NONREDUCTIVE THEORIES OF SENSE-PERCEPTION IN THE PHILOSOPHY OF KALĀM

Abstract In this article, I will argue that various scholars of kalām unanimously agree that sense-perception is something beyond the physical processes in the sense organs. There may be something happening in our eyes when we see a red apple, but seeing a red apple is not tantamount to it. We will see that some scholars of kalām argue that sense-perception is akin to being aware or conscious of the object of perception, and, hence, distinct from the physical process in the sense organs. One group will go so far as to accept that sense-perception is not even dependent on any physical processes in the body. Another group will accept that sense-perception presupposes that various physical conditions obtain, yet still regard sense-perception as something distinct from the occurrence of those conditions. I am suggesting that these nonreductive theories of sense-perception are the reason why Arabic-Islamic philosophers, starting from the eleventh century CE, consistently reject the Aristotelian-Avicennian theory of sense-perception.

Much of what has been written about the history of the philosophical theories of sense-perception focuses on the Aristotelian account of senseperception and the ways in which the medieval philosophers understood it.One of the main questions in the discussion, famously a matter of debate between Myles Burnyeat and Richard Sorabji, is what Aristotle meant by saying that sense-perception implies a change in the senseorgan. 1 This question is part of a bigger issue whether sense-perception is a material or an immaterial process.For instance, one could wonder whether seeing a red apple amounts to the material change in the eye of the observer, or there is some immaterial information received through observation, and whether either kind of change is necessary or sufficient condition for the occurrence of the phenomenon of seeing red in an apple.
Against this common trend, I am suggesting looking at some non-Aristotelian theories of sense-perception in medieval philosophy.In this article, I will focus on the tradition of philosophy in the Islamic world called "kalām."Modern scholarship has just discovered that senseperception was a widely discussed issue in the philosophy of kalām, with the recent contributions by David Bennett and Laura Hassan. 2 Both contributors, albeit from different perspectives, suggest that the scholars of kalām opposed the traditional Aristotelian understanding of sense-perception.
In this article, I will argue that all the main theories of senseperception in kalām involve a nonreductive understanding of senseperception.By "nonreductive," I mean that sense-perception, for the scholars of kalām, is something beyond the mechanistic processes occurring in the sense-organs. 3Seeing a red apple involves (but may need not, depending on the theory in question) a physical change in the sense-organs, but seeing a red apple does not reduce to it.Developing further the thesis of Bennett and Hassan, I will suggest that this nonre-1 See, for instance, the papers collected in Martha Nussbaum and Amélie Rorty, Essays on Aristotle's "De anima" (Oxford: Clarendon Press, 1992) and Dominik Perler (ed.), Ancient and medieval theories of intentionality (Brill, 2001). 2 David Bennett, "Sense-Perception in the Arabic Tradition: The Controversy Con-ductive understanding of sense-perception leads the Arabic-Islamic philosophers to a refutation of the Aristotelian-Avicennian theory of sense-perception, which they understand as a reductive theory of senseperception, identifying the phenomenon of sense-perception with the physical change in the organ.
In support of my thesis, I will draw a systematic map of different theories of sense-perception in kalām.We will see that some theories of sense-perception in kalām separate sense-perception from anything happening in the bodily organs, to the extent that seeing a red apple can happen in whichever physical conditions.This kind of approach is common, for instance, to the Ašʿarites.Another group of the scholars of kalām, the Basrian Muʿtazilites, opposes the independence of senseperception from the physical conditions.Still, we will see that even this group understood sense-perception in nonreductive terms.This will become particularly clear in the new brand of the Basrian Muʿtazilism, initiated by Abū l-Ḥusayn al-Baṣrī (d.1044).For this group, seeing a red apple involves all right physical conditions and changes in the organs (at least for organic creatures), but, still, it is something distinct from the occurrence of those changes.I will show that the Basrians develop an active theory of sense-perception to support their nonreductive approach.
I will base my analysis mainly on the sources from the eleventh century CE, the time when kalām reached its best as an independent philosophical tradition, and before it was heavily influenced by the philosophy of Avicenna (d.1037).For the Ašʿarites, my main sources will be Salmān b.Nāṣir al-Anṣārī (d.1118), the student of ʿAbd al-Malik al-Ǧuwaynī (d.1085), alongside Muḥammad b. ʿAbd al-Karīm al-Šahrastānī (d.1153), a student of al-Anṣārī.For the Basrian Muʿtazilites, my main sources will be ʿAbd al-Ǧabbār al-Hamaḏānī (d.1025), but even more so the later representatives of his school, such as Ibn Mattawayh (11th century), Mānkdīm Šašdīw (d.1034) and Abū Saʿd al-Ǧišumī (d.1101).Finally, the main source for the doctrines of the school of Abū l-Ḥusayn al-Baṣrī will be Maḥmūd b.Muḥammad al-Malāḥimī (d.1131).

