It is a widely accepted idea that we are accountable for what we do. In Muslim thinking it is even deeply rooted in a worldview that is shared by the two other Abrahamic religions, where moral accountability is vitally linked to the belief in the Last Judgement. Man's individual responsibility for his actions is consequently regarded as a crucial feature of the relationship between God and His creation. It is however difficult to accept – and even more difficult to plausibly establish – individual responsibility for our doings without presupposing that our actions are determined by our very own self. Yet human self-determination can be conceived in various forms. The mutakallimūn, that is, theologians who attempted to rationalize and systematically explain their doctrines, also developed different approaches to account for why our acting should be self-determined in some way.Footnote 1
I. SELF-DETERMINATION AS FREEDOM OF ACTION
Many of us would intuitively affirm that moral responsibility is related to freedom of action: if we are individually accountable for what we do, we assume that we have control over our actions, and we think that it is up to us to decide whether and how we act. We then presuppose that morality is only possible if freedom of action is true.
Essentially, this was also the prevailing consensus among Muʿtazilite theologians. For them, God's justice was a fundamental pillar of their teaching and they believed that His judgement of man's actions is founded on objective principles and values. Based on these assumptions, they argued that in no way does God create and determine human acts; for if we are accountable for what we do, if we justly deserve reward or punishment, we must be the originators of our acts and have the capacity to behave otherwise than we do.Footnote 2 Accordingly, for the Muʿtazilites human self-determination implies that our acts causally depend upon us, and that the capacity by virtue of which we produce our acts is a power over alternatives: it enables us to perform two contrary acts and also allows us not to act at all (al-istiṭāʿa […] hiya qudra ʿalayhi wa-ʿalā ḍiddihi wa-hiya ġayr mūǧiba li-al-fiʿl).Footnote 3
While there was much agreement about man's capability to causally determine his acts on account of his power (qudra), early Muʿtazilite theologians did not necessarily share a unanimous conception of what precisely is meant by human power.Footnote 4 Another question they discussed was whether our acting can be explained on the sole basis of our capability to act or whether our freedom of action – that is, our autonomous choice between various possible options of behaving – depends on something else.Footnote 5
During the fourth/tenth century that primarily concerns us here, the Muʿtazila was no longer an intellectual endeavour of merely independent thinkers, but consisted rather of various well-established sub-schools. The predominant teaching was that of the School of Baṣra, which fundamentally relied on doctrines developed, structured and systematised by Abū ʿAlī al-Ǧubbāʾī (d. 303/915) and his son Abū Hāšim (d. 321/933). The school's theory of the human act is expounded in the works of one of its chief theologians, ʿAbd al-Ǧabbār al-Hamaḏānī (d. 415/1025), and further detailed by a number of his students.Footnote 6 In their teaching, human capability to act also fulfils a crucial purpose. By the notion of qudra – or more precisely, by its plural, qudar – they referred to entities (maʿānī) of “power” subsisting in the human body on account of which we are capable of acting. They maintained the fundamental principle, already formulated by the earlier Muʿtazilites, that our capability of acting in no way implies any necessity to act. Instead, it empowers us to choose between various alternatives: if we act, it is also our option whether to do the opposite or not to act at all. Any human action is consequently the agent's autonomous decision and therefore determined by his very own self. They argued that this is firmly established by our common experience that we act in accordance with our intentions (qaṣd).Footnote 7
According to the Baṣran Muʿtazila, our intentional acts do, however, not occur by virtue of our will (irāda) or motivation (dāʿī). They even believed that the actual performance of our actions cannot depend on our motivations without violating the idea of our actions being free. They argued that if human actions were not solely grounded in one's capability, but causally depended on something supplemental to one's power – such as motivations –, man could no longer be considered as an autonomous agent.Footnote 8
In fact, the Baṣran Muʿtazilites developed a concept that rather appears to contradict the principle outlined above: they posited the existence of motivations that are so strong that the agent cannot but act in a certain way. In the technical language, they framed this idea with the term ilǧāʾ.Footnote 9 The actual purpose behind the Baṣran theory was, however, that motivations account for why some actions are more likely or more reasonable to expect (awlā) than others.Footnote 10 In other words, motivations are the condition for our doing something deliberately and not just randomly.Footnote 11 The notion of ilǧāʾ does therefore imply a fairly high degree of likelihood but no necessity. However, our acts do not need to be purposive: according to the Baṣran School, they can even lack any rational foundation and consequently be pointless (ʿabaṯ).Footnote 12
Finally, the Baṣran Muʿtazilites even posited that our autonomous acts can be entirely non-voluntary: they argued that a sleeping or unconscious man still determines what he does. According to them, we must be in a state of awareness and consciousness whenever we generate motivations. Referring to the sleeper and the unconscious, they could then explain that this is not always the case with human agents, and so conclude that the actual performance of our acts cannot depend on the presence of motivations.Footnote 13 The upshot of this theory was that, according to the Baṣran School, we can self-determine our behaviour through exercising freedom of action without deliberation. For them, freedom of action is consequently even possible without voluntariness.
The Baṣran Muʿtazilite theory was not uncontroversial and posed a fundamental question: if freedom of action means that we have various alternatives of action, and if it is true that these alternatives are possible and become actual only by virtue of our power to act, would this not lead to the inconceivable conclusion that two contrary acts occur at the same time? The objection was not only raised by such detractors as Abū al-Ḥasan al-Ašʿarī (d. 324/935), the founder of the Ašʿarite school, and his follower Abū Bakr al-Bāqillānī (d. 403/1013),Footnote 14 but was also identified as a serious problem by a number of Muʿtazilite theologians. These critics from inside the Muʿtazilite school therefore pointed out that freedom of action actually presupposes voluntariness. For example, Abū al-Qāsim al-Kaʿbī al-Balḫī (d. 319/931), who was the head of the Muʿtazila of Baghdad and opposed his contemporary Abū ʿAlī al-Ǧubbāʾī in a number of theological issues, is said to have affirmed that the occurrence of our actions depends on our will in that our will causes our acts to happen.Footnote 15 We have no sources that provide us with further details how he supported his theory. Yet some other Muʿtazilites adopted a similar standpoint, including ʿAbd al-Ǧabbār's student Abū al-Ḥusayn al-Baṣrī (d. 426/1044) and his later follower Rukn al-Dīn Ibn al-Malāḥimī (d. 536/1141), whose teaching has survived in several important theological works.
