CHAPTER EIGHT Harvesting Garden Semantics in Late Medieval Anatolia Nicolas Trépanier It sometimes seems like discussions of political history are the only fruits that grow in the field of late medieval Anatolian historiography,1 and that this ground is simply not fertile enough for the cultivation of social history. It is true that fourteenth-century Anatolia did not leave us with the wealth of archival documents that characterises later Ottoman history. Yet while some naively await the unearthing of new sources, other fields have proven over and over again that new questions, rather than new source material, are the best tools for historiographical development. This chapter, just like many other contributions to this volume, seeks to expand the range of questions one can raise in relation to late medieval Anatolian history by showing that extant sources do carry more information than a cursory look would suggest. The range of topics these sources allow us to investigate is almost unlimited, but gardens, and garden-related vocabulary in particular, are ideally suited to showcase this untapped potential2 for a number of reasons. For one, there exists a manageable variety of terms (˙adÈqa, junayna, bågh/ba©, båghcha/ba©çe, bËstån/bostan) that present a good example of the wealth of nuances hidden between words otherwise indiscriminately translated into English as ‘garden’. Furthermore, the same set of terms is used in Turkish- and Persianlanguage sources, while a separate set appears in Arabic-language ones, allowing us to investigate the nature of the porosity of boundaries between the three written languages that Muslim authors of that period used in different circumstances.3 While Anatolia did not have any significant Arab population, the Arabic language remained the language of Islamic law, which was the primary legal framework of the region. For this reason, Arabic was the language used in documents related to real estate ownership, including some that describe gardens. Politically and culturally, Anatolia was still dominated by a Persian-speaking elite that had established itself in the region in connection to the Seljuk court culture. This elite, which included the famous poet Jalal al-Din BLESSING 9781474411295 PRINT.indd 185 22/08/2016 16:13 186 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 Rumi, was actively engaging with the vibrant Persian literary tradition (in which gardens often appear) and spoke the language in their daily life, including in their interactions with gardens as places of entertainment and as agricultural investments. Turkish was the spoken language of a large number of immigrants from Central Asia that were in the process of replacing this Persian-speaking elite as the ruling class of Anatolia. As often happens, increased political power led to increased cultural legitimacy, and the fourteenth century is the moment when we see the first written examples of Old Anatolian Turkish (the language that would evolve into Ottoman and then modern Turkish) in poetry and prose, an early literary production that selectively incorporated some of the conventions of written Persian. This survey of the various languages used by Anatolian Muslims also allows us to witness the variety of functions that gardens performed and, therefore, to sample in a limited number of pages a range of late medieval Anatolian experiences that goes from interactions between social classes to agricultural work and geographical perceptions. While works of poetry frequently present gardens as metaphors for paradise or, more generally, places of beauty, the gardens discussed in this chapter are those planted areas (for purposes ranging from fruit or vegetable production to socialisation and relaxation) that physically existed in late medieval Anatolia. There are a number of reasons for this choice, among which two stand out. First, the vocabulary designating physical gardens is almost certainly more clearly defined than the vocabulary used for gardens as poetic images, thanks to the functional (economic, social, legal) implications to which physical gardens are bound, and from which poetic flourishes are free. Just like the words ‘bungalow’, ‘rowhouse’ and ‘apartment building’ refer to clearly different physical structures in a way that the terms ‘home,’ ‘abode’ and ‘dwelling’ do not, this chapter focuses on vocabulary terms that have a specific physical counterpart. A second reason to leave aside works of poetry is that whereas the literary culture that shaped the Anatolians’ poetic use of gardens was largely shared with regions as diverse as Iran, India and Central Asia, the ecological conditions, economic structure and agricultural traditions varied quite a lot between these regions. In other words, court poets in Konya and Delhi might have used the same word when waxing lyrical about a garden, but the cultivated land they saw when looking out of the window would have been dramatically different. It is significant, for example, that Anatolian sources present no evidence, direct or indirect, that the gardens they describe were organised along the principles of the iconic chahårbågh (quadripartite garden) of Iran.4 I have therefore limited my source material to texts describing Anatolia by authors who saw the region with their own eyes, rather than following the time-dishonoured traditions of approaching literary texts as direct representations of daily life and BLESSING 9781474411295 PRINT.indd 186 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 