How to Be a Feminist Muslim

Abstract

Can Muslim values be reconciled with a feminist outlook? The question is pressing on both an individual level—for Muslim feminists—and on a political level—for the project of making Islamic practice compatible with the ideals of a just and liberal society. A version of this question arises specifically for the central Muslim text, the Quran: Can the message of the Quran be reconciled with a feminist outlook? There have, broadly speaking, been two approaches to this more specific question. I argue that both are inadequate. I then develop a novel approach to reconciliation that does not threaten the objective and universal normative force Muslims attribute to the Quran. My approach is revolutionary rather than apologetic and carves out a central role for moral understanding in Islam-as-practiced.

They (your wives) are clothing for you and you are clothing for them.

– Quran 2: 187

Muslim values define the place I call home—a place at once familiar and safe from a world that is increasingly hostile toward Muslims. Yet those same values also seem to limit my ability to live a life that I conceive of as a good one—a life defined in part by feminist values. I am certainly not the first to grapple with this apparent tension between Muslim values and feminist ideals.¹ Muslim women—like other women who subscribe to religions with patriarchal structures—are often forced to choose to identify either as feminists, or with their religion, but not both.² Is it possible to reconcile these seemingly inconsistent sets of values? Can I be a Muslim feminist, or does patriarchy have God on its side?³

This essay is as much political as it is personal: resolving the tension between a feminist outlook and the fact that Muslim theology appears to advocate gender inequality is crucial not only for Muslims who consider themselves feminists, but also for bridging the apparent gap between Islam and the ideals of a just and liberal society. I will restrict the scope of this paper to religious texts, and in particular, the Quran. Whether Islamic tradition and Islam-as-practiced can be reconciled with a feminist outlook is a broader question, the scope of which goes beyond the narrower target of this paper.

The problem of reconciling Islam-as-practiced with a feminist outlook demands that we focus on religious texts for two reasons. First, Islamic texts exert great influence on mainstream Islam as it is currently practiced. And to the extent that Muslim communities often rely on texts in order to justify religious norms and clarify misconceptions, the most promising way to repudiate certain sexist norms is by showing that they are not in fact supported by the text. Second, Muslim women in particular need a way to resist the pressures they face within conservative Muslim communities. Almost paradoxically, it is particularly difficult for Muslim women to defy religious authority in liberal societies because such societies tend to practice religious tolerance, and that tolerance extends to permitting sexist religious practices to flourish within minority groups. Muslim women in such minority communities frequently do not have recourse to outside resources—whether legal or social—for effecting change within a community.⁴ A natural first step in effecting change from within in such a way that women acquire leverage with those in charge is to show that current practice is not in fact supported by the central religious text. It is thus unsurprising that feminists and Muslim scholars have largely focused on religious texts when addressing the question of how Islam might be reconciled with a feminist outlook. Of course, some may worry that the focus on the message of the Quran, as opposed to the feminist values that that message is seemingly in tension with, presupposes without argument that it is Islam that bears the burden of proving that it is compatible with feminist values. I address this worry in the paper's final section.

Extant approaches to the question of how the message of the Quran might be reconciled with a feminist outlook proceed either by reinterpreting particular verses in more egalitarian ways, or by situating verses in their historical context in such a way that their message retains force only in a particular historical time and place. However, as I will argue, each approach undermines either the objectivity or the universality of norms derived from the Quran.⁵ Let us say that a norm derived from the text that prescribes x is ‘objective’ just in case its recognition (by us) has not been influenced by values that both pertain to whether one ought to x and are acquired independently of the text.⁶ For example, a norm that prescribes how women ought to behave in a particular situation is objective just in case its recognition has not been influenced by independently acquired values that pertain to how women ought to behave in that situation. A religious text-based norm is ‘universal’ just in case it holds for all times and for all places. Because Muslims take the norms prescribed by the Quran to have both objective and universal normative force, neither extant approach can succeed in reconciling the message of the Quran with a feminist outlook while at the same time preserving the place of the Quran for Muslims as a legitimate source of religious norms.

I will argue that reconciling the message of the Quran with feminist ideals requires that we reconceptualize the relationship that the Quran bears to Islam-as-practiced. This reconceptualization is a philosophical project whose aim is to preserve the normative force of the Quran while making possible an egalitarian feminist Islam. Instead of a straightforward relationship between text and practice in which the entire text functions as an ‘instruction manual’ for how Muslims ought to live their lives, I will show that there is a way to draw a principled distinction between those verses that have normative force for us here and now and those that do not. I will argue that this distinction between verses maps onto a distinction between Quranic verses that govern how we should be and those that govern what we should do. Let us say that the former verses prescribe ‘thick norms’, while the latter prescribe ‘thin norms’.⁷ I will argue that while the thick norms have unconditional normative force, the thin norms have a normative force that is conditional on certain social circumstances. But the thin norms also serve as a guide for how to apply the thick norms. A central upshot of this reconceptualized relationship between the Quran and Islam-as-practiced will be that it carves out a significant role for moral understanding.

I proceed as follows. In §1, I discuss some relevant background. In §2, I present the two extant approaches to reconciling the Quran with a feminist outlook and show that each approach undermines either the objectivity or the universality of religious norms. In §3, I defend my alternative approach to reconciliation and demonstrate how it avoids the problems that arise with the approaches discussed in §2. In my concluding remarks, I address some pressing methodological questions and conclude. My criticisms of the extant approaches and my alternative approach are specific to Islam, but the arguments in this paper might extend to the problem of reconciliation as it arises for other scripture-based religions. While I do not discuss the problem of reconciliation as it arises for other religions, there is reason to think that the challenge of reconciling scripture and feminism is particularly difficult for Islam because the central Muslim text—the Quran—is widely taken to be authored by God, and is thus infallible.