As we are told by Ibn al-Malāḥimī, Abū al-Ḥusayn posited that acting is impossible without motivation. He supported his theory by referring precisely to the quandary of man's capability equally to produce two contrary acts. Human power, he argued, makes two opposite acts possible to exactly the same degree. This principle was the foundation of the Muʿtazilite belief in freedom of action. If our capability favoured alternative A over alternative B, our freedom of action would be seriously threatened. Yet only one possibility can be actualised, and so Abū al-Ḥusayn concludes that something else is needed for us to exercise our freedom of choice – something that accounts for why either possible action happens. Abū al-Ḥusayn identifies this something with the agent's motivation (dāʿī) for the act of his choice or – in the case of various conflicting motivations – the preponderant motivation (taraǧǧuḥ). Against his teacher ʿAbd al-Ǧabbār, he maintained that this is even true for the sleeper and the unaware: although their actions do not follow the same logic as if they were conscious, what they do is still coherent within their actual experience – such as their dreams – and has therefore a purpose. We might forget our motivations once we regain consciousness, we might even act under false assumptions and, as a result, fail to achieve an outcome we actually intended by our actions – in any case, however, the occurrence of our acts depends on the presence of at least one motivation to do what we do, irrespective of how it comes about.Footnote 16
Abū al-Ḥusayn's claim for the need for motivations also affected his conception of the agent (fāʿil). According to the definition found in Taṣaffuḥ al-adilla, the agent causes his acts in a way that is not necessarily effective (al-muʾaṯṯir ʿalā ṭarīq al-ṣiḥḥa). Essentially, this is the Muʿtazilite principle outlined above expressed in other words. What distinguishes Abū al-Ḥusayn's conception is how he explains the non-necessity of what we do. For him, being not necessarily effective means to make an effect take place according to one's abilities and motivations. However, if we affirm that the occurrence of our actions does not depend on our being motivated to do them, we have to concede the possibility that agents do not cause things to happen in accordance with their motivations. Yet, Abū al-Ḥusayn concludes, this would corrupt the agent's very being.Footnote 17
Abū al-Ḥusayn's discussion of the real nature of the agent is found in the context of the question whether God has power (qādir) to do evil. The issue was highly controversial, not only between different theological strands, but also within the Muʿtazilite school itself, since it touched upon the veracity of two fundamental doctrines: the principles that God is almighty and that He is good. Abū al-Ḥusayn's solution to the problem reveals much about his understanding of free agency and the process of decision-making involved. He affirms that God actually has the capability to do evil, thereby doing justice to His all-encompassing power. Nonetheless, he says, it is impossible that God turn His ability to do evil into real action. The reason behind this is His reluctance and lack of motivations to do so (min qibali al-ṣawārif wa-intifāʾ al-dawāʿī).Footnote 18 This is indeed consistent with Abū al-Ḥusayn's analysis of what it means to be capable: namely that one has the ability not only to act but also to omit the act.Footnote 19
Abū al-Ḥusayn compares the function of motivations with our need to have tools at our disposal allowing us the performance of certain acts: we need a needle to sew and a pen to write. Even if we have the ability to sew and to write, it is not possible for us to do so (lā yaḥṣulu maʿahu al-ṣiḥḥa) as long as we do not have access to these tools. The fact that we lack these tools, however, does not affect our ability in itself. In line with this idea, Abū al-Ḥusayn stresses the difference between stating that an act is impossible (yastaḥīlu) for somebody and that somebody has no capability to do it. Based on this assumption, he goes on arguing that, similarly, an agent who is capable of performing a certain act depends on his motivations. Yet in no way does this conflict with his actions being self-determined. Abū al-Ḥusayn supports this argument by comparing somebody capable of actions with somebody who is not (ʿāǧiz). Whoever is reluctant to do that which he is capable of doing will act once his motivations change. In contrast, it is not up to us to cause acts that are beyond our capacities, irrespective of whether we want them to occur or not.Footnote 20
In conclusion to this brief sketch of Muʿtazilite positions, it appears that their theologians agreed on basing human self-determination on freedom of action. Beyond this consensus, however, they proposed different explanations for why our actions should be free. In the fourth/tenth century, the Baṣran Muʿtazila represented a trend that identified freedom of action with a power over alternatives. For them, this capacity is sufficient for our actions to be free. Others pointed out that exercising freedom of action also depends on the agent's decision-making, something that is not implied by the mere capability to behave in different ways. They argued that free agency always requires intentionality, which accounts for why one of several possible acts happens. Abū al-Ḥusayn al-Baṣrī followed this logic and therefore considered that our actions depend on motivations without which exercising freedom of action would be impossible.
II. SELF-DETERMINATION AS VOLUNTARINESS
1. Al-Aš ʿarī: Moral Responsibility in the Absence of Freedom of Action
Freedom of action was not only subject to internal debates between theologians who believed in that principle. The idea as such was highly controversial and categorically rejected by many opponents. The Muʿtazilites had, however, advanced a very clear explanation for why we have control over, and consequently are responsible for, our acts. Whoever wanted to question their theory on the same level of rational plausibility would therefore have to put forward an alternative conception of human self-determination. An important position against freedom of action was developed on the basis of the doctrines of Abū al-Ḥasan al-Ašʿarī, a former Muʿtazilite whose teaching laid the foundation for the Ašʿarite school of kalām.
