6 — Commands and Prohibitions
Are commands performative speech act or indicative evidence?
Finally we leave behind speculation about the nature of knowledge and language, and we turn to a practical interpretive question: how to understand imperative verbs. Not surprisingly, however, al‑Juwaynī has to start with a theoretical statement about what commands are and how they relate to imperatives.
An imperative has the general form ifʿal. That means that if we take the verb “to do,” which has the three root letters f ʿ l (ʿ is the letter ʿayn, a strangled sound like a German r, with no equivalent in English), then faʿala will mean “he did,” yafʿalu will mean “he does,” and ifʿal will mean “do!” Similarly fataḥa means “he opened,” yaftaḥu means “he opens,” and iftaḥ means “open!” The imperative, then, is a general form that can be given to many specific verbs.
Notice, however, that an imperative verb is not itself a command; it is only a verbal form that indicates the existence of a command. The command itself is a request; it is expressed verbally (rather than by winks or gestures or longing glances), but it is not itself a word. Contrary to my earlier observation that al‑Juwaynī regards revelation as essentially informative and independent of its historical context, this definition of command suggests that he regards speech as a historical event that takes place in the context of a particular interpersonal relationship and performatively brings about a new state of affairs: a command can occur only in a hierarchical relationship of authority, and it brings about a moral obligation. That makes law something that arises from one person’s authority over another, not a matter-of-fact statement about the consequences of actions.
The idea that revelation is a performative speech act that takes place within an interpersonal relationship and brings about new responsibilities was developed most fully by the Ḥanbalī Abū Yaʿlā (d. 458/1066). This view allowed Abū Yaʿlā to argue that a command brings about much more than its verbal form would suggest: it brings about an obligation to obey immediately and continuously, and to perform whatever other actions are needed in order to obey, and not to perform any action that would impede obedience. The verbal form ifʿal was not coined to denote all these things itself, but the interpersonal act of commanding brings about all these obligations at once, because they are part of what an authoritative person is obligating his inferior to do when he utters a command.
Al‑Juwaynī, however, cannot exactly say that God’s speech is an act of requesting, because that would make it part of God’s temporal activity—his creation—whereas al‑Juwaynī belongs to the Ashʿarī school of theology, which insists God’s speech is one of his eternal attributes. In keeping with this Ashʿarī doctrine of inner speech, we must understand al‑Juwaynī to be saying that a command is not a historical act of requesting, but an eternal inner meaning of request and an intent to obligate those under one’s authority, which is outwardly expressed using an imperative verb.
This changes the picture considerably. Now the word ifʿal is not a historical interpersonal event whose consequences for the present and for other relationships must be determined by the interpreter, but merely a piece of evidence indicating that there exists in God’s mind an eternal command which makes something obligatory. The interpreter’s task, therefore, is to translate the imperative verb into a statement that a certain act is obligatory. For a moment it sounded like al‑Juwaynī was starting to treat speech as performative and hence context-specific, but in fact he is still working on the assumption that words are indicative evidence of timeless legal values.
The flexible interpretive default of obligation
For the first time, al‑Juwaynī gives us a rule for how to interpret verbal evidence to determine God’s intended meaning: unless something in the context suggests otherwise, an imperative verb indicates that God has commanded the action denoted by the verb. Given his definition of command, this means that the act is not just recommended but strictly obligatory.
Not all legal theorists were so bold. The Muʿtazila argued that an imperative indicates only that God wills that an act be done, so the most we can be sure of is that the act is beneficial and that it will be rewarded, which makes it “recommended.” We cannot infer that God will necessarily punish failure to act, so the imperative does not indicate obligation. A few Muʿtazila were even more reticent, arguing that the most an imperative indicates is that something is beneficial and hence at least permissible. On the other hand, the Ashʿarī theologian al‑Bāqillānī argued that we cannot be sure whether an imperative indicates a command or not, and even if it does, we cannot be sure whether that command is an obligation or a recommendation. Hence al‑Bāqillānī said interpreters should suspend judgment on the meaning of an imperative, until they find sufficient evidence to clarify what God intends. All of these interpretive principles, however, made it quite difficult to establish that something was obligatory: they required finding specific evidence for each action showing that God had promised to punish those who did not do it. Most legal theorists therefore settled on obligation as the default meaning of an imperative. This was part of a general trend toward strong default interpretations: legal theorists started to pack as much meaning as possible into the words of revelation, so that they could affirm strong and broad legal rules with minimal interpretive effort, and say they came directly from revelation rather than from human reasoning.