187 piecing together material of various origins as if they all came out of a homogeneous Orient. Definitions The sources covered here can be divided in two main categories, with overlapping contrasts in genre and language. Waqfiyyas, or deeds of pious endowments, constitute the first category of sources. These documents, typically one to four pages in length, were drawn up as the founding documents for pious endowments (waqf) established under Islamic law and include descriptions of the real estate property that is to provide the revenue for the endowment, especially agricultural land (mostly cereal fields, but also gardens).5 While these texts were not, strictly speaking, produced by the state, the endowers are often associated with local ruling families. More importantly, the formulaic structure and administrative-legal purposes of these texts allow us to classify them as ‘archival documents’. They are, in fact, the only extant contemporary texts to fall into this category until the earliest extant Ottoman cadastral surveys and court records were produced in the second half of the fifteenth century. Unlike most of their later counterparts, which use Ottoman Turkish, fourteenthcentury Anatolian waqfiyyas are – with a few exceptions – composed in Arabic, and designate gardens using the words karm, ˙adÈqa and junayna. The second category of texts considered in this chapter comprises Persian- and Turkish-language narrative sources, primarily hagiographies and, to a lesser extent, other religious texts and chronicles.6 No doubt in large part because the Turkish language was then taking its first steps in the literary realm, there is a very high degree of overlap in the vocabulary used in both of these languages to refer to gardens, both preferring the words bågh, båghcha and bËstån.7 The reader should remember that the two categories of sources are very different not only in literary style, but also in their contents and the intent of their respective authors. Thus, whereas mentions of gardens in waqfiyyas are primarily intended to describe the source of the endowment’s revenues, authors of narrative texts generally use them as the backdrop for a scene where the characters are the main focus.8 For this reason, waqfiyyas never refer to the social uses of gardens, while narrative sources give much less precise detail about such topics as garden locations and agricultural revenues. Arabic: Waqfiyya Vocabulary Arabic-language waqfiyyas, as stated above, use the words karm, ˙adÈqa and junayna.9 Of these, karm (vineyard) is the easiest to define because it is the only one for which dictionaries offer a specific meaning. Indeed, this meaning seems to be the way in which BLESSING 9781474411295 PRINT.indd 187 22/08/2016 16:13 188 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 the authors of waqfiyyas understood it, given that on two occasions a document specifies that a karm is endowed ‘along with the press (mi‘ßar) that is in it’.10 Such established, narrower definitions do not exist for junayna and ˙adÈqa,11 and the waqfiyyas do not associate either of these terms with a given type of production (e.g., vegetable gardens or orchards). While the possibility exists that such association existed without having left visible traces in the documents, a more productive approach for the historian is to look at hints that do appear – and there are such hints, more specifically in relation to the location of these pieces of property. For example, the word ˙adÈqa occurs in the case of gardens located in urban settings, whereas none of the five junaynas they mention that can be localised as set inside a city. It seems logical, then, to suggest that the semantic divide between the words ˙adÈqa and junayna may have run along the urban–rural one. This impression is strengthened by the fact that not a single one of the six cases in which the word ˙adÈqa is used is provided with a list of bordering features,12 a characteristic that squares well with the clarity of the borders of an urban piece of land, surrounded by buildings and therefore requiring much less precision in description than a rural plot (for the same reason that cereal fields sometimes require stone boundary markers). Thus, we can conclude that while vineyards are called karm, waqfiyyas use the term junayna to refer to any form of economically productive garden in the countryside, and ˙adÈqa for the urban equivalent. Ba¯ gh: Persian Gardens designated as ‘bågh’ frequently appear as the setting for anecdotes in Persian sources, especially in the Manaqib al-‘Arifin, a hagiography of Jalal al-Din Rumi and the early Mevlevi masters. These passages make it clear that båghs could be used as the location for social interaction and entertainment among both religious and political grandees and by both genders (although men and women would visit them separately),13 as well as quiet, individual relaxation14 and religious practices.15 Some references are also made to the consumption of food on these occasions, such as one moment in which guests arriving directly from a trip to a bågh are offered ˙alwå’ and another in which they consume figs.16 Yet significantly, none of the full meals and feasts that do appear in these sources is described as being held in a bågh. There is no detailed description of båghs in Persian (nor, for that matter, of any other type of garden) in any one of the sources covered here, most likely because authors assumed a certain degree of knowledge from the