CONSCIOUS EXPERIENCE
Discussions of sense-perception in kalām often start with the question whether perception (idrāk) differs from knowledge (ʿilm).According to our sources, there is a disagreement both among the Ašʿarites and the Muʿtazilites regarding this question.Among the Ašʿarites, Abū Isḥāq al-Isfarāʾīnī (d.1027) identifies sense-perception as a kind of knowledge, while Abū Bakr al-Baqillānī (d.1013) rejects it, with Abū l-Ḥasan al-Ašʿarī (d.936) himself remaining undecided on the issue. 4Among the Muʿtazilites, al-Kaʿbī and his Baghdadians argue that sense-perception is just knowledge, while the Basrian Muʿtazilites reject it. 5he earliest representative of a Muʿtazilite identification of senseperception with knowledge might have been Abū l-Huḏayl (d.c. 842).According to the account of al-Ašʿarī, Abū l-Huḏayl argues that "Perception inheres in the heart, not in the eye; it is necessary knowledge." 6e find a very similar account, but this time not ascribed to anyone, in the doxography of al-Kaʿbī: Some of them said: the subject of the inherence [of perception] is the heart.[Perception] is the knowledge of the perceived.The pupil of the eye does nothing beyond being set up opposite to the perceived, if the person receives it through [the pupil]… 7 This position is opposed to the following: Some of them said: the perception of the colour happens in the pupil itself; it is identical to sensing it (ḥissuhu).Knowledge, however, is in the heart, not anywhere else. 8 Although the evidence is scarce, these quotations reveal that, from the very beginning, those who identify sense-perception with knowledge intended that sense-perception is something distinct from the physical processes in the sense organs.The act of seeing something does not happen in the eye when a reflection of the seen object appears in it. 9 seeing happens in the heart, the seat of knowledge.In the notions of kalām, idrāk (perception) is distinct from whatever happens in the ḥassa (a sense organ). 10he reason why sense-perception is identified with knowledge, to distinguish it from organic processes in the sense organs, is that knowledge is something independent from those processes as well.So, according to the report of al-Anṣārī: Those who agreed with the position of the Teacher [Abū Isḥāq al-Isfarāʾīnī] say: even when we say that perception is of the same kind as knowledge, we still say that it is different from those kinds of knowledge that are not sense-perception.However, one and the same notion (maʿnā) or name connects them, the same way as we said in the case of five senses that the special character (ḫāṣṣiya) of perception is common to them, even if hearing is different from sight, and neither of them is limited to the other.Nevertheless, perception is just like knowledge (ka-ʿilm) insofar as it does not require any [physical] connection (iṭṭiṣāl) or opposition (muqābala) or an impression in a sense-organ (inṭibāʿ ḥāssa). 11-Anṣārī explains in this passage, on behalf of the proponents of the identity of sense-perception with knowledge, that "knowledge" is a generic notion.Different types of knowledge fall under that notion, just like different types of sense-perception fall under the same notion of sense-perception.As this kind of generic notion, "knowledge" means a mental act that has no necessary connection with the physical processes in the body.Thus, sense-perception, as a type of knowledge, requires neither physical connection (in the case of vision, for example, through the ray of light between the observer and the observed), 12 nor opposition (again, probably, in the case of vision; meaning the opposition between the seen object and the pupil of the eye), nor an impression of the sensed in a sense organ (probably referring to hearing, unless an intramissionist theory of vision is meant here).In any case, sense-perception is something different and independent from whatever happens in the senseorgans.That is why sense-perception is called "knowledge." Al-Anṣārī develops a similar line of thought on behalf of al-Kaʿbī: If al-Kaʿbī says: When you say that perception does not require taking on the shape (tašakkul) of the perceived object by the perceiver, nor any opposition or the perceiver's being in a location, [perception] just amounts to knowledge!
We say: We have already said that any person of sound mind perceives a difference between [knowledge and perception].
If they say: That difference that you are talking about is tantamount to the impression of the perceived in the sense organ (inṭibāʿ al-ḥāssa bil-maḥsūs), which we indicated.We believe, however, that perception is the awareness of the soul (šuʿūr al-nafs); and that is just knowledge.
We say: Knowledge, which is included in perception, inheres in the heart, according to us, or in the brain, according to the ancients.Perception, however, inheres in the organ of vision and in the pupil.Every person of sound mind distinguishes between them necessarily and asserts with certainty that [perception] is something beyond the reflection (taḫayyul) and the impression (inṭibāʿ).Whenever somebody perceives with one of the sense-organs, perception inheres in the sense-organ, and, furthermore, sensation (iḥsās) and perception (idrāk) include knowledge in the heart. 13e passage starts with restating, on behalf of the proponents of the identity of sense-perception with knowledge, that sense-perception is independent from any physical changes in the body.The logic of the argument presupposes that everyone in this debate agrees to this fact (even if, as we will see further, the Basrian Muʿtazilites actually do not).To this, al-Anṣārī replies that there is an obvious difference between knowing something and perceiving something.This is a recurring statement in this kind of discussion.It usually refers to the idea that there is an obvious difference between seeing something, and still thinking about the same object, after closing the eyes.The former is sense-perception, the latter is knowledge. 14o this, the "dialectical al-Kaʿbī" replies that sense-perception may indeed involve a certain process in the body, that is, the impression of the perceived in the organ of perception.And that is how sense-perception is different from knowledge.Still, that process does not constitute senseperception as such.Sense-perception is the "awareness of the soul," not the physical process in the body, corresponding to it. 15n my understanding, what the "dialectical al-Kaʿbī" is talking about is something we know under the notion of conscious experience.In other words, those who identify sense-perception with knowledge do so because for them both knowledge and sense-perception is about how we experience the phenomenal object of perception or knowledge.According to the proponents of this view, whether my eyes are open or not, there is no phenomenal difference between how I experience the red apple: I have all the same phenomenal information in both states.Hence, the only difference between seeing a red apple and thinking about a red apple after closing the eyes, is the presence of a state in the body, the "impression in the sense organ," which lies outside of my conscious experience.
Al-Anṣārī disagrees.For him, knowledge and perception are two different processes.One happens in the heart, another in the sense-organ.Therefore, sense-perception cannot be just identical to the conscious experience, present in the case of knowledge.
