John Hare’s God’s Command, 6.3, “Revelation and Reason”

All three of our authors have an important place for both revelation and reason, but they describe the relation between the two sources of knowledge differently. The term ‘revelation’ is a convenience, but is potentially misleading. It would be better, but cumbersome, to talk about God’s deliverances through the Scriptures and the Traditions.

6.3.1 “‘Abd al-Jabbar”

Al-Jabbar makes a distinction between necessary knowledge and acquired knowledge. Necessary knowledge, unlike acquired knowledge, is known immediately and is known by all sane adult human beings. This includes knowledge from sense perceptions and rules of logic and knowledge of one’s own mental states. The most important for present purposes are certain moral truths and reliable reports. An adult with sound mind necessarily knows the evil of wrongdoing, the evil of being ungrateful to a benefactor, and the evil of lying if it is not intended to bring about benefit or to repel harm. One also knows the goodness of compassion and giving. These moral principles are the basis for rational obligations. Knowledge of reliable reports is also necessary for knowledge, and is required for religious obligation, which is a part of obligation not known by reason—like the obligation to pray and fast.

So al-Jabbar gives a kind of priority to reason over revelation. Neither revelation nor reason makes something right or wrong. But the right that revelation indicates, reason sees is instrumental towards a right that reason already knows. We know by necessary knowledge that we should choose our duty, and revelation tells us that prayer is conducive to this end. There is a difference between intrinsic wrongs and things that are wrong by relation to their consequences, such as the wrongs of the Law, which are only wrong inasmuch as they lead to the performance of a rational wrong or ceasing to perform certain duties. This does not mean, however, that revelation is redundant. One may not know, before being told, how to achieve the end in question, and one also may be insufficiently motivated.

The opponents of the Mu’tazilites tended to object that the moral principles that are supposed to be necessarily known, and so known to all sane adults, are in fact not known by all. There is, in fact, widespread disagreement. For example, the nomadic Bedouins of Arabia approve the practice of plunder. But this does not mean that there is disagreement here about the principle that injustice is prohibited; it’s just that Bedouins have a different conception of private property. But it’s hard to see that the objection from disagreement can be overcome in this way. To be sure, ‘injustice’ is named together with the wrong and so anyone who agrees that some act is unjust is going to agree that it is wrong; but the relevant disagreement is surely about what kinds of act are unjust.

So take lying. One might object that this is not something about which all sane adults agree, and indeed many of al-Jabbar’s own opponents disagreed with him, holding that it would be right to lie to save the life of a prophet. But the case of lying is anomalous here, and it may be that the operative conception of lying is already evaluatively laden. There is nothing implausible, Hare writes, about holding that there are very general principles that are very widely shared across human cultures, as long as one does not insist that they generate absolute prohibitions.

According to al-Jabbar, we need to distinguish rational worship and religious worship. Both kinds involve obligations that are assigned by God. This seems to imply that we can worship rationally by obedience to the principles that are necessarily known, even if we do not know about God, and even if the obedience is not consciously directed towards God. On this view, it is only in relation to religious worship that God must be “described with every and each action,” to use al-Maturidi’s terms.

John Hare’s God’s Command, 6.2.1, “Human Freedom”

Hare now takes up the same three figures, but in relation to the different question whether human beings have freedom of choice in what they do, or whether our actions are only the product of divine causation. This question is the subject of prolonged discussion by all three, but we will focus on material that has implications for the relation between divine command and human obligation.

6.2.1: “’Abd al-Jabbar”

Al-Jabbar starts from the premise that it is irrational to assign an obligation to perform an act, unless the addressee is capable, or has the power to perform it, in order to be considered truly his action. The maxim that it is bad or irrational to impose unbearable obligations is taken from the Qur’an. What kind of power are we talking about? Two things are important to say about it: it has to precede the act and it has to be a power over opposites—that is, a power to perform an action or its opposite. It’s humans who do the wrong that create it. God does have the power to do wrong, but it is impious to think He does it, and there is no reason to think He does it.

Al-Jabbar uses a distinction here that descends from Aristotle’s discussion of the “mixed” cases of voluntary action in the Nicomachean Ethics, which was available in Arabic, though he reaches a slightly different conclusion from Aristotle about praise and blame. Aristotle holds that an action is involuntary if it is done either by force or by ignorance, and it is done by force if the origin of the action is outside the agent. But there are three kinds of mixed cases of “force.” One is where the action is done from fear of greater evils—like acting under a threat to one’s family. Such a case is “mixed” because it resembles both the voluntary and the involuntary, but Aristotle says it’s more like the voluntary. The second case is where one receives not praise or blame, but pardon, when one does what he ought not under pressure which overstrains human nature, and which no one could withstand. The third case is where the action is so base that no one could be forced to do it, like matricide.