part of their readers. Still, it seems clear that these spaces tended to be at least large enough that two people could BLESSING 9781474411295 PRINT.indd 188 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 189 easily be in the same bågh at the same time without being aware of each other’s presence.17 Furthermore, the concept also entailed the presence of walls,18 and the mention of a bench in one of them strengthens the idea of a place used for social intercourse.19 On the other hand, it is clear that båghs could also have an agricultural function. There are many direct or indirect mentions of fruit production,20 quite possibly in a specifically designated section of the bågh.21 It also seems that complementary productions such as honey and the transformation of grapes into syrup (dËshåb) could take place on location.22 As far as their location is concerned, båghs are often described as being situated at some distance from central urban elements (religious buildings or residential areas),23 an idea strengthened by the apparently common use of riding animals to reach them and a few passages describing characters sleeping at the garden rather than returning home for the night.24 Furthermore, it seems fairly well established that båghs would most likely be located next to each other rather than scattered between other types of land use.25 One passage from Bazm u Razm, a biographical chronicle of central Anatolian statesman Qadi Burhaneddin (r. 1380–98), clearly locates båghs in between the city and the countryside, a detail that mirrors earlier observations that a thick network of irrigation channels and walled gardens surrounded the city of Konya at the turn of the thirteenth century.26 It is, however, not impossible that this type of garden could also form ‘islands’ out in the countryside, entirely surrounded by cereal fields and/or by uncultivated land, in areas selected for the availability of water (the importance of which is known to have been crucial in selecting the gardens around Alanya).27 All of this, put together, suggests that the Persian word ‘bågh’, when used in its literal sense, referred to something that could be called a ‘garden complex’, a relatively large area outside the city whose surface was divided between sections devoted to entertainment and relaxation and to agricultural production areas.28 Bag˘ : Turkish Occurrences of the word ba© in Turkish-language sources are few in number, and provide us with very little that could be used to create a coherent picture of what the word would have meant to the authors that used it during this time period. In some cases, that picture appears to have been compatible with use in Persian sources. In terms of social functions, for example, one reference seems to agree with the proposition that ba©s could be used as places for the entertainment of the political elite.29 Furthermore, Turkish sources confirm the presence of fairly tall walls around (at least some) ba©s,30 as well as food production, although they offer no detail on the nature of this production.31 BLESSING 9781474411295 PRINT.indd 189 22/08/2016 16:13 190 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 Turkish sources use the word ba© to refer to gardens in a variety of locations. Yet, far from creating confusion, this apparent lack of a precise definition may in fact be quite enlightening if we take it as an indication that the word simply had a general meaning.32 Thus, when some unspecified characters of the Vilayat-nama, a hagiography of the early Bekta∞i leaders set in the thirteenth century, claim in the manner of a Greek chorus that ‘a tekye [modern Turkish tekke, that is, a dervish convent] without a ba© and a ba© without water are impossible’,33 their statement should be interpreted as a judgement on the landscaping requirements around a tekye, and one should not assume from this passage that ba©s had some exclusive relation to religious buildings (any more, at least, than water had an exclusive relation to gardens). There is consequently no doubt that the word ba©, as used in Turkish, had a general meaning akin to the English word ‘garden’, rather than to any particular type of garden as in the Persian use of the same word, or to vineyards/orchards as it does in modern Turkish. Ba¯ ghcha: Persian As far as their usage in Persian is concerned, there are two main differences in the meanings of the words ‘bågh’ and ‘båghcha’.34 The first lies in the location of the gardens to which they refer. Unlike the peri-urban or rural bågh, the båghcha is generally appended to a religious building (school or convent),35 and is never depicted as either near another garden nor in the countryside.36 The other important difference concerns agricultural production. There is, in fact, not a single reference, even indirect, to food production (nor, for that matter, of the presence of flowers) in a båghcha when the word is used in Persian. The descriptive elements found in two anecdotes point to the presence of doors as well as, indirectly, walls.37 One case involves a well, presumably used to provide water to the madrasa to which the båghcha is appended, and another centres around a drain clogged by ‘thorns and weeds’ (khår va khåshåq), which incidentally also suggests the presence of plants, even if not commercially grown.38 On the other hand, social interaction seems to have happened in båghchas, although it is worth noting that all three instances supporting this claim hint at a much more spontaneous usage