The Study of Religions in Premodern Muslim Civilization: Some Distinctions Concerning Its Disciplinary Status

: Scholars have made contesting claims about the nature and scale of works on religions by Muslim scholars before modern times. The present paper explores various primary and secondary sources, especially the classical bibliographical indexes that the scholarly tradition under scrutiny itself produced, and classiﬁes these works into three types: (a) polemics, (b) works that present authentic knowledge about various faith traditions or introduce methodological novelties but carry some degree of apologetic undertone, and (c) descriptive writings on religions which resemble the modern-day academic study of religion. Based on these distinctions and an assessment of the number of works in each type, the paper maintains that a sprouting tradition of descriptive studies of religions existed in the pre-modern Muslim societies, which introduced certain methodical novelties such as comparative method, historiography, and, last but not least, textual criticism, which seems to have heralded the modern biblical studies in some respects. However, this tradition could not mature into a full-ﬂedged discipline at par with many other branches of knowledge that ﬂourished in the heyday of Muslim civilization. These ﬁndings imply that the descriptive study of religions other than one’s own is not necessarily a modern Western phenomenon. It can take root in multiple cultural settings.


Introduction
A bulk of writings about religions other than Islam has reached us from the premodern Muslim civilization, more specifically from its classical (Hodgson 1977;Lassner and Bonner 2010;Peters 1994) or medieval (Lasker and Stroumsa 1996;Saunders 1965;Von Grunebaum 1961) period that lasted from the middle of the seventh century up to the end of the fifteenth century (Waardenburg 1998, p. 18). Scholars have made conflicting claims regarding the exact nature and magnitude of this body of literature. On the one hand, several academics view that Muslims had pioneered the study of religions before it emerged as a discipline in modern times. For instance, Franz Rosenthal opines that "the comparative study of religions has been rightly acclaimed as one of the great contributions of Muslim civilization to mankind's intellectual progress" (Rosenthal 1976, p. 5). Steven M. Wasserstrom holds a similar opinion. However, he sees this development as a creation of what Marshall Hodgson (1922Hodgson ( -1968 calls the "Islamicate" civilization, by which the latter means cultural products of the societies dominated by Muslims but not necessarily religious (Hodgson 1977, pp. 57-58). Wasserstrom writes: There is general agreement among historians of the history of religions that Islamicate civilization produced the greatest premodern historical studies of world religions. Indeed, Western scholarly approbation of this literature has been sustained and enthusiastic, based on the observation that that historical science was pioneered by Muslims (Wasserstrom 1988, p. 408).
In the same vein, many contemporary Muslim scholars confidently claim that a fullfledged science of religions existed during Muslim civilization's heyday (Al-Sharqawi 2000; The modern discipline of religious studies is perplexingly heterogeneous, with a wide range of approaches and diverse methodological preferences. In the face of continuous methodological rifts, some important scholars used the platform of the International Association for the History of Religions (IAHR) to promulgate the basic minimum methodological criteria for this field in 1960. The crux of their joint statement was the rejection of any political, religious, or ideological agenda and acceptance of objective and scientific inquiry as to the sole purpose of the academics working in this field (Schimmel 1960;McCutcheon 1996). However, concerns like a poor representation of different cultural perspectives and regional contexts (Pye 1989;Geertz 2000;Wiebe 2012) and male dominance of the field (Joy 2005;Reed 2015) could not be pacified. Similarly, the purported connection of religious studies to orientalism (Whaling 1995), colonialism (Hedges 2008;Tayob 2018), and Christian theology (Neville 1993;Milbank 2004) continued to haunt the field. Some scholars even challenged the conceptualization of "religion" itself as a valid analytical category for its inability to portray the cross-cultural aspects of human life accurately (Fitzgerald 2000, p. 10;Smith 1998, p. 269).
Thus, a positivistic view of religious studies as a science with precisely defined methodologies is contested at multiple levels (Gross 2005). Instead, a culturally inclusive, methodologically plural, self-critical, and reflexive conceptualizing of this field is more plausible, which would imply keeping the methodological criterion as minimal as possible to allow the representation of the hitherto unrepresented groups. Thus, for the present undertaking, we propose a single point criterion. Those works on religions from Muslim history that are mostly descriptive and not explicitly polemical or apologetic can be considered as the study of religion even if they differ from the modern religious studies in some respects.
