Religion-Based Cultural Communities in the Pre-Modern Balkans

This work was supported by Ghent University, Belgium. Competing interests: the author is a member of the scientific board of this journal. Publisher: Institute of Slavic Studies, Polish Academy of Sciences. This is an Open Access article distributed under the terms of the Creative Commons Attribution 3.0 PL License (creativecommons.org/licenses/by/3.0/pl/), which permits redistribution, commercial and non-commercial, provided that the article is properly cited. © The Author(s) 2020. Raymond Detrez Ghent University https://orcid.org/0000-0002-8055-9829

In my previously published studies, I have examined the Orthodox Christian community in the Ottoman Empire in the pre-national era (sixteenth to early nineteenth centuries) from different perspectives. My major assertion was that, due to various historical circumstances, this community went through a process of cultural convergence. By the end of the eighteenth century, this resulted in the emergence of a (rudimentary and momentary) supra-ethnic Orthodox Christian proto-nation, which -mainly for reasons of convenience -I have called "Romaic" . The presence of Greek as a shared literary language played a distinct part in this process, both as an eloquent expression of the cohesion of the Romaic community and as a tool that facilitated the cultural exchange between the intellectual elites of the various ethnic groups. The use of only one language, by both consumers and creators of a shared written culture, resulted in the formation of a "Romaic literary system". This consisted of a shared "high", "scholarly" literature and, on behalf of the uneducated, a number of "low" religious and edifying texts in the various vernaculars.
In this contribution, I intend to examine to what extent this model is applicable to the other religious communities in the Balkans and to illustrate how religious appurtenance and the use of a common supra-ethnic "high" language decisively contributed to the formation of scholarly networks. On the one hand, this generated dynamic multi-ethnic communities of intellectuals who were influenced by religion, but on the other hand it segregated even conational intellectuals who followed the same religious lines.

A segregated society
Life in the pre-modern Ottoman Empire was marked by religion in many ways. All intellectual, scholarly and artistic activities were permeated by faith in an omnipresent and almighty God. This was the case almost everywhere in premodern Europe. In the Balkans however, identification with a religious group was enforced by the specific way the multi-ethnic population of the Ottoman Empire was administratively divided into confessional communities or millets. The very organization of the Ottoman state made religious affiliation and allegiance extremely important, especially where intellectual life was concerned.1 There were three millets: the Orthodox Christian or Rum millet-i, the Armenian millet and the Jewish millet.2 All millets belonged to the Judeo-Christian 1 As a matter of fact, the term millet in the sense of a religious community is attested for the first time in late eighteenth-century sources. Previously, non-Muslim religious communities in the Ottoman Empire were tayife's -a general term meaning 'group'. As genuine administrative units, millets were created in the nineteenth century in the framework of the Ottoman state reform program known as the Tanzimat. However, in Balkan historiography the term millet was already used to denote non-Muslim religious communities from the fifteenth century onward. The standard reference work on the millet system is Braude & Lewis, 1982; is an abridged and updated version. For basic information on the Armenian millet, see ; on the Jewish millet, see . More information about the Armenians and the Jews in Bulgaria in Кръстева, 1998, рp. 138-154, 221-242, respectively. About the Jews also  2 Rum is an Arabic-Turkish derivation from Greek Rōmaios meaning 'Roman', as the Byzantines called themselves. The term also acquired the additional meaning of 'Orthodox Christian', which was preserved in the Ottoman period. It then referred to (ethnic) Greeks and to all Orthodox Christians in the empire. Spain and Portugal, succeeded much better in exploiting the limited cultural opportunities the millet-system offered.
Typical of the millet-system was the disregard of ethnic distinctions. The Rum millet-i housed Albanians, Bulgarians, Gagauzes, Greeks, Karamanlides, Serbs, Vlachs, and others.3 The majority of them lived in dioceses belonging to the Patriarchate of Constantinople, but other Orthodox churches in the Balkans (the Autocephalous Archbishopric of Ohrid and the Serbian Patriarchate of Peć) were just as multi-ethnic. In spite of their ethnic diversity, all Orthodox Christians were treated by the Ottomans as belonging to one single community and they saw themselves as such. Although some ethnic awareness continued to exist, it did not occupy the dominant moral and aesthetic position it eventually acquired among people in the nationalist nineteenth century.