than was the case with båghs. In other words, whereas the bågh is a destination in itself, people gather in the båghcha because they happen to be passing by.39 All these examples suggest that, in Persian, the word båghcha would have referred to a section of the urban space, perhaps best defined as the bordering zone of monumental architecture, with no apparent agricultural function.40 This entails that what Persian narrative texts designate as ‘båghcha’ were different from any garden BLESSING 9781474411295 PRINT.indd 190 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 191 appearing in waqfiyyas (since the latter only mention land producing revenue). More generally, it also shows that the word båghcha had, in late medieval Anatolia, a specific meaning different from uses of the same word in other historical contexts, where it sometimes generically referred to any small garden, and sometimes to vegetable patches. Bag˘ çe: Turkish The word ba©çe, as used in Old Anatolian Turkish, appears to suffer from the same semantic inconsistency as does ba© in the same language. As a matter of fact, with the possible exception of the famed early Ottoman chronicler Å∞ıkpa∞azåde (1400 – after 1484) – who might actually have understood the word in a way similar to that of the Persian-language authors – its use in Turkish-language sources suggests a catch-all term for gardens rather than a word that refers to any particular type thereof. For example, Å∞ıkpa∞azåde situates his ba©çes within city walls,41 whereas in both the Vilayat-nama and the Gharib-nama (c. 1330), a religious treatise widely considered the first extant text written in Old Anatolian Turkish, they occur in the countryside.42 Generally, however, information is simply too scarce to provide us with a consistent picture. Thus, a few mentions of the social use of a ba©çe suggest something akin to the Persian use of the word ba©, that is to say, a destination for social interaction located away from the residential areas.43 Likewise, it is impossible to determine the typicality of a statement designating a well-irrigated plain as an ideal location to establish a ba©çe.44 Perhaps the only common characteristic on which the various Turkish sources other than Å∞ıkpa∞azåde seem to agree (and, therefore, disagree with Persian ones) is with respect to the fruit production which, they claim, was taking place in ba©çes.45 It thus seems that, for lack of a better definition, the word ba©çe was understood by Å∞ıkpa∞azåde in the same way as by the Persian authors. Other Turkish-language authors, on the other hand, may have taken it to refer to ‘a place where fruits are grown’, that is, any garden comprising an orchard component, with no particular limitations as to its location or other characteristics. Bu¯ sta¯ n: Persian The word bËstån is relatively rare in Persian sources, at least when it comes to describing physical locations. One occurrence of the word comes from Bazm u Razm, saying that ‘båghs and bËståns’ located between the city of Erzincan and its countryside were destroyed in an apparent effort to force the besieged city’s authorities into negotiations with the incoming army.46 While this does not say much BLESSING 9781474411295 PRINT.indd 191 22/08/2016 16:13 192 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 about the difference between båghs and bËståns, it gives for both an indication of location as well as a strong hint that they had some relationship to food production. Two other passages, this time in the Manaqib al-‘Arifin (1353), give more detail. One describes the bËstån as the annual endeavour of a nearby fortress’ commander, where cucumbers (khiyår, presumably among other vegetables) are grown on ‘slates’ (takhthå).47 It is worth noting that, though the episode in which it appears involves outdoor social interaction, this interaction ostensibly occurs outside the bËstån. Another case interestingly locates the bËstån inside a bågh, giving a strong impression that the former was the section of the latter where a gardener would perform his duties.48 These passages are, to be sure, limited ground upon which to establish a meaningful definition. Nevertheless, the information available here makes these Persian uses of bËstån perfectly consistent with their Turkish counterparts, as will be seen below.49 Bosta¯ n: Turkish The only Turkish source that uses the word bostån in a clearly nonmetaphorical way is the Vilayat-nama, in which the terms appears in only two, albeit fairly substantial, anecdotes.50 Both take place in the countryside: one, just as in the Manaqib al-‘Arifin, in the vicinity of a fortress; the other in an unspecified location along a major road. Once again, intensive gardening work taking place there on an annual basis is hinted at through the only production mentioned, melon (kavun) – and the apparently common practice of selling the produce directly on location. Considering that central Anatolia’s relatively dry climate requires irrigation for the cultivation of fruits and vegetables, it therefore seems that Turkish-language authors used the term bËstån/bostån in a fashion similar to that of their Persian-language counterparts from whom they had borrowed the term, that is to say, in reference to a labour-intensive commercial operation located in the countryside where vegetables were grown and sold, without any apparent social or entertainment functions. Conclusion A number of conclusions can be derived from the previous pages. At the most basic levels, this