Al-Anṣārī addresses the role of conscious experience in senseperception in his report of a debate that happens between al-Isfarāʾīnī, al-Ǧuwaynī and al-Baqillānī regarding the question whether senseperception necessarily implies knowledge.The cases that they discuss include how children and animals experience sense-perception, the case of feeling pain and the case of a sleeping person.Al-Ǧuwaynī and al-Baqillānī argue that there is no necessary connection between sense-perception and knowledge, even if they usually (fī l-ʿāda) come together.For instance, a person in pain usually knows about it, but may also fail to know about it if they are unconscious or if another pain is covering the first pain.Likewise, a sleeping person may be woken up by a sound, but they do not have knowledge of that sound because they are asleep. 16nowledge is used in this context synonymously with consciousness. 17The discussed question is whether it is possible to have senseperception without experiencing it consciously, for instance, whether it is possible to feel pain without experiencing it consciously.In this context, the position of the proponents of the identity of knowledge and sense-perception is presented as follows: For those who believe that perception is a type of knowledge, that kind of knowledge subsists in the sense-organ, which is the subject of the inherence of perception, being the awareness and the consciousness (al-šuʿūr wa-l-istišʿār).Furthermore, that knowledge includes [another] knowledge in the heart. 18 other words, the question is about conscious experience, and the proponents of the identity of sense-perception with knowledge argue that there are two levels of experience.One is the experience in the sense organ, another is the experience of the heart.Apparently, the idea is that an unconscious person in pain still experiences their pain, even if not in the heart, that is, not consciously.Note that the position here is slightly different from the one ascribed to al-Kaʿbī.In al-Kaʿbī, we had one item of conscious experience, whether we call it knowledge or senseperception.Now, there are two items of knowledge.Thus, al-Anṣārī's opponent (probably al-Isfarāʾīnī here) concedes that there are two processes, one in the heart and another in the sense organs.But both processes are "awareness" (even if on different levels) and both are distinct from the physical state of the sense organs.Now, al-Baqillānī, as reported by al-Anṣārī, clearly disagrees.He argues for instance, that a sleeping person who has been woken up by a sound does perceive it, even if they do not know it. 19Equally, a person with two different pains perceives both in reality even if she knows only of one of them. 20Thus, al-Baqillānī's position is that sense-perception does not need to imply conscious experience.
Al-Baqillānī's position is unacceptable for the proponents of explaining sense-perception as knowledge.Al-Isfarāʾīnī replies to him that a sleeping person does not perceive the sound that wakes them up.Rather, there is a part of a sleeping person that is not asleep (and hence, is conscious) that perceives that sound. 21Interestingly, al-Ǧuwaynī, as reported by al-Anṣārī, agrees with al-Isfarāʾīnī, although he has just agreed with al-Baqillānī regarding the case of unconscious pain before.
Whether sense-perception is conscious experience (al-Isfarāʾīnī and al-Kaʿbī), or it is something else (al-Baqillānī), all aforementioned authors appear to be in agreement that it is not something reducible to the physical state of the sense organs.In the passage quote above, al-Anṣārī explicitly acknowledges that perception is "something beyond the reflec-18 Al-Anṣārī, Šarḥ al-Iršād, vol.2, 277.1-3. 19Al-Anṣārī, Šarḥ al-Iršād, vol.2, 277.9-11. 20Al-Anṣārī, Šarḥ al-Iršād, vol.2, 276.14-17. 21Al-Anṣārī, Šarḥ al-Iršād, vol.2, 277.12-15.tion (taḫayyul) and the impression (inṭibāʿ)." 22This is also perfectly in line with what al-Ǧuwaynī says in the context of the discussion of the senses of taste, touch, and smell.For him, all of these senses are distinct from sense-perception as such.That is why somebody can say: "I smelled an apple, but I did not perceive its smell." 23Al-Anṣārī explains: "Taste" and "touch" stand for connections (ʿibārāt ʿan ittiṣālāt) between bodies.They are neither perceptions (idrākāt), nor conditions for them. 24 means that sense-perception, conversely, does not "stand for connections between bodies."In other words, sense-perception is not reducible to the physical states of the sense-organs.
There is, however, a problem with this position.The position of the proponents of the identity of sense-perception and knowledge, the position of Abū Huḏayl, al-Isfarāʾīnī and al-Kaʿbī, argues that senseperception is not reducible to the physical states of the sense-organs based on the identification of sense-perception with conscious experience (whether that experience is just the same as the one in the heart or not).Indeed, it is plausible to assume that experiencing the vision of a red apple is something distinct from whatever happens physically in my eye and in front of it.But those who deny that sense-perception involves experience, such as al-Baqillānī, must have a hard time to prove that sense-perception is distinct from the physical state of the sense organs.This difficulty might have been the reason why, for instance, al-Šahrastānī, in his presentation of the debate, does not feel any need to insist that sense-perception is not knowledge in any sense.Instead, he takes al-Isfarāʾīnī's position as he finds it in al-Anṣārī, says that sense-perception is a type of conscious experience, even if not the same type as knowledge, and ascribes it to al-Ašʿarī himself. 25Whether al-Ǧuwaynī or al-Anṣārī would agree with that move remains an open question.