At the same time, al‑Juwaynī does not commit himself to interpreting every imperative as evidence of an obligation. He reserves the right to interpret any imperative as merely a recommendation, or as a mere granting of permission (as when God said in Q 62:10 to “disperse in the land” after gathering for Friday prayers), or as one of the other speech acts or speech-meanings he lists at the end of this section: threatening (as in “go ahead, make my day!”), giving alternatives (as in “speak up or shut up, it’s all the same to me”), or creating (as when “God says to a thing ‘be!’ and it is,” Q 2:117, 3:47, 19:35, 40:68). Any of these interpretations is legitimate, and can even be called a literal interpretation (remember that legal theorists do not like to call their interpretations figurative), as long as it is supported with some kind of evidence.
That evidence can be a qualification that accompanies the imperative, or some other evidence from its immediate context, or even evidence from some entirely unrelated piece of revelation. In fact, most legal theorists considered every piece of revelation to be part of the context of every other piece, so all revealed evidence is considered contextual evidence. Al‑Juwaynī is not saying that the specific historical and interpersonal context within which an imperative was originally revealed affects its meaning; he is saying that the corpus of revelation forms a context within which the present work of interpretation takes place, and that the interpreter can use elements of that corpus of revelation to modify the default meanings of other elements.
If al‑Juwaynī were to define a command as an imperative verb, he would have to include all these alternative meanings of the imperative in his definition of command, and then he could not assume that commands create obligations by default; or else, if he wanted to keep on interpreting commands as strict obligations, he would have to say that all imperatives create obligations, which would take away his flexibility to interpret some imperatives in other ways. By saying that a command is an inner meaning, however, and that imperatives are just verbal forms that may be used to express any number of meanings besides commands, he is able to justify interpreting imperatives as obligations by default, while retaining the flexibility to choose other interpretations as he sees fit.
A classic example of this flexibility is afforded by the Prophet’s command to take a full bath before Friday prayer, which by default should make the bath obligatory. Most jurists, however, consider this merely a recommended practice, and they justify this departure from the default meaning of the Prophet’s words by citing as evidence a non-Prophetic report about an early Muslim (the future caliph ʿUthmān) who joined in Friday prayers without bathing despite a rebuke from ʿUmar, the second caliph. This weak piece of evidence—an inconclusive report about the practice of someone other than the Prophet—is sufficient to overcome not only the default rule that imperatives imply obligation, but even the Prophet’s own statement (in a separate ḥadīth) that a bath before Friday prayer is “obligatory.” This, one jurist explains, only means “optionally obligatory,” as a matter of etiquette and cleanliness.1 Needless to say, not all jurists found this interpretive maneuver convincing, but ironically those few who insisted on taking the Prophet’s command at face value lost the argument. Their legal school, the Ẓāhiriyya, is now extinct, largely because their hermeneutic was not flexible enough to justify the practices most Muslims wished to justify.
Further pragmatic interpretative defaults
The same tendency to claim strong default meanings, while allowing flexibility to change those defaults by appeal to other evidence, can be seen in the other conclusions al‑Juwaynī draws from imperatives.
Some of the most traditionalist legal theorists, including the Ḥanbalī Abū Yaʿlā, said that an imperative makes it obligatory to obey immediately and continuously. For example, by default the Qurʾān’s command to pray would make it obligatory to start praying immediately, and to do so as continuously as possible until the day of one’s death. Of course no one thinks this is required, so the jurist must supply some evidence as to why prayer is to be performed five times a day at specific times. That evidence is not to be found in the Qurʾān, but it is readily available in ḥadīth and in the consensus of the Muslim community. Thus the default of immediate and continuous obedience does not really affect the results of interpretation; it just affects how much evidence is required to defend one’s interpretation.