vocabulary study offers at least some general guidelines as to the meanings associated with the words discussed for the particular context of late mediaeval Anatolia. Thus, the difference between two of the terms used in waqfiyyas, ˙adÈqa and junayna, seems to relate to their respective urban and rural location, both terms (because of the nature of the documents) referring to commercial fruit- and/or vegetable-growing operations. Conversely, BLESSING 9781474411295 PRINT.indd 192 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 193 Persian sources use the term bågh to refer to garden complexes that were used for both social and agricultural functions (and, therefore, probably to the same reality as the word junayna in the waqfiyyas), båghcha for patches of greenery surrounding an urban building, and bËstån for truck gardens. As for Turkish sources, the limited material they offer us suggests looser definitions that are at least compatible with those of the words they borrowed from the Persian language, ba© and ba©çe probably being used as generic terms (just like the word ‘garden’ in modern English), and bostån more specifically pointing out a labour-intensive, commercial growing operation that might be the closest equivalent of the word karm as it appears in waqfiyyas. But beyond this basic level of analysis, this semantic discussion can lead us to broader conclusions about the social status of these three languages among the Muslim population of Anatolia in the fourteenth century. It is, for example, striking to see that Persian and Turkish share all three garden-related items of vocabulary with each other, and none with Arabic.51 This suggests that Arabic was a socially marginal language whose use was limited to elite legalistic use.52 Furthermore, it gives us some insights into the first literary forays of Turkish, a language that had been a strictly oral language in the region just a century prior. Rather than characterising this phenomenon as a mere migration of spoken words on to the page, one should probably better see it as a process of cultural bridge-building between the popular culture of Turkish-speakers and the Persian literary culture, the former borrowing not only an alphabet, but also vocabulary and, hesitantly, concept definitions from the latter. The social history of fourteenth-century Anatolia is, by the brightest prognostic, still in its infancy. Much groundwork remains to be done, and much of it entails rereading sources that have too often been dismissed as unsuitable for research into social history. We should not forget, after all, that when properly tilled and irrigated, the same soil can yield fruit year after year. Notes 1. The question that attracts the bulk of historians’ attention is, of course, that of the rise of the Ottoman Empire (for a survey of the historiography on this issue, see Kafadar, Between Two Worlds, pp. 29–59). Among the rare exceptions, one should cite Vryonis, Decline of Medieval Hellenism, as well as a handful of studies on Sufism during that period and some recent publications by the contributors to this book. 2. Gardens, and especially elite-owned garden complexes of the late Seljuq era near the Mediterranean city of Alanya, form the subject of one of the most important archaeological studies of late medieval Anatolia, in which Scott Redford states: ‘A well-differentiated nomenclature did not exist either for gardens or buildings in medieval Islam. RËm Seljuk documents and texts often employ one or more of several words of Arabic and Persian origin meaning garden (˙adÈqa, bËstån, BLESSING 9781474411295 PRINT.indd 193 22/08/2016 16:13 194 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 3. 4. 5. 6. BLESSING 9781474411295 PRINT.indd 194 junayna, bågh), seemingly without differentiation. Only on occasion is a more specific word such as “karm”, denoting specifically a vineyard, employed in describing an irrigated agricultural landscape. Otherwise, the gardens examined in this study seem to have been as much orchards as gardens, with fruit trees the only element consistently mentioned therein’ (Redford, Landscape and the State, p. 93). Likewise, in a study of references to agricultural structures in late medieval Spain, based on narrative sources and fatwa collections, Vincent Lagardère suggests that the semantic differences between the various Arabic terms to refer to gardens (janna, karm, bËstån and raw∂a; he does not mention ˙adÈqa) are either inexistent or impossible to ascertain to any significant degree (Lagardère, ‘Structures agraires et perception de l’espace’). The very existence of the following pages obviously entails a certain degree of disagreement with Redford and Lagardère. My claim that some measure of semantic differentiation exists between the various words is based on a number of factors, including the use of a different (though, in the case of Redford, partly overlapping) pool of textual sources, a different chronological and geographical scope, and a somewhat more adventurous approach to the textual material on my part. The statements of both these scholars, however, should serve as a warning to the reader who might be tempted to take the conclusions suggested here as clear-cut definitions rather than as the general semantic orientations that they are. The choice to concentrate on the languages of the Anatolian Muslims (and not, say, Greek or Armenian) stems partly from this author’s linguistic limitations. But also reflects the internal structure of