‘FOREIGN BOOKS’ IN ARABIC LITERATURE: DISCOURSES ON BOOKS, KNOWLEDGE AND ETHNICITY IN THE WRITINGS OF AL-ǦĀḤIẒ

Al-Ǧāḥiẓ is one of the ʿAbbāsid era’s most celebrated bibliophiles, and his praise of books and championing of ‘writerly culture’ in 3rd/9th-century Iraq are well documented. However, he also expressed distinctly negative appraisals of books that have hitherto received much less scholarly attention. This paper will examine the curiously paradoxical views of alǦāḥiẓ by considering his opinions on non-Arabic books in the context of scholarly debates in his contemporary Iraq. Al-Ǧāḥiẓ’s conception of such books intersected debates regarding (a) the suitability of books to transmit knowledge, (b) rivalries between Arabs and non-Arabs in early ʿAbbāsid Iraq, and (c) the merits of translating scholarly writings from pre-Islamic civilisations. Al-Ǧāḥiẓ’s opinions on these issues led him to develop a particular conception of the ‘perfect book’ whereby he could unreservedly praise his own writings and extol ʿAbbāsid literary culture, but at the same time subordinate foreign literary cultures to the non-literate pre-Islamic Arabians. Al-Ǧāḥiẓ’s theories reveal that 3rd/9th-century Iraq had not yet become entirely a ‘civilisation of the book’, and that conceptions of language, ethnicity and knowledge influenced the formation of Muslim bibliophilia.

In the wake of the Islamic conquests of the 1st/7th century which amalgamated the various pre-Islamic civilisations of the Near East and Mediterranean into one empire, and with the subsequent emergence of Arabic as the region's new lingua franca of cultural production, scholars in early ʿAbbāsid Iraq (mid 2nd/8th to 3rd/9th centuries) experienced an environment responding to significant social and cultural change. From the later 2nd/8th century, the intellectual status quo was also confronted by technological developments which made the production of relatively inexpensive paper possible and facilitated a hitherto unprecedented opportunity for commercial book publication. 1 Scholars who formerly had relied largely on the aural acquisition of knowledge via lectures now could study from a growing library of books and disseminate their research and ideas to a wider reading public. The unique advantages of books to store and transmit knowledge were apparent to 3rd/9th-century intellectuals who praised writing specifically for this ability to communicate across time and space in ways that the oral/aural teacher/student relationship was physically unable to match. 2 Akin to many such momentous epistemological and technological changes throughout history, the introduction of books and the integration of various ethnicities into the new social order of the 3rd/9th century were controversial and stimulated spirited debates which propelled Muslim civilisation along new trajectories. From the 4th/10th century, these eventually led to the bibliophilia and cultural unity for which medieval Muslim civilisation is famous. 3 But in the 2nd/8th and 3rd/9th centuries, scholars were in the midst of debate over whether a mute, inanimate book could sensibly replace a speaking, living lecturer as an authoritative source for knowledge, while the different ethnicities of the Muslim world vigorously jostled for status. 4 These discourses collided with a third intellectual challenge when Arabic-speaking scholars began to translate and reflect on the books of Greco-Roman, Sāsānid Persian, Indian and other pre-Islamic cultures.
The 'translation movement', which began in the early 2nd/8th century (perhaps even before), 5 made Arabic translations of pre-Islamic 2 See, for example, the praise of ḫaṭṭ (writing) in both al-Ǧāḥiẓ's R. al-Muʿallimīn (Rasāʾil,3: 27), and Ḥayawān (1: 49-52). 3 Regarding the cultural bibliophilia, S. Toorawa identifies the late 4th/10th century as a time when 'reliance on books would become pro forma' (Ibn Abī Ṭāhir,24). This was the period of Ibn al-Nadīm, whose lengthy Fihrist stands as a monument to the writerly culture of his generation. The cultural unity of the medieval Islamic world is well known, neatly encapsulated by a verse Badīʿ al-Zamān al-Hamaḏānī composed for his al-Maqāma al-ʿilmiyya (Maqāmāt,203): Alexandria is my home, Should I settle there; But in Syria I spend the night, and in Iraq, my day. 4 Extant evidence for the controversies surrounding these debates can be adduced from bans on selling certain books in 279 and 284 reported in al-Ṭabarī (Tārīḫ, 10: 27, 54) (also discussed by S. Toorawa Ibn Abī Ṭāhir,20); the heated debate over writing Islamic traditions (M. Cook 'The Opponents'); and the discordant pro-Arab/anti-Arab Šuʿubiyya-style polemics so commonly encountered in 3rd/9th century adab texts, including those of al-Ǧāḥiẓ. L. Behzadi's summary is apt: 'ongoing discourses of [al-Ǧāḥiẓ's] time must have been much more colourful than we usually assume' (Sprache und Verstehen,175). 5 Traditionally, the translation movement has been associated with al-Maʾmūn (r. 198/813-218/833), but more recently, scholars have identified the urge to absorb and translate foreign knowledge from an earlier date, during the caliphates of al- , and perhaps the late Umayyad period (R. Rashed, 'Greek into scholarly texts widespread by the early 3rd/9th century. 6 The translated texts differed starkly from the nascent Islamic sciences and Arabic philology, not only in terms of content, but also in form. Significantly, the translated texts were rooted in a manuscript tradition that contrasted with the Islamic and Arabic sciences where aurality was asserted as a key component of authority and authenticity. Furthermore, they were originally written in the distant past and in non-Arabic languages by peoples with neither geographical connection to Arabia, nor temporal proximity to Islamic history. Muslim scholars in the late 2nd/8th and 3rd/9th centuries thus simultaneously confronted issues of integrating diverse intellectual traditions of the Near East, theorising the acceptability of books as a means of transmitting knowledge, and accommodating multiple ethnicities into their social order. I suggest that these debates were interrelated and led 3rd/9th-century writers to adopt ambivalent positions which can be seen as steps on the way towards the more defined literate, bibliophilic and culturally inquisitive outlook of subsequent centuries. On the one hand, the budding bibliophilia of the 3rd/9th century would aid the favourable reception of both Arabic books and translated 'pre-Islamic' manuscripts, but the tensions inherent in the process of Arabicising a multi-ethnic society thrust non-Arabic writings into debates about how the polyglot heritage of the Muslim civilisation should be navigated. This paper will explore the way in which the Iraqi polymath, ʿAmr ibn Baḥr al-Ǧāḥiẓ (d. 255/869) conceptualised 'the book' in the context of his views on non-Arabic peoples and the production of knowledge. Al-Ǧāḥiẓ may not be a squarely 'typical' scholar of his time, if such a notion should indeed exist, 7 but his writings on these debates are extensive and demonstrate their interrelatedness. Arabic',G. Saliba,. The earlier beginnings of translation activity are also evidenced in the Islamic tradition: al-Fihrist of Ibn al-Nadīm (written 377/938) notes the role of al-Manṣūr and al-Rashīd (Fihrist,(304)(305)333) and even suggests that the Umayyads commenced part of this venture, citing Hišām's (r. 105/724-125/743) and al-Ḥaǧǧāǧ's arabising of public records and translating some ʿilm (knowledge) into Arabic (Fihrist,303). 6 'Foreign books' commonly encountered in 3rd/9th century Arabic literature are Greek mathematical, scientific and philosophical writings, Sāsānid Persian historical and legendary court literature and collected aphorisms of a devotional and philosophical nature from Sāsānid Persia and India. 7 S. Toorawa criticises the trend in modern scholarship to identify al-Ǧāḥiẓ with the ethos of the 3rd/9th century (Ibn Abī Ṭāhir, 124-7), however, I believe it may be difficult to ascribe to anyone the attributes of a 'standard' citizen of Al-Ǧāḥiẓ was a prolific writer 8 and avid reader, identified in both medieval and modern writings as one of Islam's most famous bibliophiles. 9 Some later biographers even reported that books caused his death, crushing him around his ninety-fifth year under a collapsed bookcase! 10 But in contrast to reports of his bibliophilia, al-Ǧāḥiẓ himself expressed ambivalent opinions on the utility of books. On the one hand, his well-known love of knowledge seems to have engendered his respect for books as vital carriers of knowledge and led him to adopt a markedly bibliocentric view of the world, whereby he appraised foreign peoples in correlation with their book production. This facet of al-Ǧāḥiẓ has been often cited in modern scholarship, 11 but little attention has been given to a paradoxically contrary trend in his writings where he expressed doubts about books as symbols of knowledge and societal achievement, and even disparaged books and 'foreign book-producing' peoples. Resolving the contradiction of al-Ǧāḥiẓ's ambivalent his time: how would we go about defining one of these for the 20th century? Al-Ǧāḥiẓ expressed views which were accepted by his patrons and contemporaries, and, as such, represents an entirely valid, while certainly not the only valid, point of reference in our understanding of this period. 8 According to Ch. Pellat, ('Nouvel essai', 119), he wrote 245 works. Many are better described as epistles, running some dozens of pages or less, though others are firmly 'books' in the 'modern' sense (G. Schoeler, 'Writing for a Reading Public', 52-3, 62-3): stand-alone texts with a set structure, and two in particular, al-Ḥayawān and al-Bayān wa-l-tabyīn have survived in multi-volume works. 9 Yāqūt relates the statement of Abū Hiffān: 'I have neither seen nor heard of anyone who loves books and the fields of knowledge more than al-Ǧāḥiẓ' (lam ara qaṭṭu wa-lā samiʿtu man aḥabba l-kutuba wa-l-ʿulūma akṯara mina l-Ǧāḥiẓ; Muʿjam,4: 474). See also 5: 104) and 130,208). Al-Ǧāḥiẓ's love of books is almost universally cited in modern writings about him, see Ch. Pellat 'al-Djāḥiẓ' on his wide readings; N. Anghelescu for his status as champion of literate culture (Langage et Culture, 63); and more generally S. Günther 'Praise to the Book!'. 10 This famous but apocryphal-sounding account of his death is reported in Ch. Pellat The Life and Works, 9. Neither al-Baġdādī's nor Yāqūt's biographies of al-Ǧāḥiẓ mention it, recording instead that al-Ǧāḥiẓ was in his ninety-sixth year around the time of his death and that his physical condition was extremely poor, suffering from semi-paralysis (fāliǧ) and gout (niqris), and thus not likely in a fit state to browse bookshelves in his last days Tārīḫ,12: 214,Yāqūt,Muʿjam,4: 492, bibliophilia must take into account the developing 'writerly culture' 12 and conceptions of 'foreign peoples' in his contemporary Iraq. His writings highlight how the hallmarks of the medieval Islamic civilisation -bibliophilia, knowledge-seeking and a cosmopolitan outlook -were closely interrelated, though not definitively conceptualised, during the first ʿAbbāsid century.