Against this background, the present paper aims to distinguish various writings on religions produced by Muslim scholars before modern times. For this purpose, it surveys the classical Arabic bibliographical indexes and lexicons to trace out how different genres of writings about religions gradually surfaced. Titles of many books also give vital hints about their disputative or descriptive nature. Such heuristic techniques provide crucial clues to self-understanding of the knowledge tradition under discussion. Then, the paper takes into consideration the work of those academics who have already ventured to count the number of books on religions from the Muslim cultural history under various categories like heresiographies, refutations, and descriptive writings. Last, on noting that the percentage of the descriptive works on religions that resemble the modern-day religious studies is considerable, this paper discusses whether they signify a sustained disciplinary tradition with continued teacher-student lineages and intertextuality. In this regard, the article also considers if the classical Muslim mappings of various sciences mention the study of religions as a distinct branch of knowledge.

A Typology of Muslim Writings on Religions
Various types of works from the classical Muslim civilization refer to religions other than Islam, many of which only indirectly and in passing. For instance, descriptions about different religions are found in the multivolume histories of the world, like those produced by Muh . ammad ibn Jarīr al-T . abarī (839-923) and 'Alī ibn , in the travelogues of the famous travelers and explorers such as Muh . ammad al-Idrīsī (1099-1165), and Muh . ammad ibn 'Abd Allāh ibn Bat . t .ū t . ah (1304Bat . t .ū t . ah ( -1369, and also in the fiction like Alf Layla wa Layla (One Thousand and One Nights) (Al-T . abarī 1967;Al-Mas'ūdī 1985;Ibn Bat . t .ū t . ah [1853] 2020Ibn al-Muqaffa' 1935;Waardenburg 1998, pp. 21-23). However, numerous works specifically dealt with religions, which appeared under a variety of titles and generic nomenclatures such as al-rudūd (refutations), maqālāt al-firaq (propositions of various religious factions), and al-milal wa al-nih . al.
A convenient yet authentic source of information about these genres and writings on religions is the bibliographical indexes produced in the classical Muslim civilization, which meticulously cataloged titles of books on various subjects and classified knowledge into different branches with their peculiar nomenclatures. These indexes include, for instance, al-Fihrist (The Index) Kātib Chalabī (1609-1657). Referring back to these bibliographical indexes and classifications of various branches of knowledge can give us crucial clues to the self-understanding of the knowledge tradition whose nature and status is disputed about and avoid the pitfall of imposing the modern concepts and categories on the ideas and notions from the distant past without considering their proper context.
A common genre about religions found in the available bibliographical indexes is alrudūd, which means "refutations". One of the initial bibliographical indexes is al-Fihrist by Abū al-Faraj Muh . ammad ibn Ish .ā q al-Nadīm (d. 995 or 998), which includes an extended list of al-rudūd works, including those against the other religions (Al-Nadīm 1971, pp. 158, 207, 210, 215, 230). Closely associated with this genre are two branches of knowledge that appear in the bibliographical indexes, namely, 'ilm al-munāz . arah (knowledge of debating or polemics) and 'ilm al-jadal (knowledge of disputation) (Chalabī n.d., pp. 1: 13-18). As the nomenclatures are indicative, al-rudūd, 'ilm al-munāz . arah, and 'ilm al-jadal are primarily a disputative body of literature. Most of the disputative writings enlisted in these bibliographical indexes were against Christianity, followed by those against Judaism, and sparsely addressed non-Semitic religions. It is critical to note that al-Nadīm devotes one out of ten chapters of al-Fihrist to describe the subject of religions, which consists of his descriptive account of beliefs of various nations. However, the chapter does not record any specific books (Al-Nadīm 1971, pp. 383-414). However, it is vital to note for the present discussion that he omits al-rudūd in the chapter on religions, a clear hint that he distinguished between the descriptive interest in religions and polemics. From the point of view of the present undertaking, a more important genre found in a bit later indexes is maqālāt al-firaq (propositions of various religious factions), also known as 'ilm maqālāt al-firaq (knowledge of the various religious factions) that deals with Muslim sects and by extension also religions other than Islam. Kashf al-Z . unūn records maqālāt al-firaq as a branch of knowledge and mentions under it some works that deal with Muslim sects as well as non-Islamic religions, like al-Maqālāt fī Us .ū l al-Diyānāt by Abū al-H . asan 'Alī ibn Hussayn al-Mas'ūdī (d. 346), al-Maqālāt by Abū Mans .ū r Muh . ammad ibn Muh . ammad al-Māturīdī (d. 333), by Zufar ibn Huzayl, by 'Abd Allāh ibn Mah . mūd al-Ka'abī (d. 319), and by Ah . mad ibn Muh . ammad al-Samnāni (d. 736) (Chalabī n.d., p. 2:1782). In his bibliographical index, T .ā shkubrīzādah classifies 'ilm maqālāt al-firaq under the category of rational sciences and as a branch of 'ilm al-ilāhī (theology). He defines it in the following words: "The branch of knowledge which studies and classifies various false madhāhib (sects/religions)" (T .