The existence of an Armenian and a Jewish mono-ethnic millet did not result from the fact that Jews and Armenians represented separate ethnic groups. Jews clearly constituted a distinct religious community; the Armenians, being Christians, adhered to a Christological doctrine differing from the Niceno-Constantinopolitan Creed and had an ecclesiastical organization of their own, so they were considered as a distinct millet. On the other hand, without being formally recognized as millets, Muslims and Catholics constituted multi-ethnic religious communities in the same way as the Rum millet-i.
In spite of all the differences between Muslims and Christians as far as their legal status was concerned, the intellectual life of the Rum millet-i, the Islamic ummah and the Catholic communitas christiana displayed a number of striking similarities, specifically concerning the use of literary languages and spoken vernaculars. They are the subject of the following exploratory investigation.

The Rum millet-i
From the Middle Ages onward, the Greek language played a particular role among the Orthodox Christians in the Balkans. The introduction of Cyrillic script did not eliminate Greek influence, but it elevated the cultural level and increased the craving for knowledge. The main source of scholarship, Byzantine literature, was accessible only to those with a command of Greek. In addition, Greek already functioned in the medieval Balkans as an interethnic tool of scholarly communication among Orthodox Christians and continued to do so under Ottoman rule. Greek was considered a sacred language: the gospels, the writings of the Eastern Church Fathers, and the decisions of the ecumenical councils were all written in Greek. The Patriarchate of Constantinople used Greek as a liturgical language and as the language of church administration. In all dioceses dependent on the Patriarchate of Constantinople in Bulgaria, Macedonia, and Albania, where the majority of the population was not Greek, the patriarchate nevertheless appointed Greeks or Graecized non-Greeks as bishops. Since the language of the church was Greek, education, which for many years had been limited mainly to the training of future clerics, was in Greek too.
The same situation existed within the jurisdiction of the Autocephalous Archbishopric of Ohrid, which was Graecized to the same extent as the Patriarchate of Constantinople. Also, the Romanian metropolitanates of Ungro-Wallachia and Bogdan (Moldova), both of which belonged to the Patriarchate of Constantinople, increasingly used Greek in addition to Romanian after they abandoned Church Slavonic as a liturgical and administrative language in the late seventeenth century. Only the Patriarchate of Peć, abolished after the Ottoman conquest of Serbia in the fifteenth century and re-established in 1557 under Süleyman the Magnificent, firmly maintained the Church Slavonic tradition.
Undoubtedly, until the eighteenth century, Church Slavonic services were still customary in villages and towns in Bulgaria and Macedonia. Liturgical books in Church Slavonic and religious literature in Russian were imported from the eighteenth century onwards (Павлова, 1979, pp. 163-165). However, Russian books only started to enter Bulgaria on a large scale in the 1830s (Генчев, 2002, pp. 82-83). The vast spread of Greek outside of the ecclesiastical sphere resulted from the emergence of a new upper class -a well-to-do petty bourgeoisie -consisting of Greeks, who had an age-old tradition of seafaring and trade, and Graecized non-Greeks, mainly Slavs and Vlachs. They used Greek for professional reasons and as a social marker that distinguished them from the poor and uneducated peasantry. This multi-ethnic, Graecophone bourgeoisie insisted on divine services and education in Greek, even when church authorities themselves preferred to pursue a more flexible language policy in order to keep their flock united.
In addition, from the eighteenth century on, Greek was the language that -especially among the fledgling bourgeoisie -also emblematized Enlight-enment. Greek travelers and merchants in Western Europe were the first to get in touch with the new philosophical and political ideas. Due to the wide-spread knowledge of Greek among the bourgeoisie, Greek translations of Enlightenment authors and Enlightenment literature written by Greek authors easily found their way all over the Balkans. To have a command of Greek now became an indication of progressivity and modernity. Again, also within the Enlightened and increasingly secularized Orthodox Christian community, Greek enabled intellectuals of various ethnic backgrounds to communicate and to create new intellectual networks.