Al-Jabbar extends Aristotle’s treatment of the third kind of case (like matricide) to cover all actions wrong in themselves (a category Aristotle does not have). He agrees with Aristotle’s assessment that in some mixed cases we do not receive praise and blame, but he says this not about cases of pressure that overstrains human nature (where Aristotle says we receive pardon), but about all cases where we are motivated by self-preservation. Again, this is because he has a category Aristotle does not have, that of actions to benefit others without reference to oneself, which do deserve praise. Finally, he reflects Aristotle’s point about the pleasant and the noble (which for Aristotle are ingredients of the agent’s own eudaimonia), but he says, not that we can’t be compelled by them, but that we should not be praised for pursuing them as our own advantage. Each of these three changes to Aristotle is highly illuminating about the structure of the Mu’tazilite’s thought as a whole, which denies eudaimonism and embraces the view that we can be moved by what is good in itself, independent of our own advantage.

Al-Jabbar has a complex picture of desire, motivation, and will. The central point for our purposes is that he is concerned to deny that there is any determining cause of our actions, either external or internal. He does not have, just as Aristotle does not have, a Kantian sense of “will,” in which it is the center of agency. If he had thought in the Kantian way, he might not have been so reluctant to posit an internal determining cause. But his notion, though rendered “will,” is closer to wanting than what Kant would call “willing.” One final point is that al-Jabbar holds that it is obvious that we have the relevant kind of power over our actions (a power that precedes the act, and that is a power both to act and not to act). In this way the Mu’tazilite resembles Scotus, and the resemblance is a deep one; the power over opposites is something we know from ordinary experience.

Image: "Quran" by Urganci. CC license. 

John Hare’s God’s Command, 6.1.1, Intrinsic Value, “‘Abd al-Jabbar,”

Hare begins by roughly translating al-Jabbar’s language of “hasan” and “qabih” as “right” and “wrong,” respectively, but this will introduce a strain in certain contexts. Hare then makes two qualifications: al-Jabbar doesn’t distinguish between two normative families of terms (value and obligation) the way Hare does. But he does have an account of obligation. The second qualification is to distinguish qabih (wrong) from “zulm,” meaning injustice, something a bit narrower.

Al-Jabbar defines “wrong” by connecting it to that which deserves blame, but there are two qualifications. An act can be wrong without a person being blamed—if the action is such that, if certain conditions held (like the person was awake), performing the action would have been blameworthy. Also, sometimes the wrongness of an action can be overridden by a greater or equal right-making property. The second qualification is that there is no such neutralizing or overriding right-making property in the act that deserves blame.

The contrary of wrongness is obligation, where the person who omits the act (if he’s able to do it) deserves blame. Distinguished from both wrong and obligation are two other kinds of right, the “merely right” and the “recommended.” Cases of the “merely right” are breathing the air or eating harmless food, where the agent doesn’t deserve praise or blame. But they’re not simply neutral, for they are good things to do (as we’d put it in English). Cases of the “recommended” are praying or fasting beyond what’s required, where the agent is praised for the act but would not be blamed for the omission. [Supererogatory?]

Al-Jabbar holds that the right and wrong acts distinguished in his system are evident to human reason in their right and wrong character. They are “known immediately,” independently of revelation. Revelation does indeed inform us of the obligations we already have, but these truths are known by reason when they are revealed, and this knowledge by reason is primary in justification. These standards that we learn from reason apply also to God. “The Eternal Glorious One is able to do what would be wrong if He did it.” Because God in fact only commands and does what is right (though he could do what is wrong), we can use these standards to judge what God is and is not commanding us to do.

Al-Jabbar claims that there are “aspects” by which wrong acts are wrong and right acts are right, and that we can discern these aspects with our reason. “Lying” and “wrongdoing” are aspects that necessarily bring wrong with them, on his account, unlike “injury,” which may bring wrong or right depending on the situation. He distinguishes between the aspect of an act and the genus of an act. The genus does not make an act wrong. Entering a house is a genus of act, as is bowing in prayer. But neither is necessarily right or necessarily wrong. But “injustice” is not a genus of act, because injustice is named together with the bad. But lying is an aspect, not a genus. Al-Jabbar holds that lying necessarily brings wrong with it, but he also holds that a small lie may be exempt from blame, on account of the good past deeds of the speaker and the amount of praise he has earned.