ā shkubrīzādah 1985, p. 1:298). T .ā shkubrīzādah relates this genre to heresiography and construes it with an explicit apologetic tone that is the exposition of false religions and sects. One of the earliest works from this genre is Kitāb al-Maqālāt wa al-Firaq by Sa'd ibn 'Abd Allāh al-Qummī (d. 914) (Al-Qummī 1963 A book about this genre from the beginning of the twenty-first century concludes that there had been two branches of maqālāt al-firaq, one dealing with the Muslim sects and the other relating to different religions. Thus, Abū al-H . asan al-'Ash'arī (d. 935) wrote two books which fall under the genre of maqālāt al-firaq: the first with the title Maqālāt al-Islāmiyyīn, which means propositions of the Muslims, and the second titled Maqālāt Ghayr al-Islāmiyyīn, which means propositions of non-Muslims. In the modern period, an established norm among Muslim scholars is to name the knowledge about Muslim sects as 'ilm al-firaq, the science of sects, and the branch that deals with other religions as 'ilm al-adyān, that is, the science of religions (Al-Tamīmī 2002, pp. 7-8). It is important to note that al-maqālāt was also a broad mode of presentation, a form of writing which encompassed a wide variety of topics, some of whom had nothing to do with religious sects or religions. Consider, for instance, the following titles: Maqālah fī al-Bāh (An Essay on Semen), Maqālāh fī al-H . isāb (An Essay on Mathematics), and Maqālāh fī al-Nawm wa al-Yaqz . ah (An Essay on Sleep and Awakening) (Chalabī n.d., p. 2:1783).
The essential genre of writings related to religions, however, had been al-milal wa alnih . al as several significant extant works from the Muslim cultural history bear this phrase in their titles. Mostly, the study of religions in Muslim history is referred to with reference to this genre. For instance, a recent book in the Urdu language titled Muslim Tradition of the Study of Religion (Mut .ā la'ah-e-Madhāhib kī Islāmī Riwāyat) maintains that during the Abbasids era, Muslims established a specialization called al-milal wa al-nih . al (religious denominations and schools) whose objective was to describe and analyse different religions and the Muslim sects (Qāsimī 2019, p. 13 It is also important to note that some crucial works on religions continued appearing without bearing the generic titles like al-radd, maqālāt al-firaq, and al-milal wa al-nih . al. Some of such books with peculiar titles can be classified under one or the other genre without difficulty, while others cannot be.  (Fānī 1843). The term al-milal wa al-nih . al is sometimes used in a more general sense to allude to all of such descriptive books, even though they bear unique names.
It seems that the writings about religious sects and religions started appearing in the earliest Islamic centuries. However, the genres of maqālāt al-firaq and al-milal wa al-nih . al crystalized a bit later. It is interesting to note that al-Fihrist, which is the earliest bibliographical index, does not mention al-milal wa al-nih . al and maqālāt al-firaq as literary genres or knowledge branches. Furthermore, the index includes no book with the title of al-milal wa al-nih . al. The term maqālah occurs in it several times but not clearly in connection with religions. These facts indicate that both of these genres had not appeared until the fourth Islamic century when al-Fihrist was written. On the other hand, a much later bibliographical index by T .ā shkubrīzādah mentions al-Shahrastānī's book al-Milal wa al-Nih . al under the maqālāt al-firaq genre (Ibid., pp. 1:298-99), which means that for him, al-milal wa al-nih . al and maqālāt al-firaq were not two separate genres. These examples imply the fluidity of the genres about religions.
Genre is traditionally considered "a literary classification alluding to a specific style of work with its appropriate form and conventions" (Joy and Neumaier-Dargyay 1995, p. 4). Production and interpretation of a text within a generic system is mediated through its intertextual connections to similar previous texts (Bauman 1999, p. 84). Some scholars even argue that "it is impossible to produce texts which bear no any relationship whatsoever to established genres" (Chandler 1997, p. 6). However, Thomas O. Beebee has developed the notion of "generic instability", which implies that "a text's generic status is rarely what it seems to be, that is always already unstable" (Beebee 1994, p. 27). We have noticed above that not all the writings on religions by the classical Muslim writers appeared with an explicit connection to one or the other genre and that the genres of these writings also changed over time. Thus, we may agree with Patrice Claude Brodeur, who opines that Beebee's notion of generic instability helps understand the Muslim works on religious others (Brodeur 1999, p.17) In a concise summary, the above survey of traditional bibliographical indexes reveals the following: (a) From the earliest Islamic centuries onwards, both descriptive and polemical writings about religions existed concurrently; (b) the descriptive genres like maqālāt al-firaq and al-milal wa al-nih . al became popular after the fourth Islamic century; (c) quite often, the writings on other religions occurred together with descriptions of Muslim sects, an observation that is especially relevant to the genre of maqālāt al-firaq; and (d) there has been a generic fluidity concerning books on religions.