Greek Enlightenment authors increasingly resorted to demotic Greek in order to reach an audience as wide as possible.4 Prior to the eighteenth century, intellectuals used various forms of archaizing Greek, varying from New Testamentic koine Greek to classical Attic. To keep things easy, all these varieties have been labelled here "literary Greek". Uneducated people who knew only demotic Greek had practically no access to texts written in literary Greek. Sociolinguists characterize such a linguistic situation as "diglossia without bilingualism" . Within many societies there are two radically different variants of one single language or two languages. Linguists call these "codes": a "high code", a predominantly written language that fulfils "high" functions in worship, legislation, the media, education, literature, etc.; and a "low code", the spoken vernacular that is used for routine colloquial functions.
As literary Greek was widely known among the elites of all ethnic groups in the Balkans, and since other Balkan languages -with the exception of Church Slavonic in Serbia -were not used to perform high functions, the linguistic situation within the Orthodox Christian community might be characterized as an "extended diglossia". Literary Greek was used for all high, literary functions, while the vernaculars -Albanian, Aromanian (the language of the Vlachs), Bulgarian, (demotic) Greek, Karamanli and Gagauz Turkish -were used for low, colloquial functions. This particular "extended diglossia" emerged as follows. In Venice in 1557, the Greek scholar Damascenus Stoudites published his Thesaurus (Biblion onomazomenon Thēsauros), a collection of 36 undemanding sermons, hagiographies, and edifying texts in demotic Greek, meant for common people who did not know literary Greek and consequently had no access to the guiding principles of Christian faith. In the introduction to the chapters in his Thesaurus, he points out that he uses "common language" (eis tēn koinēn glōssan), "colloquial language" (logos pezēi phrasei) or "simple language" (logos idiōtikēi phrasei), to distinguish it from the literary language in which he wrote his other books (Дел'Агата, 1999, pp. 42-43). Damascenus did not call his language "low", but that is precisely what he had in mind.
Chapters from the Thesaurus, translated and called damaskins, acquired huge popularity in Bulgaria and Macedonia and even developed into a particular literary genre (Петканова-Тотева, 1965). Until far into the nineteenth century, damaskinars -writers of damaskins -copied damaskins and created new damaskin-like texts in their own hand. The earliest translations were still in Church Slavonic, but from the beginning of the seventeenth century the translators used the Bulgarian vernacular. Obviously, the damaskinars faced the same problem as Damascenus: Bulgarians were no longer able to understand "high code" Church Slavonic. To define their language, the damaskinars adopted Damascenus' terminology: they claimed to write in "common language" (po obštem ezycě, na obštij ezik, obštim ezykom, obštim skazaniem) or "simple language" (prostim skazaniem, prostyim tlăkom, na prostom ezykom, na Bălgarsky po prosto, na prost jezyk) (Дел'Агата, 1999, pp. 42-43). The famous damaskinar Josif Bradati (Joseph with the Beard) explains the function of his "low code" even more explicitly, pointing out that he translated "into the simple Bulgarian language to teach and edify the simple and ignorant Bulgarian people" (Петканова-Тотева, 1965, p. 202). And Sophroniy, the bishop of Vraca, writes that he translates "from Church Slavonic and the profound Greek language into Bulgarian and simple language in order to be read in the churches on Sundays so that simple, uneducated people and women and children could understand God's law" (Ангелов, 1994, p. 233).
One may wonder which language the damaskinari actually considered "high code": Church Slavonic or Greek? It appears, however, that "low code" Bulgarian related more closely to "high code" literary Greek than to Church Slavonic. In eighteenth-century Bulgaria and Macedonia, from the many "high" functions that Church Slavonic exerted in the Middle Ages, only the liturgical one remained. In the sixteenth century, the žitiе (vita) of the Bulgarian new martyr, Georgi of Sofia, was written in Church Slavonic, but afterward monks and priests mainly produced copies of Church Slavonic texts. This was at variance with what happened in Serbia as hardly any new texts in Church Slavonic were created by Bulgarians (Николова, 2006, p. 42). An exception is Parteniy Pavlovič who wrote his autobiography in Church Slavonic in the early eighteenth century. However, Parteniy spent almost all his life in a Serbian intellectual environment where Church Slavonic had retained most of its functions. The very few divine services in Church Slavonic, like that dedicated to Teodosij of Tărnovo by Spiridon of Gabrovo (1814), belong to the liturgical sphere (Богданов, 1983, p. 258). By the end of the eighteenth and at the beginning of the nineteenth centuries, in Bulgaria and Macedonia virtually all "high" functions of the language were fulfilled by literary Greek (Георгиева, 1989, pp. 18-19; Гутшмит, 1973, p. 100).