The point of departure for al-Ašʿarī's reasonings on human acts was the doctrine of divine omnipotence. He claimed that God's absolute power is in no way restricted: it encompasses every creation and so all happenings in the world, even those proceeding from others, must depend on Him. Were this not the case, God would have to be conceived as weak and powerless.Footnote 21
Al-Ašʿarī's understanding of God's omnipotence directly affected his conception of human acts. Since they belong to the temporal world, they must also be determined by God: in this sense, al-Ašʿarī posits, acts of disobedience are created and foreordained by God.Footnote 22 Similarly, if piety was not created by God and the believer could also disbelieve, God would necessarily be unable to impose His power upon His creation, and this would in turn fundamentally violate the idea of His omnipotence.Footnote 23 Believers and unbelievers therefore act without any possibility for them to act differently: “If the unbeliever were capable of believing,” al-Ašʿarī says, “he would believe.”Footnote 24
Al-Ašʿarī consequently shared the Baṣran Muʿtazilites’ view that divine determinism is incompatible with human freedom of action. Departing from a common premise they arrived at diametrically opposed conclusions, however. Since the Muʿtazilites’ primary concern was not to violate the principle of God's justice, they assumed that He refrains from exercising His omnipotence in the realm of human acts, thereby giving us full control over how we act. On the other hand, al-Ašʿarī's main preoccupation was God's omnipotence, and so he argued that there is no creator (ḫāliq) and no agent (fāʿil) apart from God.Footnote 25 For him, the idea that everything occurs by the divine will means that human free agency cannot be true: according to his view, God alone creates our actions and so al-Ašʿarī denied that we have the capacity to act otherwise than we actually do.Footnote 26 He consequently was what we would nowadays call a hard determinist.
By adopting this line of reasoning, al-Ašʿarī inevitably faced a fundamental objection. As a former Muʿtazilite he knew all too well that determinism poses a serious threat to morality, since morality involves individual responsibility for our own actions. Therefore, what we are held responsible for must in some way be within our control. But how then could moral responsibility be true if, according to al-Ašʿarī, it is not up to us whether or not we act as we do? In order to confirm the validity of determinism, he had to disprove the principal assumption that moral responsibility really depends on our actions being free. He therefore developed an alternative way of understanding human self-determination, a way which differs from its Muʿtazilite conception as free and genuine creation of our acts.
Appealing to our experience, al-Ašʿarī argues that human actions are not all of the same kind. Common sense shows us that there is a difference between such motions as trembling and walking: we feel that our trembling occurs necessarily (iḍṭirāran ), while this is not the case with our walking.Footnote 27 These two motions are, however, not distinguished on account of who produces or initiates the act. Both have a temporal existence: they come into existence after their non-existence and at some point cease to exist. As previously outlined, their creator therefore cannot be other than God in al-Ašʿarī's view. But what then distinguishes the “necessary” act from other acts? For al-Ašʿarī the necessity of our trembling involves our weakness (ʿaǧz). As a logical corollary, the opposite must be true for all non-necessary actions: they involve our power (qudra or quwwa). Such acts, which, in al-Ašʿarī's words, “occur on account of a created power” (waqaʿa bi-quwwatin muḥdaṯa) are denoted as kasb or iktisāb, usually translated as “acquisition”.Footnote 28
Al-Ašʿarī was not the first to refer to human acts as kasb and iktisāb. Rather, he built on the ideas of earlier thinkers who tended to minimise or even completely deny a human capacity of creation. It is possible that these theologians believed the appropriateness of this terminology was supported by its frequent occurrence in the Qurʾān, where the verbs kasaba, iktasaba and their maṣdars are used in the meaning of “to do” or “to perform an act”. More precisely, these terms are mainly employed when referring to us acting whilst being held accountable for fulfilling or neglecting obligations, duties and prohibitions imposed by God.Footnote 29 In the theological context of describing and analysing human acts, the terms eventually appear to have been introduced into the technical vocabulary by Ḍirār b. ʿAmr (d. c. 200/815). He developed the theory that all our acts are created by God, while our role is restricted to “acquiring” them. This led him to the much-debated conclusion that every human act has two agents, namely God who produces the act and man who “acquires” it. Ḍirār's theory was modified only a little later by al-Ḥusayn b. Muḥammad al-Naǧǧār (d. c. 220/835). Al-Naǧǧār also posited that we “acquire” acts created by God, but for him only man himself is the agent of any such actions. Ibn Kullāb (d. c. 240/854), who, in many respects, was a precursor of al-Ašʿarī's teaching, also used the term kasb, but we do not have a clear account of precisely how he employed it.Footnote 30
Al-Ašʿarī took these reflections as points of departure for his own theory. Being concerned to do justice to God as the all-encompassing creator, the terminology derived from the root k-s-b helped him to speak about human acts and, at the same time, to avoid asserting that these acts are brought into existence by man himself, as implied by such verbs as faʿala or – the even more controversial – ḫalaqa. In addition, the Qurʾānic connotation of kasb/iktisāb introduced the very central aspect of morality. This semantic nuance was crucial for al-Ašʿarī's theory, since for him, human beings are only responsible for “acquired” acts and not for “necessary” acts.Footnote 31
In order to properly understand al-Ašʿarī's position, it still needs to be explained on what basis he holds us accountable for such acts, if it is not us who actually originate them. Although al-Ašʿarī's extant writings do not directly answer this question, a passage from his Lumaʿ helps us to make sense of his line of reasoning. In the following quotation, al-Ašʿarī analyses the relation between our actions and our will:
[W]hen an unwilled act of a man takes place, it must be the result of unmindfulness, or weakness and feebleness, or failure to attain his desire. […] That is so because the reason which enforces the man's weakness and failure to attain his desire, when he knows what proceeds from him but does not will it, is that what he wills does not take place and that he did not will what does take place. For if what he wills takes place, he is not overtaken by weakness and feebleness; but if it does not take place, he is overtaken by feebleness and failure to attain his desire, because it proceeds from him while he knows it but does not will it.Footnote 32
The discussed scenario of somebody failing to do what he actually wants describes a pattern which is echoed in al-Ašʿarī's portrayal of the “necessary” act. In this context, “necessity” (ḍarūra) is defined as:
that to which the thing is constrained and compelled and forced, and from which it can find no way to get free or to escape, even though it strive to be freed from it and want (arāda!) to escape from it and exhaust its endeavors to do so.Footnote 33
It is true that in the above-quoted passages al-Ašʿarī does not refer in consistent terms to the “weakness” or “feebleness” that accounts for why we may act against our willing and wanting. Whereas in the first case he appeals to our “weakness” (ḍaʿf) and “feebleness” (wahn) to explain why in some cases we consciously do things that oppose our will, he denotes the “weakness” involved in “necessary” actions by the term ʿaǧz. Nonetheless, he obviously refers to the same circumstances of acting: despite terminological inconsistency, one could hardly consider how the act discussed in the first quotation should be perceived by its agent as a non-necessary act. Yet it is precisely on the basis of the agent's individual experience of acting under compulsion that al-Ašʿarī establishes the “necessary” act as opposed to “acquired” acts.