1. Sexism in the Quran? Some Background

The two core sources of Muslim religious norms are the Quran and Sunna. According to Muslim theology, the Quran was revealed by God to prophet Muhammad between 610 and 632 CE through the angel Gabriel in 6,219 verses. These verses were collected and recorded soon after the prophet's death in 632 CE, though not in the order in which they were revealed. By contrast, the Sunna is a written record of the prophet's verbal explanations of Quranic verses, verbal guidance given to others, and the prophet's own actions. Unlike the Quran, the Sunna was not documented until the eighth and ninth centuries CE (cf. An-Na'im 2011: 43–44).

The Quran and Sunna together form the basis for Sharia—i.e., Islamic law. But Sharia, and Muslim religious norms more generally, is also heavily influenced by two other sources. The first is ijma, the consensus of the Muslim community. The second is qiyas, which is reasoning by appeal to precedent. These sources come into play in matters where the Quran and Sunna are silent.

I focus on sexist norms that are seemingly grounded in the Quran as opposed to those that are derived from the Sunna. Because Muslims treat the Quran as the infallible word of God, verses in the Quran that seem to support sexist norms pose the most intractable obstacle to the project of reconciling feminism with Islam. Sexism that stems from the Sunna is easier to combat in principle (if not always in practice), for one might always point to the Quran and insist that the sexist practice or attitude in question was not decreed by God, but was instead introduced through human error.

To say that the Quran's text is sexist is itself a controversial claim. After all, that is precisely what feminist interpretations of the Quran contest. But putting aside such analyses for the moment, there are at least three respects in which—at least on mainstream interpretations of the Quran—the Quran's text is sexist:

Agency: the Quran often denies basic agency to women without also denying it to men. This is most salient in the norms that govern marital relations.

Testimony: the testimony of women is generally taken to be less valuable than the testimony of men, both in the case of contractual agreements and in the case of crime and punishment.

Property: the Quran appears to advocate inequality in wealth and possession, as in the norms that dictate inheritance following a person's death.

I will focus on these three aspects in which the Quran is putatively sexist while leaving open the possibility that there are others. That the Quran is putatively sexist in these ways might seem to be hardly worth documenting. However, the particular ways in which this (apparent) sexism manifests itself in Quranic text is relevant to both whether and how the Quran can be reconciled with a feminist outlook. Which parts of the Quran the (putatively) sexist verses come from is also relevant. In the rest of this section, I discuss these aspects of the Quran in more detail.

1.1 Agency

The Quran apparently supports an asymmetry between the obligations a woman has toward her husband and those he has toward her. A woman has a seemingly unqualified obligation to obey her husband. If she fails in discharging this obligation, the Quran prescribes a series of punishments of increasing severity to be administered by her husband:

Men are in charge of women by [right of] what Allah has given one over the other and what they spend [for maintenance] from their wealth. So righteous women are devoutly obedient, guarding in [the husband's] absence what Allah would have them guard. But those [wives] from whom you fear arrogance—[first] advise them; [then if they persist], forsake them in bed; and [finally], strike them. But if they obey you [once more], seek no means against them. (4:34; all Quran verses quoted in this paper are from the Sahih International Translation.)

By contrast, a husband has no such duty of obedience toward his wife. I later discuss how Islamic scholars have tried to reinterpret this verse in a more egalitarian way.

1.2 Inequality in Testimony

Another respect in which Muslim norms that seem to stem from the Quran are sexist is in their discounting the testimony of women. Consider, for instance, verse 2:282:

O you who have believed, when you contract a debt for a specified term, write it down. And let a scribe write [it] between you in justice. Let no scribe refuse to write as Allah has taught him. So let him write and let the one who has the obligation dictate. And let him fear Allah, his Lord, and not leave anything out of it. But if the one who has the obligation is of limited understanding or weak or unable to dictate himself, then let his guardian dictate in justice. And bring to witness two witnesses from among your men. And if there are not two men [available], then a man and two women from those whom you accept as witnesses—so that if one of the women errs, then the other can remind her.

The verse, which concerns witness norms for certain commercial contracts, seems to suggest that women make weak witnesses for they are prone to error, whereas men are not. And the implication is that a woman's testimony is only half as valuable as a man's testimony.

Many Muslim societies have integrated this implication of the verse into law. And while some such laws (such as the law that requires four male witnesses to testify to rape) are not obviously based in the Quran, it cannot be denied that the Quran clearly supports an asymmetry in value between a man's and a woman's testimony.⁸

1.3 Inequality in Inheritance and Property

Finally, there is obvious inequality in the norms that govern inheritance. For example, consider verse 4:11:

Allah instructs you concerning your children: for the male, what is equal to the share of two females. But if there are [only] daughters, two or more, for them is two thirds of one's estate. And if there is only one, for her is half. And for one's parents, to each one of them is a sixth of his estate if he left children. But if he had no children and the parents [alone] inherit from him, then for his mother is one third. And if he had brothers [or sisters], for his mother is a sixth, after any bequest he [may have] made or debt. Your parents or your children—you know not which of them are nearest to you in benefit. [These shares are] an obligation [imposed] by Allah.

The verse is ambiguous in at least one instance: the case where one has no children and the parents are the only family to inherit. In this case, while the mother inherits one third, it is unclear whether the father inherits the same amount or more or less (since the remaining two thirds might also be divided amongst nonfamilial inheritors). However, the inheritance norms prescribed here otherwise appear to be sexist.