This "extended diglossia", consisting of literary Greek on the one hand and vernacular Greek and Bulgarian on the other, may be extended even more to include other languages spoken by Orthodox Christians in the Ottoman Empire and so become a genuine "Romaic diglossia", which was at the base of the "Romaic literary system". Using Greek script, the Karamanlides wrote a language philologists called karamanlidika in Greek and karamanlıca in Turkish; however, the authors of the hundreds of religious and edifying texts (including damaskins) in that language called it sate tourktze (sade türkçe) or atzik tourktze (açık türkçe) -"simple", "open", "accessible" Turkish (A survey of karamanlidika literature in . They obviously regarded karamanlidika as "low code" with regard to literary Greek, which was widely used by educated Karamanlides as "high code". Karamanlidika texts were read during the liturgy, but a complete liturgy in karamanlidika most probably did not exist (Clogg, 1999, pp. 119-122). In the introduction to the 1803 second edition of the first printed book in karamanlidika, Gkioulzari imani mesichi (1718), the translation of the Greek Apanthisma tēs christianikēs pisteōs [Flower Garden of the Christian Faith], the author/translator notes that the book is meant for Orthodox Christians in Anatolia who do not understand Greek and "found themselves in profound ignorance" about "our Holy and Orthodox faith" (Clogg, 1999, pp. 122-123). Karamalidika books were also used by the Gagauzes; they too read "high literature" in Greek (Kapalo, 2010, pp. 2-20;Мошков, 1901, pp. 42-44).
Aromanian and Albanian functioned in the same way as Bulgarian, demotic Greek and karamalidika, as "low code" in relation to "high code" literary Greek, but prior to the nineteenth century the literary production in both languages was extremely limited. However, the most important of the very few texts in Aromanian at our disposal that date from the eighteenth and early nineteenth centuries perfectly fits into the diglossic model I have sketched. The Liturghier aromânesc [Aromanian Missal], written in Greek script and dating from the middle of the eighteenth century, contains translations of sermons and other religious texts. The Codex Dimonie from the beginning of the nineteenth century, discovered in Ohrid, consists of translations into Aromanian of edifying texts by various Greek authors, including Damascenus Stoudites (Kahl, 2006, pp. 24-42;Lazarou, 1986, pp. 126-141).
In Albanian, the most convincing cases are the Anonimi i Elbasanit [Anonymous [manuscript] from Elbasan] from the middle of the eighteenth century, which contains fragments of the Gospel translated into Albanian and written in an adapted Greek script. The anonymous author is probably Grigorios of Moschopolis, the future bishop of Durrës and author of a number of hagiographies in literary Greek. Todhri (Theodoros) Haxhifilipi produced a translation of the Bible in Albanian which unfortunately got lost. His younger contemporary Kostandin Berati is reported to have written a collection of Biblical and liturgical texts and a religious poem, both in Albanian, and a Greek-Albanian dictionary. A part of his collection was included in an overwhelmingly Greek manuscript produced in 1822 by Kostandin Cepi (Elsie, 1995, pp. 122-130).