Very few if any commands actually have to be fulfilled continuously; most if not all of the things humans do unceasingly, such as breathing, are involuntary and therefore can hardly be the object of a command. To insist on a default of continuous obedience therefore seems a pointless nuisance, and only a handful of very strict legal theorists upheld it, and those few did so only as a symbolic acknowledgement of the great debt of obedience owed to God.
The idea that by default imperatives require immediate obedience was upheld by a greater number of legal theorists; but this likewise did not really affect their view of the law. They still recognized that most obligations are due at specific times (like prayer and fasting), or can be performed at any time in one’s life (like the pilgrimage, or the expiation that must be made by feeding poor people if one breaks a religious vow). All the default of immediate obligation affected was how they justified the law: they had to find evidence of when each obligation was due, or else they had to argue that even an immediate obligation (such as expiating a broken vow) could still be fulfilled later even if one did not do it the moment it became obligatory. Thus the defaults of immediacy and continuity had little practical bearing on the law; they were mainly symbolic claims of piety, or technicalities about how evidence should be used. Of course, that should be no surprise, since most of what al‑Juwaynī has said so far has been less about how to interpret than about how to imagine the law and its relationship to the evidence upon which it is supposed to be based.
Al‑Juwaynī did not think that insisting on immediate and continuous obedience would enhance the power or flexibility of his hermeneutic, because so little of the law actually has to be fulfilled immediately or continuously. Besides, on its surface an imperative verb does not appear to say anything about the timing or number of actions it demands, nor does the notion of command seem to involve a specific timing or number of acts. As one commentator points out, the timing and number of acts required by a command seems to vary from command to command: when a thirsty master requests a drink his servant presumes he wants it right away, but if he tells him to pay the rent the servant will presume he just means to do so sometime before the rent comes due. In and of itself the idea of “requesting” does not seem to imply anything about timing or repetition, and al‑Juwaynī did not think it advantageous to assume that it did.
There are other things, however, that it would be quite convenient to be able to infer directly from imperatives, without having to prove them every time. One is that the command to pray (for example) makes obligatory the act of purifying oneself for prayer, as well as the act of prayer itself. As long as the interpreter has established that purity is a condition of a valid prayer, he does not have to argue that purification becomes obligatory at the time of prayer; the command to pray makes it obligatory, and no other argument or proof is needed.
Similarly, part of the meaning of a command is that if one obeys one will have fulfilled the command. This seems so obvious that it should go without saying, but a few legal theorists—mainly Muʿtazilī theologians—were so reluctant to read anything into revelation that it did not explicitly state, that they said the only thing a command communicates is the legal value of an act. Whether simply performing that act (e.g. praying in some unspecified way) will actually fulfill the command, or whether there are other special conditions (such as purity and proper timing) that must be observed in order to fulfill the command, is another question entirely, and requires a separate investigation using other evidence. Al‑Juwaynī was much more pragmatic, and sought to minimize the amount of proof texting one had to do: the command to pray already contains implicitly all the conditions and prerequisites God intends, so the command itself is sufficient evidence that if we obey it in accordance with all those implicit conditions, we will have fulfilled the command.
The same goes for the legal value of an act contrary to what has been commanded: if God commands me to stand, I should be able to affirm without any further proof that I am forbidden to sit down.
Notice that al‑Juwaynī has not assigned all the strongest possible default meanings to imperatives, since he has not required immediate and continual obedience. But he has followed a consistent pattern: he has chosen precisely those default interpretations that can be accommodated to the law with the least interpretive effort. And he has always allowed room for exceptions to his principles: an imperative can always be interpreted differently, as long as one can find some shred of evidence to support a meaning other than the default one.
This combination of interpretive ease and flexibility maximizes the interpreter’s power to construct the meaning of revelation as he sees fit, while claiming that he is simply discerning the meaning of revealed language in accordance with established rules of interpretation. This is what I meant when I said in the Introduction that al‑Juwaynī’s hermeneutic does not force a jurist to hear in revelation something he does not want to hear. If he wants everyone to pray, and to do it right, and to close their shops when it is time for Friday prayer, his hermeneutic allows him to claim that the Qurʾān’s command to pray itself imposes a strict obligation not only to pray, but also to purify oneself and to leave off contrary acts such as buying and selling. But if he does not want to bathe before every Friday prayer, his hermeneutic allows him to dig up some evidence that justifies going against the plain sense of the Prophet’s command.