What al-Ašʿarī here suggests against the Muʿtazilites’ conception of self-determination as freedom looks consequently very much like basing morality and responsibility on voluntariness: whenever performing an “acquired” act, we act as morally responsible agents, because we do things according to our willing and wanting. On the other hand, we cannot blame people for actions they do on account of their weakness and against their willing and wanting them to happen. The Baṣran Muʿtazilite principle that freedom does not presuppose voluntariness is thereby turned upside down: for al-Ašʿarī we can act voluntarily without having freedom of action – i.e. without any possibility to act otherwise than we do.
The fundamental assumption that our actions can still be voluntary even if we cannot omit them might be far from straightforward and even be in apparent contradiction to our common thinking. That both ideas can be harmonised is, however, well illustrated by a famous example that helped the 17th-century philosopher John Locke to demonstrate that voluntariness and necessity are not opposed to each other. That example concerns a man who stays in a room of his own volition, while unbeknownst to him the door is locked. Although he is not able to do otherwise and leave the room, he is still acting according to his volition. Therefore, his staying in the room is, in this sense, a self-determined act exercised through voluntariness.Footnote 34
If we interpret al-Ašʿarī's accounting for moral responsibility along this line of reasoning, we still face a number of unresolved questions. Most of these issues turn around the precise conception and function of our “power” (qudra) within his theory. Since for al-Ašʿarī all human acts are God's creation, there is no causal connection but only conjunction between the qudra and the “acquired” act. But how do we then have to understand his definition of “acquired” acts as “that which occurs on account of a created power” (mā waqaʿa bi-qudratin muḥdaṯa)? What, in particular, is the meaning of the particle bi-, considering al-Ašʿarī's denial that man is in any way capable of causing his own acts? If we perform “acquired” acts whenever our will conforms with how we act, what is al-Ašʿarī's ontological conception of the will – and why does he rely on the concept of qudra at all?Footnote 35
We cannot be sure whether or not al-Ašʿarī ever addressed these questions. Even though they are not discussed in his extant writings, we have to bear in mind that the vast majority of his extensive work is missing while only a handful of treatises have survived. What we can reconstruct on the basis of some few original writings and the additional later accounts of his theology is therefore not a coherent theory. However, later representatives of al-Ašʿarī's school took his thoughts as their point of departure for further analysis of human acts. I want to focus in the following on a third-generation Ašʿarite, Abū Bakr al-Bāqillānī, who can rightly be described as one of the major protagonists of the school's scholastic consolidation.
2. Al-Bāqillānī: The Effectiveness of Human Power
Essentially, two extant works of al-Bāqillānī's theological writings provide us with substantial information as to how he drew on and further developed al-Ašʿarī's theory of “acquired” acts. The first of these two works, the Kitāb al-Tamhīd, bears witness to al-Bāqillānī's attempt to systematically compile and coherently organise the teachings of his predecessors.Footnote 36 It has been convincingly argued that this book was in fact one of al-Bāqillānī's early works, possibly written around 360/970.Footnote 37 Our second source, which must have been one of al-Bāqillānī's last works, is his magnum opus in theology, Hidāyat al-mustaršidīn.Footnote 38 Within the four fragments that have survived from this multi-volume summa, we find substantial parts of the discussions related to human actions.Footnote 39 Al-Bāqillānī's solutions of specific theological questions were not consistent throughout his life. It is in particular in the Hidāya that he looked for alternative approaches to such issues and thereby revised some of his earlier positions.Footnote 40 Our particular interest in the Hidāya consists here in the fact that al-Bāqillānī addresses a number of those abovementioned questions left unresolved in al-Ašʿarī's theory of human action. Yet the incompleteness of the text and damages to the manuscriptsFootnote 41 sometimes make it difficult to reconstruct his position in its full details.