Interpreting al-Ǧāḥiẓ on ethnicity: a scholarly advisory Akin to other volatile substances, analysis of al-Ǧāḥiẓ must come with appropriate caveats. He has been described as sarcastic, witty, rambling, emotive, unsystematic and elusive, more lyrical than meticulous, 13 neatly summarised by one modern scholar: 'who is to say what [al-Ǧāḥiẓ's] true intentions are -perhaps not even [al-Ǧāḥiẓ] himself'. 14 Al-Ǧāḥiẓ makes strong arguments, though his opinions can appear to shift from one text to the next, leading readers to question whether he possessed strong opinions at all, or whether his ambivalence is a mixture of his own confusion and/or scholarly interest in debating multiple, and conflicting angles of an argument. As such, al-Ǧāḥiẓ's work does not lend itself to macrosynthesis of anecdotes from his vast oeuvre. Rather, each quotation demands microanalysis to infer its meaning in the context of the text in which it is contained. 15 Despite these difficulties, I concur with some modern opinions that al-Ǧāḥiẓ may be more systematic than traditional Western research has assumed. 16 Nonetheless, the complications of al-Ǧāḥiẓ are manifold and evident in his discussions of books and ethnic groups which are the focus of this paper. As for peoples of the world, in some cases we find al-Ǧāḥiẓ praising the merits of a people, while in a different text he sharply lampoons them. 17 Political agendas and patronly tastes may have coloured al-Ǧāḥiẓ's analysis in particular epistles; 18 and we ought to exercise further caution in interpreting his texts at face value, for, in addition to the sarcasm and tongue-in-cheek sometimes apparent in his style, al-Ǧāḥiẓ wrote in polemical genres which fostered exaggeration and embellishment at the expense of sober discourse.
It has been noted that al-Ǧāḥiẓ was singularly influenced by discourses emanating from the genre al-Maḥāsin wa-l-Masāwiʾ (good versus bad traits) which used dialectic as a means of analysis whereby everything is imagined to be relative and all ideas could be called into question. 19 Gériès has suggested how al-Ǧāḥiẓ used this methodology to good effect in his theological writings 20 and it appears that al-Ǧāḥiẓ may have engaged in a similar logic regarding ethnicities. He is known to have written several contradictory pieces about peoples of the world: consider for instance a (now lost) work in praise of the South Arabian Qaḥṭān tribal group 21 , and an antithetical text praising the specific merits of their rivals, the North Arabian ʿAdnān over Qaḥṭān. 22 Al-Ǧāḥiẓ's slaves) as 'disorganised rabble' (Bayān, 1: 137) and belittled their condition visà-vis other peoples (Bayān, 1: 384). In a separate, polemical essay entitled Faḫr al-sūdān ʿalā l-bīḍān, however, he argued their merits, describing the Zanj in entirely positive terms (Rasāʾil,.
18 Noted by J. Lassner regarding Manāqib al-Turk (Virtues of the Turks) which al-Ǧāḥiẓ dedicated to the powerful Turkic wazīr al-Fatḥ ibn Ḫāqān (Rasāʾil, 3: 163; J. Lassner, The Shaping, 119-120); and Pellat proposed that al-Ǧāḥiẓ's al-Radd ʿalā al-naṣārā (Rebuttal against Christians) was written for the Caliph al-Mutawakkil during a period of official anti-Christian sentiment (Le Milieu Baṣrien, 231). 19 I. Gériès, 'al-Maḥāsin wa-ʾl-Masāwī,' EI 2 , v, 1223-4. 20 Ibid. 21 In 3rd/9th-century Arabic writings, Qaḥṭān was identified as the legendary ancestor of all Southern Arabians (tribes who claimed a Yemeni origin). A rivalry between 'Northern' and 'Southern' Arabs fueled polemical treatises which tend towards a binary division of Arab tribes into these two groups. Qaḥṭān and ʿAdnān became virtual by-words for these two-halves of the Arab people. 22 Rasāʾil, 1: 225. See also his Faḫr al-sūdān which defends the merits of al-Sūdān (lit. 'the blacks' from sub-Saharan Africa), but also contains a condescending approach and reserved praise: for example he concedes that African slaves in Iraq lack intelligence, but ascribes this to their status as slaves and their origin from a part of Africa where people are 'devoid of beauty and sagacity (jamāl wa-ʿuqūl). Al-Ǧāḥiẓ 'promises' that in other parts of Africa, beautiful and more cultured peoples (jamāl wa-kamāl) could be found (Rasāʾil, ambivalence may thus represent his method for achieving a deeper understanding of his subject matter. Notwithstanding the twists and turns of al-Ǧāḥiẓ's views on ethnicity, Lassner made an important observation regarding al-Ǧāḥiẓ's epistle Manāqib al-Turk (Virtues of the Turks). He ventured that a search for compatibility between the disparate elements of ʿAbbāsid society lay at the root of al-Ǧāḥiẓ's writings on foreigners whereby al-Ǧāḥiẓ attempted to devise an integrating model in which non-Arabs, including al-Ǧāḥiẓ himself, could find a place in the social order. 23 The desire to formulate an integrating model certainly accords well with the trend in ʿAbbāsid civilisation towards constructing a less divided social order. But the issue is more complex: as Pellat noted, al-Ǧāḥiẓ considered himself 'very much a member of the Arab community' and a 'passionate defender of the Arab heritage'. 24 Hence, al-Ǧāḥiẓ's interest in merging various ethnicities into one social order, or perhaps a 'cultural order' unified by adab, conflicted with an Arabian particularism in his writings. As discussed above, issues of ethnicity, knowledge extracted from non-Arabic sources and the authoritativeness of books were being debated simultaneously in al-Ǧāḥiẓ's Iraq and al-Ǧāḥiẓ's conflicting leanings regarding different peoples of the world interact closely with his paradoxical opinion of books.

'Foreigners' and 'foreign books' in al-Ǧāḥiẓ's writings
In the shadow of the burgeoning translation movement in 3rd/9thcentury Iraq, the interplay of books and ethnicity in al-Ǧāḥiẓ's thought coalesce in his writings about 'foreigners'. His conceptions of ethnicity and the relationship between 'peoples' of the world (whom al-Ǧāḥiẓ generally labels umam) 25 are complex and worthy of deeper study, 26 but 1: 211-12). 23 J. Lassner,The Shaping, Ch. Pellat,The Life and Works,3;idem,387. 25 Al-Ǧlḥiẓ generally designates umma as the largest distinct grouping of a people. The basis upon which Al-Ǧsḥiẓ conceived ummas to be distinct is not always apparent, but his hierarchy of terminology which identifies umma as the largest group can be found in Manāqib al-Turk (Rasāʾil,3: 213). See also his usage of umma to distinguish the Kurds, Berbers, Africans, etc (Bayān, 1: 137); or to connote the constituent 'peoples' of the world generally, Bayān, 3: 12. 26 Al-Ǧāḥiẓ's opinions on ethnicity have been considered by modern scholars in varying degrees of detail. In addition to J. Lassner's study of Manāqib al-Turk, Pellat considers some of al-Ǧāḥiẓ's reactions to his multiethnic milieu (Le Milieu Baṣrien,(224)(225)(226)(227)(228)(229)(230)(231)(232)(233)(234) and 'foreign literature' in 'Djāhiẓ et for our purposes, a brief outline of al-Ǧāḥiẓ's conception of his community and the 'outside world' indicates how his worldview was in part formed through a 'bibliocentric lens'.
Common to any analysis of identity and foreignness, al-Ǧāḥiẓ's actual genealogy (which may have been black African (aswad) or at least not Arabian 27 ) is less of a concern than the community to which he expressed his belonging and upon which he based his conception of the 'outside world'. The answer to this is nuanced. Al-Ǧāḥiẓ was certainly a partisan of the Arabs and took up their defence against those whom he called šuʿūbīs (his contemporaries who lauded the past glories of non-Arabian pre-Islamic peoples and argued for their superiority over the Arabs). 28 But he usually refrained from identifying his own community as generically 'Arab'. At times, al-Ǧāḥiẓ divides the 'Arabs' temporally and geographically into pre-Islamic (ǧāhiliyyūn), Islamic (islāmiyyūn), desert-dwelling (badawiyyūn) and settled (ḥaḍariyyūn). 29 These distinctions separate al-Ǧāḥiẓ's generation of urban Arabic speakers in both time and space from the 'pure Arabians' (al-aʿrāb al-ḫullaṣ) 30 who inhabited desert spaces in pre-Islamic and early Islamic times. 31 Al-Ǧāḥiẓ usually identifies his contemporary community as 'us' (naḥnu), 32 'this nation' (hāḏihi al-umma) 33 or 'our religious community' (millatunā). 34 In debates with šuʿūbīs, al-Ǧāḥiẓ does assume the position la littérature comparé', although in the latter article Pellat does not discuss the parameters of 'foreignness', assuming that al-Ǧāḥiẓ treats the Persians, Greeks and Indians as foreign peoples. See also S. Enderwitz, 'Culture History and Religion' for al-Ǧāḥiẓ's view on foreigners and adab culture. 27 On his non-Arabian origins, see Š. Ḍayf (al-Fann wa-l-Maḏāhibuhu, 154) and for mention of his 'aswad' roots see al-Baġdādī (Tārīḫ Baġdād, 12: 209), Yāqūt (Muʿǧam,4: 473). On the other hand, ʿA. Arḥīla defends al-Ǧāḥiẓ's Arabian origins 29).