CAUSATION AND SCEPTICISM
The Ašʿarites of the eleventh century have another reason to believe that sense-perception is distinct from the physical processes in the sense-organs.They argue that sense-perception is not even caused by those processes in the sense-organs.As David Bennett shows,26 senseperception is directly caused by God, according to the Ašʿarites, not by what happens in the sense-organs.This debate is put in terms of the infamous notion of maʿnā. 27The Ašʿarites argue that perception (idrāk) is a separate maʿnā, that is, a separate "something," beyond the physical conditions.Hence, a person can be described as being in the state of perceiving just due to the presence of the maʿnā of perception, irrespective of any physical conditions.The opponents of this position are the Basrian Muʿtazilites, for instance, ʿAbd al-Ǧabbār and Ibn Mattawayh.Following the core figure of their school, Abū Hāšim al-Ǧubbāʾī (d.933), against another core figure, Abū Huḏayl, whose position is akin to that of the Ašʿarites, the Basrian Muʿtazilites argue that sense-perception depends on certain conditions: the subject of perception must be a living being and there should be no hindrances to perception.The "absence of hindrances" in this definition of sense-perception implies the presence of the required physical conditions, both inside the body of the perceiver and outside it.As al-Anṣārī puts the position of his opponents: "The condition for the perceiver to be a perceiver in this world is that he has a specific [corporeal] structure (binya maḫṣūṣa)" and "Each of us can only perceive through the instruments and organs (adawāt wa-ālāt), such as the sense-organs (ḥāssa) etc." 28 The Ašʿarite resistance to a connection between the occurrence of sense-perception and the presence of the physical conditions is part of their occasionalist approach to causation.According to a view predominantly accepted by the Ašʿarites, everything that happens in the world is directly caused by God.Sense-perception is just one thing among many. 29Therefore, the Ašʿarites do not present any specific argument in favour of making sense-perception causally independent from the physical processes in the body.Instead, al-Anṣārī, for instance, refers Work of the Maʿnā in Early Muʿtazilite Theory," in N. Germann and M. Najafi (ed.), Philosophy and Language in the Islamic World (Berlin, De Gruyter, 2021), 1-20, with a helpful list of the previous studies on maʿnā. 28 to a general point based on the atomistic ontology of kalām: if an accident, such as perception (idrāk), can inhere in an elemental substance (ǧawhar), nothing from outside of that substance can impact whether that accident can or cannot inhere in it.Whether we take a substance in combination with other substances or in isolation from them, the accident of perception will be able to belong to that substance.Therefore, God can just create an accident of perception in us, irrespective of any further conditions. 30udging by the state of our sources, the burden of proof in this debate was on the side of the Basrian Muʿtazilites, who tried to prove that sense-perception must involve physical conditions.Their main argument is based on the sceptical consequences of the occasionalist position.The argument goes as follows: If the perceiver could perceive through perception (bi-idrāk), inevitably, it would be possible for a healthy and sound perceiver to perceive a person in front of him, since a [correspondent] perception is created for him, but he would not see elephants playing in front of him or cattle grazing [in front of him], since no correspondent vision would be created for him.Likewise, inevitably, it would be possible that he sees a person far away from him without seeing somebody who is in front of him.Likewise, inevitably, it would be possible that he hears a quiet sound from far away while there could be trumpets blowing in front of him and he would not hear them. 31 other words, if we allowed that sense-perception did not depend on the presence of any physical conditions, then we would never be able to claim with certainty that our perception corresponds to reality.It would all depend on God.In some cases, He might create correct perceptions for us.But in other cases, God could create false perceptions in our minds, different from reality.It would be possible that there are elephants dancing around us but we would not see them.Overall, perception would not be a reliable source of information anymore.As Ibn Mattawayh says, "any certainty about what we see would be gone.The occasionalist response to this argument is based on the notion of ʿāda ("common phenomenon" or "the habitual course of events"), central to their occasionalist metaphysics and epistemology.Al-Anṣārī says, for instance, that ʿāda is the only reason why we judge that the above possibilities are improbable (mustabʿad). 33Usually, whenever we see an elephant, it is there.So, if we do not see an elephant, it probably means that there is none in front of us.So long as God follows the habitual course of events, He creates for us an undeniable item of knowledge (ḥalaqa lanā al-ʿilm iḍṭirāran) that things are as we see them. 34amously, this is the same answer that Abū Ḥāmid al-Ġazālī (d.1111) gives to those who argue that occasionalism leads to scepticism in his Tahāfut al-falāsifa. 35As suggested by Frank Griffel, the Ghazālian solution to the sceptical problems implied by occasionalism is a new definition of knowledge.Knowledge is what actually is the case, not what could or could not have been the case. 36When I see an elephant, I assent to the proposition "There is an elephant in front of me."But I cannot assent to the proposition "There is necessarily an elephant in front of me" because there is a possibility that God did not create the vision of the elephant in me despite the presence of the elephant.Our knowledge is knowledge of the matters of fact, not of modalities.As we can see from al-Anṣārī's analysis of sense-perception, al-Ġazālī's solution in the Tahāfut is simply derived from the traditional occasionalist way of dealing with the epistemic problems for their occasionalist universe.