Al-Ašʿarī's starting point for explaining our responsibility for certain acts is also central to al-Bāqillānī's approach: he adopts the distinction between “necessary” and “acquired” acts. Originally, the Hidāya contained a definition of the concept of “acquisition”, which is however lost and only referred to as occurring in a previous passage.Footnote 42 Anyway, the Tamhīd also includes a definition that draws on al-Ašʿarī's theory whilst introducing a new idea. According to this definition, “acquisitions” are acts, which, in contrast to “necessary” acts, are performed by agents who possess a power in the substrate of and simultaneously with the act (taṣarruf fī al-fiʿl bi-qudratin tuqārinuhu fī al-maḥall).Footnote 43 This is merely a reformulation of al-Ašʿarī's view. But what is particular about al-Bāqillānī's approach is that, apparently for the first time in the Ašʿarite literature, he ties the notion of “acquisition” to the agent's choice (iḫtiyār) and also once to the agent's intention or purpose (qaṣd).Footnote 44 Al-Bāqillānī's claim that power over “acquired” acts involves the agent's choice of this particular act is also repeated in the Hidāya as part of his extensive refutation of the so-called theory of tawallud, that is the idea that human beings have capability of acting outside their own body through causal chains:
[T]he “acquisition” is distinguished from the “necessary” [act] by virtue of the fact that power over [the act] subsists [in the agent] and that he in whom [the power] subsists is capable of [acting] and choosing it.Footnote 45
Al-Ašʿarī's original assumption that our moral accountability does not depend on freedom of action being true is consequently followed up by al-Bāqillānī. Suppose I choose to walk from A to B and act according to my choice. Although I am not the producer of this act in the Ašʿarite view, I still have decided to move intentionally, unlike when shivering from fever, about which I have no choice at all. Al-Bāqillānī's preference for the notion of choice might look like only a minor and arbitrary modification, unless we consider some further explanations, which help us to understand why he avoided the notion of the will that al-Ašʿarī had still used in the Lumaʿ.
Al-Bāqillānī belonged to a generation of Ašʿarites that made significant contributions to the consolidation of the school's teaching.Footnote 46 The permanent challenge by rival systems of thought certainly raised the theologians’ concern to achieve greater consistency in their doctrines. One result of the ongoing attempts at systematisation was that, against al-Ašʿarī's original suggestion, the conceptualisation of the will was not – or no longer – compatible with the notion of kasb. Accordingly, al-Bāqillānī rejects – first in the Tamhīd and later in the Hidāya – the notion that whether or not we “acquire” an act actually depends on our will being involved.Footnote 47 One explanation for this claim is al-Bāqillānī's view that the will itself is an “acquired” act.Footnote 48 Little imagination is therefore needed to anticipate the almost inescapable objection that if the “acquisition” of an act of will requires another act of will, this would lead to the unacceptable claim of an infinite chain of acts of will.
But al-Bāqillānī makes an additional point, which is based on his reflections on how the human will relates to reality. Our will (and similarly aversion), he argues, does not necessarily relate to the possible or to possible “acquisitions” (mā yaṣiḥḥu ḥudūṯuhu aw ḥudūṯuhu wa-iktisābuhu) but in some cases to the impossible (mā yastaḥīlu). Suppose somebody wants to perform – or, to keep the Ašʿarite terminology, to “acquire” – an impossible act. His will is then based on his conviction (iʿtiqād) and assumption (ẓann) that the actually impossible act is possible. This conviction or assumption is, however, not knowledge (ʿilm), since real knowledge only encompasses that which is actually true (ṣaḥīḥan ḥā [diṯan]). In contrast, convictions and assumptions can be wrong and consequently extend to the impossible. Indeed, the upshot of these premises is that whenever we know that something is impossible, we cannot will it (istaḥāla taʿalluq al-irāda wa-al-karāha bi-mā yuʿlamu al-ʿālim istiḥālat ḥudūṯihi). But as long as we do not know, but only assume that an act is possible, the act we want to perform is by definition not an object of knowledge (maʿlūm), and so may consequently be impossible.Footnote 49
Accordingly, al-Bāqillānī denies that our will accounts for how, or even necessitates that, we “acquire” acts (laysat bi-ʿilla li-wuǧūdihi wa-lā sabab mūǧib lahu). And because we sometimes fail to exercise our will, he says, there can only be one possible conclusion: the non-occurrence of an act we want at a given moment can only be explained by a lack of power (ʿadam al-fiʿl maʿ al-irāda dalīl ʿalā anna man lam yaqaʿ minhu ġayr qādir ʿalayhi).Footnote 50
As previously outlined, al-Bāqillānī's recourse to the notion of “power” (qudra) in the context of “acquisitions” was, in itself, not a new idea: al-Ašʿarī had already used it to distinguish between “necessary” and “acquired” acts. However, al-Ašʿarī's conception of human power remained obscure in various respects. Before I turn my attention to al-Bāqillānī's approach to these unresolved questions, it is worth pointing out first that his conception of the human qudra is built on the major tenets established by the school's founder.
There are essentially three features of human power by which al-Ašʿarī sharply distinguished his conception from that of the Muʿtazilites. For al-Ašʿarī's opponents, man's power must already exist before he acts. Otherwise, they argued, we would act out of necessity and not contingently as they claimed. Al-Ašʿarī, in contrast, posited that man has power only simultaneously with his act. While the Muʿtazilite principle of freedom of action implied that man's power enables him to two contrary acts or to an act and omission, al-Ašʿarī denied this doctrine. For him, human power only relates to a single act. Finally, al-Ašʿarī rejected the Muʿtazilite idea that human power can continue to exist for several instants of time.