Distinguishing the Study of Religions from the Polemics
As we need to make sense of the genres and writings from the distant past in a crosscultural context, the distinction of etic and emic standpoints propounded by the American linguist and anthropologist Kenneth L. Pike (1912Pike ( -2000 can be a helpful heuristic framework. According to this distinction, the emic standpoint relates to insiders' understanding of a cultural or linguistic system, while the etic stance refers to outsiders. Knowledge of a new cultural system initially relies on the categories and conceptual constructs that are alien to it. This approach is called the etic standpoint. Gradually, the unfamiliar culture unfolds its concepts, categories, nuances, and systemic dynamics, just like it appears to the insiders. Being able to see a cultural system like insiders do is called an emic standpoint. The etic perspective is necessary to begin the understanding of a new cultural system, but without moving towards the emic perspective, the knowledge remains superficial (Pike 1967;Headland et al. 1990;Ekstrand and Ekstrand 1986). A system's emic categories can function as etic categories for another one, and vice versa. Thus, the dialectic of etic and emic standpoints can help relate the premodern Muslim scholarship on religions to modern religious studies.
Relying on the work of some previous academics, we can initially classify the abovestated emic genres into two broader etic groups: descriptive and disputative. Al-rudūd and al-munāz . arāt fall in the first group, in form as well as in content. The maqālāt al-firaq genre often bears a lesser polemical tone and does not address a particular disputant by definition. It deals mainly with the Muslim sects and their subdivisions, but occasionally also extends to non-Muslim dominations. Finally, works appearing under the genre of al-milal wa al-nih . al pertain to a more systematic and descriptive textual tradition dealing with religions.
Consequently, two categories of Muslim works on religions become apparent: one descriptive and informative and the other normative and argumentative. Such distinction is already recognized by Zafar Ishaq Ansari (1932Ansari ( -2016 in the following words: Quite often the major underlying motive to know about other religions was to demonstrate that they were flawed and that Islam was superior to all of them. The above, however, was not the only model produced by Muslim scholars. For also we find another model for studying religions other than one's own, a model wherein the impulse to refute other religions is hardly evident (Ansari 2014, pp. v-vi).
He is also of the view that even in the polemic writings, Muslim scholars used to collect and state the facts about other religions honestly (Ibid., p. v). However, this contention needs to be qualified, as al-Bīrūnī has noted a tendency among some scholars of his age to attribute views to the followers of other religions that they did not hold (Al-Bīrūnī 1910, p. 5).
In line with the above classification, the genres of al-milal wa al-nih . al and maqālāt al-firaq, and some other books with unique titles constitute a cluster of descriptive writings on religions. In contrast, the genres of al-rudūd, 'ilm al-jadal, 'ilm al-munāz . arāt, and a large number of similar works with peculiar titles exhibit explicit polemical and disputative spirit. Nonetheless, exceptions to this general classification exist. Theological presuppositions, even judgments, can be located in the former class of works, while methodological novelties and valuable descriptions of facts can be found in the latter type.
The argumentative literature can be further divided into those works which simply restate the familiar religious disputes and those which bring to light new sources or introduce novel methodologies in the field. Perhaps the best example of the latter category is the work of 'Alī ibn Ah . mad ibn H . azm (994-1064), who devised the method of textual criticism quite similar to the one which Baruch Spinoza (1632-1677) and the following generations of Biblical scholars would employ later on (Al-Sharqāwī 1993). Though Spinoza is considered the father of modern biblical criticism, scholars have established that most of his points and even language drew heavily on Ibn H . azm's book al-Fas . l (Freedman 1989, p. 38). Thus, Hava Lazarus-Yafeh concludes that Muslims "developed a kind of Bible criticism very close in nature and detail both to earlier pre-Islamic Bible criticism and to the beginnings of later scholarly European Bible criticism" (Lazarus-Yafeh 1992, p. 130).