The Basrian Muʿtazilites are perfectly aware of this solution, and it does not convince them at all. 37Ibn Mattawayh says for instance that it contradicts our real-life experience.I may say to an occasionalist "Go inside the house, there is some money in it."If he does so but does not see the money, he says to me: "There is none."And if I repeat my request, he will reply: "If there were any, I would have seen it!" 38According to Ibn Mattawayh, this conversation demonstrates that our knowledge what actually is the case is secondary to our knowledge what could and could not have been the case. 39This idea becomes matter of extensive argumentation in Abū l-Ḥusayn al-Baṣrī, according to al-Malāḥimī.Following this argumentation, our knowledge that p, when it is based on sense-perception, is derivative from our knowledge that "if it were not p, then we would not perceive it." 40Abū l-Ḥusayn insists that even children and animals form their knowledge based on sense-perception this way.In the first place, they assent to the modal proposition "If it were there, I would perceive it."And in the second place they conclude "I do not perceive it; hence, it is not there."If the opponent does not want to ascribe this logical inference to animals and children, they should just accept that neither group has any knowledge based on sense-perception at all. 41bū l-Ḥusayn's attempt to argue that sense-perception can occur only under certain conditions presupposes that we can prove it.In the notions of kalām, our knowledge that sense-perception depends on the physical conditions may be acquired (muktasab). 42Al-Malāḥimī reports, however, that Abū l-Ḥusayn's own preferred view was that it is an item of necessary undeniable knowledge (ʿilm ḍarūrī). 43Abū l-Ḥusayn argues that we immediately learn from experience (iḫtibār) that it is impossible that somebody touches hot iron and does not perceive the heat.And experience does not play the role of a proof here.Rather, it is a reminder (tanbīh) of what we already know. 44hus, the Basrian Muʿtazilites deny the occasionalist approach to sense-perception, common to the Ašʿarites and to some of their early predecessors.For the Basrian Muʿtazilites, sense-perception presupposes certain conditions, both inside the sense-organs of the perceiver and outside them.To be precise, the Basrians distinguish between the producer (muʾaṯṯir) of sense-perception and further conditions (šarāʾiṭ).All the 39 Ibn Mattawayh, Taḏkira, 703.14-15. 40 physical states of the sense-organs, the absence of any hindrances, and the presence of the object of perception are merely conditions for senseperception.The real cause of sense-perception is just the fact that the perceiver is alive (ḥayy). 45We will come back to this point in the next section.
The Basrian understanding of the causation of sense-perception is based on the idea of conditional necessity.The common formula for sense-perception is that "if it is possible, then it is necessary." 46In other words, when all conditions are satisfied, the living being must perceive, it has no choice. 47A similar relation holds between senseperception and knowledge.In the Basrian notions, sense-perception is a ṭarīqa ("way") to knowledge.What it means is that sense-perception does not necessitate knowledge by itself.Children, for instance, have sense-perception but they do not have knowledge.Still, if all other conditions, such as the completeness of the intellect (kamāl al-ʿaql), are satisfied, sense-perception inevitably leads to knowledge. 48This is the usual Basrian response to anyone objecting that their suggested dualism of sense-perception and knowledge leads to similar sceptical problems as the dualism of sense-perception and physical conditions, suggested by the occasionalists."No, it does not," reply the Basrians, "because knowledge necessarily follows upon sense-perception if all conditions are satisfied." 49

BASRIAN SUPERVENIENCE
The question whether perception (idrāk) is a maʿnā, discussed in the last section, is not identical to the question whether sense-perception is reducible to the physical states of the sense-organs.Admittedly, if we agree with the Ašʿarties and the early Muʿtazilites that idrāk is 45 Ibn Mattawayh, Taḏkira, 699; Šašdīw Mānkdīm, Šarḥ al-Uṣūl al-ḫamsa, ed.
ʿA. al-Karīm ʿUṯmān (Cairo: Maktabat wahba, 1965), 169-171.Bennett ("Sense-Perception in the Arabic Tradition") argues that sense-perception has no cause for the Basrian Muʿtazilites.Bennett possibly means "no external cause," since the Basrians say explicitly that being a ḥayy is the muʾaṯṯir of being a mudrik. 46Al-Ǧišumī, Šarḥ ʿUyūn al-masāʾil, 627.7-8; cf.Ibn Mattawayh, Taḏkira, 699.4 and ʿAbd al-Ǧabbār, Muġnī, vol.4, 39. 47 Regarding the involuntary character of sense-perception see Ibn Mattawayh, Taḏkira, 705.1-5 and al-Ǧišumī, Šarḥ ʿUyūn al-masāʾil, 628.3-6. 48Ibn Mattawayh, Taḏkira, 707.11-14. 49For instance, Ibn Mattawayh, Taḏkira, 701.16-20 and ʿAbd al-Ǧabbār, Muġnī, vol.4, 40.9-19.a maʿnā, a nonreductive theory of sense-perception follows automatically.The occasionalists establish an extensional distinction between sense-perception and the physical states of the organs.For them, it is possible that sense-perception occurs without those physical states.If that is the case, clearly, sense-perception cannot be reducible to the states of the sense organs.And yet a denial of the idea that idrāk is a maʿnā, common to the Basrian Muʿtazilites, does not amount to a reductive theory of sense-perception.The Basrian Muʿtazilites insist on the extensional identity of sense-perception and the physical states of the sense-organs (in the case of corporeal creatures), but they still agree that sense-perception is something intensionally different from those processes.
There is plenty of evidence that the Basrian theory of senseperception is nonreductive.Before Abū l-Ḥusayn al-Baṣrī, this evidence revolves around the causation of sense-perception.As I have mentioned in the end of the last section, the Basrians insist that the only real cause of sense-perception is the fact that the perceiver is alive.Whatever else must occur in the moment of perception, such as the soundness of the sense organs and the presence of the object of perception, are not the causes of sense perception, they are merely conditions.Ibn Mattawayh puts this doctrine as follows: tawayh, sense perception belongs to the whole, not to the sense-organs properly speaking.He develops this notion in the following passage: The sense-organ and its soundness cannot be the causal producers (muʾaṯṯira … taʾṯīr al-ʿilal) of [perception], since whatever qualifies (ḥukm) the sense-organ goes back to the parts.The perceiver, however, finds that his being a perceiver goes back to the whole (al-ǧumla). 52aving aside the nuances of the ontology of kalām behind the notions used in this passage, the core idea is that the sense organs cannot be the causes of sense-perception because they are not the subject of sense-perception.Rather, when a person sees an apple, we say that this whole person sees an apple, not any of her organs.The subject of senseperception is always one and the same, irrespective of which perception we are discussing.Therefore, it is the whole unified subject of perception that is the cause of its own perception, not the sense-organs.