Al-Bāqillānī agrees with al-Ašʿarī on each of these issues. He even presents many identical arguments to support these claims, first in the Tamhīd and later in the Hidāya. Against the Muʿtazilite thesis that man must be capable of action before he acts, al-Bāqillānī replies that we then have to concede the possibility of actions being done by incapables (ʿāǧiz) who lost their capacities at the moment they act.Footnote 51 If man had simultaneous power over two alternative acts, this would not only pose the logical problem that two opposed acts would necessarily happen at the same time: both contraries would even occur in the same substrate, which, al-Bāqillānī says, is inconceivable.Footnote 52 Finally, he rejects the continued existence (baqāʾ) of human power on the basis of the Ašʿarites’ categorical denial that accidents (aʿrāḍ) exist longer than one instant of time.Footnote 53
For a better understanding of how al-Bāqillānī attempted to solve some of the above-mentioned questions raised by al-Ašʿarī's theory, we have to ask about the function of human power within the framework of his theory of acts. While al-Bāqillānī's Tamhīd does not offer any further reflections on this question, he seeks to address it in the Hidāya. In a passage of this text, which defends the claim that man's power and his action must be simultaneous, al-Bāqillānī posits that the occurrence of acts depends on the existence of power (yaḥtāǧu al-fiʿl fī wuqūʿihi wa-wuǧūdihi ilā wuǧūd al-qudra ʿalayhi). This dependence is described by al-Bāqillānī as being analogous to the dependence between a predicate (ḥukm) and the ground (ʿilla) on account of which a predication is made or said to be true. In contrast to the relation between ʿilla and ḥukm, however, it is not man's power which necessitates his acts (al-qudra ġayr mūǧiba; lā yaḥtāǧu [al-fiʿl] ilayhi [=ilā wuǧūd al-qudra] li-yūǧada).Footnote 54
But if our actions do not occur by virtue of our power, how can it be true that man's acts depend on his power? Al-Bāqillānī's answer to this apparent contradiction was to claim that denying that our power causes the existence of acts does not prevent it from affecting our acting in some other way. Accordingly, he argued that the human power of “acquisition” actually has an effect (taʾṯīr). While al-Ašʿarī himself does not even use the term taʾṯīr in relation to man's power in his extant writings, some later Ašʿarites report that he completely rejected the idea of any such effectiveness.Footnote 55 According to al-Šahrastānī (d. 548/1153), this position was only later revised by al-Bāqillānī. He is said to have formulated the theory that acts are qualified by a property or attribute whenever their performance is conjoined by the existence of power in the agent.Footnote 56 Al-Bāqillānī's conceptualisation of the effectiveness of human power was, however, more complex, if not ambivalent. In the Hidāya he suggests three approaches to understanding how power affects our acting. One of them proposes that the agent himself is affected, while the two others attempt to examine the correlation between agents and their acts.
Al-Bāqillānī first takes into consideration how agents themselves are affected by their qudra. He departs from the notorious scenario on which the distinction between “necessary” and all other acts is based. As we have seen, this distinction was commonly used by Ašʿarites to prove that whenever we act necessarily, we do not possess the power by virtue of which we describe agents of “acquired” acts as powerful (qādir). Now al-Bāqillānī developed a different conception than al-Ašʿarī of that which is expressed, or referred to, by such affirmations as “he is powerful”. For al-Ašʿarī, this predication only refers to the existence of the qudra, that is an entity conceptualised as an accident (ʿaraḍ) that inheres in the agent. In contrast, al-Bāqillānī says in the Hidāya that such predications as “being powerful” refer to a real feature – a ḥāl (“state”) in the technical language – of the subject described as powerful. For him, the ḥāl of “being powerful” (kawnuhu qādiran ) and the existence of power (qudra) are not identical but reciprocally entail each other: the existence of power is evidenced (madlūl) by the ḥāl, which, in turn, is grounded in, and becomes actual by virtue of, the existence of power.Footnote 57 It is precisely in this sense that al-Bāqillānī describes the effectiveness of human power in relation to the agent himself: whenever man performs an acquired act, the qudra is the ground for a ḥāl or attribute (ṣifa) of “being powerful” and thus for a qualification that distinguishes him from compelled agents (taʾṯīruhā kawn al-qādir bihā qādiran ʿalā an yataḥarraka wa-yaskuna wa-yurīda wa-yanẓura wa-yaʿlima wa-yafkira. fa-yakūnu bi-kawnihi qādiran ʿalā ḥāl man lahu hāḏihi al-ṣifāt mufāriqan li-ḥāl al-muḍṭarr allaḏī laysa bi-q[ā]dir ʿalā an yataḥarraka wa-yaskuna wa-yurīda wa-yaʿlima).Footnote 58
This conception of the effectiveness of human power, however, does not account for why acts should be considered as ours. The proponents of human freedom of action, in contrast, provided a rather simple solution to this question: for them, our acts only occur by virtue of our power. They consequently argued that we are responsible for our acts because their occurrence is causally connected to us. From his Ašʿarite perspective, al-Bāqillānī found this explanation unacceptable, since for him it violates the claim of God being the all-encompassing Creator. He therefore rejects the notion that our acts are brought into existence by virtue of our power, however not at the cost of denying any relation (taʿalluq) between our power and our acts.