2. Extant Approaches to Reconciliation

Recent attempts to reconcile the message of the Quran with egalitarian feminist values fall under one of two—or a combination of both—approaches. The first approach seeks to reinterpret Quranic verses in a more egalitarian way. The second situates Quranic verses in their historical context so that these verses retain their force only in a particular historical time and place. Let us call the former the ‘reinterpretation view’, and the latter the ‘contextual view’. I will argue that both undermine the normative force of the Quran for Muslims.

2.1 The Reinterpretation View

The reinterpretation view focuses on reinterpreting apparently sexist verses in more egalitarian ways by exploiting the ambiguity and polysemy inherent in many Quranic verses. To show how this works, consider 4:34 (on obedience, discussed above)—a verse that has proven difficult to explain away for many feminist scholars. One version of the reinterpretation view proceeds by showing that various Arabic words in the verse can be interpreted differently. Amina Wadud argues, for example, that the word qanitat that has been rendered in 4:34 as ‘obey’ can sometimes refer to human obedience toward God, and that the Quran ‘never orders a woman to obey her husband. It never states that obedience to their husbands is a characteristic of the “better women”’ (Wadud 1999: 77). Likewise, Wadud argues that the reference in this verse to daraba, which has generally been understood as ‘to strike’ (thus rendering the verse in a way that gives sanction for striking one's wife), can also mean ‘to set an example’, and contrasts with darraba, which means ‘to strike repeatedly or intensely’. Rafi Ullah Shahab argues that daraba can mean ‘to prevent’, suggesting that the verse simply instructs husbands to leave their wives alone in bed and keep them from leaving the house (Shahab 1993: 231; Barlas 2002: 188). And according to Hassan (1999: 355), the word daraba in a legal context means ‘holding in confinement’.

While the above version of the reinterpretation view proceeds by showing that particular Arabic words or phrases can have more than one meaning, a different version proceeds by showing that other verses on the relevant topic convey a different meaning—one inconsistent with an inegalitarian understanding of the original verse. Thus, for example, with respect to the case of women's testimony, Barlas (2002: 190) argues that other verses in the Quran give equal weight to women's and to men's testimony. According to Barlas, there are a total of five cases of testimony giving in the Quran, and only one suggests that a woman's testimony is worth less than a man's. For the problematic case, Barlas says that there are ‘very specific social reasons’ for the gender disparity, though she does not specify them (see 24:6–9 for another case of testimony giving that does not make a separate provision for women).

The reinterpretation view does not seek alternative meaning for specific verses in the Quran in a piecemeal and arbitrary fashion, but is motivated in at least two ways: by the intratextual method and by the tawhidic paradigm (the term ‘the tawhidic paradigm’ is coined by Wadud [2006]). The intratextual method proceeds by focusing on a ‘governing prescription’ that then guides how other verses are interpreted. A central example of this method treats the creation story in the Quran as evidence for gender equality (Cf. Hidayatullah 2013: 91).⁹ Hassan (1991) does this by focusing in part on verse 4:1: ‘O mankind, fear your Lord, who created you from one soul and created from it its mate and dispersed from both of them many men and women’. Hassan argues that there is no textual or semantic justification for attributing maleness to the soul mentioned in this verse nor femaleness to its mate.¹⁰ There is instead ground for thinking that God's first creation was undifferentiated in terms of gender. Verses 33:35 and 9:71 are also frequently taken to provide explicit or implicit evidence for gender equality. The first suggests a parity between men and women in their capacity as moral agents, whereas the second suggests that men and women are each other's moral protectors and guides. Together, these verses provide evidence for gender equality, and evidence against devaluing women as moral and political agents. This core governing prescription then guides how other verses are interpreted. In particular, it motivates egalitarian interpretations of otherwise ambiguous terms and phrases (as in the case of verse 4:34).

Like the intratextual method, the tawhidic paradigm also motivates a feminist reading of Quranic verses that are otherwise open to multiple interpretations. The tawhidic paradigm employs the Islamic concept of tawhid, or God's unity, incomparability, indivisibility, and justness to then argue that interpretations that imply that women owe obedience to men or that men have dominion over women are mistaken (Hidayatullah 2013: 94–97; 2014: 110–122). A prime target of this method is verse 4:34 that on the standard reading suggests that wives have a duty of obedience toward their husbands—a duty whose violation may be punished through increasingly harsh measures, as necessary. But the tawhidic paradigm renders this interpretation implausible on the grounds that it attributes Godlike authority to husbands, and thereby violates the concept of tawhid, according to which unqualified obedience is owed only to God (see Barlas 2002: 108).

I am not convinced that the reinterpretation view can succeed on its own terms. Even if some verses can be given less inegalitarian interpretations, those interpretations—like in the case of 4:34—are not egalitarian enough.¹¹ But even if the view were to succeed on its own terms, it still faces serious obstacles. To begin with, the reinterpretation view threatens to be dialectically ineffective. Unless one also shows that a feminist interpretation of the Quran is the right one—where the right interpretation is the one that captures God's intended message—there seems to be no reason why someone who is not already inclined to adopt a feminist reading should interpret the Quran in that way. When faced with two epistemically available interpretations, one sexist and another feminist, the proponent of the feminist interpretation would have no traction with someone who insists on following a sexist interpretation. To show that an interpretation is right, they must go beyond showing that it is merely an epistemically available interpretation. For example, they might rule out other candidate interpretations by appeal to further criteria, such as consistency with God's nature. And this is where the personal and political may come apart: it may be that on a personal level, I have reconciled the message of the Quran with a feminist outlook if I have accepted a feminist interpretation of the Quran. But unless I have also shown that a feminist interpretation is the right one (rather than merely an available one), I have not succeeded in showing that a feminist interpretation ought to be endorsed by Islamic societies and so ought to dictate Islamic tradition and way of life. In addition, until a feminist interpretation is shown to be the right one and not merely an available one, it is just a possible interpretation. But the guiding question of this paper is not whether it is possible to provide a feminist interpretation of the Quran, but whether the message of the Quran is in fact compatible with feminist values. And this latter question cannot be answered until the proponent of a feminist interpretation shows that it is the right one.

