Scholarship that makes efforts to reconstruct a typically ‘Muslim’ or ‘Islamic’ view of the world generally neglects to reflect upon the methodological intricacies of reconstructing patterns of perception characteristic of human collectives or societies on a macro-historical scale. It accords exaggerated attention to stereotypes and fails to acknowledge that the various texts subject to analysis often contain much more material than only the most blatant statements about ‘the Other’.
Texts record perceptions on different levels. On a first level, they record what the authors themselves perceived. The author figures as the ‘subject of perception’ while the text contains elements that can be labelled as ‘objects of perception’. On a second level, authors reproduce and present the views of others. In such cases, it is necessary to consider if the author has concealed his own views behind the (fictitious) utterance ascribed to someone else. On a third level, authors describe the interaction of persons or groups without explicitly defining the perceptions involved. Such descriptions may allow inference as to how different actors viewed each other. Such evidence is implicit and rests on shaky methodological foundations, but often represents the only key to the perceptions of those who did not safeguard their vision of the world in writing.
Approached in this way, source material on Muslim perceptions of Latin Christianity gains depth considerably. Serial analysis of a larger corpus of texts and the inherent explicit and implicit evidence allows for the reconstruction of patterns of perception characteristic not only of individuals, but also of groups and even societies. However, selection and categorization can produce unacceptable distortions if they are employed without prior reflection. By selecting passages from Arabic- Islamic works that clearly define a European or Latin-Christian ‘Other’ and additionally characterize it in a negative way, one reconstructs only one among several existing patterns of perception. By imposing this pattern on all representatives of the Islamic world at all times and in all places, one artificially constructs monolithic, unchanging, static subjects and objects of perception while failing to acknowledge the large variety of much more complex historical situations. The validity of scholarly approaches that reduce a multiplicity of perceptions to a paradigmatic image of ‘the Other’ was already questioned by Maxime Rodinson in 1980:
II n’y a pas d’Autre en soi et il est plutot vain de chercher a theoriser sur cette base; le resultat ne peut etre que mince et d’une trop grande generalite pour etre bien utile. Il y a des situations differentes, multiformes d’ailleurs, ou deux societes portent l’une sur l’autre un regard. Ce regard est toujours different lui aussi, toujours multiple, toujours soumis au changement. Sans doute les orientations dominantes de la societe qui regarde comportent-elles plus ou moins de receptivite, de capacite d’attention et de comprehension, d’empathie, virtualites que les relations de force entre elle et celle qu’elle considere mobilisent, freinent ou contredisent. Mais ces lignes generales, abon- dantes en facettes deja, sont loin d’epuiser l’inepuisable polychromie du reel.    
Several centuries of contact in an area reaching from the Iberian Peninsula to Central Asia inevitably produced a diversity of relations between a multitude of subjects and objects of perception.!53 Subjects of perception were part of differing political, economic, and social constellations and developments and thus formulated a variety of views at different times and places. Corresponding constellations and developments affected the objects of perceptions and their different manifestations, which consequently failed to produce uniform sensations in the minds of observers. The large array of differing constellations makes an exhaustive enumeration impossible. We must consider, however, that individuals fulfilled various roles when maintaining military, political, economic, intellectual, religious, personal, emotional, and other forms of relations. Soldiers, diplomats, merchants, scholars, believers, siblings, lovers, etc. personify a different spectrum of ‘functional’ behaviour. It is impossible to determine such behaviour, which is necessarily dependent upon the individual context. Moreover, defining a context with corresponding roles does not automatically determine a specified set of perceptions as would the application of a mathematical formula. However, generating such an artificial typology forces us to consider a broader range of possible relations and perceptions than the selective analysis of a textual corpus containing explicit statements on, or a specific terminology characteristic of ‘the Other’. 154 We can question the legitimacy of propagating the notion of a single dominant ‘Muslim’ perception even further by pointing to the existence of ‘third spaces’ and ‘hybrid’ or ‘transcultural’ phenomena in the areas connecting both cultural spheres.155
With this in mind, it is possible to approach the extant sources from a different point of view. A method used to master the intricacies of early medieval Latin hagiography,i56 i.e. the comparison of variants, corroborates that the perceptions accessible to us via texts are never uniform. Comparing source material dealing with different subjects of perception, e.g. Muslim believers, authorities, merchants, intellectuals, slaves, etc., shows that these Muslims perceived Latin Christians differently in accordance with their respective ‘functional’ role within the general context of encounter. Contrasting source material in which subjects and/or objects of perception take on similar or even identical functional roles permits to identify differing contexts, context-dependent relationships, and, in consequence, variants of perception. Analysing how one single subject of perception, e.g. a productive Muslim author or a Muslim traveller through Christian territory, described Latin Christians, shows that many different factors were at play that influenced the respective author’s views at a given time and place.^7
What is valid for one single author is also valid for an entire group of authors writing in the period between the seventh and the fifteenth century. Arabic-Islamic scholars, expressing their thoughts in the same language, often had recourse to each other’s works and thus drew from and contributed to the same intellectual culture. Notwithstanding, they were of various geographic, ethnic, and cultural backgrounds, lived and worked in differing historical and social contexts, formed part of different variants of a changing intellectual landscape, and were involved differently in relations with the Latin-Christian sphere. The following chapters will show that a single ‘Muslim’ perception of Latin-Christian Europe did not exist. Rather, different contexts produced different relationships, which, in turn, gave rise to different perceptions. What we can reconstruct are ranges of perception that apply to specific subjects of perception as regards their—by no means consistent— views on a well-defined object of perception in a given period, place, and context.