28 Bayān wa-l-tabyīn in particular refers to these partisans of pre-Islamic, non-Arabian peoples as al-šuʿūbiyya (see 1: 383; 3: 5, 29, 31, 89 Manẓūr,Lisān (11: 631) where it is defined as dīn wa-šarīʿa ('religion and religious law') and al-Ḫalīl, ʿAyn (8: 324) where it is related specifically to the communal religious of 'the Arab', 35 but his general reticence to label his community as simply 'Arabs' is evident and logical given the diverse ethnic backgrounds in 3rd/9th century Iraq. Thus, al-Ǧāḥiẓ's umma could perhaps be considered Arabicised without being Arabian, 36 maintaining a link to the 'aʿrāb' Arabians to the extent of its preservation of their 'correct' Arabic language. 37 The maintenance of 'proper Arabic' emerges in al-Ǧāḥiẓ's writing as the privilege of scholars, betraying an intellectual elitism, much remarked upon in modern literature. 38 In brief, this restricted his community to the educated 'reading public' 39 and he expressed little regard for the uneducated, 40 and even less for the group he labelled aʿārīb: inarticulate Arabic speakers of vile origin. 41 Al-Ǧāḥiẓ's 'us' accordingly connotes a narrow band of literate, educated, Arabic speakers who inhabited the urban centres of the Muslim world and share 'our religious community (milla), our religion (dīn), our language, our education/manners (adab), and our ethics (aḫlāq)'. 42 Ties of religion, language, education and ethics appear more determinative than strict genealogy.
In terms of the rest of humanity, al-Ǧāḥiẓ often presents a two-fold law of a group of people. 35 See, for example, his hypothetical dispute with the šuʿūbīs where he and the Arabs are addressed collectively with the second person plural pronoun, antum (Bayan,3: 14). 36 By the term 'Arabian' I intend the Arabic-speaking peoples who inhabit the area now identified as the Arabian Peninsula. They are to be distinguished from Arabic speakers of the urban centres of the ʿAbbāsid period. The urban Arabic speakers were also aware of this difference, commonly (although not exclusively) applying the term ʿarāb to connote the desert-dwelling Arabs. To use al-Jāhiẓ's terminology, I mean by 'Arabians', al-Jāḥiẓ's badwaiyyūn of the pre-Islamic and Islamic periods. The 'long-standing integration' of non-Arabians like al-Ǧāḥiẓ into an Arabised identity is discussed in Ch. Pellat, Le Milieu Baṣrien, which Pellat considers the cause for the non-Arabian al-Ǧāḥiẓ to side with Arabians in contemporary racial debates  He notes that the Umayyads preserved the praiseworthy traditions and qualities of the pre-Islamic Arabs, whilst al-Ǧāḥiẓ remarks that these traditions suffered a decline during the ʿAbbāsid period. As such, the 'Arabs' are relatively historically remote in much of al-Bayān wa-l-tabyīn and many of the more excellent aspects of their culture seem, in al-Ǧāḥiẓ's view, to have passed (Bayān, 3: 366-367). 44 Bayān, 1: 137, see also Bayān, 1: 384, Ḥayawān, 1: 53. The term 'Rūm generally designates contemporary Byzantines in Arabic texts, but can also refer to the Greco-Roman civilization. What we refer to today as the Ancient Greek civilization is usually identified as al-Yūnān. However, there is occasional overlap in Arabic writings of the 3rd/9th and 4th/10th centuries where the relationship between Rūm and Yūnān were variously interpreted, sometimes al-Yūnān were deemed as descended from al-Rūm. Al-Masʿūdī helpfully summarises the various opinions and notes how the later al-Rūm lived in the same lands as al-Yūnān and adopted their language and ways (madhhab) (Murūj 2: 664). Al-Jāḥiẓ cites al-Yūnān as a 'disappeared peoples' (umma bā'ida) (Bayān 1: 188), although in his epistle al-Radd `alā al-Naṣārā, both al-Rūm and al-Yūnān are cited, indicating a perceived continuity between these two peoples in his worldview. 45 Bayān, 1: 137, Hamaǧ derives from flies or gnats which cluster around sheep and donkeys (Ibn Manẓūr, Lisān, 2: 393). It is applied to people by analogy on account of the diminutive size of gnats and disorganisation of their flight (ibid, al-Zamaḫšarī, Asās al-Balāgha, 706). Rabble or riffraff could act as translations. 46 They are described as al-umam allatī fīhā l-aḫlāq wa-l-ādāb wa-l-ḥukm wa-l-ʿilm (Bayān, 1: 384). 47 Bayān, 1: 88. 48 Bayān, 3: 13, Ḥayawān, 1: 53.
As regards the outside world, therefore, al-Ǧāḥiẓ adopted a distinctly bibliocentric lens by which 'foreign' book producing peoples were accepted to join the 'Arabs' in the global hierarchy, while those whom al-Ǧāḥiẓ believed lacked literary traditions were excluded. Precisely why 'the book' could be utilised as an arbiter between madhkūr (worthwhile') and hamaj ('worthless') peoples and the precise workings of this worldview in al-Ǧāḥiẓ's writings, can be understood in the context of the conceptions of books and knowledge in al-Ǧāḥiẓ's society.

The nexus of 'book-knowledge-civilisation' in Muslim thought
Al-Ǧāḥiẓ's 'intellectual elitist' worldview indicates a profound respect for knowledge (ʿilm) and a conceptualisation of books (kutub) as representing physical embodiments of ʿilm. 49 The three 'foreign' peoples, qua book producers, generated ʿilm and so earned the right to exist alongside the Arabs whose ʿilm al-Ǧāḥiẓ vigorously defended in his writings. This seems to harbinger a model of 'universal bibliophilia' encountered in later medieval Arabic writing where literary output and the worth of foreign peoples are unambiguously connected. 50 This bibliophilia of the medieval period has led scholars to label the Muslim world a 'civilisation of the book' 51 and the role of books in al-Ǧāḥiẓ's worldview seems to be an early expression of this 'global' aspect of Muslim bibliophilia. In seeking the origins of the Muslim partiality to books, Western scholars have traditionally considered that the prototypical respect of kitāb and ʿilm emanate directly from the Qurʾān. 52 The Qurʾān does contain a literate-intellectualised conception 49 Al-Ǧāḥiẓ's esteem for knowledge is famous (Ch. Pellat,Le Milieu Baṣrien,68), and al-Ǧāḥiẓ's dual conception of books-knowledge has been noted by ʿA. Arhīla, al-Kitāb (see particularly 16, 66, 142) and N. Anghelescu, Langage et Culture, 59-59. Al-Ǧāḥiẓ wrote specifically on the topic of knowledge, composing texts entitled Faḍl al-ʿilm, Risālat al-muʿallimīn, al-ʿĀlim wa-l-ǧāhil and three separate texts about maʿrifa. Ch. Pellat,'Nouvel essai',130,141, For an archetypal expression of this bibliocentrism, see Ṭabaqāt al-Umam of Ṣāʿid ibn Aḥmad al-Andalusī (d. 462/1070). See note [65] below. 51 G. N. Atiyeh, The Book in the Islamic World, xiv. J. Pedersen's The Arabic Book is the classic exploration of the paradigmatic 'bibliophilia' in Islam. See also J. Bloom Paper Before Print (116-23) for a description of the outpouring of 'book culture' and S. Günther's 'Praise to the Book!', 126 for the scholarly backing of this enterprise. 52 J. Pedersen opens his classic work with the phrase 'The Arabic book owes its origin to Islam' (The Arabic Book, 3) and identifies the Qurʾān as the first 'proper' Arab book (ibid., 12-16). As for ʿilm (knowledge), F. Rosenthal of human existence: it makes myriad citation of kitāb and ʿilm, it teaches that the kitāb will 'release the people from darkness into light' (Qurʾān 14:1), and it closely equates ʿilm with the ideal human condition, describing Muslims as those who have or seek ʿilm, in contrast to nonbelievers who act 'without it' (Qurʾān 31:20). 53 However, the inference that the Qurʾān is the basis of later expressions of Muslim intellectualised bibliophilia risks anachronism. 54 While the Qurʾān, al-Ǧāḥiẓ and later Muslim writers all seem united in the same knowledge-seeking bibliophilic chorus, current scholarship is revealing that the acceptance of books as authoritative depositories of ʿilm was a protracted process, the stages of which ought to be separated. First, in an insightful monograph, Madigan demonstrated that the Qurʾānic conception of the 'enlightening kitāb' was not a 'book' in the modern sense of a closed, definitive, authored text. Madigan argued that the Qurʾānic 'kitāb' is a symbol for God's authoritative knowledge, representing the totality of His guidance to mankind. It thereby transcends terrestrial, time-bound texts, 55 and, in fact, the Qurʾān states considered the history and importance of ʿilm throughout Muslim thought, proposing that prior to Islam the Arabians had a very limited appreciation for ʿilm (conceived as scholarly knowledge), and were left in a somewhat 'dark age' where knowledge was restricted to desert landmarks and practical matters of survival (Knowledge Triumphant,(9)(10)(11)(12)(13)(14)(15)(16). According to Rosenthal, the revelation of the Qur'ān, with its particular emphasis on ʿilm (ibid., 20) ushered in the advanced theoretical epistemology (ibid., 2). See also ʿA. Arḥīla for the commonly held view that wide-ranging knowledge seeking is an integral aspect of Islamic belief 19,67). 53 In the same vein, the Qurʾān chastises those who ignore ʿilm when it is taught/revealed to them (Qurʾān,2: 145;13: 37). 54 Promulgators of this conception, such as Rosenthal, do note the multifaceted meaning of ʿilm in Muslim thought. However, Rosenthal's analysis implies the Qurʾān has retained a determinative role in shaping ʿilm's parameters (Qurʾān,. This analysis primarily relies on texts from the later 3rd/9th century, leaving the first 250 years of Muslim intellectual history as a stasis in which the Quranic ideal seemingly was little changed. 55 For D. Madigan's elucidation on the meaning of kitāb in the Qurʾān see The Qurʾān's Self Image, 52-4; 70-2; 105; 145. According to Madigan, in the language of the Qurʾān, a printed copy of the text should not be referred to as kitāb, and for this reason, he proposes, the term muṣḥaf was adopted (ibid., 36-37). When the Qurʾān refers to written documents it eschews the verb kataba for physical writing (ibid., 108-9) and refers to physical 'books' with words like ṣuḥuf and qirṭās (ibid., 122-3). that no human-authored book can approach the power of the Kitāb. 56 While the Qurʾān does establish a discursive framework in which kitāb and ʿilm are connected in a tremendously positive manner and the possession of kitāb symbolises 'correct guided' life, it is unlikely that the earliest Muslim audiences associated this symbolic grandeur with terrestrial kutub (understood as human authored texts or anything 'written' 57 ). The elevation of terrestrial kutub to the centrepiece of later Muslim bibliophilia is a separate phenomenon achieved via the gradual maturation of the writerly culture.