The demand that an interpretation must be shown to be the right one is not unique to feminist interpretations. Even standard (and, often, sexist) interpretations must meet it, insofar as the project is that of discovering the message of the Quran. Moreover, that a sexist interpretation is the standard one does not make it the right one, even if standard interpretations are perhaps more persuasive and thus dialectically effective than feminist reinterpretations simply because they represent the status quo.¹² Thus, the reinterpretation strategy threatens sexist interpretations as much as it threatens feminist ones. Unless a sexist interpretation has been shown to be correct, and not merely an epistemically available interpretation, there seems to be no reason why someone who is not already inclined to adopt a sexist reading should interpret the Quran that way. My concern in this paper, however, is with the viability of a feminist interpretation (given the reinterpretation view) rather than a sexist one, and so my focus will be whether a feminist interpretation can be defended through reinterpretation.

I will argue that proponents of feminist interpretations of the Quran have failed to show that such reinterpretations are right, which has both rendered these interpretations dialectically ineffective and prevented them from establishing that the message of the Quran is in fact compatible with feminist values. To see why, let us begin by noting that for Muslims, the Quran is not a mere object of reverence or a mere devotional text whose sole purpose is to generate religious feeling or to bring one closer to God. It is also a rulebook for how one should live here and now. The message of the Quran thus has normative force for Muslims. I will argue that the reinterpretation view threatens to undermine this normative force. I do this by showing that the view faces an epistemic version of Euthyphro's dilemma.

In Plato's Euthyphro (1997), Euthyphro proposes that an act is pious if and only if the gods love it. Socrates then responds, ‘Is the pious loved by the gods because it is pious, or is it pious because it is loved by the gods?’ This is a question about what is metaphysically prior: the fact that something pious is pious or the fact that the gods love that same thing? A parallel dilemma arises for the Quran as follows: Is something a norm because the Quran prescribes it or does the Quran prescribe it because it is a norm? And in this case, the answer for Muslims is straightforward: the norm is prior to the text. The Quran is a document that conveys the word of God rather than determines it. On an epistemic version of the dilemma, by contrast, the question concerns how Muslims recognize the norms concerning how they ought to live: Do Muslims recognize these norms because the Quran prescribes them or does the Quran appear to prescribe certain norms to Muslims because they already (independently) recognize those norms? Put differently, is the ultimate ‘detector’ of these norms the Quran or Muslims?

If the ultimate detector of norms is the Quran, then in order to discover those norms, we must be able discern what the message of the Quran really is. But, as I will show, this has not—and arguably cannot—be done without illegitimately bringing independently acquired feminist values to bear on the message of the text: the message is understood through the lens of these values.¹³ If Muslims are the ultimate detector of norms, then the Quran is redundant: we already know how we ought to live our lives, and so what point is there in looking to the Quran? In what follows, I discuss each horn of this dilemma in more detail.

Let us begin with the first horn. According to Muslims, a primary purpose of the Quran is to prescribe how one ought to live. The reinterpretation view exploits the ambiguity inherent in the text in order to yield a more egalitarian interpretation. Reinterpretation is thus agenda driven: egalitarian feminist values determine how the text is interpreted. Interpretation is, of course, influenced by background values in a number of ways. But this influence is problematic when the values that play a role in choice of interpretation are the very values we then take the text to prescribe. When values problematically influence interpretation in this way, the objectivity (in my sense) of norms prescribed by the Quran is undermined. And crucially, I contend, the reinterpretation view has not succeeded in showing that a feminist interpretation is the right interpretation without relying on independently acquired feminist values. The intratextual method and the tawhidic paradigm rely on either a governing prescription or the concept of tawhid, both of which are themselves subject to interpretation. For example, the verses in the Quran that purportedly suggest the moral and political equality of genders can be reinterpreted in a sexist way,¹⁴ as well as counterbalanced by verses that promote inequality between genders.¹⁵ Likewise, the concept of tawhid leaves room for interpretation. It is not clear, for example, that a woman's unqualified obedience to her husband violates the concept of tawhid: tawhid would be preserved if absolute obedience to God was mediated through obedience to one's husband, such that one obeys God by obeying one's husband. While distasteful to feminist sensibilities, this view does not put man on the same level as God and so does not violate God's incomparability. Thus, extant attempts to motivate a feminist interpretation of the Quran have not succeeded in showing that it is the right interpretation, for those attempts themselves depend on independently acquired feminist values.