the society described here, insofar as they were often used by the very same people (Arabic mostly appearing as a second language for religious scholars, and Old Anatolian Turkish first migrating to the written page through the quill of authors educated in Persian). It is also clear that the Muslim population, or at least the part of it educated enough to leave us with written material, was in a position of political, legal and economic dominance in the region, which entails a relationship with the land that may not have been shared by the local Christians. See Subtelny, ‘Agriculture and the Timurid Chahårbågh’, p. 116. The catalogue of Ankara’s Vakıflar Genel Müdürlü©ü (VGM) (National Directorate of Endowments) mentions as many as a hundred waqfiyyas from fourteenth-century Anatolia. Further research, however, led me to reduce this number by about two-thirds, after removing duplicates as well as entries for which the actual document could not be found. Among the three dozen or so documents that have been analysed for the purposes of this study, those directly cited will appear in the following format ‘WQ 000-000-000’, where the first number refers to the defter (volume), the second to the sayfa (page) and the third to the sıra (serial number) according to the VGM’s catalogue system. The sources covered here include most of the original texts written by Muslims in or around fourteenth-century Anatolia. In Persian, they are: Astarabadi, Bazm u Razm (hereafter noted as BR, followed by the page number); Aflaki, Manaqib al-Arifin (hereafter noted as MA, followed by the section/paragraph number); and Sipahsalar, Risala (hereafter noted as RS, followed by the page number). In Turkish, the sources used are: Å∞ıkpa∞azåde, Tavarikh-i Al-i Osman (hereafter noted as APZ, 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 7. 8. 9. 10. 11. 12. 195 followed by the chapter (båb) number); Enveri, Dustur-nama (hereafter noted as DN, followed by the line number from the Mélikoff edition); Å∞ık Pa∞a, Gharib-nama (hereafter noted as GN, followed by the volume and page number); Abd al-Karim b. Shaykh Musa, Maqalat-i Sayyid Harun (hereafter noted as SH, followed by the page number); and Vilayat-nama-yi Haji Baktash Veli (hereafter noted as VN, followed by the folio number from the facsimile). The contrast to the vocabulary choices of waqfiyyas is partly due to difference in languages, since bågh, båghcha and bËstån are Persian in origin. The latter term appears twice in waqfiyyas. Both mentions occur in the same document (WQ 611-93-79) as part of village toponyms (rather than referring to any particular garden). Likewise, waqfiyyas include a number of references to bågh and båghcha, but, with a single exception (båghåt: WQ581/2-198-300), always as toponyms for geographical markers bordering endowed property, rather than the endowed property itself. Still another word that could be translated as garden, raw∂a, appears on a number of occasions, especially in Persianlanguage sources. However, it is almost systematically associated with poetic flourishes and/or religious symbolism (often in relation to Paradise, which is also called bËstån-i fardavs), and clearly does not refer to a particular category of earthly gardens. See, for example, BR 178 and MA 8/101. Professor Scott Redford pointed out that ‘the fourteenth century is when the term (raw∂a) starts being associated with funerary gardens’ (personal communication). However, in the sources I used, even the passages that associate raw∂as with Paradise do so as part of superlative praise rather than designating earthly locations. The only exception to this rule is a passing mention in which the word raw∂a designates a location inside a funerary complex (turbah). The context of mention does not allow one to identify the exact meaning of the word (which could refer to a small enclosure around a tomb, or even the tomb itself), but makes it clear that raw∂a here did not fall in the same category as other words discussed in this chapter, that is, terms to designate relatively large areas where plants would be grown. Redford, ‘Just Landscape in Medieval Anatolia’, p. 317, points out that gardens were also used as hunting grounds in the slightly earlier Seljuk period. The few instances of hunting that appear in the sources pertaining to the fourteenth century, however, are not presented in relation to gardens or similar spaces. One waqfiyya in Turkish (WQ 612-6-8), which contains a single garden, is designated as ‘ba©çe’. Both references can be found in WQ 608/1-223-238. As noted earlier, Redford, Landscape and the State, p. 93, singles out the word karm as one of the few garden-related vocabulary items that carry a clear definition. Redford also notes the presence of a grape press in the medieval layers of the Gritille archaeological excavation, in southeast Anatolia (Redford, The Archaeology of the Frontier, p. 164). Lane, An Arabic–English Lexicon, in fact offers many cases where either word has a narrower meaning, but these are so diverse as to be mutually exclusive. Among the definitions he gives for ˙adÈqa, for example, are ‘any round piece of land surrounded by a fence or the like’ and ‘a garden, though without a wall’. Waqfiyyas include, for many but not all of the endowed real estate BLESSING 9781474411295 PRINT.indd 195 22/08/2016 16:13 196 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. BLESSING 9781474411295 PRINT.indd 196 they mention, lists of bordering plots, topographical features or other