The comparative method devised by Abū al-H . asan al-'Āmirī (d. 992) counts for another example of the methodological novelties. Being a follower of the Ya'qūb ibn Ish .ā q al-Kindī's (800-873) school of philosophy, al-'Āmirī "argues for a rational investigation of religious beliefs and praxis, and based on his claim that the ultimate purpose of knowledge is virtuous action, attempts in a programmatic comparison of Islam with other religions to show how Islam is more successful than its rivals at achieving this goal" (Rowson 1996, p. 217). The quotation shows how al-'Āmirī conceived a unique comparative method that relied on human reason. Though his approach to the comparative method is normative, his deliberations on shortcomings of "comparison of incomparable" are amazing (Al-'Āmirī 1988, p. 125;Maishanu 1999, pp. 59-65).
Thus, instead of a simple disputative and descriptive division, the following tripartite typology of the classical Muslim works on religions is proposed: (a) polemics centered on typical theological disputes, (b) works written from a confessional standpoint but characterized with methodological novelties or bringing to light crucial historical data on religions, and (c) descriptive studies on religions with no distinct apologetic objectives.
Khalīfah has already proposed a threefold typology, but his order of arrangement is diachronic, according to three suggested developmental stages. Thus, according to him, the first and the earliest stage of the scholarly tradition under discussion pertained to polemics and works loaded with theological presuppositions. Notwithstanding this shortcoming, scholars at this stage produced a substantial number of books on the origin, history, and development of religions. This stage was followed by "the period of the philosophical interpretation of religion", which implies an objective and non-apologetic interest in different religions. The third stage is represented by works that treated the phenomenon of religion(s) as a sui generis category instead of a branch of theological and philosophical speculations (Khalīfah 1976, pp. 18-22).
However, it seems hard to substantiate a linear development from polemics through non-apologetic and objective works to the phenomenological study of religion in the Muslim scholarly tradition. For instance, one of the earliest books on religions is Kitāb al-As . nām (Ibn al-Kalbī 1924), which is by and large a descriptive iconography of pre-Islamic Arabia, and it predates the major polemical works which appeared in the Muslim cultural history. Though this book is about a religion that was not practiced anymore, it still points to an early interest in the phenomenon of religion. Al-Bīrūnī's work on Indian religions and culture is rightly considered as the most objective by any classical Muslim scholar, and his magnum opus preceded the confessional works like Al-Radd al-Jamīl li Ilāhiyyat al-Masīh . bi S . arīh al-Injīl (A Beautiful Refutation of the Divinity of the Messiah by Clear Implication of the Gospel) (Al-Ghazālī 1986). The instances of such anomalies are numerous. If we take into consideration the case of South Asian Islamic culture, too, the non-polemical works like Mughal Prince Dārā Shikoh's (1615-1659) Majma' al-Bah . rain (Shikoh 1929) appeared before polemical Iz . hār al-H . aqq (Manifestation of the Truth) (Al-Kirānwī 1989). Therefore, it is maintained that the three strands of works on religions delineated above-disputative, somewhat apologetic but methodologically significant, and descriptive-continued appearing concurrently throughout the Muslim cultural history to our times. Now, the question is which of these writings and genres can be considered as the study of religion and which cannot be, according to the single point criterion adopted at the beginning of this section? To begin with, the polemical literature-regardless of its historical importance or ethical standing (Crewe 2004, p. 148;Al-Maydānī 1993)-cannot be convincingly considered as the study of religion. The fact of the matter is that this type of literature is not unique to Muslim civilization. There is little doubt that historically the work of Christian and Jewish polemicists preceded that of the Muslims. As for the Christian polemics against the Jews, even the names of their genres are comparable with al-munāz . arāt and al-rudūd. Consider, for instance, titles such as Contra Iudaeos (Against the Jews) and Dialogus Adversus Iudeos (Dialogue against the Jews) (Dhan 1991, p. 41). The same observation applies to the case of Jewish polemics against Christians (Lasker and Stroumsa 1996). Furthermore, a polemical encounter (munāz . arah), by definition, requires an imagined or actual opponent; without an opponent, it is inconceivable. Thus, no tradition can claim having produced something unique based on polemical scholarship. Jews, Christians, and Muslims all engaged in religious controversies in medieval times (Steinschneider [1877(Steinschneider [ ] 1966Ebied and Thomas 2005;Tolan 2013;Southern 1978).
The works that bear some apologetic undertone but introduce methodological innovations can have a place in the history of the study of religions to some extent. However, most academics would not accept the outright apologetic or disputative works in the modern-day university settings, perhaps even those who favor collaboration between theology and religious studies (Cady and Brown 2002, pp. 1-5;Neville 1993).