To sharpen the worry, consider a different kind of case. Suppose that I can trust your testimony only if I have grounds for taking it to be reliable. But now also suppose that my only ground for taking your testimony to be reliable is that you have said that your testimony is reliable. Circularity threatens: your testimony can provide grounds for my belief only if I can trust it, but I can trust it only if your testimony can provide grounds for my belief. Similarly, suppose I can take the Quran to prescribe x only when I bring certain values to bear when reading the text. But now also suppose that the values that I bring to bear in interpreting the text are the very values I take the Quran to then prescribe. Circularity threatens. The problem with the reinterpretation view is that in defending a feminist interpretation of the Quran, it presupposes the very values it then takes the Quran to prescribe: values that pertain to gender norms and the moral and political status of men and women in society. It thereby undermines the objectivity of norms prescribed by the text. This gets us the first horn of the dilemma.¹⁶ To be fair, the reinterpretation view is also problematic when it comes to sexist interpretations: it is far from clear that background values do not problematically influence sexist interpretations, i.e., influence them in a way that undermines the objectivity of norms prescribed by the text. But if this is right, then the problem that I have highlighted with the reinterpretation view for feminist interpretations plausibly faces all interpretations, for one might argue that no interpretation can be entirely neutral on questions that concern gender (after all, any interpretation will need to determine how to interpret the reference to daraba in verse 4:34). But if all interpretations are such that they are problematically influenced by background values in a way that undermines objectivity, then it appears that a consequence of endorsing the reinterpretation view is that the norms prescribed by the Quran are unknowable.¹⁷ I am happy to concede that the foregoing is a consequence of the reinterpretation view. Since I ultimately reject the reinterpretation view as inadequate, it is not a consequence that (on my view) threatens the status of the Quran as a prescriptive document that guides Islamic practice. I will later discuss why this worry with the reinterpretation view does not affect my positive proposal.

If the message of the Quran is not underdetermined, then the text has one right interpretation. Insofar as our understanding of the message of the text cannot be postponed (we need to know now what it prescribes so we can act accordingly), devout proponents of an interpretation of the Quran must take their interpretation to be the right one even if it has not been shown to be right (as I argued above, attempts to show that a feminist interpretation is right undermine the objectivity of the norms prescribed by the Quran). And this is where the second horn arises: in taking their interpretation to be the right one—an interpretation that is based on values held independently of the Quran—the Quran is stripped of its normative force for Muslims. What is there for the Quran to do when we already have independently acquired values (perhaps based in intuition or a ‘moral compass’) that dictate how we ought to live, and which play an essential role in determining what we take the message of the Quran to be?

I have shown that the reinterpretation view undermines the normative force of the Quran either by undermining the objectivity of the norms it prescribes or by rendering its message irrelevant. Before I go on to discuss the contextual view, I will consider two potential objections to my argument against the reinterpretation view.

First, one might argue that even if we are the detectors of the relevant norms, the Quran can still play an important role in guiding and clarifying the application of those norms. However, the Quran can only guide and clarify the application of those norms if the norms are just those prescribed by our moral compass independently of the Quran. But, as discussed above, the first horn of the epistemic version of Euthyphro's dilemma undermines the very possibility of identifying just which objective norms the Quran prescribes, and so also undermines the very possibility of identifying the norms whose application the Quran clarifies. The objector also makes the Quran relevant for Muslims for the wrong reason. For Muslims, the normative force of the Quran comes from the fact that it is the complete guide for how to live as opposed to a guide that supplements independently acquired values. The Quran is supposed to have categorical force for Muslims, not a force that is conditional on whether they subscribe to independently acquired values.

A second objection seeks to undermine the assumption that proponents of the reinterpretation view must show that theirs is the right interpretation of the Quran. Why can these theorists not hold that there is no single right interpretation of the Quran, and thus rest content with showing that their interpretations are simply available? This line of objection cannot succeed. If it turns out that there are many available, mutually inconsistent interpretations, but no right interpretation, then the Quran is underdetermined with respect to its message. But if so, the Quran has no normative force for Muslims. To see why, suppose that one interpretation of the Quran prescribes φ, but another prescribes not-φ. Then on the assumption that there is no one right interpretation and that the normative force of the Quran derives from the fact that it is God's word, the two prescriptions would cancel one another out: because there would be no unique determination of the meaning of the text, there would be no answer to the question ‘what is the word of God?’—an unacceptable result for Muslims! However, even if we do not currently have a way to tell which interpretation is the right one, it does not follow that there is no right interpretation or that it is in principle impossible to determine which interpretation is the right one.

2.2 The Contextual View

The contextual view takes into account the historical context in which a verse was revealed, including the moral universe and social conditions of the people to whom it was revealed. Going back to verse 4:34 (on obedience), this is how the contextual view might proceed. Barlas (2002: 188) argues that once we take into account the historical context in which the verse was revealed, the apparent sanction to strike one's wife is a restriction rather than a license, for, at the time, men did not need permission to strike women: the verse prescribes striking as a last resort.

Likewise in the case of polygyny (i.e., a man having more than one wife): the verse in which the Quran apparently sanctions polygyny is (in historical context) a way of providing for female orphans:

And give to the orphans their properties and do not substitute the defective [of your own] for the good [of theirs]. And do not consume their properties into your own. Indeed, that is ever a great sin. And if you fear that you will not deal justly with the orphan girls, then marry those that please you of [other] women, two or three or four. But if you fear that you will not be just, then [marry only] one or those your right hand possesses. (4:2–3)

This verse permits men to marry up to four female orphans in the absence of a social safety net that would protect female orphans from destitution. Understood in historical context, the verse prescribes a way of providing for female orphans in the absence of any other safety net and so does not constitute a universal (i.e., one that holds at all times and in all places) permission for men to marry up to four wives.

The problem with the contextual view is that it renders the Quran a historical document with a message whose normative force is conditional on the time and place in which it was revealed. This view thus undermines the universal normative force that Muslims take the Quran to have. If the contextual view is right, then Muslims today have no reason to take any prescription in the Quran to govern their present behavior.