The aspect of injustice is not to be attributed to God’s acts, according to al-Jabbar, but not because there is some difference between aspects as ascribed to humans and to God. He allows that we might seem to judge God’s acts differently from our own, when, for example, we judge that his goodness is consistent with causing pain to children. But in fact there is a difference of circumstances here, because we are assuming that God compensates the children in the next life, and so in fact the same standard is being applied. A key difference between the three authors in this chapter is that they disagree about whether God could do something wrong, even if he does not in fact do so.

Two more preliminary matters: first, previous chapters assumed an affinity between natural law theory and eudaemonism. One value of studying Islamic medieval moral theology is that we can see a school where this pairing does not obtain. The Mu’tazilites, and al-Jabbar in particular, hold that the right in all of its aspects attracts us in itself, intrinsically, not because it leads to a benefit for us as agents of the action. Al-Jabbar recognize that his opponents will claim that people do not avoid injustice and lying intrinsically, but only because of some benefit to themselves. He replies that people will do wrong for the sake of some benefit, but they will do right without any benefit to themselves. Even a heartless man would warn a blind man against falling into a well. Al-Jabbar replies that it is possible to act without thinking about one’s own interest at all. [Seems right to me, contra Piper.]

Second, al-Jabbar offers explicit arguments against divine command theory. DCT can be found in all three Abrahamic faiths, and it creates much the same difficulties in all three. Al-Jabbar offers at least seven arguments against it, and Hare presents four of them. The first is that commands do not imply obligation. Al-Jabbar quotes the Qur’an: “Surely God bids to justice and good doing and giving to kinsmen.” Al-Jabbar thinks such virtues are indicated by the command but not produced by it. This sort of objection is frequently made by those who can’t see what normativity is added by a command, even a divine one. Either, they think, the thing commanded is already right or it is not; the commanding can’t change it from one to the other, though it can inform us of a character that the act already has. (Hare had earlier rejected this view that reduces imperatives to an indicative indicating that someone wants something. Hare thinks the best response, on al-Jabbar’s own terms, is to point out that al-Jabbar has the concept of obligation, distinct from rightness, and that God’s command might make something right but not obligatory into what’s both. This wouldn’t involve the command making the action right, because it already is.)

On al-Jabbar’s second objection to DCT, the account of wrong as what is forbidden by God does not fit our normal language. We don’t say it’s forbidden of God to do evil, for example, even though it would be evil of him. Moreover, there are things that are virtuous and would still be virtuous even if God told us not to do them. [Here I think al-Jabbar’s mistake is rejecting DCT instead of God’s ability to issue such hideous commands. Hare’s response is similar but a bit different, saying God’s commands are based on what’s good. I resist that because on occasion it seems to me God’s command might be predicated on what’s less evil, not what’s good. Perhaps even God chooses to break a tie.]

Third, if DCT were right, we couldn’t know our obligations without knowing they were commanded by God. But al-Jabbar says the sane man knows his obligation even though he doesn’t know that there is a commander. (Hare’s reply is to punt to Adams’s reminder that we can distinguish between what a term for a characteristic means and what makes a thing have that characteristic.)

Fourth, DCT has a problem understanding the goodness of God. If we say God’s acts are not wrong because God is not commanded, we can’t say God’s acts are right either. But we need, and the Qur’an gives, standards of value intelligible to us in terms of which we can praise God for doing right. [Hare says one reply is to say that ‘good’ means “attracting us and deserving to attract us” (where both conditions are necessary), and that we can say that God and God’s acts are the paradigm case of what is good in this sense. My own reply to these last two objections would also punt to the ontology/epistemology distinction and their different orders.]

John Hare’s God’s Command, Chapter 6, “Divine Command in Some Medieval Islamic Thinkers,” Introduction

This chapter is about the concept of divine command theory (DCT) in three medieval Islamic thinkers: al-Ash’ari, al-Maturidi, and ‘Abd al-Jabbar. Hare will argue that al-Maturidi takes a plausible mediating position between an extreme form of DCT (in al-Ash’ari) and an extreme form of natural-reason theory (in the Mu’tazilites, especially ‘Abd al-Jabbar). Despite reservations, Hare took up this part of the book because the concept of DCT is central outside the Christian tradition as well as within it, and there is a great deal to be learned from the comparison. Within medieval Islam, and within contemporary Jewish appropriations of medieval Judaism, there is very much the same range of options in understanding the relation between a sovereign God who gives us commands and our own reason, as we try to determine how to live our lives.