"High code" Greek literature was the common intellectual property of the entire Orthodox Christian community in the Ottoman Empire. Tellingly, in Bulgaria between 1750 and 1840, 1,115 different books were printed in Greek, whereas only 52 were available in Bulgarian (Стоянов, 1957, pp. 471-472, 1978, pp. 47-168). The majority of these books were on history, science, theology, philosophy, jurisprudence, etc. They are found in private libraries of educated Bulgarians, and in the libraries of Bulgarian monasteries, schools, and čitališta (reading rooms) (Danova, 2002(Danova, -2004. Although to my knowledge no figures are available about books circulating among Albanians and Vlachs, there is no reason to assume that the situation there was different. Albanian, Bulgarian, Karamanli and Vlach intellectuals had an adequate command of literary Greek and as Greek books perfectly satisfied their intellectual needs there was no demand for books in their own language. Although the majority of the writers of these Greek books were (ethnic) Greeks, the Albanians, Bulgarians, Karamanlides, Vlachs and others were no mere consumers of this literature but also contributed to it as authors. In the eighteenth and the first half of the nineteenth centuries, about 30 Bulgarian intellectuals (Алексиева, 2010; Николова, 2006, p. 109; Стоянов, 1957, pp. 455-475), more than 40 Vlachs (Mackridge, 1981, p. 71;Έξαρχος, 1994, pp. 51-53), six Albanians (Elsie, 1995, p. 71) and a number of Karamanlis (Mackridge, 2009, pp. 64-65) wrote in literary Greek, addressing not only the Greeks but the entire Orthodox Christian community in the Ottoman Empire.

The Muslim community
I have elaborately described the Romaic diglossia and literary system, assuming they may be used as models when studying similar sociolinguistic and literary phenomena within the Balkan Muslim and Catholic communities. Since I am neither an Orientalist nor a Latinist, my analysis of the Muslim and the Catholic communities will necessarily be cursory. It is meant rather as an invitation to experts in the aforementioned fields to make further comparative investigations.
Just like the Orthodox Christian community, the Muslim community was based on religion and was multi-ethnic. "As a result of the process of Islamization (in the broadest sense)", writes Rossitsa Gradeva, A community united first and foremost by religion emerged in the Balkans. Through different in origin and identity, its members became part of Muslim culture and contributed to its development in a Balkan mould. Using different languages (Turkish, Bulgarian, Albanian, Serbian and Croatian) in both their "high" and "popular" literature, they all employed Arabic script. This they used in most cases to write literary, academic and theological works, and they even wrote down folklore in the alien Arabic and occasionally Persian or Ottoman Turkish languages. Some corners of the empire also witnessed the emergence of local literature written in the languages of the local Serbo-Croat or Albanian Muslims, but also in Arabic script. Arabic, the language of the Qur'an, was most closely associated with the profession of Islam, and so became, at least in literature, the language uniting all Muslims. Its unifying role was a factor in the "descent" of many topics into popular writing, and vice versa, and it additionally boosted interest in literature at even the lowest layers of Ottoman society. (Gradeva, 2004, pp. 159-160) Within the Balkan Islamic community, no less than three languages (Arabic, Persian and Ottoman Turkish) exerted the "high" function that Greek exerted within the Christian community. Arabic, the language of the Quran, of Quran comments and exegeses, and religious literature, was generally and undoubtedly the most important of the three, and its role in Muslim society most resembled that of literary Greek among educated Orthodox Christians. Since all Muslims were supposed at least to be able to read Arabic, this language was par excellence the tool of interethnic intellectual communication. Classical (New) Persian was the language of mystical and allegorical love poetry and belles-lettres, the command of which was limited to a small number of highly educated people. Finally, Ottoman Turkish, the language of administration, which differed considerably from vernacular Turkish, was used as a chancellery language, but also with lit-erary aims. Moreover, texts written in one of these languages were also the cultural possession of all Muslims, both as consumers and producers. All popular Oriental literary genres -various forms of poetry (gazel, mesnevi, rubai, kasida, ilahi), religious treaties, historiography, philological commentaries, etc.) -were practiced, not only by Turks but also by autochthonous Balkan Muslims like Albanians, Bosniaks, Pomaks and others. Balkan cities like Plovdiv, Sofia, Vidin and Stara Zagora in Bulgaria, Sarajevo and Travnik in Bosnia, Prizren in Kosovo, Gjirokastër in Albania, Yoannina in Greece, and many others possessed rich libraries with books in Arabic, Persian and Ottoman Turkish. (For the libraries in Bulgaria, see Стайнова, 1983; Събев, 2017.) In the sixteenth century, Ahmed Sudi Bosnevi translated Saadi's famous Persian poem Gulistan into Turkish. Bosniak literates also wrote original mystical poetry in Persian. Known to the entire Muslim world were the commentaries in Arabic by the early seventeenth-century Abdullah Bosnevi, 'the Bosniak' on the Fuṣūṣ al-Ḥikam (The Bezels of Wisdom) by the thirteenth-century Andalusian mystical Ibn al-ʿArabī. In the same period, Ibrahim Pečevija, originating from Pécs in Hungary, wrote a history of the Ottoman Empire in Ottoman Turkish, making use of Western and, more specifically, Hungarian sources. A long list of more than 150 Bosniak authors writing in Arabic, Persian and Ottoman Turkish has been compiled by Smail Balić (1992, pp. 208-264).