The concept of taʿalluq was frequently used in the terminology of kalām. Both Ašʿarite and Muʿtazilite theologians applied the notion of correlation to a number of properties or attributes (ṣifāt) and also to accidents (aʿrāḍ) in which these properties are grounded.Footnote 59 More precisely, the term was employed whenever the subject qualified by such a property is in some way related to another object or a target: assumptions or statements are made about something, and we perceive objects distinct from ourselves by sensual perception. The kalām theologians therefore reasoned that our making assumptions or statements and our perception imply a correlation with something else. As explained by al-Bāqillānī, the extent of that which relates to a particular property completely depends on the property in question. For example, an assumption (ẓann) is not based on sufficient evidence for being qualified as knowledge (ʿilm). While real knowledge implies that that which is known is true, assumptions can be wrong. As a logical corollary, that which we assume to be true may in fact not be real, and so al-Bāqillānī concludes that that which relates to assumptions is much more wide-ranging (awsaʿ, aʿamm) than that which can be known: unlike knowledge, that necessarily relates to the known as it really is (lā yataʿallaqu bi-al-maʿlūm illā ʿalā mā huwa bihi), we may also assume that something is other than it really is (ʿalā mā huwa bihi tāratan wa-ʿalā mā laysa bihi uḫrā).Footnote 60 The range of possible objects of perception is, in turn, even more restricted, since we can only perceive specific things that actually exist (al-idrāk fa-innahu lā yataʿallaqu illā bi-kāʾin mawǧūd).Footnote 61
With regard to human power, al-Bāqillānī claims that, in principle, our qudra may relate to whatever can be created and “acquired” (tataʿallaqu bi-mā ya ṣiḥḥu an yuḥdaṯa wa-yaṣiḥḥu [an] yuktasaba), further specifying that this excludes the eternally and the continuously existent (lā yaǧūzu taʿalluquhā bi-al-qadīm wa-bi-al-bāqī) – that is God and atoms – or two contraries (al-ḍiddayn) at the same time. Therefore, any considerations about how power affects our acting merely applies to actions of our limbs or mental acts (afʿāl al-ǧawāriḥ wa-al-qulūb), including motion, rest, acts of will and knowledge. In order to do justice to man's inability to create what he “acquires”, al-Bāqillānī adds the remark that the performance of such acts involves two powers, each of which relates to a given act in different respects (yataʿallaqu bihi qudratān ʿalā waǧhayn muḫtalifayn): God's power accounts for its creation (iḫtirāʿ) and man's power for its “acquisition”.Footnote 62
According to al-Bāqillānī, man's power must be suitable (ṣāliḥa) for a specific act, and there can only be a correlation on condition that we have power when our act comes into existence.Footnote 63 However, whenever these conditions are fulfilled, our power – or our “being powerful” – necessarily relates to the “acquired” act. Otherwise, al-Bāqillānī argues, acts performed by a powerful, a sick person and a powerless were alike.Footnote 64 As for the question about the precise nature of the relationship between man's power and his acquired acts, al-Bāqillānī provides two possible solutions.
In the first of his two approaches, al-Bāqillānī argues that positing the effectiveness (taʾṯīr) of human power does not necessarily imply the meaning that man's “acquired” acts are created and exist (iḥdāṯuhu wa-wuǧūduhu) by virtue of his power. Nor does it mean that the coming into existence of such acts entails that they come to have an attribute by virtue of man's power (taǧaddud [MS: taǧdīd] ṣifa tatbaʿu ḥudūṯahu wa-wuǧūdahu ṣāra al-muktasab ʿalayhā bi-al-qudra). Rather, al-Bāqillānī illustrates his view by drawing a parallel with the relation between knowledge and the known or sensual perception and the object perceived. Both have a real and knowable correlation with a specific object, even though knowledge and perception do not cause their objects to exist or to possess an attribute. Al-Bāqillānī concludes that it is precisely in this sense that the relation between man's power and his “acquired” act should be understood (ǧārin maǧrā […] fī taʿalluq al-ʿilm bi-al-maʿlūm wa-al-idrāk bi-al-mudrak fī annahu taʿalluq ṯābit maʿlūm maḫṣūṣ fa-in lam yakun maʿnāhu wa-taʾṯīruhu fī ǧaʿl al-maʿlūm wa-al-mudrak mawǧūdan aw ḥādiṯan aw ʿalā ṣifa tatbaʿu al-ḥudūṯ wa-ka-ḏālika al-qudra mutaʿalliqa bi-al-maqdūr taʿalluq maʿlūm maḫṣūṣ wa-in lam yakun maʿnāhu wa-taʾṯīruhu ǧaʿl al-maqdūr mawǧūdan aw ḥādiṯan bihi wa-ǧaʿlahu ʿalā ṣifa tābiʿa li-ḥudūṯihi).Footnote 65
Yet al-Bāqillānī's first response to the question how acts created by God are related to their human agents does not resolve the more fundamental issue, namely that of our individual moral responsibility. For the Muʿtazilites, we are accountable for what we do and fail to do because the creation and omission of our acts is determined by our very own selves. Accordingly, they argued that the Ašʿarite theory of action makes nonsense of morality. How could it be true that we are rightly and fairly praised and blamed for acts if it is not us who create them? And would not God oblige man beyond his capacities and unjustly reward and punish him for what is actually divinely created?Footnote 66
To answer this problem, al-Bāqillānī develops a different understanding of what is specifically subject to moral assessment in our acting. His solution is found in his second approach to conceptualising how man's created power affects, and relates to, his acts. Surprisingly, al-Bāqillānī appears, however, to contradict himself when he addresses the question of our individual moral responsibility: despite his previous denial, he now affirms that the human act comes to have an attribute on account of man's created power. As he further explains, it is to this very attribute that God's command, prohibition, promise, threat, praise, blame, compensation and punishment relate (maʿnā taʿalluq al-qudra al-ḥādiṯa bi-maqdūrihā wa-taʾṯīrihā fīhi an yaṣīra [MS: taṣīra] bihā ʿalā ṣifa tābiʿa li-ḥudūṯihi wa-bi-tilka al-ṣifa yataʿallaqu al-amr wa-al-nahy wa-al-waʿd wa-al-waʿīd wa-al-madḥ wa-al-ḏamm wa-al-ṯawāb wa-al-ʿiqāb).Footnote 67
Al-Bāqillānī's line of argumentation is consequently based on the assumption that individual moral responsibility does not mean that man is accountable for the very existence of his acts. He thereby neutralizes a central argument of the proponents of freedom of action. In affirming that “acquired” acts come to have an attribute by virtue of man's power, al-Bāqillānī then provides a second major component in his attempt to reconcile moral responsibility with divine determinism. Appealing to this attribute, al-Bāqillānī could argue that a real feature of “acquired” acts is determined by man's very own self although he does not create them. In line with the argument that compelled agents have no power, he could furthermore argue that it is precisely this attribute that distinguishes “acquired” from “necessary” acts.Footnote 68
The idea that acquired acts come to have an attribute by the effect of human power was already discussed in Gimaret's seminal Théories de l'acte humain en théologie musulmane. The earliest evidence for this thesis he was able to consult was an account found in Abū Ǧaʿfar al-Simnānī's (d. 444/1052) al-Bayān ʿan uṣūl al-īmān. In this work al-Simnānī posits that whenever we “acquire” actions, their coming into being is accompanied by an attribute that relates to man's power and will. This attribute is subject to God's command and prohibition, praise and blame. Al-Simnānī was al-Bāqillānī's student, but he nowhere credits these assumptions to his teacher. Gimaret could only suppose on the basis of later reports that this was also al-Bāqillānī's position.Footnote 69 These later sources include most importantly the works of al-Šahrastānī. He attributes to al-Bāqillānī the thesis that God creates our acts – say an act of moving – and leaves us control over the precise modalities (wuǧūh) of our acting. That is, whenever we “acquire” a movement created by God, we determine by our own self whether it is rising, sitting down, praying, etc.Footnote 70 Yet the passages from the Hidāya examined above confirm what Gimaret already suspected: that al-Šahrastānī does not reproduce al-Bāqillānī's original theory but rather adapts it to his own understanding of human agency.