One might attempt to resist this conclusion by maintaining that even though specific prescriptions of the Quran are dependent on historical context, these prescriptions are instantiations of general prescriptions that possess normative force here and now (a prescription such as ‘provide for orphans’). However, which general prescription we ‘abstract away’ from the more specific prescriptions will likely be determined at least in part by our background values. The epistemic version of Euthyphro's dilemma would threaten again.

Yet perhaps it is possible to draw a distinction between verses that have universal normative force and those that do not. Any such distinction would have to rest on a standard that is independent of our values, for otherwise the problems that plague the reinterpretation view will re-emerge for the contextual view. In the next (and final) section, I propose a novel approach to reconciliation that appeals to a standard independent of our own values for distinguishing those verses that have universal normative force from those that do not. I will show how my approach secures the place of the Quran as a prescriptive document with objective and universal normative force, while pointing the way forward for reconciliation between feminist ideals and the message of the Quran.

3. Mecca and Medina: A New Proposal

My arguments against the approaches discussed in §2 make room for a different type of strategy for reconciling the putatively sexist verses in the Quran with a feminist outlook. Crucial to my argument for this strategy is a widely recognized distinction between two types of verses in the Quran: those that were revealed in Mecca, where the Muslim community was initially smaller and where Mohammed started his ministry, and those that were revealed in Medina, where Mohammed eventually moved in 622 CE in order to escape an assassination plot, and where the Muslim community was larger and had a more complex demographic (cf. Gilliot 2006: 42). The Meccan verses differ from the Medinan verses not just in where they were revealed, but also in their content: the Meccan verses tend to be generic (i.e., not pertaining to particular situations) and egalitarian, whereas the Medina verses are often inegalitarian, sometimes promote violence, and often specify norms that govern very specific situations. The Meccan ‘message’ was rejected by the people in Mecca, prompting—along with Muhammad's move to Medina—an eventual shift in the content of the message revealed by Muhammed. (Cf. An-Na'im [1990] for a detailed characterization of this distinction; An-Na'im develops the view of his teacher Mahmoud Muhammed Taha, who was executed in 1985 in Sudan for sedition and apostasy.) The verses also differ in style: the Meccan verses tend to be short and belong to relatively short surahs (chapters), whereas the Medina verses tend to be longer and belong to longer surahs. All the putatively sexist verses discussed in this paper belong to the Medinan message.

This distinction between the Meccan message and the Medinan message does not, on its own, suggest that the Medinan message should be ignored in favor of the Meccan message. After all, from a Muslim's point of view, both were revealed to Mohammed and both constitute ‘God's word’. An-Na'im and Taha have argued that an Islamic reformation ought to be based on the Meccan message (cf. An-Na'im 2008), but it would seem that such a reformation could succeed only if we have good grounds for putting aside the Medinan message. The question then is this: Does the difference in the location of revelation track a deeper difference in type of message—a difference that might affect the normative force of the relevant messages? My answer is yes.

I contend that the distinction between Meccan and Medinan verses maps onto a distinction between norms that govern how to be, and norms that govern what to do. Some examples of norms that tell us how to be are those expressed by ‘be just’, ‘be kind’, and ‘be pure’. By contrast, some examples of norms that tell us what to do are those expressed by ‘take an eye for an eye’, ‘do what maximizes greatest happiness for all’, and ‘refrain from consuming that which is haraam (forbidden)’. Let us call the norms that prescribe how to be ‘thick norms’ and the norms that prescribe what to do ‘thin norms’. From verse 4:34 (on obedience, quoted above), we may then derive a thin norm that says, ‘It is permissible to strike your wife when she disobeys you’. Verse 4:34 is a Medinan verse. By contrast, consider the following verse:

O Children of Adam, take your adornment at every masjid, and eat and drink, but be not excessive. Indeed, He likes not those who commit excess. (7:31)

This verse prescribes a thick norm—a norm that says ‘be not excessive’. And it is a Meccan verse.

I claim that the two types of norms differ in that Muslims should treat thin norms as having a normative force that is conditional on the social milieu in which the verses prescribing them were revealed, and thick norms as having unconditional normative force. Thin norms thus have normative force in the social milieu of seventh-century Arabia, but need not have normative force in any other social milieu. The verses that express these norms largely consist in pronouncements about what (specifically) would be right, wrong, permissible or forbidden in seventh-century Arabia. By contrast, thick norms have a normative force that holds at all times and in all places, regardless of when and where the relevant verses were revealed. They thus have universal normative force.¹⁸ The verses that express these norms tend not to take the form of pronouncements about what (specifically) is right or wrong or permissible or forbidden, but rather promote certain virtues and discourage certain vices. Yet these norms will not (on their own) be very informative with respect to appropriate practice in a given time or place; they will be action guiding only once they are understood in light of facts about the conditions that prevail at a time and place. The thin Medinan norms are expressions of how best to implement the thick norms in a particular time and place. They thus depend on thick norms for their legitimacy.

Compare standard thick norms such as those expressed by ‘be courageous’ or ‘be wise’. These norms provide some general standard of right conduct, but successfully guide conduct only when understood in light of prevailing material conditions. So, we might ask: ‘How should I be courageous?’ And a legitimate reply might be expressed as a thin norm: ‘By doing this and that at this time and in such and such a way’. How one should fill in that schematic answer will depend upon the context in which one is trying to be courageous. A consequence of this view is that the Quranic norms that prescribe how to be (i.e., thick norms) are explanatorily prior to the Quranic norms that prescribe what to do (i.e., thin norms).¹⁹ In particular, the thin norms derive their legitimacy from the thick norms.