Then, we have the third type of works that pertain to descriptive genres of al-milal wa al-nih . al and maqālāt al-firaq and some other books with a similar disposition, which plausibly appear to be the pioneering efforts in the worldwide history of religious studies. Consider, for instance, the following piece of methodological reflection of al-Bīrūnī: "If such an author is not alive to the requirements of a strictly scientific method, he will procure some superficial information which will satisfy neither the adherent(s) of the doctrine in question nor those who really know it" (Al-Bīrūnī 1910, p. 1:4). The undertone here clearly demonstrates what is known as objectivity in today's academic world, a hallmark of the scientific enterprise.
Admittedly, it is not difficult to unearth the underlying religious or cultural biases even in the descriptive works like those of al-Bīrūnī and al-Shahrastānī. However, one may not forget here that beyond a certain level hardly any human science is value-free, not even contemporary religious studies. Being entirely objective is one thing, and trying to be objective another. The former is not one of the human possibilities; the latter is (Pye 1974, p. 116). Therefore, if a scholar seems trying to be objective and fair as far as possible, then some degree of cultural and theological baggage does not blemish the scholarship.
As regards the volume of writings, which can be considered as the study of religion according to our criteria, fortunately, some scholars have already ventured to distinguish various types of classical Muslim works on religions and/or count them. In this regard, Lazarus-Yafeh (1992) and Adang (1996) have focused on the medieval Muslim writings on Judaism. (Anawati 1969;Brodeur 1999, pp. 33-34) and Thomas (2008Thomas ( , 2018 have analyzed the Muslim works on Christianity. Most importantly, Guy Monnot has surveyed and counted the Muslim studies on non-Biblical religions up to 1882 and classified them into general heresiographies, specific refutations, descriptive books, and miscellaneous writings that appeared as a part of encyclopedias and bibliographies etcetera (Monnot 1986;Brodeur 1999, pp. 34-35). Based on such previous studies, Patrice C. Brodeur has presented an accumulative table of books on religions produced by the premodern Muslim civilization from the second to thirteenth Islamic centuries, which roughly overlap with the eight to nineteenth centuries CE. The table shows the number of various types of books that appeared across these centuries, which is given below as Table 1 after the correction of a mistake with the calculation. The number of "particular refutations" in different centuries totals 59, not 60. The total of all categories is 122 books, not 123 (Brodeur 1999, p. 128). The above numbers help put into perspective the volume of various types of writings on religions produced by the premodern Muslim civilization. In this table, "general heresiographies" refer to those polemical books which deal with three or more Muslim sects, and "particular refutations" are argumentative works against one or two non-Muslim religions. In comparison, "descriptive works" are historical books that describe one or more religions, which can also have some occasional apologetic undertone.
However, the classification used in Table 1 needs to be related to the threefold typology that we have propounded above. In our typology, general heresiographies and particular refutations are classified together as polemical or disputative works. Thus, according to our criterion, keeping aside the seven miscellaneous works, we have 88 disputative books. In contrast, 29 books are descriptive or not explicitly apologetic, and therefore can be viewed as the study of religions.
The proportion of the mainly descriptive books-around one-fourth-is considerable and cannot be explained away as isolated cases. Thus, a tradition of the study of religion did exist in the Muslim cultural history. At the same time, it is also evident that this scholarly activity flourished together with the polemics against different religions and the purported heresies in Islam.

The Question of Disciplinary Status in the Hierarchy of Sciences
We have noted above the ambiguity and fluidity of genres and writings about religions in Muslim cultural history, and that often works on religions other than Islam appeared as an extension of the discussion of various Muslim sects and as a part of the al-kalām (Muslim theology) discourses (Thomas 2008(Thomas , 2018. Therefore, the question arises about the disciplinary status of the study of religion in the knowledge tradition under discussion. To get at a plausible answer to this query, let us explore (a) the available traditional classifications of various branches of knowledge; (b) intertextual relations of different writings on religions; (c) the existence of the science-of-the-science or the principles of studying religions, as we have in the cases of established branches of knowledge such as tafsīr, hadīth, and fiqh; and (d) the institutional status through curricula and the teacherstudent genealogies.
Classification of knowledge into various branches called disciplines, fields, subjects, and sciences is crucial for the systematic production of authentic knowledge and its onward transfer to the coming generations, and sometimes to indicate the relative importance of some particular subject matters over the others. The terms like disciplines, fields, subjects, and sciences are sometimes used in a broad and interchangeable sense and with unique nuance for each of them at the others (Parker 2002, pp. 373-75). When used in the specific sense, a discipline implies theoretical refinement, methodological rigor, and welldefined boundaries of a discursive activity (Shumway 1991, p. 202), while subject and field are methodologically inclusive, mainly defined only by their particular subject matter and knowledge base and terminology. Fred D'Agostino maintains that well-defined disciplinary matrices are better analytically, but less-defined disciplines can help the growth of knowledge by allowing variance in approaches and thinking (D'Agostino 2012, pp. 347-48).