The chapter on human “power of acquisition”, which is found in the Hidāya, does not end with a conclusion. This opens some room for speculation as to which of the three solutions al-Bāqillānī ultimately considered the appropriate answer to the problem of how our power affects our acting. Even if al-Bāqillānī's first answer does not necessarily conflict with the two others, the second and the third answer appear to be inconsistent, if not contradictory, and therefore demand some further clarification. A possible explanation of the problem could be that al-Bāqillānī only distinguished the three levels of effectiveness of human power from a logical point of view: the first level would then concern the agent himself in that he acts intentionally by virtue of his power of “acquisition”, the second level of effectiveness would connect him with his act, and the third would affect the act inasmuch as it is distinguished by some property from “necessary” acts.
A fragment of a sentence, which, due to manuscript damage, is decontextualized, seems to favour a different interpretation, however: it rather suggests that, according to al-Bāqillānī, either of these three modalities excludes the two others.Footnote 71 It seems, however, that the discussion of the three options is not meant to test out alternative hypotheses, of which only one can be confirmed while the two others have to be rejected categorically. Instead, al-Bāqillānī apparently supposes that, depending on some circumstances, human power is effective in any of the described ways. This understanding is confirmed by a passage from the section on God's attributes (Kitāb al-Ṣifāt) contained in the Hidāya, which explains possible meanings of the effectiveness of power (qudra):
Power (al-qudra) either relates [a] to the creation of an entity (maʿnā) or [b] to the “acquisition” [of this entity] in that [power] relates to the entity whilst it exists or [c] in that [the entity] comes to have an attribute that accompanies its coming into existence.Footnote 72
From al-Bāqillānī's perspective, only God has power to create things (and, consequently, His power alone can relate to their creation); therefore, only the second and the third modality described in this quotation are relevant to human acts, as is also underlined by the use of the term “acquisition”. These two modalities can be easily identified with the second and the third explanation of the effectiveness of human power as discussed in the chapter on human “acquisitions”. Since they are presented here as real alternatives, al-Bāqillānī must have assumed that human power can be effective in more than one way. This largely answers the question how to understand al-Bāqillānī's contradictory affirmations – that human power relates to an act without causing it to have an attribute (second answer) and that “acquired” acts come to have an attribute by virtue of man's power (third answer): both scenarios are possible options, which we can only assume occur under different circumstances. In the extant parts of the Hidāya, I cannot find, however, any solution for the question what precisely determines which of the possible alternatives will happen.
Muʿtazilite and Ašʿarite theologians used to explain the link between human actions and moral responsibility in terms of man's capability or power (qudra). According to the Muʿtazilites, we are free agents by virtue of this very capability. It enables us to act otherwise than we do, but it in no way necessitates any action. Whether or not we act and whatever we do is completely up to us. Within the school, it was however debated whether our power over alternative acts is sufficient for us to control what we do, or whether our decision-making requires something else: some Muʿtazilites argued that exercising freedom depends on intentions or motivations in order to turn our abilities into real actions.
For al-Ašʿarī, human acts that occur on account of man's power are voluntary acts. Man does not create his actions, but his acting is self-determined because he does things according to his willing and wanting. Yet it appears that al-Ašʿarī believed that human capability or power has no effect whatsoever. This raised the question of the function of human power. It would seem that al-Bāqillānī attempted to solve the problem when he drew on al-Ašʿarī's theory: where the latter speaks of mere conjunction between man's power and his act, al-Bāqillānī speaks of correlation. For him, human agents who are held responsible for what they do have a real property or attribute on account of their qudra. It is by virtue of this specific feature that they are distinguished from whoever is powerless (ʿāǧiz) and therefore irresponsible for actions he cannot refrain from performing. In addition, al-Bāqillānī suggests that man is related to his acts by virtue of his power, even though he does not create them. Consequently, man cannot be praised and blamed for the existence of his actions. Al-Bāqillānī therefore concludes that man assumes the responsibility for something else: he argues that God's command and prohibition relate to an attribute of acts that is caused by human power.
Early drafts of part II of this paper were presented at the workshop “Fresh Perspectives on Early Islamic Thought” at Göttingen University (March 2014); and the BRAIS conference (April 2014) in Edinburgh; I am grateful for the feedback I received on these occasions. I would like to thank Sabine Schmidtke for providing me with copies of the manuscripts of al-Bāqillānī's Hidāya, and two anonymous reviewers for their comments and suggestions. This research was supported by the Gerda Henkel Foundation's M4HUMAN programme.