure 4). As has been noted, from the tenth century onwards, the space outside the walls of madīnat Qurt uba was the stage for impressive urban development. 83 At Cercadilla, archaeological work documented a new phase of occupation defined by the construction of an extensive suburb with urban characteristics. The feature that best defines the taxonomic composition of the 82 Specifically, kill-off patterns indicate the consumption of animals of different ages (including a group of osteologically immature individuals killed during their first year); skeletal distribution is defined by the presence of all the anatomical parts (although the dominance of the cranial region is clear); and the biometric evidence suggests that we are dealing with specimens that very possibly belonged to domestic pigs, rather than much bigger wild boar. García García, "Explotación y consumo de los animales," 388-90, 406-10, 420-28. 83 Acién Almansa and Vallejo Triano, "Urbanismo y estado islámico." archaeofaunal assemblages dated to the caliphate period is the generalised absence of pig remains. Similar results were also obtained in a neighbouring urban area (Figure 4). 84 In the case of Cercadilla, the difference between the samples from the emirate and the caliphate periods could be the consequence of the abandonment of local breeding of these animals and the consumption of their meat during the tenth century.
In light of this last result, it is worth asking whether the composition of the population of Cercadilla was stable between the emirate and caliphate periods, or whether a new group settled in this area in the tenth century that replaced the earlier one. Unfortunately, with the data available, a conclusive answer to this question cannot be given. However, it should be remembered that the survival of the necropolis until the beginning of the eleventh century seems to indicate that at least part of the population living in this area was buried according to Christian ritual, despite the fact that the data reflect an Islamised social context of food consumption. This apparent contradiction might reflect the difference between two phenomena that were not necessarily synchronous, namely social Islamisation-understood as the adoption of Muslim cultural and social manners-and religious conversion.
The drastic disappearance of pig remains in the caliphate rubbish dumps at Cercadilla can be considered the materialisation of the process of social Islamisation resulting from this urban area becoming a part of the Islamised social model that triumphed in Qurt uba during the tenth century. The greater frequency of large bowls (ataifores) in the caliphate period at Cercadilla seems to indicate the same phenomenon, perhaps related to the spread of communal forms of food consumption that is typical of Islamised contexts. 85

Concluding remarks
This paper has attempted to offer some insights on two key concepts for understanding the emergence of al-Andalus, namely the process of social Islamisation and the construction of the symbolic and material structure of the Umayyad state. As has been suggested, both phenomena must be considered part of the same germinal process that led to the formation of Andalusi society, and therefore they must be included in the same analytical package. The archaeological study of food consumption habits makes it possible to evaluate the rhythms and progress of the configuration of an Islamic society in the west. 86 The data derived from the zooarchaeological record illustrate the informative potential of using the relative abundance of pig remains in early Andalusi sites as a diagnostic marker for assessing the degree of social Islamisation. The frequency of pig remains can be used as accurately as other markers, such as pottery (and perhaps even more accurately). 87 The fact that there are no pigs in most of the case studies seems to reveal that, from the earliest times of al-Andalus, the consumption of pork had already taken on symbolic connotations that defined the demarcation of the cultural boundaries between the different ethno-religious communities that made up Andalusi society.
In addition to a generalised pattern of absence of pig remains, there is also a small number of cases which reveal the existence of exceptions to this general observation. On the one hand, the archaeofaunal record of a few rural sites in the centre of the peninsula could reflect the survival of non-Islamised socio-alimentary food consumption contexts and, perhaps, the maintenance of lifestyles that were closer to those of Late Antiquity. On the other hand, in only two urban contexts from the early Andalusi period (Conimbriga and the emirate phase of Cercadilla) do we find sufficiently representative evidence that indicates pork consumption. In the case of Conimbriga, 88 the authors propose two possible interpretations of the results regarding the ethno-religious identity of the inhabitants: that it was a Christian group or that these remains reflect the transgression of Islamic dietary norms by a socially Islamised population. Considering the fact that there was a major Christian dhimmi community in Coimbra and in the Beira region, 89 and also considering the information derived from Cercadilla, the first hypothesis is more convincing.