Towards a 'writerly culture': the concept of the 'book' in 3rd/9th-century Iraq Muslim 'writerly culture', which began to emerge about one hundred years after the Qurʾān's revelation, would eventually champion the human-authored book and provide the necessary theoretical backdrop to use the book in appraising peoples of the world. 58 But in al-Ǧāḥiẓ's day, 'writerly culture' had not entirely matured and the status of books remained debated. Until the latter 2nd/8th century, scholars were primarily praised for their capacity to memorise. 59 In contrast, book 'publication' was extremely limited, 60 formal writing was restricted to 56 See the Qurʾān 2: 79 for its extreme censure of the human act of claiming their writings to be like God's kitāb. 57 A definition drawn from the earliest Arabic dictionary Kitāb al-ʿAyn, which identifies kitāb as the 'verbal noun' (maṣdar) of the verb kataba 'to write' 5: 341). 58 Madigan notes that the impetus to ascribe terrestrial written texts with kitāb was aided by the written codification of the Qurʾān in a written muṣḥaf during the 1st/7th century which began to elevate respect for the written word (The Qurʾān's Self Image,23,. 59 Anecdotes recording the lampooning of traditionists in the 2nd century who relied on written notes and praising those who allegedly knew all their material from memory are frequently cited in debates about the authenticity of the ḥadīṯ, the permissibility of writing them and the development of a written ḥadīṯ tradition. Conversely, Schoeler stresses the important role of notebooks (hypomnemata) from early times (The Oral and the Written, 114-128). Irrespective of the private use of such notes, the public display of knowledge from memory was important, witnessed by the scale of anecdote in the Islamic tradition.
60 Both Nabia Abbot (Studies in Arabic Literary Papyri) and Fuat Sezgin (Geschichte des arabischen Schrifttums, 1) gathered evidence for scholarly writings in the latter first and second centuries of Islam. Subsequent research has cast doubt on these attempts to identify primordially early texts in the Islamic tradition (G. Schoeler, The Oral and the Written, 40). bureaucratic matters and scholarly writing was limited to informal notebooks for personal use or shared between students. These writings, sometimes identified in the sources as kutub, should not be construed as 'books' in the sense of formally published closed-ended texts. 61 Knowledge transmission was likely aided by written notes, and scholars did not only rely on their powers of memory. However, presumptions of 20th-century scholars in the tradition of Goldziher who sought to prove that the transmission of knowledge relied on writing have ignored the staunch opposition to writing as detailed by Cook and they lack evidence given the limited numbers of surviving papyri and other writing fragments from the period. 62 The word of the scholar possessed greater value than his writings, 63 and the authority of human written texts seems to have been somewhat mistrusted across the Islamic world, and particularly in al-Ǧāḥiẓ's hometown of al-Baṣra. 64 In this environment, the first translations of the 'foreign' texts from the Sāsānid Persian, Greek, and Indian traditions would have circulated primarily in the palaces and administrative centres and not ventured far into the circles of the wider Muslim scholarly community. 65 Overall, recourse to written notes was largely outside of public view, 66 books lacked authority as standalone repositories of knowledge and consequently there was therefore almost no scope in the first two centuries of Islam to accept either the notebook kutub or translations of non-Arabic texts as epitomes of authoritative knowledge transmission.
As noted above, the introduction of paper and perhaps a greater familiarity with the textual traditions of pre-Islamic Near Eastern 61 This theory was first proposed in the 19th century by Alois Sprenger and has been carefully developed in The Oral and the Written and The Genesis of Literature by Schoeler who identifies these writings as 'notebooks' / hypomnema. For a summary of the difference between hypomnema and syngramma (the published book) see G. Schoeler cultures via the growing translation movement nudged the writerly culture forwards in the latter 2nd/8th century. 67 At this time Sībawayh (d. 180/796) 'published' al-Kitāb: one of the very first Arabic books definitively produced by its author and released to the public in written form. 68 Following Sībawayh's model, books began, slowly and rather falteringly at first, to be 'published' to a growing reading public, 69 and during the lifetime of al-Ǧāḥiẓ, the book was beginning to establish itself as a definitive repository of knowledge that could be read on its own. 70 With the human-authored text finally familiar and widespread in society, Muslim writerly culture could begin to conceptualise the humanauthored book as synonymous with ʿilm and right-guided living in their vein of the Qurʾānic kitāb. 71 As tangible objects, they became closely associated with their authors 72 and as abstract symbols of knowledge, 67 The role of translated materials is discussed by ʿA. Arḥīla, al-Kitāb 66-7 and the effects of increased paper production by J. Bloom,Paper Before Print,91,. Other factors cited for the growing importance of books include the rise of administrative writings (S. Toorawa, Ibn Abī Ṭāhir, 1-2, 9), a gradual development from increasing reliance on the scholarly notebooks (G. Schoeler, The Genesis of Literature), the role of the Muʿtazilite sect's preference of reasoned thought over memorised facts (S. Günther, 'Praise to the Book!', 131) and a broadening of the literate public who sought books outside of a formally professional context (S. Toorawa, Ibn Abī Ṭāhir, 1). It would seem logical that all of the above played a part. 68 G. Schoeler, The Genesis of Literature, 88-9 who identifies this book as a syngramma -akin to the modern concept of an authored, published 'book'. 69  Abbott compiles a list of the surviving 'books' produced in the 3rd/9th century, indicating the growing frequency of formal publication ('A Ninth-Century Fragment', 147-9). 70 Noted by S. Günther, 'Praise to the Book!' 139-140, who cites the earlier article of A. Ghersetti for evidence of the third-century opinion of the book as primary means of preserving knowledge ('L'utilità della scrittura e la lode del libro' [The Usefulness of Writing and Praise of the Book]). 71 Al-Ǧāḥiẓ explicitly makes this connection, to be considered below. 72 The close association of authors and their books begins to appear in the 3rd/9th century and is even more apparent the succeeding period. The early identification of books with their authors can be noted with Mālik ibn Anas and al-Muwaṭṭāʾ, al-Buḫārī and Muslim with their ḥadīṯ collections, and in the 4th/10th century we note Abū al-Faraǧ al-Iṣfahānī became popularly identified as ṣāḥib al-Aġānī indicating the association between author and his most famous work. In a similar vein, nations would similarly become identified by their they became a readily deployable means to recognise cultured life. The possession of books therefore was directly linked to praiseworthy social status, opening the door for the application of a bibliophilic model to appraise other societies and past civilisations.
The writerly culture's maturation in the 3rd/9th century accords well with al-Ǧāḥiẓ's bibliophilic worldview, and the contrast with the seemingly retrenched orality of the earlier period has understandably led modern researchers to identify the 3rd/9th century as literate, 73 and writers such as al-Ǧāḥiẓ and Ibn Qutayba as veritable champions of the writerly culture. 74 The enthusiastic appraisal of this period's literacy, however, ought to be tempered: writerly culture and the lofty status of kitāb were in a formative stage during al-Ǧāḥiẓ's lifetime. Authors still relied on aural sources, even into the 4th/10th century, 75 and the degree of autodidactism (from books), anecdotally noted in the 3rd/9th century does not appear to have entirely superseded aural study. 76 Genres such books, and expressions such as qālat al-Rūm ('the Rūm say') or fī kutub al-Rūm ('in the books of the Rūm') are noted in texts of Ibn Qutayba (G. Lecomte,Ibn Qutayba,190). The notion of authors acquiring a proprietary right in their book is a vast and under-explored ramification of the development of the writerly culture, however, its origins in the 3rd/9th century along with the development of the critique of plagiarism are introduced in S. Toorawa,Ibn Abī Ṭāhir,. 73 76 The emergence of the autodidactic basis for self-study from books is noted by S. Toorawa,Ibn Abī Ṭāhir,[15][16]. While this clearly formed the basis for much of the intellectual activity of the period, in the 3rd/9th century, Schoeler's evidence suggests that the autodidactic process and wiǧāda ('finding' information in books) still lacked the authority of learning from formal lessons as philosophy and medicine may have been less fettered by oral legacies, but this should not distract us from conceiving the 3rd/9th century as one of transition. 77 Published texts were certainly widespread, but the concept of the human-authored book as an authority for ʿilm was novel and remained an open question. This seems to have influenced al-Ǧāḥiẓ, and closer analysis of his contrary opinions on books and, consequently, foreigners reveals a more complex discourse.

Al-Ǧāḥiẓ: an ambivalent bibliophile
Al-Ǧāḥiẓ the 'book-praiser' is most evident in al-Ḥayawān and a shorter epistle on teachers (Risālat al-Muʿallimīn) where, in an elaborate analysis of the literate traditions of past civilisations, he marshals 'the book' in a markedly bibliocentric manner. Al-Ǧāḥiẓ as 'book-censurer', on the other hand, emerges in al-Bayān wa-l-tabyīn and other texts where he expresses different intentions which impacted upon his esteem for the bibliophilic literate epistemology and the relative merits of book producing peoples.

Al-Ǧāḥiẓ: lover of books and bnowledge
The bibliophilic tenor of al-Ḥayawān and Risālat al-Muʿallimīn is well known. 78 al-Ǧāḥiẓ extolled human authored books for their utility, durability and dependability which make them an easy reference, a more efficient store of information than memory (seemingly a direct critique of the aural Islamic tradition), 79 and the most robust method to preserve information against the ravages of time. 80 In short, he writes: Were it not for the book, the stories of the past would become corrupted and the sayings of those absent would be cut off. Your tongue [can only or maǧālis (sessions) with other scholars (G. Schoeler, The Genesis of Literature,. 77 For the more 'writerly' context of the 'foreign sciences' see S. Toorawa,Ibn Abī Ṭāhir,9. He cites Rosenthal's The Technique and Approach of Muslim Scholarship which stresses that 'all branches of literature relied for their preservation on written fixation' which, dating from 1947 seems to overstate the rapidity of the writerly culture's advance in the 3rd/9th century (S. Toorawa,Ibn Abī Ṭāhir,11). 78 In particular, S. Günther ('Praise to the Book!') and N. Anghelescu inform] those present, while the pen [can inform] the absent -those who came before you and those who will come after. Thus, the benefit of the pen is greater, and the public administration (dawāwīn) in greater need of it. 81 With his flamboyant description of the 'book' as 'a vessel, full of knowledge, a container stuffed with cleverness, and a receptacle of mirth and sagacity', 82 al-Ǧāḥiẓ portrays books as quintessential carriers of ʿilm, akin to the Qurʾānic kitāb, and in reporting that the Qurʾān and other books of revelation are the best kutub, 83 he implicitly groups all books, terrestrial and divine, in one conceptual category, the Qurʾān now being the 'best book', and not the 'only book'.