The Albanian Priştineli Mesīhī or Mesīhī from Prishtina, who used to be a șehir oğlanı, 'city boy' or 'bohemian' in early sixteenth-century Istanbul, wrote love poetry in Turkish. A younger contemporary of his was Dukagjin-zāde Taşlıcalı Yahyā bey, a member of the Dukagjin clan who ended up in Istanbul in the framework of devșirme or child levy. He authored a collection of gazels and a famous mystical mesnevi, in which the relations between two lovers are allegorically described as that between a king and a beggar. In the eighteenth century, Nezim Frakulla from Berat, who also spent most of his life in Istanbul, wrote in Turkish, Persian, and probably also in Arabic Norris, 1993, pp. 61-64).
In addition to the Islamic "high codes" that were used by the Muslim cultural elites of all Balkan ethnic groups but which were accessible only to the educated, there was also a "low code" Bosniak literature called aljamiado, which was intended primarily for those who had no command or only a poor knowledge of Arabic.5 Like their Orthodox Christian counterparts, these works of literature in the vernaculars had an overwhelmingly religious character. However, by the end of the eighteenth century, worldly subjects were also dealt with -a tendency that can be similarly observed in Orthodox Christian Enlightenment literature. The genre that was practiced most often was the ilahis. These mystical poems were written in Bosniak in Arabic script by Sufi dervishes and were chanted during their weekly Sufi rituals. The authors followed the Arabic and Persian models as closely as possible and abundantly included Arabic and, in particular, Persian words into their poems.
In Albanian too there is a corpus of poems written in Arabic script by bejtexhinj (singular bejtexhi, author of bejts, 'couplets'). Most of them are didactic, edifying, or meditative poems. The above-mentioned Nezim Frakulla compiled an anthology of Albanian poems, which in his view were as valuable as the best Persian poems. Hasan Zyko Kamberi, who lived in the second half of the eighteenth century, participated in the battle of Smederevo in 1789 as a soldier in the army of Ali Pasha and died as a Bektashi dervish. He wrote mystical, but also intimate, satirical and critical poetry and a rhymed account of the battle of Smederevo.
I did not succeed in discovering texts written by Bulgarian Muslims (Pomaks) in Arabic script, although such texts must have existed. The Greek physician and local historian Polys Mylonas (Μυλωνάς, 1990, p. 25) mentions a Bulgarian official of the Foreign Ministry to whom (in the 1980s) a Pomak hodja from Tikveš read "a half-philological poem which was written in pure language but in Arabic script" (misofilologiko poiēma, to opoio ētan grammeno se katharē glossa me arabikous omōs charaktēres).

The Catholic cultural community
Within the borders of the Ottoman Empire there were Catholics in Northern Albania (more specifically in the region of Shkodër), in Bosnia, and in the region of Čiprovci in northwest Bulgaria, where many Croat families had settled. The Catholics too possessed a common "high code", namely Latin, which was used for "high" functions such as the liturgy, and in theological and scholarly literature and in administration. Within the multi-ethnic Catholic community, Latin -like Greek in the Orthodox Christian community and Arabic in the Muslim one -was the language that enabled communication among intellectuals belonging to various ethnic groups and nations. At the same time, the Latin characters served as a symbol of Catholicism in the same way as the Greek characters did among the Orthodox Christians and the Arabic ones among the Muslims.