To see the relationship between thick and thin norms at work in the Quran, consider a Meccan verse that prescribes mercy:

And not equal are the good deed and the bad. Repel [evil] by that [deed] which is better; and thereupon the one whom between you and him is enmity [will become] as though he was a devoted friend. (41:34)

This verse expresses an unconditional norm. By contrast, consider this Medinan verse that prescribes punishments for various wrongs:

And we ordained for them therein a life for a life, an eye for an eye, a nose for a nose, an ear for an ear, a tooth for a tooth, and for wounds is legal retribution. But whoever gives [up his right as] charity, it is an expiation for him. And whoever does not judge by what Allah has revealed, then it is those who are the wrongdoers. (5:45)

This verse prescribes what would be the ‘right’ or just thing to do in the case of various wrongs while also explaining how one might show mercy by giving up the right to retribution in these cases. The norm, however, is particular to seventh-century Arabia: what qualifies as legal retribution in any given case is context dependent. Thus, while the Meccan verse promotes a general norm—namely to be just and merciful—the Medinan verse tells us how the norm should be implemented in seventh-century Arabia. We, of course, do not live in seventh-century Arabia, and so are not bound by the thin norm expressed by the Medinan verse. At this time and in this place, the way to be merciful is not by giving up the putative right to ‘an eye for an eye’ (for extrajudicial punishment is now illegitimate) but (for example) by attempting to mitigate judicial sentences through victim testimony.

Yet how can we determine how to act while relying only on thick norms? On their own, those norms tell us how to be not what to do. Doesn't this approach to the Quran leave too much to us, and so undermine the Quran's ability to govern how we ought to live? I argue that it does not. The Quran provides us with an example of what the relationship between thick norms and thin norms ought to be. It therefore makes it possible for us to grasp the right relationship between these norms. And once we have grasped this relationship, we will be in a position to see how the thick norms of the Quran can be implemented in our sociopolitical context.

The application of thick norms to our sociopolitical context requires an exercise of mature moral understanding. While the example given by the Quran provides us with the basis for acquiring this understanding, it does so only in conjunction with our appreciation of the circumstances in which the Quran was revealed (i.e., those of seventh-century Arabia). The moral understanding we acquire is not autonomous: one cannot move from thick norms to thin norms without also knowing something about the relevant material facts. Full moral understanding is therefore not timeless on my account (at least not for Muslims): it requires constant maintenance in light of our evolving circumstances. As a result, my proposed approach is revolutionary rather than merely apologetic. It secures a place in Islamic practice for acquiring and maintaining a context-sensitive moral understanding. Islamic practice requires not absolute and uncritical obedience to an inert text, but obedience informed by a moral understanding that is responsive to changing social conditions.

At this juncture, one might raise the following objection. On my proposal, verse 4:34 prescribes that in the social milieu of seventh-century Arabia, a husband is sometimes permitted to strike his wife. But surely (the objection goes), striking one's wife is unacceptable simpliciter, both now and in seventh-century Arabia. I believe that this is the most threatening objection to my proposal. Yet a response is available. Let us draw a distinction between those norms that are ideal and those that are ‘stepping-stones’ to the ideal norms. Armed with this distinction, we might insist that the thin norm prescribed in 4:34 is not ideal, but a norm that paves the way toward an ideal norm. Specifically, we can construe the norm prescribed in 4:34 as making a small moral gain. Suppose, as Barlas (2002: 188) claims, that the pre-existing norm in seventh-century Arabia was to strike one's wife as a first resort in the event of disobedience. By contrast, 4:34 prescribes that striking should not be a first resort, but the last one. If the ideal norm is feminist, then as a ‘stepping-stone’ prescription, the seventh-century prescription is a means to a feminist norm in the following sense: in order for a population (e.g., that of seventh-century Arabia) that is morally regressive by the standards of contemporary egalitarian societies to be in a position to follow the ideal norm, it must first follow a nonideal norm—one that signifies a moral improvement given the status quo—of the of the kind prescribed by 4:34. This move does not presuppose that the norms endorsed by contemporary egalitarian societies are ideal: the ideal norms may be inaccessible, or further along on the path of moral progress.

Still, one might worry that the force of the objection remains: How could it ever be correct to strike one's wife in any circumstance? But here we must draw a distinction between a norm that is correct and a norm that is morally ideal. We can hold that it would be morally wrong to strike one's wife in any circumstance, but still correct to do so in a particular context. An analogy with strategic voting might help here. A strategic voter votes for candidate B over the ideal candidate A, because B is the candidate most likely to win (where the goal is to prevent candidate C from winning). Likewise, we might imagine God to be a ‘strategic norm giver’: while a thin norm A might be the ideal norm, norm B might the one most likely to be accepted (given prevailing social conditions) and the norm that can then pave the road for the ideal norm.²⁰ In these circumstances, norm B would be the correct norm, even if morally suboptimal.

It is worth noting that the strategy I have proposed above is not new to Islamic thought. In particular, it is adopted by Ibn Tufayl (1105–1185), a mentor and friend of Averroes.²¹ In his philosophical tale Hayy ibn Haqzan, the protagonist Hayy arrives at moral truths through reason but fails in the attempt to communicate those truths to others. As Ibn Tufayl (2003: 164) writes, ‘Hayy now understood the human condition. He saw that most men are no better than unreasoning animals, and realized that all wisdom and guidance, all that could possibly help them was contained already in the words of the prophets and the religious traditions.’ The upshot of Hayy's story and Ibn Tufayl's conclusion is that most people are not in a position to accept moral truths as they really are, and must be guided along the path of moral progress by ‘the prophet's teachings and religious traditions’.

