would have to be like for the sentences expressing them to be literally true. On this view, it follows that cognition of meaning in the Qur'an is first and foremost a matter of understanding the truth and the literal character of divine utterance. If meaning and thought are modelled in this fashion, epistemic access to figurative meaning is asymmetrically dependent on the cognition of literal meaning.
Ibn Taymiyya put Ghazali's theory under critical stress by arguing that all that hearers of divine discourse need know is how the divine discourser intended His speech to be taken. That is, one only has to grasp its illocutionary force arising from contexts (qari'in) of use (isti'mal), plus the intention revealed in God's habit of address ('adat al-mutakallim). Hence, the apprehension of figure in the Qur'an resides in the apprehension of force. But if this is the case, Ghazali's view that epistemic access to figurative meaning is asymmetrically dependent on cognition of literal meaning is seriously undermined and the distinction between haqiqa and majiz is utterly erased. Thus, for Ibn Taymiyya the question of epistemic access to figural speech does not really arise. Hermeneutics is the only matter of concern, that is to say, the interpretation of the pragmatic force of the divine utterance.
This chapter will sketch some of the main features of this debate. The issues it raises lie at the heart of modern discussions between Muslim traditionalists, who often side with Ghazali, and fundamentalists whose champion is Ibn Taymiyya. But the question of how Muslims are to understand verses in the Qur'an that refer to God "sitting" or ''descending'' or having a particular spatial locus are at base matters of linguistic epistemology. Or rather, they concern the relation of epistemology to divine discourse.
AL-GHAZALi's VERBAL EPISTEMOLOGY
In a work written towards the end of his life, The Essential in Legal Theory (al-Mustasfi min 'ilm al-usUl), Ghazili defines divine discourse (kalam) as ''either something a prophet hears from an angel or an angel from God or a prophet from God or a saint [wali] from an angel or the Muslim community from the Prophet''.4 That is, only to an appropriately qualified audience does divine speech bear significance. Hence, knowing what God means and how He means it when He speaks depends on who hears His voice. There is a difference, in other words, between the way prophets and saints hear the divine voice and the way the Muslim community (umma) including its scholars ('ulama') hears it.
The umma hears letters and sounds reported by the Prophet that signify the meaning of God's eternal speech (dall 'ala ma'na kalam Allah) insofar as it has entered time. That is, what they hear is the expression of God's thoughts, since for Ghazall God ''is called a speaker from two aspects, from the aspect of sounds and letters and from the aspect of inner speech [hadith al-nafs] devoid of sound and letters''.5 The latter corresponds to what Ash'arï called ''interior discourse'' (kalam nafsi). It is regarded as eternal, but what it expresses ('ibara) is not. Moreover, they grasp it ''by prior cognition of its assigned'' or agreed-upon meaning or wad' al-lugha (prior imposition of language).6 For Ghazala believed that in the final analysis it makes no difference whether language originates from divine inspiration (tawqaf) or from a convention (istilah) agreed upon by a community of primordial Arabic speakers.7 He nevertheless saw language as essentially transcending the fate of the mortals who speak it. Deviations in particular contexts therefore represent a departure from a common standard. That is, they are deviations from a language envisioned as an established social institution or set of conventions.
By contrast, when prophets and saints hear the divine voice, God makes what He intends (al-murad) ''known by creating in the hearer a necessary knowledge ('ilm darun)'',8 that is, a knowledge which the hearer has no choice but to accept when it is presented to his or her mind.9 When the prophet Moses heard God speak, his hearing had ''no letter nor sound nor language established in such a way that one knows its sense (ma'nahu) through prior cognition of its assigned'' or agreed-upon (muda'a) meaning. Instead, God creates the object of cognition or ''what is spoken'', the act of hearing His speech as well as the meaning ''intended by His speech''. For ''every speaker needs to posit a sign to inform [others] of the content of his mind except God [who] is able to create a necessary knowledge without positing a sign; for His speech is not of the same genus as human speech.'' So ''the act of hearing [His speech] that He creates for His servant is not of the genus of hearing sounds''.10
But for Ghazaalal, prior cognition of what expressions signify according to the conventions laid down in wad' al-lugha was not enough. The original speakers of Arabic did not establish conventions without having some reason for doing so. The motivation behind linguistic conventions is the communication of truth. Language, in other words, is founded in order to convey truth, not falsity. Now, the aim of logic is to fix the use of expressions by analysis of their contribution to determining the truth-value of judgements in which those expressions figure. So, it became
Ghazali's view that logical norms ought to govern the way qur'anic expressions are used.11
But adherence to a logical analysis of qur'anic signification brought with it certain presuppositions, epistemological as well as metaphysical. In the latter case, it presupposed adopting an Aristotelian framework where all objects are seen as possessing universal essences allowing them to be defined in terms of genera and species. Epistemologically, this meant that the first moment in cognition, forming concepts or tasawwur, is based upon knowledge of definition. Still, concepts by themselves have no truth-value until combined as subjects and predicates of judgements expressed by propositions (qadaya) and sentences. Knowledge manifests its second element as tasdaq or takdhib (the cognition of truth or falsehood).