The pattern observed in the case studies from Qurt uba demonstrates the link between Christian identity and pork consumption during the period in question. This seems to be indicated by the presence of abundant pig remains in the dumps from the emirate period excavated at Cercadilla-a settlement with a clear Christian dhimmi component-85 Fuertes Santos, La cerámica medieval de Cercadilla. The relationship between the use of ataifores and the existence of Islamised social contexts of food consumption in Cercadilla still needs further research. 86 Guichard, For studies that demonstrate the value of pottery for gaining insights into the formation of early al-Andalus, see Amorós Ruiz, "El Tolmo de Minateda;" Carvajal López, "After the Conquest;" Carvajal López and Day, "Cooking Pots and Islamicization;" Carvajal López and Jiménez Puertas, "Cuisine, Islamisation and Ceramics;" Gutiérrez Lloret, "La arqueología en la historia;" Kirchner Granell, "La arqueología del campesinado." 88 Detry, Cardoso, and Correia, "What Did the Romans and Moslems Eat in Conimbriga (Portugal)?," 103. 89 Aillet, "El Monasterio de Lorvão," 92. particularly when compared with the contents of the dumps dated to the same period at Šaqunda, which are characterised by the absence of such remains. On the other hand, the differences observed regarding the presence of pigs between the emirate and caliphate periods at Cercadilla suggests that this area of Qurt uba did not fully immerse itself in the Islamised social model until the tenth century. Therefore, even in the capital of al-Andalus, the survival of eating habits that reflect the socio-cultural horizon of Late Antiquity exemplifies the fluid, gradual nature of social Islamisation as a decisive process of historical change in the shaping of the practices and behaviour in everyday Andalusi society.
Amongst the multiple consequences that the prohibition of the consumption of pork may have had for the indigenous populations, the focus here has been placed on the socio-political consequences of the Islamisation process, by using an approach similar to what Arjun Appadurai has called gastro-politics. 90 The relevance of this type of analysis lies in the fact that it enables the times of Islamisation to be understood and calibrated "from the bottom up." This level of understanding can also be applied to the construction of Islamic social order, considering that eating habits possibly reflected the ethno-religious identity of social actors within the multicultural spectrum that defined early al-Andalus. If, as has been argued, being a Muslim implies living like a Muslim-which entails the adoption of the orthopraxy of Islam of which the interdiction of pork was an early part-the observance of this dietary norm can be regarded as one of the main mechanisms used by the Umayyad state to establish its power as the main promoter of Islamic social order.
In conclusion, if by "power" we simply understand the ability to make others do what you want them to do, according to the loose notion of the concept proposed by Barclay,91 the exclusion of pigs from the food system inherited from the late antique period must be conceived as a form of imposition of the power of Islam, which is the same as saying of the Andalusi state. Obviously, the implications of this phenomenon are not limited to what has been discussed in this paper, the aim of which has been to establish a line of analysis for further exploration. In particular, in terms of resistance to Islamisation, the implications of the ban on the use of pigs for the maintenance of late antique socio-alimentary patterns and its consequences for the livestock economy and family subsistence strategies are two important aspects to be addressed in the future. a "Juan de la Cierva-Formación" Postdoctoral Researcher at the University of Alicante (Ref. FJC2020-044358-I) sponsored by the Ministry of Science and Innovation (MCIN/AEI/10.13039/ 501100011033) and the EU NextGenerationEU/PRTR.

Notes on contributor
Marcos García García holds a BA in History (2011), MSc in Environmental Archaeology and Palaeoeconomy from the University of Sheffield (2012), and PhD in History and Arts from the University of Granada (2019). Between 2019 and 2021 he was a Postdoctoral Research Associate at the University of York as part of the international project Landscapes of (Re)conquest (Arts and Humanities Research Council). He is currently a Juan de la Cierva-Formación Postdoctoral Researcher at the University of Alicante. His research focuses on the fields of medieval history and archaeology, specialising in the zooarchaeological analysis as a way to contribute to the study of food systems and the examination of social processes between the Late Antiquity and the modern period. Works Cited Secondary Sources