In the past two decades, scholarship has underscored that the Arabic-Islamic sphere always perceived medieval Western Europe from various angles and has occasionally even made the effort to categorize different kinds of Muslim perceptions. Tarif Khalidi (1995) made the rather unconvincing distinction that ‘it was the geography and society of Europe that interested Islam rather than its history or its culture’.^8 Although dedicating much attention to the influence of normative religious frameworks on the perception of ‘the Other’, Muhammad Nur al-Din Afaya (2000) pointed to the multiplicity of perception patterns and their dependence on the historical context in his analysis of a few Arabic-Islamic authors. ^ Shams al-Din al-Kllanl (2004) distinguished between ‘religious’ (dint), ‘civiliza- tional’ (hadari), and ‘geographical’ (jughraf) Arabic-Islamic perspectives in a monograph that remains rather descriptive and too sterile in its categorization.    In a study with a clear penchant for the modern period, Christopher Nouryeh (2005) defined Arabic-Islamic authors writing on ‘the West’ as ‘neglected bridge- builders’.i6i Nizar Hermes (2012) advanced an analysis of ‘the [European] Other’ that firmly criticizes the traditional approach outlined above. Centred on the ninth to twelfth centuries and focusing on fewer sources, his study provides valuable support to the present study and its alternative methodological approach.^
Only recently, Brian Catlos (2014) has provided a nuanced categorization of Christian-Muslim perceptions in a study dedicated to Muslims under Christian rule. Catlos distinguishes between a macro- or ‘ecumenian’, a micro- or local, and a meso- or corporate mode of perception. The macro- or ‘ecumenian’ mode relates to dogmatic-informed religious identity that, because it claims a monopoly on truth, is rigidly defined and mutually exclusive. The micro- or local mode applies to an ill-defined, because very flexible, range of perceptions that reflects the large variety of individual and collective relations, including personal bonds, cross-communal friendship and solidarity, intermarriage, syncretism, but also instances of inter-communal violence. The meso- or corporate mode of perception applies to the sphere of formal collectives. These adhere to the ‘ecumenian’ mode in theory but do not necessarily put it into practice. In an effort to cope with complex realities, e.g. in the legal and economic spheres, they tend to pursue a pragmatic approach that ensures their own survival.163
Catlos’ distinction is enormously helpful in that it systematizes different levels of perception that co-existed in social relations between Christians and Muslims. The present study, however, does not focus on these social relations, but tries to understand how a highly diversified Arabic-Islamic scholarly elite acquired and processed information on Latin-Christian societies over the centuries. The following chapters will show that Arabic-Islamic scholars were not necessarily involved in direct relations with Latin-Christian societies. This had an effect on their depictions of the Latin-Christian sphere in that the meso- or corporate mode of perception was generally irrelevant whereas the micro- or local mode of perception mostly entered their works as the result of long and complex processes of transmission. Notwithstanding, their portrayal of the Latin-Christian sphere or some of its aspects does not automatically fit into Catlos’ category of ‘ecumenian’ perceptions. Given that the scholarly works used here endeavour to make sense of the world in historical, geographical, and ethnographical terms, the dogmatic-informed religious identity of their authors was often of secondary importance. Although not devoid of ideological overtones, this ‘scholarly mode’ of perception primarily is characterized by the wish to understand particular historical, geographical, or social phenomena.