Having established the intellectual value of books, al-Ǧāḥiẓ describes their utility in developing a successful and right-guided society. He explains that 'spending on books indicates a respect for ʿilm, and a respect for ʿilm indicates the nobility of the soul and its integrity from the intoxication of faults.' 84 He explicitly lauds this 'bookish' ʿilm, casting it in opposition to jahl (ignorance/passion), as the 'pillar of the soul', the 'origin of all good things', 85 and the basis for social order: Taken together, al-Ǧāḥiẓ's statements demonstrate that books transcend mere depositories of information: they are symbols for the pursuit of ʿilm and prima facie evidence for the existence of culture and learning.
In al-Ḥayawān, al-Ǧāḥiẓ develops this bibliophilic reasoning into a world-historical vision whereby books become the sine qua non of humanity's intellectual development across time and space, which is the basis of his 'universal' bibliocentric outlook noted above. To prove it, he explains that humans, as created by God, are unable to live selfsufficiently and are dependent on one another. 87 From this principle, he argues that maintaining this necessary contact with neighbours is not always possible, hence the logical necessity of writing to communicate with those who are not immediately present. 88 Al-Ǧāḥiẓ asserts that groups of people (umam) similarly rely on the passage of knowledge from past societies to advance their own learning and avoid mistakes of the past. 89 Arguing that books are the only remaining tangible link with the past, 90 al-Ǧāḥiẓ concludes that books vitally maintain the venture of knowledge on earth. 91 The cycle is also continuous: writing is an 'intellectual duty' for the present in order to edify future generations and allow them to develop ʿilm into new horizons. 92 The elevation of books into vessels of ʿilm, and the portrayal of ʿilm as the unifying force underwriting the sweep of human history from its origins and into the future neatly generates a global bibliocentric worldview whereby Muslim civilisation can locate itself as a participant in the historical endeavour of knowledge where each civilisation is a link in a chain soldered by books. Here al-Ǧāḥiẓ presents one of the earliest formulations in Muslim writing of an intellectualised and bibliophilic worldview at its humanistic apogee: inclusive and cosmopolitan. 87 Ḥayawān, 1: 34-5. The Qurʾān frequently considers mankind's lack of self sufficiency, although this is adduced as evidence of their inferiority to God (e.g. 92: 8, 96: 6-7). Al-Ǧāḥiẓ echoes this principle, but has shifted the emphasis to the need for humans to look for mutual help from each other. 88

Al-Ǧāḥiẓ: respect for foreign book-writers
His intellectualised worldview weaves together the contemporary status of books, opinions on the merits of past peoples and the contemporary translation project into a discourse asserting the primacy of books in knowledge acquisition which enables him to use 'the book' as a means to appraise other peoples, automatically elevating the status of book producers and validating his own 3rd/9th-century adab culture which, via the translation project, was benefitting from past literary heritages. 93 Since many of the 3rd/9th-century intellectuals were not Arabian, it may seem unusual to refer to the intellectual heritage of Persians and Greeks as 'foreign', however, in light of al-Ǧāḥiẓ's intellectualised conception of world history, issues of ethnicity, as noted above, were less concerned with 'blood relation', and more with intellectual achievement. As such, al-Ǧāḥiẓ's discourse on ethnicities had a rearward looking aspect -the 'Persians' could be viewed as a past civilisation, and understood as contributors in the story of human knowledge production, assessable by the volume of their scholarly heritage. Societies that produced books earned a place for themselves in 'history', while those lacking literate traditions neither learned anything from those before them, nor could bequeath anything to posterity, and so had no place in al-Ǧāḥiẓ's view of history. 94 In al-Ḥayawān, al-Ǧāḥiẓ thus commends his book writing intellectual predecessors, expressing his gratitude that Indian astronomy was preserved in their 'scripts' (ḫuṭūṭ) 95 and professing a high opinion of Aristotle, Plato, Ptolemy and Democritus as pioneers of science and learning. 96 In light of the close association of books with their authors in the emerging writerly culture, this intellectual, bibliocentric lens rendered foreign societies synonymous with their books. The wise Greek-thinkers and just Persian kings who emerge in Arabic adab as veritable stereotypes of Greeks and Persians 93 Al-Ǧāḥiẓ relied on these sources himself: Ch. Pellat (The Life and Works, 4-5) notes al-Ǧāḥiẓ's exposure to Persian and Indian influence in al-Baṣra (where al-Ǧāḥiẓ was born) and increasing access to Greek influence in Baghdād (where he lived a long segment of his adult life). 94 The role of the book in establishing the merits of past literate cultures has been similarly considered by N. Anghelescu, Langage et Culture 55-6, 59. She has, however, neglected to consider the negative opinions which al-Ǧāḥiẓ expressed about books, and to attempt a synthesis of these paradoxical strands in his thought. Such an exploration and a possible synthesis will be offered below. 95  appear to have stepped off the pages of the translated Greek and Persian books and into the imaginations of ʿAbbāsid writers. It is perhaps not coincidental therefore, that medieval Arabic literature generally gives more detailed accounts of the history of Greek books than it does the Greco-Roman political history. 97 The 'inclusive' aspect of the writerly culture of al-Ǧāḥiẓ's day has been identified as the beginning of an increasingly 'secular' 98 (perhaps better labelled cosmopolitan) conception of knowledge in ʿAbbāsid circles as a widening audience consumed the knowledge of past peoples by reading their books. Based on the discourse in al-Ḥayawān, it seems straightforward to conclude that al-Ǧāḥiẓ squarely identified book production with worthy culture and that his bibliophilia transferred smoothly to xenophilia in the case of non-Arabic book-producing peoples. 99 But this discourse on the centrality of books and praise for book producing people also displays a more pointed self-serving element underlying his seemingly effusive bibliophilic cosmopolitanism.
Given that the Muslim civilisation of al-Ǧāḥiẓ's day had built its intellectual edifice upon the collective traditions of Arabians, Persians, Indians and the Rūm, it is entirely logical that al-Ǧāḥiẓ would commend these peoples on the basis of books. His discourse asserts that their books were 'worth reading' and so argues for the usefulness of their further translation and study in ʿAbbāsid Iraq. While the original authors were praised as a by-product of this argument, perhaps more importantly for al-Ǧāḥiẓ's purposes, his equation of reading with the concept of social progression over time enabled him to theorise that his culture, which both consumed books of the past and wrote new books, stood at the pinnacle of human progression: the legitimate and worthy successor to 97 While fourth-century Muslim world histories such as al-Ṭabarī, al-Masʿūdī or al-Maqdīsī give fairly sketchy accounts of Roman Emperors and Greek kingdoms, their contemporary Ibn al-Nadīm narrates in fine detail the transmission of Greek texts from their ancient origins to their Arabic translations (e.g. for medical texts see Fihrist, 345-6). Further analysis of this discrepancy would be interesting to explore, but it is beyond the scope of this paper. 98 S. Toorawa, Ibn Abī Ṭāhir, 129. Toorawa notes that the term 'secular' is of questionable application here, but intends by this the development of an adab culture in distinction to the scholarship specialised in the Islamic traditions. 99 Perhaps this accounts for al-Ǧāḥiẓ's mention in his epistle on the 'Virtues of the Blacks' that the term for the written text of the Qurʾān, muṣḥaf, derives from 'Ḥabāshī' origins which thereby attempts to delineate some literate element in African culture, and thus a point of merit (Rasāʾil,1: 202, its polyglot past. This has been identified as one of the centrepieces of al-Ǧāḥiẓ's thought and the theme of al-Ḥayawān. 100 Al-Ǧāḥiẓ, in the guise of a bibliophilic xenophile could thereby claim that the ʿAbbāsid scholars, via their translation of Persian, Greek and Indian sources had collated the entirety of humanity's knowledge and could legitimately consider themselves the most erudite nation yet. His outward cosmopolitan 'humanism' 101 may thus be more inward looking and selfserving, linking with al-Ǧāḥiẓ's discourses on ethnicity as an Arabian partisan in al-Bayān wa-l-tabyīn. Pellat observed this tendency and related it to the character of adab culture to 'prendre de tout un peu' and to borrow from foreign cultures without accepting their superiority to Arab culture. 102 This is an accurate observation, but these leanings of al-Ǧāḥiẓ are tied to a more multifaceted theory about books and knowledge in general. The anti-book, xenophobic trend in his writing now calls for examination.

Al-Ǧāḥiẓ the book cynic
Al-Ǧāḥiẓ's bibliophilic introduction to al-Ḥayawān was a polemical treatise. He draws our attention to this, noting that he wrote it in selfdefence against those whom he described as unjust critics of his writings. 103 Furthermore, the influence of al-Maḥāsin wa-l-Masāwiʾ genre has been noted as operative in al-Ḥayawān more generally, which cautions an uncritical acceptance of the content of al-Ḥayawān at face value. 104 His extravagant description of books as 'the most humble teacher; the most capable companion; the least boring and least grating friend…the most ready support', and his florid analogy of the book as a tree having the 'longest lifespan and sweetest fruit which is most easily 100 Al-Ǧāḥiẓ is typically associated with this perceived cosmopolitan trend in Muslim culture of the 3rd/9th century (Ch. Pellat,387;N. Anghelescu,Langage et culture,. Enderwitz proposed, based on analysis of al-Ḥayawān, that adab represents a fusion of the Arabian with the non-Arab cultural heritage and that al-Ǧāḥiẓ considered adab books as the highest forms of historical human expression, behind only the Qurʾān (S. Enderwitz,'Culture History and Religion',[235][236][237] picked' 105 are all specifically intended to rebut the critic of his books. 106 In short, by promoting the virtues and social utility of books in general, al-Ǧāḥiẓ could defend his reputation as a writer and argue that his books ought to be read in the most positive possible light. 107 Al-Ǧāḥiẓ certainly saw himself as a worthy participant in an intellectual heritage mediated by books, but he neither claims that all books are of equal merit nor that all writers are deserving of equal esteem. Even in al-Ḥayawān, al-Ǧāḥiẓ notes that other forms of communication are potentially as useful as books (depending on the circumstance), and that the pen and tongue are balanced in virtue, making no explicit statement of the written word's greater merit. 108 He expressly states that his only intention in writing al-Ḥayawān's introduction was to 'expound the virtues of books'. 109 The fuller exploration of communication and knowledge hinted in al-Ḥayawān is taken up in al-Bayān wa-l-tabyīn where his treatment of books is not quite so effusive.