Coming back to the question of the disciplinary status of the study of religion in Muslim cultural history, Osman Bakar has studied the classification of various branches of knowledge in Islam relying on the works of three crucial classical Muslim scholars, namely, Abu Nas . r Muh . ammad ibn Muh . ammad al-Fārābī (872-950), Muh . ammad ibn Muh . ammad al-Ghazālī (1058/59-1111), and Qut . b al-Dīn Mah . mūd ibn Mas'ūd al-Shīrāzī (1236-1311. Though three scholars take into account the broader theological sciences of 'ilm al-Ilāhīyyāt and 'ilm al-kalām, none of them include genres like al-milal wa al-nih . al and maqālāt al-firaq or any specific branch of knowledge related to the study of religions in their mappings (Bakar 1998). We have already noted above the marginal treatment of al-milal wa al-nih . al and 'ilm maqālāt al-firaq genres in the pre-modern Muslim bibliographical indexes. Another noteworthy point is that these indexes never mention the most crucial descriptive work in the field by al-Bīrūnī. These historical facts point to the marginality of the study of religion in the knowledge tradition under discussion.
The next possible clue about the question under discussion can be the textual continuity and intertextuality of the works on religions. Scholars produce new texts out of the sea of already existing texts that surround them, and the relation a text bears with the other ones is called intertextuality (Bazerman 2004, pp. 83-84). Through intertextual connections, we create genres and develop discourses (Bauman 2004, pp. 3-4). Intertextuality helps us understand the texts in the right context and also points to textual continuity. Thus, the question arises about how often, if at all, the subsequent Muslim writers on religions cite or allude to their predecessors' works? Has there been a tradition of commentaries, summaries, and glosses associated with the seminal texts on religions, like many texts from the other branches of knowledge from the Muslim cultural history do have (Messick 1992, pp. 251-56).
Then, the existence of the associated science-of-the-science-the discussion of theoretical and methodological issues involved in the study of a subject matter-can be a significant clue to disciplinarity. In the extant Muslim literary heritage, beyond the descriptions of religions, the second-order reflections on the principles of studying religions are insignificant if not wholly untraceable. Although some major works on religions contain introductory remarks about the problems involved in the study and comparison of religions (Al-Bīrūnī 1910, pp. 1: 3-8), we are unable to trace any book-length text on the methodological issues in the field. From the emic standpoint, there is no known tradition of us .ū l al-milal wa al-nih . al (principles of the study of religions) the way we have scienceof-the-science associated with the established branches of knowledge like us .ū l al-tafsīr (principles of exegesis) (Ibn Taymiyyah 1988) attached to al-tafsīr, us .ū l al-h . adīth (principles of authentication and interpretation of the Prophetic traditions) (Al-'Asqalānī 1990) developed along with al-h . adīth and us .ū l al-fiqh (principles of jurisprudence) (Al-Shāshī 2008) along with al-fiqh (Jurisprudence).
However, the existence of the general 'ilm al-us .ū l (the science of the sciences), which includes historiography, hermeneutics, logic, and debates like the compatibility of reason and religion, assuage this lacuna to some extent (Al-Nashshār 1984;Ibn Rushd 1986;Ibn Taymiyyah 1979;Al-Shāfi'ī 1988, p. 19). These theoretical discussions can be considered as a sort of general second-order tradition of methodological reflections having bearings for disciplines such as al-kalām (theology), al-falsafah/al-h . ikmah (philosophy), al-tārīkh (history), and, last but not least, the study of religions.
Last, disciplines by definition refer to particular areas of studies entrenched in educational institutions (Post 2009, pp. 751-12). From the available historical materials, it appears that the study of religions has seldom been a part of the curricula of the premodern Muslim seats of learning. Even the contemporary madrasa system, which represents a centuries-old tradition, does not formally include the study of religions (Khālid 2002, pp. 365-71). Similarly, though Brue B. Lawrence maintains that scholars like al Bīrūnī and al-Shahrastānī left able students (Lawrence 1973, p. 72), the teacher-student genealogies that are the hallmark of the traditional Muslim scholarship (Makdisi 1981, pp. 153-86) seem relatively weak in the case of the study of religions. Generally speaking, the prominent scholars in the field under discussion do not appear to be aware of each other's works, nor do we know much about the continuity of their endeavors through their immediate students.