In addition to Latin, Catholics also had books in various vernaculars for those who ignored this language. However, in the Catholic world the distinction between "high code" Latin and "low code" vernaculars was less clear-cut since many vernaculars also increasingly fulfilled "high code" functions. As Catholics in the Western Balkans enjoyed the guidance and the protection of the Italian ecclesiastical authorities, and many Balkan Catholics studied in Italy, Italian was also used for some "high code" functions.
In On behalf of those who did not know Latin, most of these authors also produced original works and translations from Latin and Italian into "Illyrian", a mixed language based primarily on (the Dalmatian and Bosnian variant of) Croatian that had many Bulgarian particularities and which used Latin and, more rarely, Cyrillic characters. The term "Illyrian" may refer to Croat (and more specifically Croat as spoken in Dalmatia or Bosnia) or to some other variety of (South) Slavic, including Bulgarian; alternatively, it may be used as a synonym of Slavic in general. It was, in other terms, not a standardized literary language but rather a kind of South Slavic koine with a large diversity of concrete realizations, depending on the native language of the authors who claimed to use it (Милтенова, 2008, p. 746; see also the discussion of the term "Illyrian" in  In 1638  Roughly from the seventeenth century onwards, among the Catholics in the Ottoman Empire, Italian also started to fulfill some of the "high code" functions, at least outside the liturgical sphere. Along with Italian, the various concrete realizations of Illyrian also gradually acquired some features of a "high code" as it was used in literature of a "learned" nature, and not only on behalf of the uneducated. Since this process started much earlier among the Croats outside the Ottoman Empire and their printed books were also used by Catholic Slavs within the empire, e. g. the Catholic Bulgarians, the model of a shared "high code" (Latin) and various "low code" vernaculars in their case is blurred.
Albanian Catholics were less numerous and less productive than their Bulgarian or Croat coreligionists.

Conclusion
In the pre-modern Ottoman empire, intellectual life developed along religious lines. The Rum millet-i, the Muslim ummah, and the Catholic communitas christiana represented separate communities with their own cultural traditions and institutions. All three made use of a literary language which fulfilled "high" functions in worship, legislation, scholarship, and literature, and which was known to the intellectual elites of the various ethnic groups that constituted the community. This common language undoubtedly strengthened the unity and solidarity within each community, but at the same time it isolated the community from the other communities, with which -literally and metaphorically -no "common language" for intellectual communication was available. In addition to being reluctant to communicate with the "other" on religious grounds, the fact that they largely ignored each other's literary languages rendered communication all but impossible.
Within each community, vernacular languages fulfilled low, colloquial functions. They were used in writing only on behalf of the unschooled, who had no access to "high" literature. Written texts in vernacular language were mainly of a religious, edifying nature and were used during the liturgy (in sermons) or in Sufi rituals. The striking similarities between the "literary systems" existing within each community were due to the similar social and cultural conditions in which these communities, in spite of their unequal legal status, coexisted. Certainly, the occurrence of a "high" and a "low" code and the particular use of them in "high" and "low" literature can be found in many societies and is not typical of the Ottoman Empire alone. However, its density and complexity render it a phenomenon that can be considered as "typically Balkanic" in the sense Civ'jan attributes to this term when claiming that universal features materialize in the Balkans "in a clear, powerful and extraordinarily Balkan way" (Цивьян, 2006, p. 67).
The most important conclusion to be drawn is that the lack or scarcity of "high" literature in some Balkan languages is in no way an indication of the "culturelessness" of their speakers. In the three cases described, the speakers of the vernaculars shared a rich "high" culture with their coreligionists, albeit not in their very "native language". Identifying themselves with a religious rather than an ethnic community, they considered the shared literary language not as a foreign language but as a means of intellectual communi-cation that was part of their collective identity. The use of literary Greek did not turn non-Greeks into (ethnic) Greeks, and neither did the use of Arabic turn Bosniaks and Muslim Albanians into Arabs: they only became more committed Orthodox Christians and Muslims. All this radically changed in the nineteenth century, when the rise of ethnic nationalism turned the traditional relationship between ethnic and religious belonging upside down, reducing religion to a mere component of national identity.