Timeless, contextless, universal commands
The principle of interpretive ease is likewise reflected in a little subsection that al‑Juwaynī inserts here on the scope of address of God’s commands.
Any jurist would agree that a child who has not yet reached the age of sexual and intellectual maturity is not subject to the law; in other words, his or her actions have no legal value whatsoever, and will be neither rewarded nor punished. The same is true of an insane person (literally one possessed by a jinn or genie) as long as his or her insanity lasts, and it is even true of a sane adult whenever he or she is asleep or simply inattentive or forgetful: if I sleep through the time of dawn prayer, or wake up but genuinely forget that I should pray, then the dawn prayer never actually becomes obligatory for me. That does not let me off the hook entirely: once I do remember, it will be obligatory for me to perform a makeup prayer in place of the dawn prayer; and if I deliberately chose not to set my alarm, that neglect may have been a punishable offense. But that day’s dawn prayer itself never actually had the legal value of “obligatory” for me, so I will never be punished for omitting it; I can only be punished for failing to set my alarm, or for consciously neglecting to make up the prayer later.
That particular dawn prayer’s lack of legal value for me can be proven by means of evidence, such as the famous Prophetic ḥadīth “my community has been relieved of [the consequences of] error and forgetfulness.” But al‑Juwaynī allows us to bypass such a proof, because he says that God’s command to pray was never directed to the forgetful or insane or young person in the first place. This is another hermeneutical principle that has no effect on the law, but serves merely to simplify the interpretive process by which the law can be justified. Al‑Juwaynī recognizes the role of human reasoning in constructing the law, but his hermeneutic minimizes the amount of human reasoning that is actually required, and manages to discover as many features of the law as possible in the language of revelation itself, thus bolstering his claim that the human construct of law stems entirely from revelation.
Some other legal theorists also exclude women and slaves from the scope of God’s commands. This, however, makes interpretation more difficult, because most aspects of the law are thought to apply to women and slaves as well as free men. Al‑Juwaynī therefore tacitly includes them when he declares that God’s commands are directed to “the believers,” which is a masculine noun but does not necessarily refer just to males. Again al‑Juwaynī chooses the hermeneutical principle that makes revelation fit most easily with established law. Quite a few classical legal theorists disputed the inclusion of women, not because they wanted to exempt women from the law, but because they wanted to be scrupulous about not reading more into revelation than it actually said. Because classical Arabic has distinct masculine and feminine forms for imperatives verbs, they considered it more cautious not to assume that a masculine imperative (or noun or pronoun for that matter) was meant to encompass both sexes, even though it certainly could be used to refer to mixed groups. If this exclusion were widely known today, it would be sharply contested as linguistic reification and reinforcement of the social oppression of women, but medieval legal theorists perceived it as an innocent theoretical stance motivated by hermeneutical restraint. It did not lead anyone to relieve women of their legal obligations; it just added one more step to the argument required for justifying a law’s application to women. Al‑Juwaynī himself does not take sides in that debate here; instead he avoids this purely theoretical issue, leaving room for his readers to choose as they wish, yet tacitly siding with the more pragmatic position of inclusion, which spares us from having to give a proof every time we want a masculine imperative to apply to women.
Very few legal theorists excluded slaves, because there really was no practical or linguistic motivation to do so; classical legal theorists accepted slavery as a legitimate feature of society, but unless one wanted to make a point of assigning a lower spiritual status to slaves—something that the Prophet had denounced—there was not much to be gained from having to cite specific evidence for every branch of law one wanted to apply to slaves.
There was also no linguistic ground for excluding unbelievers from the scope of God’s commands, but a few legal theorists did so. This made little practical difference, since Muslims were accustomed to having Jews and Christians (who were generally considered unbelievers despite their protected status as People of the Book) living among them and following their own laws on internal matters. But al‑Juwaynī’s point here is apologetic: the unbelievers will not be rewarded for following their own laws, because reward for any action depends on faith, which to his mind meant Islamic faith. They will be punished for failing to follow the law of Islam. The concession that allowed Jews and Christians to follow their own laws in internal matters did not indicate that their actions had any religious merit. The Qurʾān speaks of Jewish and Christian believers whom God will reward in paradise for their deeds; but al‑Juwaynī was agreed with most Muslims of his age in regarding Islam as the sole true universal religion. Concessions were made for some non-Muslims in this life, but recall that his legal system was oriented toward enforcement in the afterlife, where no concessions would be made.