In light of the four criteria adopted in this section, it appears that despite a considerable number of books, the study of religions remained a relatively marginal discursive tradition in Muslim cultural history. It had not acquired a standing comparable to the main branches of knowledge like al-tafsīr (exegesis), al-hadith (prophetic traditions), al-fiqh (jurisprudence), al-sīrah (biography of the Holy Prophet), al-mant . iq (logic), al-nah . w (syntax), al-tārīkh (history), al-t . ib (medicine), and al-kalām (theology). The Muslim study of religions surfaced at the peripheries of al-kalām and was often viewed as a branch within it, which reduced its radar signature for the classifiers of various branches of knowledge.
As hinted at the beginning of this paper, some scholars opine that the emergence of the modern religious studies owes to certain developments in Christian theology. The premodern Muslim study of religions also had a connection to Muslim theology. It seems that some strands of al-kalām succeeded in acquiring a level of objectivity, which facilitated the production of the descriptive writings on religions.
Just like religious studies depend on the construct of "religion", classical Muslim scholars used categories like al-dīn (religion), al-milah (nation), and al-ummah (faith community) for the conceptualization of cultural and religious others. These terms had their roots in the Quran and the Prophetic traditions and Muslim theology (Aasi 1986;1999, pp. 1-15). However, various historical factors had also been at play in constructing these terms as universal categories applicable to different belief systems and faith communities. Some scholars believe that the concept of otherness in contrast to which the Muslim self-identity was constructed played a vital role.
For instance, Roxanne L. Euben, while commenting on the journeys of Ibn Battuta, criticizes his vision as narrow "borne of encounters with multiple "Others", where otherness is defined both against and through regional, racial, religious, ethnic, linguistic, and sexual differences somewhat closer to home" (Euben 2006, p. 88). Some other scholars see the rhetorical literature produced by the Muslims as ideological confrontations to consolidate a majority view against "the disruptive presence of the "alien" (El Cheikh 2015, p. 10). Similarly, by comparing the two important works produced in the Mughal and British rules in India, Morgenstein Fuerst asserts that religious categories and identities were constructed not only by "foreign imperial elites, but autochthonous elites as well" (Fuerst 2012, p. 47). Carl Ernst also opines that the translations of the Hindu texts by medieval Muslims were motivated by particular political interests rather than defined "by any internally generated sense of the coherence of Indian religious traditions" (Ernst 2003, p.174). Thus, some of the theoretical and methodological debates concerning modern religious studies-like the representation of the unrepresented, gender bias, and the role of power in knowledge creation-might as well be relevant to the classical Muslim study of religions.

Conclusions
Did the study of religions exist in premodern Muslim civilization? The answer to this question depends on which writings one precisely refers to out of many types because distinctions and nuances are critical here. Any imposed homogenization would be distorting. The literary tradition under discussion produced at least three distinct strands of works about religions: (a) polemics, (b) works that presented accurate data or introduced innovative methodologies but carried some degree of apologetic undertone, and (c) more or less descriptive studies. If one were to speak about all that Muslims wrote on the subject under question, including polemics, the appellations like "perceptions of other religions" or "writings on religious others"-as used by Waardenburg and Brodeur, respectively-seem to be appropriate. However, if one were to focus exclusively on the descriptive writings-around one-fourth of the total-then it would be plausible to hold that a bourgeoning scholarly tradition existed in the premodern Muslim civilization that resembles modern-day religious studies to a certain extent. Nonetheless, the development remained peripheral in the hierarchy of various sciences and branches of knowledge in the Muslim intellectual culture. It could not acquire the status of a regularly taught discipline in the traditional Islamic education systems.
Several Some present-day Muslim scholars attempt to develop further certain tenets of their forefathers' legacy, like textual criticism and historiography, and present an indigenous Muslim perspective in modern religious studies. It is a constructive move that can potentially enrich the field and pave the way for its cross-cultural reorientation, provided that the study of religion is not confused with polemics and apologetics. For, disputative scholarship undertaken from the standpoint of a particular confession might be a legitimate practice in its own right; however, in the modern university settings, most religious studies practitioners would detest embracing such activity as a part of their field.
The existence of descriptive studies of religions without a direct link to the cultural experience of modern Western societies implies that this field can take roots and flourish in multiple historical and cultural settings. This conclusion is pregnant with significant bearings for religious studies as a worldwide academic pursuit in our times.
Funding: This research received no external funding.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
No new data were created in this study. Data sharing is not applicable to this paper.

Conflicts of Interest:
The author declares no conflict of interest.