geographic markers that identify their exact location and limits. BR 207, MA 2/3, 3/21, 3/56, 3/245, 3/298, 3/378, 3/542, 3/590, 4/43, 6/5, 6/25, 8/12, RS 93. On gender separation, see MA 4/43. MA 3/56, 3/245, 3/556, 3/585, 4/39, 4/91. This includes, e.g., teachings from the masters and the MavlåvÈsamå’. MA 3/21, 3/56, 3/121, 3/245, 3/378, 3/459, 3/542, 6/7, RS 93. Respectively, MA 3/378 and 3/457. MA 3/121, 3/245, 3/328. BR 207, MA 3/571. Other references make it clear that setting up a bågh entailed some construction work beyond the mere tilling of the soil (see MA 5/23 and RS 93). Another passage links a bågh with the nearby presence of a water mill (MA 3/298), although it remains unclear whether the latter was indeed located within the bågh itself. BR 207. MA 3/245, 3/301, 8/18, RS 93. MA 3/457 mentions the growing of figs in the ‘karËm’ (‘RËzÈ az yårån-i kiråm yakÈ az karËm-i ikhvån bi-˙a∂rat-i Mavlånå anjÈrÈ avarda bËd . . .’ (‘One day one of the noble disciples had brought a fig from the karËm of the brothers to his Highness Mavlånå . . .’), then goes on to say that the man who brought them goes back, but cannot manage to find the gardener (båghbån) in the bågh. It is true that the word karËm – plural of karm, which is unusual in this source – does create a passage of strongly rhyming prose and might thus have been a synonym of bågh selected for purely stylistic reasons. Yet I prefer to suggest that båghs were divided between areas intended for social interaction and others specifically meant for agricultural production (e.g., karËm). Identifying what Persian-language sources call bågh to the spaces that waqfiyyas call junayna (garden outside the city) would explain the otherwise bizarre formulation found in a waqfiyya (WQ 608/2-63-52) of a ‘junayna that has vineyards and trees’ (junayna dhåt al-kurËm wa alashjår). Social interaction does not preclude the presence of fruit trees, whose shade was apparently appreciated (MA 3/521). Yet it is also possible that some ‘orchard sections’ were so heavily planted as to be rather impractical for such a purpose, as can be assumed from a scene where a man goes from one tree to the next without touching the ground (MA 3/301). Respectively, MA 6/7 and MA 8/18. A single mention of roses (MA 3/79) suggests that flowers were part of the experience of those coming to the bågh to enjoy themselves. Yet the unique status of this reference among passages presenting gardens as physical spaces stands in striking contrast to the emphasis often put on the sight and scent of flowers in poetic descriptions of gardens. This is indeed an excellent example that warns against the temptation to take poetic conventions as reflections of even the subjective experience of their authors. MA 3/298, 8/12. Riding animals: MA 2/3, 3/34, 3/590, 6/25, 7/14. Sleeping in the garden: MA 3/56, 3/121, 3/459. MA 2/3, 3/298, 4/39, 4/91, 8/18. Among waqfiyyas, see WQ 608/2-6352, 581/2-198-300 and 581/2-20-11. BR 159. Redford, Landscape and the State, pp. 63–6, draws the observation about Konya from narrative evidence (the chronicles of Choniates and Ibn Bibi). At least a significant part of such structures must 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 197 have remained under cultivation until the time discussed here and beyond. See Redford, ‘Just Landscape in Medieval Anatolia’, p. 322. Both the social functions and the apparent presence of vegetables set this historically specific use of the word apart from the general meaning of the word, as defined in the Encyclopaedia Iranica: ‘an enclosed area bearing permanent cultures – that is, all kinds of cultivated trees and shrubs, as opposed to fields under annual crops’ (Bazin, ‘Bågh: ii. General’, vol. III, p. 393). APZ 118. Another one, more heavily influenced by poetic and symbolic concerns and thus much less reliable for our purposes, rather emphasises the presence of flowers (GN I 139b 3/4). VN 122a. The Dustur-nama concurs (DN 100), although in reference to a garden located in Thrace. SH 23a, VN 121b. One passage (GN I 94b/8) suggests the production of ‘honey and oil’, although this may have less to do with actual production activities than with the strong symbolic load of the ‘sweet and fatty’ duet (on which, see Trépanier, Foodways, p. 95). One or two other passages (GN I 139b/3 and possibly GN I 103b/9) also refer to flowers, although (just like in Persian) in a way that makes it unlikely that they could have been the central focus of those gardens. Not to mention that most of these examples are set outside the Anatolian scope of this chapter (although, at least in the case of Å∞ıkpa∞azåde, the central Anatolian background of the author preserves their relevance, since this is a discussion of definition – and thus of perception – rather than one of description): near a fortress on the European side of the Dardanelles (APZ 39); within the walls of the newly conquered – and quite depopulated – Istanbul (APZ 124); and near a city in Thrace (DN 100). VN 146a. From a grammatical point of view, ‘båghcha’ is, of course, the diminutive form of ‘bågh’ (just like ‘cigarette’ is to ‘cigar’), but this analysis concentrates on the specifics of semantic differences. Madrasa (MA 3/555, 8/9), zåwiya (MA 8/31) or khånqåh (potentially the same type of building as a zåwiya, MA 3/163). However, one passage mentions a båghcha somewhere in the inner citadel of Kayseri (BR 510), while another one is located beside what appears to be an individual’s residence (RS 90-1) Only in