One might, however, object in the following way to the strategy I have employed in responding to the original objection that striking one's wife is unacceptable simpliciter: the acknowledgement that striking one's wife is wrong simpliciter and the accompanying reply that the permission to strike one's wife is merely a stepping-stone to the ideal norm illegitimately presupposes feminist values not derived from the Quran, and thus falls victim to the very objection that was levelled against the reinterpretation view. That objection, recall, was that extant attempts to show that a feminist (re)interpretation is the right one presuppose the very values that the Quran is then taken to prescribe, thereby undermining the objectivity of the norms prescribed by the Quran.²² My response is twofold. First, it is worth pointing to a significant difference between the objection raised against the reinterpretation view and the objection that the present strategy illegitimately appeals to feminist values not derived from the Quran. In the first case, independently acquired background values affect what one takes to be the meanings of particular verses. In the second case, independently acquired background values do not affect what one takes to be the meaning of particular verses, but instead influence the status of the norms prescribed by particular verses—i.e., influence whether we take the norms in question to be ideal or merely ‘stepping-stone’ norms. And I contend that this second kind of influence, unlike the first, does not undermine the objectivity of the norms prescribed by the text. Even if background values affect whether we take a norm to be the ideal norm or a ‘stepping-stone’ norm, the content of the norm remains unaffected, and so the objectivity of the norm prescribed by the text is not threatened. Second, insofar as it is the objector who insists that striking one's wife is morally wrong in all circumstances, it is the objector who, dialectically speaking, presupposes a feminist value that does not arise from the Quran. And so, if it is indeed illegitimate to bring feminist values to bear on whether a norm prescribed by the Quran is ideal or merely a stepping-stone to an ideal norm, then the original objection itself illegitimately presupposes such values, and so fails as an objection on those grounds.

My proposed approach maintains the status of the Quran as a document that has the authority to dictate how Muslims ought to live their lives, while making possible a feminist Islam. This authority remains because certain core norms—those that tend to be expressed by the Meccan verses—have unconditional normative force. It is important to my proposal that the Medinan verses still have normative force albeit a conditional one: any thin norms will apply when some specific conditions are met. Yet the proposal does not collapse into the contextual view. Whereas, on the contextual view, the Quran in its entirety has a message that is context dependent, on my approach the message of the Quran is fundamentally timeless, even if some verses retain their normative force only relative to a certain set of material and historical conditions. The normative force of the Quran is thus secure.

Also, unlike the reinterpretation view, my approach does not fall victim to the epistemic version of the Euthyphro dilemma: we have an independent standard—namely, whether the verse is part of the Meccan message or the Medinan message—for determining which Quranic verses have an unconditional normative force and which have a conditional normative force. As mentioned previously, all the putatively sexist verses discussed at the beginning of this paper belong to the Medinan message. As a result, the worry with the reinterpretation view highlighted earlier—that the first horn of the epistemic version of Euthyphro's dilemma undermines the very possibility of identifying just which objective norms the Quran prescribes—leaves my positive proposal unaffected: only the Meccan verses, or more generally the verses that prescribe thick norms, have unconditional normative force on my view, and so bind us here and now. And because the putatively sexist verses all prescribe thin norms and tend to all belong (and as far as I know, do all belong) to the Medinan message rather than the Meccan message, it does not matter that we cannot reliably identify these thin norms, for those prescriptions do not bind us here and now.²³ Muslim feminists can thus embrace a feminist Islam while leaving intact the place of the Quran as a text that tells Muslims how they ought to live.

Concluding Remarks

I began by showing how extant approaches to reconciling the message of the Quran with feminist values have gone wrong. I then argued for an alternative approach to reconciliation. This alternative approach—unlike the reinterpretation view and the contextual view—has the advantage that it makes room for the possibility of a feminist Islam without stripping the Quran of its normative force. It also secures an important role for moral understanding within Islam-as-practiced, and therefore constitutes a revolutionary (rather than apologetic) approach to how Muslims ought to engage with the Quran.

I conclude with a brief note about methodology. Saba Mahmood (2003, in discussing the apparent tension between Islam and the ideals of a liberal democracy, raises the following worry: ‘[w]hat's curious to me is that in these explorations by Muslim scholars Islam bears the burden of proving its compatibility with liberal ideals, and the line of question is almost never reversed.’ As mentioned at the outset, an analogous worry arises for my project of reconciling the message of the Quran with feminist values. In response, I contend that my project does not assume that in the event of a conflict between the message of the Quran and feminist values, it must be the message that must ‘bend’ and accommodate feminist values. The project instead questions whether there is a tension between the message of the Quran and feminist values in the first place. While there is arguably a clear tension between the Islamic tradition, which dictates how Islam is practiced in many Muslim countries and feminist values,²⁴ it would be question-begging to assume at the outset that the message of the Quran is in conflict with feminist values. And, indeed, a core upshot of my paper is that there is no such conflict.

Moreover, my project does not suggest that women who actively choose to support patriarchal structures lack agency.²⁵ Nor does it suggest that such women are not feminists. One may have personal agency even if one chooses to exercise it in support of nonfeminist, and even sexist, ideals. And one may qualify as a feminist even if one promotes some nonfeminist and sexist values. My concern in this paper has not been with the question of who should count as a feminist, but with egalitarian feminist values. And I hope to have shown that those values can be reconciled with the message of the Quran while preserving the Quran's prescriptive authority for Muslims.