Even before turning to his other writings, a reader of al-Ḥayawān can perceive various criticisms of books. Al-Ǧāḥiẓ censured the books of the zanādiqa (Zoroastrian Persians), lamenting their lack of ʿilm, poor style and dismissing their stories of heroes, demons and wondrous adventure (à la Šāhnāme) as 'idle, inept legends' lacking 'useful' knowledge, wisdom, witticism or anecdote. 110 In his opinion, expenditure on these books is wasteful: they are ǧāhil (ignorant), misguiding readers away from both self-improvement and religious enlightenment. 111 Al-105 Ḥayawān, 1: 33-4. 106 Ḥayawān, 1: 23: laʿalla raʾyaka inda ḏālika an yataḥawwala waqawluka yatabaddal (that perhaps your opinion [after reading this book] will transform and your [previous critique] will change).
Ḥayawān also invokes ǧahl to describe books written by Muslims which are censured as 'trashy' (kutub al-furrāġ al-ḫulaʿāʾ) or as 'diversions and banter ' (kutub al-malāhī wa-l-fukāhāt). Similarly, he relates criticism of books authored by those with bellicose agendas, shallow values or affected by the 'rancour of the ǧāhiliyya'. 112 Al-Ǧāḥiẓ narrates these opinions from the voice of a (hypothetical) critic, but it indicates what he conceived as the antithesis of his books. While we have seen that al-Ǧāḥiẓ equated some books with the Qurʾānic kitāb, i.e. as symbols of ʿilm, this was by no means a blanket endorsement for all books.
Outside of al-Ḥayawān, we find al-Ǧāḥiẓ denigrating the 'writerly culture', censuring those who read excessively and reproduce 'book language' as mere followers (tābiʿ), 113 especially criticising the kuttābthe state secretaries whom modern scholars consider to be among the first movers towards the writerly culture at the end of the 2nd/8th century. 114 These negative aspects of the writerly culture justify, for al-Ǧāḥiẓ, why God chose not to bestow skills in al-ḫaṭṭ (handwriting) on Muḥammad. 115 And so al-Ǧāḥiẓ paradoxically undermines al-Ḥayawān's framework for the transmission of knowledge across the sweep of human history by literate scholars.
Al-Ǧāḥiẓ hints at further suspicions regarding books in al-Ḥayawān in an insightful passage where he highlights the perilous journey of 'book knowledge' across time through the hands of copyists and translators. In particular, we read that the dual requirements for an ideal translator -(a) linguistically wholly proficient in the original and target language, and (b) intellectually on par with the authors whom he translates -can only exist in theory. 116 Consequently, al-Ǧāḥiẓ notes that translations even in the 'straightforward' fields of geometry and philosophy (al-handasa wa-l-falsafa) can be found lacking, while errors are almost guaranteed in religious sciences, where precision and knowledge are paramount. 117 In short, books emerge as twisted, corrupt and unreliable conveyors of past knowledge as they pass through the 'criminal hands' of copyists. 118 The above passage is enigmatic: it is not the direct speech of al-Ǧāḥiẓ, but again a 'quotation' from a hypothetical critic defending the Arabic oral poetic tradition. However, al-Ǧāḥiẓ does not refute any of these arguments and in fact adopts them himself elsewhere. 119 Thus, with his signature ambivalence, al-Ǧāḥiẓ leaves to the reader the task of resolving the question of whether books can or cannot accurately transmit ʿilm. Akin to the linguistic quandary that bedevils translation, al-Ǧāḥiẓ also cites the negative effect of writerly culture on language generally. He explains that book culture tends towards takalluf (unnatural mannerism) 120 and readers who fashion their speech after books develop stiff and artificial language. 121 Surprisingly for a bibliophile, al-Ǧāḥiẓ deems written language an inappropriate guide for good rhetoric! Al-Ǧāḥiẓ's well-reasoned bibliophilia and praise of books as symbolic embodiments of intellectual progress are therefore checked by his apparent belief that reading books neither guarantees accurate transmission of knowledge nor necessarily enables self-improvement, and that writerly culture can lack creativity and vitality. This sentiment closely mirrors a negative opinion of foreign civilisations and their books, which emerges from al-Bayān wa-l-tabyīn and other Ǧāḥiẓian epistles.
119 See his discussion of problems with translations from Persian and the Rūmī language and problems with Christian theology, discussed in the next section and note 150, below. 120 Bayān, 3: 29. 121 This even includes reading books authored by eloquent and/or wise writers (kutub al-bulaghāʾ…wa-dawāwīn al-hukamāʾ): Rasā'il, 3: 40-1 (Muʿallimīn). 122 The only article of which I am aware which specifically considers al-Ǧāḥiẓ's negative opinions of Persian, Indian and Rūmī books is Ch. Pellat's 'Djāhiẓ et la littérature comparée' where Pellat attributes these comments to al-Ǧāḥiẓ's method of grappling with the adab culture's integration of foreign ideas. This shall be further considered below.
heritage. He considered their books exhibit a laboured style, which he attributed to the Persian authors' lack of natural brilliance and linguistic spontaneity, as, in his view, they copied from each other. 123 Further, he cast doubt on the authenticity of the translated Persian books circulating in ʿAbbāsid Iraq, insinuating that they may have been, in part, forged by translators in the early ʿAbbāsid era. 124 In this argument, he adopts the reasoning of that hypothetical 'book critic' from al-Ḥayawān: mistrusting books capacity to faithfully convey ʿilm across time, and denying the 'Persian book' as a basis for praising the past Persian heritage.
Having discredited Persian books, al-Ǧāḥiẓ censures his contemporaries who based their knowledge upon them in preference to the Qurʾān and ḥadīṯ. 125 Repeating his critique of the book as ǧahl, al-Ǧāḥiẓ laments that the knowledge gained from Persian books is in fact ǧahl 126 and sharply rebukes Ibn al-Muqaffaʿ, the epitome of the 2nd/8thcentury Persian ʿAbbāsid scholar/translator who championed Sasanian books and culture. 127 Evoking the Qurʾān (62:5), al-Ǧāḥiẓ relates an anecdote comparing Ibn al-Muqaffaʿ to a donkey weighed down with books carrying much ʿilm, but not benefiting in the least: 'his knowledge made him weak, his reason baffled him, his wisdom blinded him and his insight confused him'. 128 Persian books in al-Ǧāḥiẓ's estimation are thus a far cry from authoritative, enlightening sources and there is little praise of Persian culture here.
Al-Ǧāḥiẓ's censure of the ancient Greeks again focuses on a criticism of their writing style. He claims that their knowledge of philosophy and logic did not translate into elegant expressions, and that despite their theoretical understanding of language, the Greeks (al-Yūnāniyyūn) did not produce well-formed speech in practice (pace Demosthenes et al, of whom al-Ǧāḥiẓ makes no mention). 129 Similarly, al-Ǧāḥiẓ claimed that 123 Bayān, 3: 28. 124 Bayān, 3: 29. 125 Examples he gives of these Persian books are the collections of anecdotes from the Sasanian king Ardašir, sayings of the vizier Bozorgmehr, Kalīla wa-Dimna and the religious books of Mazdak. See Rasāʾil, 2: 191-2 (Kuttāb).
126 Ibid.,[2][3][4][5]2: 192,195. Also, al-Ǧāḥiẓ lists Ibn al-Muqaffaʿ's Arabic language book al-Adab (which is commonly known today as al-Adab al-kabīr) among the Sasanian books of 'substandard' ʿilm. 128 Rasāʾil,2: 195 (Kuttāb). 129 Bayān, 3: 27-8. He specifically notes that Ṣāḥib al-manṭiq (a sobriquet for Aristotle, see G. Lecomte,Ibn Qutayba, was a poor speaker. Ancient Greek society as a whole, notwithstanding the intelligence of their scholars, failed to make any practical application of their theoretical knowledge and wisdom. 130 Knowledge has no civilisational benefit if unapplied, al-Ǧāḥiẓ argues, and accordingly, when comparing the Greeks to other peoples (umam) such as the Chinese, Persians, Turks and Bedouin Arabs (Aʿrāb), he concludes that the Greeks, for all their wisdom, do not deserve a higher status. 131 Contemporary Byzantine (Rūmī) civilisation faired worse. Since Ancient Greek (al-Yūnān) writers lived in the distant past, al-Ǧāḥiẓ rejects the possibility that their books and intellectual heritage could be claimed by contemporary generations of Byzantines. According to al-Ǧāḥiẓ, contemporary Rūmī literary production was negligible and their language was so different from Classical Greek that they could not possibly invoke its past glories for their benefit. 132 As for the Indians, al-Ǧāḥiẓ less frequently discusses them in his surviving writings, 133 but his extant appraisal of their culture similarly contains unenthusiastic evaluation of their books, which he describes as lacking both rhetorical power and creative spirit. They 'only contain ancient meanings, not attributable to one scholar…they are merely heritage since time immemorial, well known, well rehearsed'. 134

Towards a resolution
Al-Ǧāḥiẓ's paradoxical style, the relative obscurity of his life and uncertain chronology of his writings 135 complicate a reconciliation of his 'book loving' and 'book hating' tendencies. But patterns in his seemingly contradictory statements indicate a certain coherency within a complex web of issues. His opinions about books varied, but al-Ǧāḥiẓ was clearly preoccupied with books, indicating the maturation of the writerly culture in the 3rd/9th century. His analysis of foreign peoples also frequently cited books: whether al-Ǧāḥiẓ wished to praise or 130 Rasāʾil,[3][4][5][6][7][8][9][10][11][12][13][14][15]. 131 Ibid., 216-18. In al-Ǧāḥiẓ's opinion, here, these peoples excelled in a limited number of fields, but they failed to become all-round achievers. 132 Rasāʾil, 3: 314-15 (al-Radd ʿalā l-Naṣāra). 133 He apparently wrote books in which he considered their religious beliefs and idolatry which he cites in Ḥayawān (1: 8-9). These are now lost and al-Ǧāḥiẓ's conclusions are unknown. 134 Bayān, 3: 27: ammā l-Hindu fa-innamā lahum maʿānin mudawwanatun wa-kutubun muḫalladatun lā tuḍāfu ilā raǧulin maʿarūf... wa-innamā hiya kutubun mutawāriṯatun wa-ādābun ʿalā waǧhi l-dahri sāʾiratun maḏkura. 135 Ch. Pellat,The Life and Works,[10][11][12][13][14] denigrate groups, he marshalled 'the book', categorically denigrating those peoples whom he believed possessed no literary culture, 136 and while privileging book producers, he cited shortcomings of their books as a means to criticise them. His critiques and praises of both books and foreign peoples also revolve around questions of language. Closer consideration of al-Ǧāḥiẓ's conception of the praiseworthy book and the position of language in his worldview points towards a possible explanation of his paradoxical statements on literary culture. Such a profound appreciation for good style and appropriate language consequently had important ramifications in many of al-Ǧāḥiẓ's intellectual preoccupations, and his promotion of style and clarity of expression as a primary component of the 'worthy book' seems the product of his deep-rooted interest in bayān. Here the specific booklanguage nexus will be further explored to understand how al-Ǧāḥiẓ's opinions took their particular shape and to demonstrate how his seeming ambivalence regarding the book and foreign peoples is in fact part of a more coherent discourse.