Hare is assuming, without arguing for it, that the three Abrahamic faiths worship the same God, though they say very different things about this God. He additionally argues that a useful side effect for a Christian of examining DCT in Judaism and Islam is that new light gets shed on areas of the Christian’s own faith that had tended to get obscured. Psalm 119, for example, acquires fresh meaning, and likewise doctrines of divine concurrence.

The chapter covers just three thinkers, and has no pretension to be talking about Islamic ethics as a whole. The scope is relatively modest, and Hare admits he’s on a big learning curve in this area. But he thinks there’s an obligation, if one thinks one has something useful to say about divine command, to relate this to the faith of over a billion people for whom divine command is a central concern. It is the confinement to a discussion of Christianity that requires justification, not the inclusion of a discussion of Islam.

This chapter locates al-Maturidi against the background of a dispute between Mu’tazilites and Ash’arites about three questions. The first is whether acts and persons have intrinsic value (or whether that value is to be understood only as a divine willing or commanding), and what kind of access we have to that value. The second question is whether human beings have freedom of choice in what they do, or whether our actions are only the product of divine causation. The third question is whether there is any proper use of human reason independent of divine revelation, or whether the proper use is only derivative from what we are given in the Qur’an and the Traditions.

There are many differences between Mu’tazilites (especially between the schools from Baghdad and from Basra), and this chapter relies mainly on the texts of ‘Abd al-Jabbar (from Basra, d. 1025), who gives the fullest account. Al-Ash’ari (d. 935) and al-Maturidi are roughly contemporaries, though there is no evidence that they met. They are both responding to Mu’tazilites, and indeed al-Ash’ari started off as a Mu’tazilite under the tutelage of al-Jubba’i of Basra (d. 915). ‘Abd al-Jabbar lived almost a hundred years after them, and they are therefore not responding to his version of the arguments (which is, in many cases, a refinement of them). Hare will start with the Mu’tazilite position, and continue with the Ash’arite response. He then locates al-Maturidi between the two, taking something from each side. But both al-Ash’ari and ‘Abd al-Jabbar also see themselves as taking middle positions, and indeed we should expect this because the Qur’an itself recommends this strategy [“Thus We have made you to be a community of the middle [road]” (2: 143).] Middle-ness is not itself truth-marking; everything depends on what the extremes are between which middle ground is being claimed. But Hare thinks it’s instructive to compare al-Maturidi’s middle ground with that of Duns Scotus.

Two other general comments will be helpful in what follows. First, understanding and interpreting law is chronologically antecedent in Islam to questions in theology (kalam) about the relation between divine command and human reason. Of the four main Sunni traditions or schools of jurisprudence (Hanafite, Shafi’ite, Hanbalite, and Malikite), al-Ash’ari comes out of the Shafi’ite school and al-Maturidi from the Hanafite school. The last of these is the school that gives the most leeway of the four to legal reasoning that is not itself derived from the Qur’an and the Traditions. The Hanbalite school, by contrast, is the most conservative in terms of the attempt to confine legal reasoning to what can be derived from the Qur’an and the Traditions. Al-Ash’ari constructs his own “middle” position as being between the Hanbalites and the Mu’tazilites.

hare god's command

The second general point before we get to the three questions is that some influential secondary sources associate DCT in Islam with fundamentalism, and oppose it to enlightenment. The project of defending the Mu’tazilites within Islam is correspondingly seen as rescuing Islam from obscurantism and hostility to the modern world. But so far as the thesis of Hare’s book is correct, there is no conceptual requirement to connect DCT with fundamentalism, Christian or Muslim or Jewish. The term ‘fundamentalism’ is itself prejudicial here, but DCT can give us an account of the ground of human dignity in a way that simply making human dignity “a truth of reason” can’t. As a meta-ethical theory, DCT doesn’t tell us what the commands of God in fact are. But it gives no grounds for inferring that these commands will be any less or any more liberal than the prescriptions generated by the various versions of natural law. Having said that, Hare notes that it’s also true that a theory that has an honored place for both revelation and reason will find conversation with other traditions easier to sustain.