Al-Malāḥimī picks it up where Ibn Mattawayh left it.In his preferred notions, being part of the living being is that which entails (muqtaḍī) that the perceiver perceives, all other things being just additional conditions. 53It is not just any kind of life inhering in the sense organs that entails perception.Rather, it must be one and the same life, belonging to the whole individual.Otherwise, Zayd could perceive with the hand of ʿAmr: there is life in the hand of ʿAmr after all. 54l-Malāḥimī makes it clear that what the Basrians have in mind when they talk about the unity of life being the muʾaṯṯir or the muqtaḍī of sense-perception is an active theory of sense-perception.Senseperception is an active process initiated by the living being itself, not a passive process of undergoing change in the sense-organs under the influence of external causes.Al-Malāḥimī explains this in his response to a dialectical opponent, who suggests that being alive is merely a necessary condition (muṣaḥḥiḥ) of perception and receives (qābil) perception from an external cause.Al-Malāḥimī disagrees.For him, it must be clear to every reasonable person that perception arises (yaṣduru) from the living being itself.That which necessitates (mūǧib) sense-perception is the fact that the living being is alive and not anything external to the perceiver. 55Al-Malāḥimī justifies his active theory of perception with the following argument in his Tuḥfa: 52 Ibn Mattawayh, Taḏkira, 700.8-9; same in al-Malāḥimī, Muʿtamad, 204.7-12 with a further explanation of how this argument works in terms of the ontology of kalām. 53Al-Malāḥimī, Muʿtamad, 204.15-16. 54Al-Malāḥimī, Muʿtamad, 203.3-8. 55Al-Malāḥimī, Muʿtamad, 207.4-23.
Suppose someone else speaks to us and we hear what he says, so that the sound made does have an effect on our organ of hearing.If the effect (taʾṯīr) of the sound in the organ of hearing were identical to our hearing what he says, then our hearing him talk would arise from the person speaking to us, since it would be him who brings into existence that which affects the organ of hearing.Yet every reasonable person knows that hearing him talk arises (ṣādir) from the one who hears, not the one who speaks.So, we know that our hearing [what he says] is something additional (amr zāʾid) to being made listen: it arises from the one who hears and is rendered necessary (wuǧiba) by the listener's being alive, through an instrument (āla), namely his organ [of hearing].That is why a speaker cannot be ordered that he makes Zayd listen to his talk, nor can he be prohibited from [making Zayd listen] or praised or blamed for it.So, we know that the hearing of speech is something additional to the effect of sound in our sense-organ.If this is true for the perception of sound, the same applies to other senses as well: there must always be an effect of the external [object of perception] with perception [still] arising from the living perceiver himself. 56-Malāḥimī argues in this passage that the cause of the presence of sense-perception is the perceiver herself, not the object of perception.Admittedly, whenever there is sense-perception there must be an object of perception.But sense-perception itself arises from the perceiver, not the perceived.Being a Muʿtazilite, al-Malāḥimī resorts to argumentation based on the intuitiveness of moral judgment. 57We cannot blame the speaker if the audience does not listen, according to al-Malāḥimī.Therefore, listening is an active act of the audience, not of the speaker.By talking, the speaker creates an effect upon the sense-organs of the audience.But the members of the audience are still the ones who produce their own act of listening, distinct from those effects in the organs.
This passage shows that the active theory of perception, present in the Basrian kalām long before al-Malāḥimī, is the building block of the Basrian nonreductive theory of perception.Al-Malāḥimī uses the active role of the perceiver to justify that sense-perception is something beyond (amr zāʾid) the effect from the perceived upon the sense-organ.Al-Malāḥimī also focuses on showing that sense-perception is something beyond whatever happens in the sense-organs in the Muʿtamad.His core argument goes as follows: When we perceive some visible object and then we know it, we know a self-evident difference between the state when we know that visible object 56 Al-Malāḥimī, Tuḥfa, 76.2-11. 57Cf. similar argumentation in favour of the identification of the human with the body in my forthcoming "First-Person and Third-Person Views in Arabic Philosophy of Mind." without seeing it and when we know it while we are looking at it.That kind of difference cannot go back to the impact of the visible object upon the eye.Even if we accepted that there is an impact, it would fail to be self-evident or known directly.We could only know it through a hidden inference.On the contrary, the difference that we find, between the state when we know a visible object without seeing it and while seeing it is, is something known directly: every intelligent person finds it in himself, and so does even every reasonable child whose intellect has not properly formed yet.But that which is known directly cannot be identical to that which is known through an inference!So, if we postulate an impact of the perceived object upon the sense organ then it needs to be a condition for the additional item (al-amr al-zāʾid) that we have already assessed [sc.active perception], and it would not be any different from other conditions proper to us, such as the senseorgan, the absence of impediments and so on. 58-Malāḥimī insists in this passage that even if we accept that there is some effect in the sense-organ when we see something, senseperception won't be reducible to it anyway.His reasoning is based on identifying different levels of knowledge.The difference between seeing something and not seeing it is self-evident to the observer.But what processes happen in the eye is not something self-evident.We can perfectly know that we are seeing something and understand how our state of seeing something is different from not seeing it without knowing anything about the physics of the vision.Therefore, sense-perception as such is something distinct from the physics of sense-perception.