one case is the emphasis shifted by showing the båghcha as large enough to dwarf the size of the building it encloses or, to put it differently, depicting the båghcha as containing the building rather than merely being appended to it (a zåwiya located in Azerbaijan: MA 8/21). MA 3/163; MA 8/9. The idea is further strengthened by the water accumulation in the clogged drain episode, as we are about to see. MA 3/555, RS 90-1. Another anecdote mentions the burial of an unsympathetic religious man in the ‘båghcha of the sultan’. Whether the use of a garden as a burial place – a unique occurrence in the sources I used – was a common practice may be open to question, as a sultanic garden could hardly have been typical in any case. Be that as it may, such funerary practice certainly does not lead the reader to imagine the said båghcha as a commercial agricultural enterprise. In one instance, travellers to Azerbaijan alight in a båghcha they apparently encounter by chance during their trip (MA 8/21). In the other two BLESSING 9781474411295 PRINT.indd 197 22/08/2016 16:13 198 ARCHITECTURE AND LANDSCAPE IN MEDIEVAL ANATOLIA, 1110–1500 40. 41. 42. 43. 44. 45. 46. 47. 48. BLESSING 9781474411295 PRINT.indd 198 cases, the place appears as a playground for a child (MA 8/9) or simply as the place where a few religious characters are relaxing (MA 8/31). It may also be worth noting that, perhaps owing to the very nature of the source (the Manaqib al-‘Arifin, a hagiography of the early MavlavÈ masters), all three of these anecdotes involve religious men, although without depicting them as involved with religious teaching or practice per se. This historically specific use of the word should be put in relation to the more generalised uses of the word, referring to a ‘small garden’ (which can indeed apply here) or to a ‘vegetable patch’ (which clearly does not, the preferred word for such a place rather being ‘bËstån’, as will be seen below), see Eilers, ‘Båg’, p. 393. Båghchas should also probably be distinguished from another type of urban garden, the commercial agricultural spaces called ˙adÈqa in the waqfiyyas, keeping in mind that ˙adÈqas may not have appeared in Persian narrative sources if they did not have a social function. Another hint that these were two separate types of urban gardens is that ˙adÈqas found in the waqfiyyas seem to be well-defined, independent spaces that were not, unlike Persian båghchas, associated with a particular building, despite the fact that mentioning such an association would have been a natural way to identify them in a legal document. Though without necessarily in relation to particular buildings: in Òznik: APZ 32, and in newly conquered Istanbul: APZ 124. Either within short distance of a village (VN 58a) or with no further detail (GN I 80b). There are associations with lofty residential buildings, but, quite unlike the occurrences in Persian sources, it is difficult to see whether the architectural emphasis falls on the garden enclosing the building or on the building that the garden surrounds (GN I 80b, II 64b; SH 19b). See, e.g., VN 58a. Another one refers to the ba©çe as the place, just inside the city walls, where a conquering ruler is welcomed – an event that could hardly have been the typical use of the location (APZ 32). GN I 80b. Another passage is most likely even less generalisable to central Anatolia, as it discusses a ba©çe on an Aegean island that includes a precious stone basin (DN 86). One can find direct references to grapes (grown and sold in GN I 37a) and apples (VN 58a), as well as unspecified fruits on two other occasions (GN I 81b-82a, II 64a). Mention of a ba©çe among the list of properties endowed to a shaykh (SH 19a) also indicates that this property was a source of revenues, presumably (though not explicitly) through the sale of fruits, as in the first example cited here. BR 159. For a more detailed discussion of this episode, see Trépanier, ‘Nicaea, the Gardens, and the Starving Enemy?’. MA 8/50. It annual character is further emphasised by the use of the expression ‘bËstån-i nåv’ (‘new bËstån’). MA 6/22. This, indeed, strengthens an earlier statement I made about båghs being composed of separate agricultural and non-agricultural sections. Yet a fourth occurrence gives much less detail, but, unlike the first three, does present the bËstån as a place of social entertainment (MA 8/41). However, both the untypical format of this passage (first person, eyewitness quotation) and the formulation that is very close to that of a large number of other passages that normally substitute the word ‘bågh’ for ‘bËstån’ suggest that the use of the latter word might be untypical of the way that both Astarabadi and Aflaki defined it. 22/08/2016 16:13 HARVESTING GARDEN SEMANTICS IN LATE MEDIEVAL ANATOLIA 199 49. And, in turn, at odds with a short comparison that Encyclopaedia Iranica makes between garden-related vocabulary items (Eilers, ‘Båg’, p. 393), which rather uses it to refer to a place where flowers are grown. 50. VN 43b and 110a. 51. The difference in vocabulary cannot be ascribed to the different nature of the documents analysed, since one of the few Turkish-language waqfiyyas of that period refers to a garden using the word ba©ça (WQ 612-6-8). 52. This is, indeed, perfectly in line with the observations of Ibn Battuta, who knew no Turkish nor, at the time of his visit to Anatolia, any Persian. See Les voyages d’Ibn Batoutah, vol. 2, pp. 326–8. BLESSING 9781474411295 PRINT.indd 199 22/08/2016 16:13