There is one last claim about the scope of God’s commands embedded tacitly in this section: God’s commands are addressed not just to the sane adult and attentive believers and unbelievers of the Prophet’s time, but to people of all times and places. When he says that the inattentive, the insane, and the young are not addressed, he does not mean that those who happen to have been asleep when a particular Qurʾānic verse was revealed are not subject to its requirements! He means that it is not directed to people while they are asleep or insane; it will be directed to them the moment they wake up or recover their sanity. This shows that he does not regard the revelation of God’s commands as a one-time speech act that was addressed at one particular time to one particular audience that actually heard the Prophet utter each command. Some legal theorists did limit revelation’s scope in that way, and that forced them to prove by means of separate evidence that the obligations imposed by the Prophet were intended to remain in effect for later generations. Al‑Juwaynī, however, joins many other classical legal theorists in imagining revelation as a trans-historical event that happened not only at one moment in time, but continues to happen today as adult sane men and women hear the Qurʾān recited: God continues to speak directly to them, as long as they are awake, of age, and of sound mind. Whether we are addressed is not a function of when or where we live, but of our own present state and ability to come under legal obligation.
This means that the interpersonal effect of God’s speech occurs in God’s relationship with each person throughout time. The speech act or speech-meaning of request brings about an obligation in the context of each individual’s relationship to God. God’s command is eternal, according to the Ashʿarī doctrine that al‑Juwaynī followed, and hence its performative effect is for all time. The words that give expression to that inner meaning of command were temporal and created at one moment in history, but they are merely indications of a trans-historical reality that brings about timeless legal values for every person across all of time.
This helps us to understand why al‑Juwaynī makes no mention of the historical context of revelation anywhere in this treatise. As we noted previously, legal theorists generally held that “what matters is the generality of a verbal expression, not the particularity of the occasion of its revelation.” This makes perfect sense if the relevant “context” of a command is not the situation in which it was first spoken by the Prophet, but rather the present context of each person’s hearing that command through the Qurʾān and ḥadīth. If the meaning of God’s speech is defined as something eternal, then its meaning does not depend on the original situation of revelation, and its meaning in the present likewise does not depend on the present situation of the reader. The new situation of the reader may raise new questions that never occurred to previous jurists, but the answers to those questions must have been timelessly implicit in the texts all along; the new situations do not bring about new meanings of those texts, any more than the original situation of revelation governed its meaning at that time.
This timeless and contextless conception of revelation has been challenged by modern interpreters like Fazlur Rahman, who argued that interpreting the Qurʾān for today required reconstructing its original context so as to infer its meaning for that context, then abstracting that meaning from its historical situation and reapplying it to one’s present situation. That is a very different way of making the law flexible. Al‑Juwaynī sought interpretive wiggle room by arguing that language had strong default meaning but could easily be reinterpreted by appealing to even the weakest forms of evidence. Fazlur Rahman and the many modern Muslims who have followed his lead, however, have paid less attention to language. Instead of a hermeneutic that allows them to reconstruct the meaning of language, they have developed a hermeneutic that allows them to reconstruct the historical context of a revelation, and thus to reconstruct that revelation’s significance for the present. The analysis of language has given way to an analysis of history. But this requires looking at revelation as a historical event, and that al‑Juwaynī was not willing to do. He saw revelation as a timeless event, or rather a timeless speech-meaning, that is still being spoken directly to people today, across the wide gaps of time, history, geography, and culture.
The only full–length overview of the legal theorists’ discourse on commands of which I am aware is a dissertation by Abd al-Rafiʿi Oyewumi Omotosho, “The Problem of al-amr in uṣūl al-fiqh” (Ph.D. thesis, University of Edinburgh, 1984), which compiles the views of many legal theorists but does not attempt to explain why the debate mattered.