This argument is al-Malāḥimī's version of what we call nowadays "Knowledge Argument."The argument is based on the conceptual difference between sense-perception as such and the physics of senseperception hiding behind it.According to al-Malāḥimī, this conceptual difference is sufficient to prove the distinction between sense-perception as such and the effect of the perceived upon the sense-organ.Al-Malāḥimī sees no difference between saying that sense-perception is distinct from the effect of the perceived upon the sense-organ and saying that it is distinct from "the knowledge about the effect of the perceived upon the sense-organ."59For him, there is no difference, or at least the one implies the other. 60n my understanding, the Basrian position presupposes a supervenience model of sense-perception.The Basrian way of proving that sense-perception is distinct from the state of the sense-organs is different from what we saw in the previous section on behalf of the occasionalists.As we saw, the Ašʿarites and some early Muʿtazilites accept an extensional distinction between sense-perception and the presence of an effect upon the sense-organ.The former can exist without the latter.The Basrian Muʿtazilites, in their turn, accept only an intensional distinction.The conceptual content of sense-perception is distinct from whatever we understand about the processes in the sense-organs.Still, the former cannot exist without the latter; they are extensionally identical (at least in the case of the corporeal living beings).As we saw, the Basrians insist that the physical conditions, including the effect upon the sense-organs, must be fulfilled in order that an appropriate item of sense-perception arises from the perceiver.Without those physical conditions, the perceiver won't be capable of actively producing any sense-perception at all.Sense-perception is conceptually distinct from the effects upon the sense-organs, but it still is supervenient upon them.
Unfortunately, neither al-Malāḥimī nor anyone else among the Basrian Muʿtazilites explains what sense-perception is, if it is not identical to whatever happens in the sense-organs.We saw above that it is not a maʿnā, meaning that it is not an independent factor in the formation of sense-perception (rather, it is caused by the living being as a whole).But what is sense-perception then?Al-Malāḥimī just continuously insists that it is something (amr) in the passages quoted above.Other Basrian authors put it in terms of a state (ḥāla) or just an attribute (ṣifa) of being a perceiver, belonging to the subject of perception. 61As we saw in the first section, one possible way to understand sense-perception is through the notion of conscious experience.Indeed, there are signs that al-Malāḥimī might be tempted to understand it that way.For instance, altogether with Abū l-Ḥusayn, he rejects that the sleeping person can perceive outer sounds without knowing them, the position we saw on behalf of al-Isfarāʿīnī in the first section.According to al-Malāḥimī and Abū l-Ḥusayn, one option is that the sleeping person knows about those sounds, and the other option is that she does not perceive them. 62Just like I suggested with respect to the similar alternatives in the analysis of al-Anṣārī, the idea might be here that sense-perception must involve conscious experience.That is why it is impossible to perceive without knowing the object of perception.Still, it needs to be noted that Ibn Mattawayh argues in favour of the distinction between knowledge and sense-perception based on the same argument of a sleeping person. 63o whether understanding sense-perception as conscious experience is common to all the Basrians remains open based on the available sources.

CONCLUSION
In conclusion, I will present a map of different theories of senseperception in kalām.They are far from being a subject to any simplistic division into, say, an Ašʿarite and a Muʿtazilite theories of perception.Rather, we can identify a few core topics discussed in kalām with respect to sense-perception and try to divide individual authors into groups in accordance with their position on those topics.
The first topic is the nature of sense-perception, for instance, its relation to conscious experience.We have seen that the position that senseperception either is identical to being aware (šuʿūr) of the subject of perception or at least implies it is common among the scholars of kalām.A typical issue discussed in the context of this topic is whether a sleeping person is aware of the sound that wakes them up, or she is not aware of it and, hence, does not perceive it at all.The authors who connect sense-perception with conscious experience are clearly al-Kaʿbī and al-Isfarāʿīnī, and possibly also Abū l-Ḥusayn al-Baṣrī.The authors who are resistant to this connection are al-Baqillānī and possibly also Ibn Mattawayh (and, hence, ʿAbd al-Ǧabbār), al-Ǧuwaynī and al-Anṣārī.
The next topic, closely related to the first one, is the question where sense-perception takes place.Two options are on the table .One is what can be called a "unified theory of sense-perception."According to this theory, sense-perception belongs to one and the same subject.It may be the heart or the brain, according to some theories, or it is the whole person (al-ǧumla) herself.A few scholars of kalām endorse the unified theory of sense-perception.Abū Huḏayl and al-Kaʿbī are clearly among those who argue that sense-perception happens in the heart.The Basrian Muʿtazilites, including ʿAbd al-Ǧabbār, Ibn Mattawayh and Abū l-Ḥusayn al-Baṣrī, in their turn, argue that sense-perception belongs to the living person as a whole.This theory originates as early as Bišr b. al-Muʿtamir (d.825). 64The alternative option is that sense-perception happens in the sense-organs.Based on the analysis of al-Anṣārī and al-Šahrastānī, we can suggest that most Ašʿarites appear to accept that 63 Ibn Mattawayh.Taḏkira, 697.11. 64Ibn Mattawayh, Taḏkira, 698.17-18.
sense-perception happens in the sense-organs.