But if essences stand behind the concepts expressed by words in the Qur'an there is no real need to appeal either to meaning by wad' al-lugha or to interpretation (haml) arising from actual contexts of use (isti'mal). Logical definition outstrips both. Consider the sentence ''A lion is in the house'', for example. Its figural interpretation, ''A brave man is in the house'', or its literal one, ''There is a carnivorous feline in the house'', is determined to apply when its context or its speaker's intention is known. But in logic neither context of use nor intention needs to be known. When you apply the concept of lion to ''Leo'', you by definition signify that ''Leo'' is also an animal. At the same time, you include thereby the act of signifying that ''Leo'' is also a mammal.
Ghazali, following the philosophers, called these modes of signification respectively ''signification by correspondence'' (dalalat al-mutabaqa) and ''signification by inclusion'' (dalalat al-tadamun). For logical purposes, he deemed them preferable to signification by implication (dalalat al-iltizam). They operate simply on what it means to be a lion, the essence expressed by ''lion''. ''Implications [lawazim]'', by contrast, ''are indefinite and unrestricted so that it leads to an expression being a sign (dalil) for an infinite [number of] meanings.''12 Such implicit meanings, Ghazali observes, are known from either linguistic,13 rational or ''situational contexts such as allusions (isharat), or symbols (rumuz)'' which ''are unlimited and unpredictable''.14
The ideas sketched so far, however, form only part of Ghazali's programme to link literal meaning to truth. After all, to him religious ''faith (iman) is tasdaq'' or assent to truth.15 Accordingly, he theorises that in whatever way the predicates of qur'anic sentences signify, that is, whether they signify in an essential (al-wujud al-dhati), sensible (al-wujud al-hissi), intelligible (al-wujud al-'aqla) or analogical (al-wujud al-shibhi) fashion, they still preserve a meaning focused on tasdiq or assent to truth. Or rather they preserve an existential sense; for the ''essential nature (haqiqa) [of tasdiq] is recognition (al-i'tiraf) of the existence of what the Messenger has reported (akhbara 'an) about its existence.''16
However, since this variation of predicates reflects only mental operations that supervene upon what amounts to a literal content, the same structure can illustrate the comprehension of figurative expression. For ''every majaz [figural sense] has a haqiqa [i.e. a literal sense] but it is not necessary that every haqiqa have a majaz''.17 That is, a literal, factual or existential sense can be grasped behind any figure such that the figure is a departure from a literal meaning, but not vice versa. Figure is always asymmetrically dependent on a literal sense where the latter is conceived in terms of the conditions under which certain sentences are true. Literalness is thus linked to truth.
The theological consequences of Ghazali's views proved in time to be controversial. For there comes with his commitment to logic a God whose mind could seem dissimilar to that of the author of the Qur'an. No longer is God pictured as communicating with the ordinary words of a human language like Arabic, words backed by human-like intentions. Rather, His words have to be backed up by essences. The language He speaks is a mental language of logical genera and species or universal natures, so His utterances reach out and decide every case of their application well in advance. For divine mental content embraces the details of every possible world.
If God prohibits the drinking of wine (khamr), for example, there must be written into the prohibition a selecting out of substances that share the essence of khamr in every possible context. The predicate ''... is khamr'' then governs its pattern of use in inferences connecting it with judgements employing other predicates such as ''... is prohibited." In this way, cases not covered in the Qur'an are ruled out not as a matter of linguistic convention or textual probability but as a matter of logical necessity.
The motivating idea, however, is that one can give an account of what it is to have a thought without appealing to a speaker's ability to use expressions in ways appropriate to conventional use. Thoughts do not need linguistic representation. The idea echoes Avicenna, for whom thought had only an accidental connection to language. And while recognising that ''it is impossible for internal reflection to put meanings into any order without imagining expressions for them'', those imagined expressions accompany thoughts they do not embody. Was the God of
Avicenna's Neoplatonised Aristotelianism surreptitiously replacing the God of the Qur'an in Ghazili's analysis?
But he did not need philosophy to tell him that God has an ''inner speech [hadith al-nafs] devoid of sound and letters''. He knew, for example, that ''every speaker except God needs to posit a sign to inform [others] of the content of his mind'' and that ''His speech is not of the same genus as human speech''. So much he had from Ash'arism. The difficulty was this. If God's intentions were linguistically formulated, the extensions of the qur'anic words expressing them would have to be indeterminate as well. As the historian of religious and philosophical doctrines al-Shahrastini (d. 1153) observed, ''We know for certain that no text mentioned ever touches on every event nor is it ever conceivable that this should be so.'' For ''texts are limited [but] facts [of life] are infinite, so the finite cannot embrace what is infinite''. For Shahrastani the Qur'an was clearly a text with gaps that needed to be filled by ijtihid (legally warranted personal interpretation). To that extent, the meaning of its words did not embrace every conceivable occasion of their application.