Al-Ǧāḥiẓ applied his belief in the supreme importance of communication directly to his intellectualised vision of world history. Whereas in al-Ḥayawān he emphasised the role of books in the historical venture of knowledge, it seems incorrect to assume that he considered books as the sole embodiment of this process. As noted above, even in al-Ḥayawān, he criticised the shortcomings of some books, and in al-Bayān wa-l-tabyīn, he more explicitly describes why. While books convey 'knowledge', al-Ǧāḥiẓ stresses that appropriate and clear expression (bayān) is necessary to faithfully transmit any learning. Badly or imprecisely written books distort the knowledge they contain and risk misunderstanding. Such books categorically fail to transmit knowledge, and hence al-Ǧāḥiẓ concluded that it is bayān, clear expression, that brings ʿilm to life. 142 Al-Ǧāḥiẓ believed that all of the different forms of communication (which he enumerates as speaking, writing, computation, gesturing and metaphor) can convey ʿilm, and their relative suitability depends on the circumstances. 143 Sometimes ḫaṭṭ, the written word, is ideal, but al-Ǧāḥiẓ maintains that clarity of expression trumps the medium: the correct conveyance of ʿilm depends on the selection of the best method for the given circumstance. No single method suits all occasions: 'the revelation of meaning occurs to the extent of semantic clarity, accuracy of expression, appropriate epitomisation [of meaning], and precision'. 144 The transfer of knowledge, therefore, cannot logically be the exclusive preserve of literary culture.
For al-Ǧāḥiẓ, the perpetuation of intellectual culture via the communication of ʿilm through clear bayān begins with the Qurʾān since its excellent bayān is the means by which God teaches His ʿilm, and ideal bayān belongs to God. 145 But al-Ǧāḥiẓ cites a definition of humans as the 'living, clear speaking (mubīn)' 146 and so allows them to achieve degrees of eloquence too. 147 Al-Ǧāḥiẓ thereby contrasts bayān with ʿiyy (inhibited speech) just as ʿilm opposes ǧahl. 148 As we have seen, he appraised books by their language and meaning together, and so it would appear, he appraised humans.
Crucially, al-Ǧāḥiẓ ventures that the concordance of sound meaning and correct expression in human communication approaches the Divine: [t]he best speech is that which is brief and obviates the need for protracted expression and the meaning of which is apparent in its expression; Almighty God grants [such speech] some of His loftiness and bestows upon it the light of His wisdom, according to the good intentions and piety of the speaker. 149 143 Bayān, 1: 76-80 and Ḥayawān, 1: 29. In both texts, writing is generally lauded as one of the most efficient means of effecting good bayān, but it is not necessarily the most eloquent.
Books and human activity thereby become both intellectually and linguistically construed. The Qurʾān represents the ideal concordance of meaning and language, but the model permits terrestrial communication to be appraised to the extent it approaches the Quranic standard.
Why al-Ǧāḥiẓ emphasised the centrality of bayān in the venture of knowledge and effectively promoted it above books may be explainable in part by the status of writerly culture in his day. As noted above, notwithstanding the growing importance of books in 3rd/9th-century intellectual circles, the 'silent' book had not replaced aural methods of knowledge transmission, and a respect for oral/aural skills and style would linger in Muslim culture for centuries. 150 As such, emphasis on oral and other non-written forms of communication is natural, and an exclusive praise of books as stand-alone authorities of ʿilm would have seemed radical and perhaps nonsensical. Consequently, writing is a component of bayān, and indeed is privileged by the bibliophile al-Ǧāḥiẓ, but the written word is not paramount, as one may expect could be the case in a more thoroughly 'literate' intellectual milieu. The as-yet immature writerly culture and writerly styles of communication may thus have curbed al-Ǧāḥiẓ's bibliophilia.
In addition to these factors, issues of ethnicity also appear which point to another intertwined agenda at work in his thoughts on bayān and, as a consequence, books in general. At the centre of this lies al-Ǧāḥiẓ's acceptance of the Qurʾān's stylistic pre-eminence, which is uncontroversial. However, he elevates this into an ethnic discourse whereby his conception of bayān shifts from pure stylistic analysis into a means to create a hierarchy of world languages. Inasmuch as Arabic is the language of the Qurʾān, al-Ǧāḥiẓ asserted that Arabic qua language, must possess the best potential for bayān. He explains: There is no language on earth more enjoyable, elegant or sweet to hear, more intimately connected with clear rational thought or more expressive than that heard from the correct-speaking, sound-minded Arabians (ʾaʿrāb) and eloquent scholars. 151 150 For the persistence of the 'oral' aspect of learning into the late medieval period, notwithstanding the penetration of books in all fields of scholarly activity, see J. Pederson, The Arabic Book, 17, 31-6. 151 Bayān, 1: 145: innahu laysa fī l-arḍi kalāmun huwa amtaʿu wa-lā ānaqu wa-lā alaḏḏu fī l-asmāʿi wa-lā ašaddu ittiṣālan bi-l-ʿuqūli l-salīmati ... wa-lā aǧwadu taqwīman li-l-bayāni min ṭūli stimāʿi ḥadīṯi l-aʿrābi l-ʿuqalāʾi l-fuṣaḥāʾi wa-l-ʿulamāʾi l-bulaġāʾ.
Al-Ǧāḥiẓ's intellectual elitism is again at work here; his conception of bayān promotes the Arabians and his own 'Arabised' scholarly community above all others and disadvantages foreign cultures from the outset. Al-Ǧāḥiẓ directed this thought to its logical conclusion: applying bayān to a global worldview, he conceded that all peoples can express themselves, even if only by 'crude expression, poor meaning and 'brutish language', 152 but in his opinion, only Arabic speakers are truly proficient in bayān, while a group of 'foreigners' (here he means the Persians) also had some, though lesser, expertise. 153 The 'four noteworthy peoples' (madhkūrūn) of book producers, thus reduce to at most two (more likely one-and-a-half) when the secondary hurdle of bayān is erected before them.
The practical effects of this were far reaching. Speakers of imperfect languages, according to al-Ǧāḥiẓ's logic, necessarily possess imperfect knowledge: their expressions cannot accurately articulate what they intend to say. Hence al-Ǧāḥiẓ attributed 'errors' of Christian theology (i.e. where Christian dogma differed from Islamic) to Rūmī linguistic deficiencies which prevented the accurate conveyance of the teachings of Jesus. 154 In contrast, al-Ǧāḥiẓ believed Arabic speaking theologians were better protected from theological error by the clarity of the Arabic language. 155 The addition of the linguistic parameter in judging knowledge starkly handicaps non-Arabic speaking peoples and enables al-Ǧāḥiẓ to acknowledge their literary heritage while maintaining that their books lack the fundamental linguistic component of the kitāb muḥkam mutqan, which he believed only Arabic could truly produce, and so justifying his negative opinions of non-Arabic books 152 Bayān, 3: 12-13: ḥattā inna l-Zanǧa maʿa l-ġaṯārati... li-tuṭīla l-ḫuṭab... wa-in kānat maʿānīhā aǧfā wa-aġlaẓa wa-alfāẓuhā aḫṭala wa-aǧhala. 153 Al-Ǧāḥiẓ does not specify particular foreign groups in his, stating: 'with [bayān] the Arabs are duly proud and with bayān some of the ʿAǧam claim precedence' (wa-bi-ḏālika [al-bayān] tafāḫarat al-ʿArabu wa-tafāḍalat aṣnāfu l-ʿAǧam, Bayān, 1: 75). Later on he specifies that the Persians are the other group who maintained an acceptable standard of bayān, although this is expressly inferior to the Arabic (Bayān, 3: 27-9). 154 See Rasāʾil, 3: 334 (al-Radd ʿalā l-Naṣārā). 155 Ibid., 337. Al-Ǧāḥiẓ did concede, alluding probably to the kalām debate in his own society, that not all Arab scholars understand the most complex points of theology because even the bayān of Arabic, as written by humans, can fall short of conveying such complex ʿilm. This, however, was not, for al-Ǧāḥiẓ, a weakness of Arabic, but rather, a warning to scholars of the almost certain failure that will befall theological speculation in any other tongue.