The next topic, again closely related to the previous one, is the issue of causation.The scholars of kalām put it in terms of the question whether perception (idrāk) is a maʿnā, an independent factor causing the perceiver to perceive.If it is a maʿnā, then it does not depend on the presence of any conditions inside or outside the perceiver.God can create a maʿnā of sense-perception at will.This position is commonly accepted among the Ašʿarites, but also among some Muʿtazilites, such as Abū Huḏayl. 65Most Muʿtazilites, however, especially the Basrian school, including ʿAbd al-Ǧabbār, Ibn Mattawayh and Abū l-Ḥusayn al-Baṣrī consistently reject this position.According to them, senseperception is caused by the living person as a whole (ǧumla) if all internal and external conditions are satisfied.I labelled their position as an "active theory of sense-perception." There is only one topic on which there is no disagreement among the scholars of kalām: whether sense-perception is reducible to whatever happens in the sense-organs on the physical level.My central thesis in this article is that all theories of sense-perception in kalām imply a nonreductive theory of sense-perception.In fact, I suggest that all the above doctrines have been designed in order to express nonreductive theories of sense-perception, one way or another.Whether we identify sense-perception with awareness, posit a unified subject of sense-perceptions who causes its own sense-perception, or make sense-perception causally independent from whatever happens in the sense-organs on the physical level: all that naturally leads to saying that sense-perception is one thing, and the physical state of the senseorgans is another thing.To see the validity of my hypothesis it suffices to open the tables of content of the sources mentioned in this article and see that the scholars of kalām develop all those theories of senseperception mentioned above in the context of their discussion whether sense-perception can belong to God.Thus, the aim of their discussion is to establish that a non-corporeal being, God, can have sense-perception, even if it does not have sense-organs.The natural way to do so is to argue that sense-perception is something different from whatever happens in the sense-organs. 66o finish this article, I would like to make a hypothesis about the place of the kalām-theories of sense-perception in the history of philosophy of mind in the Islamic world.In my recent article, I argued that Abū l-Barakāt al-Baġdādī (d.1165) develops what I call "a unified theory of perception."This theory suggests that perception belongs to one and the same subject of perception, the incorporeal "I."The main thesis of Abū l-Barakāt is that one does not require sense-organs for sense-perception.Abū l-Barakāt develops his position in opposition to the Avicennian theory of sense-perception, since the latter claims that sense-perception consists in the inherence of the representation of the sense-object in the sense-organs, according to Abū l-Barakāt's interpretation. 67 suggest that Abū l-Barakāt's theory of sense-perception has been informed by the nonreductive theories of perception in kalām.It is not entirely clear which specific theory was known to him (maybe all of them), but it is easy to see common elements between Abū l-Barakāt's understanding of idrāk (perception) and that of kalām.Abū l-Barakāt's preferred understanding of perception as awareness (šuʿūr) was already available in kalām.The idea of the unified subject of perception, constantly repeated by Abū l-Barakāt, is clearly present in kalām as well.Various scholars of kalām insist that perception belongs to one and the same subject of perception, whatever it is, and not to the sense-organs.They consistently label the sense-organs as the instruments (ālāt) and means (wasāʾiṭ) of perception, not the subjects of perception, a move highly characteristic of Abū l-Barakāt. 68Moreover, Abū l-Barakāt's core agenda, that the (alleged) Avicennian identification of perception with the effect of the sense-object upon the sense-organ is false, is clearly a topic in kalām as well.Al-Malāḥimī presents his active theory of senseperception, which we saw in the last section, as a response to the Avicennian position, just like Abū l-Barakāt did.Now, one could object that this topic might be something specifically post-Avicennian, since the dates of the composition of Tuḥfa (between 1137 and 1141), the treatise in which al-Malāḥimī goes against Avicenna, are sufficiently close to Abū l-Barakāt's Muʿtabar (between 1140 and 1155). 69But according to alself can be seen.For instance, al-Anṣārī, Šarḥ al-Iršād, vol.2, 300.17-301.4 and al-Šahrastānī, Nihāya, 308.4-14 explain that the Ašʿarite nonreductive theory of perception (identifying it with a type of knowledge, that is, with conscious experience) helps establish the Ašʿarite position that God can be seen.Malāḥimī's own report in the Muʿtamad, Abū l-Ḥusayn al-Baṣrī already problematized whether perception is anything distinct from the effect of the perceived upon the sense-organ (taʾṯīr al-mudrak fī l-ḥassa), without ascribing the latter position to anyone, just as a possible objection to the active theory of perception, endorsed by the Basrians. 70If this report is correct, we can suppose that the issue had been already discussed long ago before Abū l-Barakāt was writing his refutation of the Avicennian theory of perception, meaning that he was likely to be influenced by its discussion in kalām.
Thus, I would like to suggest the following narrative about the fate of various theories of sense-perception on the brink between pre-Avicennian and post-Avicennian philosophy in the Islamic world.When the authors of kalām encountered Avicennian treatises, they interpreted his theory of perception as a reductive theory of perception, which identifies sense-perception with the effect of the sensed object upon the sense-organs.None of the traditional kalām-theories of senseperception could accommodate this position, all of them being nonreductive theories of sense-perception.As a result, the post-Avicennian authors, such as al-Malāḥimī and Abū l-Barakāt, opposed the Avicennian theory of perception, based on the theoretical elements of the kalām-theories of sense-perception, and there are strong hints that the nonreductive approach to sense-perception prevailed. 71This, however, remains a question for further research.
Al-Malāḥimī, Muʿtamad, 441.21-22.This position is perfectly in accordance with the Bahšamite definition of knowledge that p, which needs to exclude the possibility that it is not p; see further F. Benevich, "Knowledge as a Mental State in Muʿtazilite 44Al-Malāḥimī, Muʿtamad, 434-435.On the notion of tanbīh in the Basrian epistemology see Benevich, "Knowledge as a Mental State in Muʿtazilite Kalām."