One aspect, which so far has been neglected in the efforts to define appropriate categories or modes of perception, is the factor time. Although Lewis used the term ‘discovery’ in the titles of two texts on the subject, conceding that images evolved and that the medieval period already witnessed a ‘gradual extension of knowledge’/64 he regarded the ‘feeling of timelessness, that nothing really changes’, as a ‘characteristic feature of Muslim writing about Europe’.i65 To underscore the connection between geopolitical developments in the medieval Mediterranean on the one hand and the parallel evolution of an Arabic-Islamic literary culture that produced the extant records on medieval Europe on the other, the present study puts emphasis on the chronological order in which different generations of medieval Arabic-Islamic scholars acquired, processed, and wrote down information about certain aspects of the Latin-Christian sphere.
Chapters 2 and 3 approach the main topic from a general point of view. Chapter 2 describes the ever-changing ‘information landscape’, i.e. the shifting channels of transmission that facilitated the flow of information from the Latin- Christian to the Arabic-Islamic sphere. It concludes that Arabic-Islamic scholars only constituted a relatively small group among those involved in relations with medieval Western Europe and generally occupied a place at the end of rather long chains of transmission. Since most publications on ‘Muslim’ perceptions of Western Europe seem to take for granted that the writings of Arabic-Islamic scholars contain the entire knowledge available on Latin-Christian Europe in the Islamic world, it seems particularly important to make this point. Chapter 3 will discuss problems of reception and interpretation encountered by Arabic-Islamic scholars conducting research in an age devoid of modern-day communication facilities. It responds to the belief that a kind of ‘mental barrier’, a frame of mind dominated by Islam and common to the majority of Muslims, was the main factor that hampered the flow of information. Chapters 4 to 8 comprise a collection of case studies dedicated to the question of how Arabic-Islamic scholars recorded a certain facet of the Latin-Christian sphere over the centuries. Each case study features a chronological analysis of the extant records on the respective phenomenon. Arabic-Islamic records on the Western Roman Empire, two peoples of early medieval Europe (Visigoths and Franks), the rise of the papacy, and a selection of Western European societies backing Latin-Christian expansionism stand at the centre of the respective chapter. Pursuing the aim of exposing textual filiations as well as the evolution of networks of transmission and reception, the case studies are often rather descriptive. For the reader’s convenience, the most important results are summarized at the end of the respective chapter. By providing a detailed and differentiated impression of how Arabic-Islamic scholars acquired, processed, and presented knowledge about certain facets of Western European history over the centuries, this approach shows that a wide spectrum of perceptions and emotions was involved in the description of the non-Muslim world. It also makes allowance for the diversity of Latin-Christian Europe, a cluster of societies that was subject to constant change in the period from Late Antiquity to the Early Modern Age. As a matter of course, this diversity was reflected in Arabic-Islamic records, as is discussed in the concluding re-evaluation in Chapter 9.
-  Rodinson, Fascination (1980/2003), pp. 197—8.
-  Every manual on the topic, e.g. those issued on occasion of the French governmental exams forfuture teachers in higher education (CAPES, Agregation) in 2001, introduces the reader to the diversecharacter of relations: Tolan and Josserand, Relations (2000); Guichard and Senac, Relations (2000);Jehel and Jehel, Relations (2000); Micheau, Relations (2000); Jansen et al., Mediterranee (2000); Foss-ier, Relations (2000); Balard et al., Islam (2000); UMR 5648 (ed.), Pays (2000); Aillet et al., Pays(2001); Arrignon et al., Pays (2001); cf. Abdellatif et al. (eds), Acteurs (2012).
-  Konig, ‘Perception(s)’ (2010), pp. 21—30, esp. 21—5.
-  E.g. Ruggles, ‘Mothers’ (2004), pp. 65—94; Epstein, Purity (2006); Mersch and Ritzerfeld (eds),Begegnungen (2009); Burkhardt et al., ‘Hybridisierung’ (2011), pp. 467—557; Zorgati, Pluralism(2011); Dakhlia, ‘Metis’ (2012), pp. 45-57.
-  Prinz, ‘Aspekte’ (1989), p. 183; Lotter, ‘Methodisches’ (1979), pp. 339^0.
-  Konig, ‘Perception^)’ (2010), pp. 25—30, with examples.
-  Khalidi, ‘Views’ (1995), p. 42.
-  Afaya, al-gharb (2000), p. 311. His argument is based on al-Mas udl, Ibn Qutayba, Ibn Hazm,al-Ghazall, Ibn Jubayr, Usama b. Munqidh, Ibn Rushd, Ibn Qayyim al-Jawziyya, and Ibn Khaldun.
-  al-Kllanl, Urubba (2004). 161 Nouryeh, Views (2005).
-  62 Hermes, Other (20 1 2). 163 Catlos, Muslims (2014), pp. 525—7. 164 Lewis, ‘Discovery’ (1957), p. 410. 165 Lewis, Discovery (1982/2001), p. 297.