So a dilemma arises: either the thoughts of Ghazali's God are totally devoid of linguistic representation and stand complete in every detail, or they are linguistically formulated and incomplete. Instead of speculating over Ghazali's options for escaping this dilemma, however, we should note that he had another story to tell in his esoteric work Niche of Lights (Mishkit al-anwir), written after the Revival. We will come to the picture painted there in due course.
IBN TAYMIYYA'S CRITIQUE
Meanwhile, we turn to Ibn Taymiyya's critique of the theory so far presented.18 In his view, it can only be a fiction that the signs expressing divine speech occur without norm or context. ''Meaningful discourse (al-kalim al-mufid)'' he declares, ''is only conveyed by a complete sentence jumla)'', that is, in the context of a sentence.19 So the fundamental unit of semantic analysis can never be the bare conceptual sign. To have a specific force, words must already stand in a fundamental grammatical relation to each other. For faith (imin) at the outset assumes that sentences of the Qur'an are true, and sentences, not individual words, can be either true or false.20 Thus, whoever hears or reads divine discourse must perceive it as ''al-kalim al-musta'mal'' or ''discourse in use'', that is, as sets of sentences uttered with an intended meaning, an intended force.
For Ibn Taymiyya, Ghazall's vision of a language of thought internal to the divine mind replaces ''al-kalam al-musta'mal'' with an artificial ''al-kalam al-muqaddar'' or ''hypothetical discourse''.21 The attempt to account for meaning in terms of the latter is a dead end. Since ''al-kalam al-muqaddar'' was never spoken by anyone, its existence is a matter of pure metaphysical speculation. Ibn Taymiyya thus makes an appeal to language as performance and to what people do with words in specific contexts of communication.
''Discourse in use'' bases itself on the divine speaker's habit of discourse ('adat al-mutakallim). No hidden essences lurk behind it to serve as the meaning of words. The meanings of words are immanent to the structure of their use (isti'mal). We know in a certain and decisive (qat '!) fashion what a speaker wills and intends to say simply by virtue of his habit of address. ''The mere hearing of the expression without knowledge of the speaker and his habit'', he writes, ''signifies nothing'', unless one knows ''what is necessary for the speaker to signify by them.'' This is because the ''signification of expressions is an intentional, volitional act signalling what the speaker means (arada al-mutakallim an yadulla biha) by them, [given that] expressions by themselves fail to signify''.22 To his student Ibn al-Qayyim (d. 1350), ''the signification of the expression is constructed upon the habit of the speaker which he intends by his verbal expressions''. If this were not true, no child would ever be able to learn language.
Ibn al-Qayyim argues that there is simply no scope for interpretive norms agreed upon (muwada'a) prior to the actual contexts in which children learn to speak.
After the child begins to distinguish [among sounds], he hears his parents or whoever raises him articulate a language and point to its meaning. In this way he understands that when a certain expression is used a certain meaning is intended ... [All that happens] without reaching an agreement with the child on a prior assignment (wad' mutaqaddim) in order to inform him of names' meanings.23
Ibn al-Qayyim concludes, ''We know that this expression is prim-ordially imposed only by virtue of using it ... in the sense [already imposed].''24
Both Ibn Taymiyya and Ibn al-Qayyim then totally reject the idea that ''a group of scholars met together and imposed all the words to be found in their language and from that point proceeded to use the words assigned'', that is, the theory that language originates from convention.25 ''On the contrary,'' Ibn al-Qayyim asserts, ''inspiration suffices for the articulation of language without a prior assignment of names to things'', and he adds, ''If one calls this divine inspiration (tawqif), then let it be called divine inspiration.''26
At the same time, on Ibn Taymiyya's theory, there are no essences hidden behind God's words to determine their use in any particular context. When I hear the sentence ''I saw her duck'', for example, there is no universal ''duckness'' in my mind that constrains me to think of the bird ''duck'' rather than the verb ''duck''. In other words, meaning does not come in the form of a universal that fixes the extension of a term in every future case of its expression's future application. Meaning is nothing deeper than the use of ordinary words in particular contexts.
This does not mean that the future use of words revealed in the Qur'an is wholly unshaped. It indicates only that God provides nothing better than the Qur'an itself and the hadith of the Prophet to explain it. ''The soundest method of commentary on the Qur'an'', Ibn Taymiyya writes, ''is to comment on it with the [words of the] Qur'an [itself];for what is unclear in one place is explained in another and what is abridged in one place is set forth plainly in another.''27 All items of divine discourse, in fact, can be understood by an appeal to divine words as these have been given. There is no essence behind the use of ''khamr'' (wine) in the Qur'an that underwrites the right way of applying it in any given context.
Was this article helpful?