Arabians and al-Ǧāḥiẓ's emphasis on Bayān While later medieval commentators attributed al-Ǧāḥiẓ's partisanship of the Arabic language to his defence of the Qurʾān, and while very similar arguments for the primacy of Arabic among world languages would appear in the iʿǧāz al-Qurʾān genre 156 , this is unlikely the only or indeed primary reason for al-Ǧāḥiẓ's stance. Al-Bayān wa-l-tabyīn does not particularly defend the Qurʾān -perhaps he felt it could handle such matters itself 157 -but instead, as noted by Pellat in 'Djāhiẓ et la littérature comparée' and Behzadi in Sprache und Verstehen, it marshals arguments for a cultural defence of the Arabs. Indeed, as Behzadi argues, al-Ǧāḥiẓ utilised bayān for deeper purposes than merely asserting Arabian superiority, but he certainly found bayān a very useful tool in šuʿūbī arguments, and, if we consider how his conceptions of bayān and the Arabians fit into the context of books, we can see how the pro-Arabian foundation of al-Ǧāḥiẓ's bayān is in fact paramount. The Kitāb al-ʿaṣā (The Book of the Stick), the most expressly pro-Arabian tract in al-Bayān wa-l-tabyīn, 'also contains some of al-Ǧāḥiẓ's most detailed statements on both books and bayān. Indeed, the precise way in which his conception of bayān interacts with writerly culture and the cultural defence of the Arabians seems a good indication that these debates were intimately related. ʿAbbāsid scholars believed that the Arabians lacked a 'book culture' in pre-Islamic times, 158 yet these 'bookless' Arabians 156 The study of the inimitability of the Qurʾān. See the defence of the Qurʾān by al-Bāqillānī (d. 403/1013) which utilises the same linguistic arguments as al-Ǧāḥiẓ and al-Bāqillānī similarly dismisses the merits of foreign languages and their books (Iʿǧāz al-Qurʾān, 29-32). The opinion of al-Ǧurǧānī regarding al-Ǧāḥiẓ and defence of the Qurʾān is discussed by L. Behzadi, He does expressly note the superiority of the Qurʾān's language which constitutes evidence of its Divinity (Bayān, 1: 383), but a purposeful linguistic defence of the Qurʾān is not expressed as a primary concern of al-Bayān wa-ltabyīn as it would become in the later iʿǧāz al-Qurʾān books of the 4th and 5th centuries. 158 Much has been made in Western scholarship of certain references to writing materials in pre-Islamic poetry and the odd treaty which the Islamic tradition informs us was 'written'. These oft repeated references concerning literacy in pre-Islamic times have overshadowed the lack of any admission, rightly or wrongly, by ʿAbbāsid scholars of the existence of significant literacy in pre-Islamic Arabia. See, for example, Ibn Qutayba's Kitāb Faḍl al-ʿArab wal-tanbīh ʿala ʿulūmihā (Virtue of the Arabs and an Explication on their Sciences) where he lists the major fields of pre-Islamic ʿilm. Writing is absent, the closest is a description of ḫaṭṭ defined as divination via making lines in the under the flag of Islam conquered ancient and learned civilisations. The incongruity of the ratio between Arabian military strength and their literacy would become problematic as the 'book' began to be culturally revered in the maturing writerly environment of the 3rd/9th century. When the parameters for esteem and power no longer rested on the force of arms alone, 159 the descendants of the 'literate' Persians, Greeks and Indians could marshal the new 'bookish' benchmarks to claim their heritage was superior to the Arabians whose lack of an ancient literary tradition became an obvious source of embarrassment.
By stressing the dual intellectual and linguistic foundation of the 'praiseworthy book', and by focusing on the importance of bayān in underwriting ʿilm, al-Ǧāḥiẓ articulated a firm argument for a nonbookish conception of ʿilm. He pursued this reasoning and defined the ideal bayān as epitomised in spontaneous good speech (badīha, irtiǧāl)the attributes of skilled orators; while chiding the studied affectation (takalluf) of written language. 160 He thus ventured that the best language actually has no need for books and so deftly parried any cultural disgrace attaching to the 'bookless' pre-Islamic Arabians: their lack of a literate tradition was transformed into a cultural strength. Al-Ǧāḥiẓ frequently cited the above conception of bayān to laud the Arabians and maintained that their eloquent speech was an inborn virtue, 161 a linguistic genius born from their desert environment, while he expressly denigrated the retrograde language of urban dwellers. 162 In sum, the non-bookproducing Arabians emerge at the summit of knowledgeable peoples for their lack of books was more than compensated by their excellent bayān. Al-Ǧāḥiẓ could cogently demonstrate that their ʿilm was an ʿilm of the most useful order, expressed, as it was, in what al-Ǧāḥiẓ deemed the most eloquent language. Simultaneously, the conquered non-Arabs and sand (Faḍl,143). Ḫaṭṭāṭ in pre-Islamic lore meant this sort of diviner, not the modern concept of a calligrapher (ibid.)! 159 Military power was, by the 3rd/9th century, considered the preserve of the Turks whom al-Ǧāḥiẓ noted for their martial abilities (cf. his Manāqib al-Turk), but this did not elevate them to the tier of 'noteworthy peoples'. 160 Bayān, 3: 28. 161 Ibid. 162 Al-Ǧāḥiẓ commonly alludes to the linguistic eloquence of the Bedouin over the city dweller (Bayān, 1: 13, 96-7), noting that 'city eloquence' is learned in any event and lacks the innate accuracy of the Bedouin (ibid., 145) and he specifically notes the corrupting influence of city language (ifsād), which he admits even affects his own language and the ʿAbbāsid scholarly community (ibid., 162-3). their book culture were structurally subordinated. Interestingly, al-Bayān wa-l-tabyīn is also silent on the translation project undertaken by the ʿAbbāsid caliphs, particularly al-Maʾmūn, a figure traditionally considered its prime architect. 163 Al-Ǧāḥiẓ dedicated a book to al-Maʾmūn which was apparently well received, 164 but al-Ǧāḥiẓ's writings are devoid of praise for the caliph's interest in foreign books, instead his praise for al-Maʾmūn, is on 'Arabic' grounds -concerning the caliph's eloquence. 165 Al-Ǧāḥiẓ's maintenance of this linguistic underpinning of ʿilm and support for the Arabians led him, somewhat unusually for an ʿAbbāsid writer, to express strong admiration for the Umayyads. When al-Ǧāḥiẓ enumerates Arab rulers who displayed proficiency in bayān, he emphasises the Umayyads (Muʿāwiya, Yazīd, al-Walīd and Sulaymān) as well their Arabian rival, Ibn Zubayr, whereas the ʿAbbāsid caliphs are expressly secondary. 166 Furthermore, despite the ʿAbbāsid interest in book learning, al-Ǧāḥiẓ even commends the Umayyads for maintaining the linguistic and cultural traditions of the pre-Islamic Arabians and laments that the ʿAbbāsids eschewed what al-Ǧāḥiẓ considered the more virtuous Arabian orality as they turned instead towards the urban writerly culture. 167 Can this be taken as an indirect slight on the ʿAbbāsids for too eagerly adopting the writerly culture and manners of the Persian state secretaries?

Conclusion
Al-Ǧāḥiẓ's disparate opinions of books, ethnicities and learning merge under the umbrella of his theories about the human intellectual heritage. As a member of the burgeoning writerly culture of Iraq and reader of 163 For example, Ibn al-Nadīm some 125 years after al-Ǧāḥiẓ's death recounts the portentous dream of al-Maʾmūn in which the caliph had a vision of Aristotle who urged him to find Greek books for translation (Fihrist,. 164 Al-Maʾmūn's praise for the quality of this book is reported by al-Ǧāḥiẓ (Bayān, 3: 374). 165 See Bayān, 1: 91, 115, 3: 374-8. Al-Maʾmūn's opinions about good books, containing excellent meaning and language together is also cited by al-Ǧāḥiẓ, but this concerns his interest primarily in Arabic books. See Rasāʾil, 1: 351 (al-Radd ʿalā l-Naṣarā).
166 Bayān, 1: 383. He expressly notes that the speech of Arab rulers other than the Umayyad era figures mentioned lacked what could be considered proper bayān, thereby placing the ʿAbbāsids in a secondary role. al-Maʾmūn is mentioned as an outstanding ʿAbbāsid, however no mention is made of his interest in translating foreign books. 167 Bayān, 3: 366.
translations from non-Arabic sources, he accepted books as valid transmitters of ʿilm and theorised a structure of knowledge transmission via books in which the translation project and the foundations of ʿAbbāsid adab culture could be legitimised. So, in al-Ḥayawān in particular, al-Ǧāḥiẓ appears as the bibliophilic humanist and icon of the cosmopolitan book consuming medieval Muslim civilisation. 168 But this bibliocentric worldview risked an implicit denigration of the bookless Arabians in the ethno-cultural debates of al-Ǧāḥiẓ's intellectual milieu.
To bolster the status of the Arabians in the story of human intellectual heritage, al-Ǧāḥiẓ played to their strengths, and found in his conception of bayān a cogent means to establish the parameters of ʿilm around good language which he argued was the preserve of the Arabians. In 3rd/9 th -century Iraq, books by no means dominated the process of knowledge acquisition, and communication retained an oral/aural aspect. This background played into al-Ǧāḥiẓ's hands, permitting him to refrain from unequivocally praising all written texts as authoritative sources of knowledge and to stress the importance of appropriately expressive language in the parameters of the 'praiseworthy book'. While he accepted that some books could be veritable paragons of cultured thought, not all books are equal, and he wielded this re-conceptualised 'ideal book' against the very cultures that based their own superiority on 'book culture'. Construing books in this fashion had another self-serving angle, as his theory implicitly assures that books written in Arabic (such as his own), must be the very best for all time.
Al-Ǧāḥiẓ's dilemma of lauding book culture as legitimator of adab, and disparaging books to defend the Arabian heritage continued to confront later writers. In the mature writerly culture and on account of their own esteem of books, some would adopt an unambiguously bibliocentric lens to appraise the world, scorning not the non-believing kāfir but instead illiterates, and they even laud the polytheistic Babylonians and Pharaonic Egyptians as praiseworthy peoples on account of their perceived book production. But other later scholars would retreat into Arabian particularism, particularly in defence of the Qurʾān. Later cosmopolitan bibliophilia would thus continue to be cleft by issues of language and theology. 169 168 See Enderwitz and Saliba for recent spirited arguments for a synthesising and cosmopolitan al-Ǧāḥiẓ. Ch. Pellat (The Life and Works,12) gives more weight to al-Ǧāḥiẓ's polemical and pro-Arab, pro-Islam stance, but also indicates that the cosmopolitan spirit of adab tempered this to some extent. 169 See Ṣāʿid al-Andalusī's Ṭabaqāt al-umam for a paradigmatic example of