Karen Armstrong, Fields of Blood: Religion and the History of Violence (New York: Alfred Knopf, 2014).
Can religion help to explain political outcomes? The ostensible religious zeal of Da’esh or the Taliban, the battles over abortion and reproductive technologies in several predominantly Catholic countries, and the culture wars in the US all invoke religious tropes and justifications. Yet both scholarly and popular apologists have argued that religion is neither responsible for, nor an explanation of, outcomes such as violence.
Instead, several authors, ranging from Karen Armstrong to William Cavanaugh to Rodney Stark, have made three sets of arguments: (1) that religion is difficult to define, has multiple interpretations and is thus difficult to distinguish from politics; (2) that religion is not responsible for widespread conflict or violence; and (3) that instead religiosity is not only associated with a set of highly desirable beliefs (such as the Golden Rule), but that it also produces significant spillover effects that benefit society as a whole.
In this review of Karen Armstrong’s Fields of Blood, I will argue that many of these arguments are inconsistent and do not cohere. They exemplify the ‘difficulty with doctrine’ as an explanation: even very serious examinations of religion’s role in politics do not help us to distinguish whether, how, and when religion affects politics. Using one of the most prominent works in this tradition, I show that critically, in both their conceptual and logical structure, many arguments about religion’s role in political conflict are unfalsifiable. They both assert and deny the role of religion in politics, but they do so largely as a consequence of relying on conceptualizations of religion that are unhelpful and that do not allow us to distinguish when and how religion and doctrine might matter. Instead, I argue that we should take religious rhetoric and the actors who use it seriously. Rather than asking how religion writ large matters, we should examine specific aspects of religion, such as religious doctrine, and inspect particular contexts where it may play a role and interact with political institutions and actors. I illustrate the possible role of doctrine with two examples where it played a role in both debates and political outcomes: reproductive technology policies and recent violent conflict.
KAREN ARMSTRONG’S FIELDS OF BLOOD
One of the most recent prominent works in the apologetic tradition is Karen Armstrong’s Fields of Blood. This is a magisterial account of religion’s role (or rather, non-role) in violence across the millennia of human civilization. She begins by charging that ‘in the west the idea that religion is inherently violent is now taken for granted and self-evident’ (Armstrong Reference Armstrong2014: 3), and that ‘modern society has made a scapegoat of faith’ (Armstrong Reference Armstrong2014: 3). It is against this perceived convention that Armstrong argues that religion ‘does many things’ – and there is nothing inherent in religion that produces violence. If anything, religion has historically offered a peaceful alternative to state violence where it could.
There are three critical elements to this argument. First, Armstrong argues that there is no universal way to define religion, which is both open to a variety of interpretations and difficult to disentangle from politics. Religion is difficult to define and has many interpretations. It has meant everything from a set of practices to a hierarchy of beliefs; included one god, many, or none at all; focused on the afterlife or on maintaining a political order. Religion is multifarious and open to interpretation: ‘religious tradition is never a single, unchanging essence that compels people to act in a uniform way. It is a template that can be modified and altered radically to serve a variety of ends’ (Armstrong Reference Armstrong2014: 123). As a result, often it becomes impossible to distinguish religion from politics. The lines are indistinct: both the ancient conceptualization of religion as a public practice and its close association with political order mean that the lines between religion and politics have been historically blurred. Further, ‘since all premodern state ideology was inseparable from religion, warfare inevitably acquired a sacral element’(Armstrong Reference Armstrong2014: 15). Religious authorities buttressed secular ones, and in return received protection and flourished.
Second, from the time when agricultural societies began to organize onwards, a systemic oppression was started by a small elite who exploited surplus production, gained a monopoly of violence, and dominated the rest of the population. This continuing ‘structural violence’ of the state is a source of cultural and scientific progress – the surplus capital is invested to fund science and the arts. But it is also responsible for the violent conflict that we observe across centuries. Religions here have been harnessed to justify the oppression, ‘sanctifying the structural violence that was essential to the survival of the civilization’ (Armstrong Reference Armstrong2014: 38). And it is this structural violence that creates a history of grievance, which drives secular conflict and ‘often radicalizes a religious tradition and can even push an originally irenic vision into a campaign of violence’ (Armstrong Reference Armstrong2014: 164). The structural violence of the state remains at the core of violent conflict within and between societies. In other words, religions can radicalize and appear to promote violence, but the state made them do it.
Finally, religious prophetic voices have offered an alternative to this propensity for violence. Even so, ‘they could not radically change their societies; the most they could do was propose a different path to demonstrate kinder and more empathic ways for people to live together’ (Armstrong Reference Armstrong2014: 16). Where it could, religion struggled to articulate a peaceful alternative to state violence – but rarely had the opportunity to see the peaceful aims inherent in the Golden Rule (‘do unto others as you would unto yourself’) become either the law or the practice of the land. In a wide-ranging historical analysis that crosses the millennia, Armstrong consistently finds that even if violence wears many guises, religion cannot be held responsible, whether in ancient Sumeria, the Chinese Warring State, the founding of the three Abrahamic religions and their subsequent fractioning and dispersal, or in modern-day wars and terrorism.
In several ways, these arguments represent a broader set of responses to the criticisms of religion of the ‘New Atheists’: Richard Dawkins, Sam Harris, Christopher Hitchens and others. These writers have denounced what they see as religious excess, irrational zeal, and violent tendencies. In response, scholars such as Rodney Stark, one of the founders of the ‘political economy of religion’, wrote about the benefits of religious belief not only to the believers but to society as a whole (his America’s Blessings of Reference Stark2012 argues that believers exhibit more sociotropic behaviours than nonbelievers, including lower rates of criminal activity, more contributions to charity, lower divorce rates, etc.). In a nuanced argument, the echoes of which can be found in Armstrong (Reference Armstrong2014), William Cavanaugh in The Myth of Religious Violence has argued that religion itself is a changing and diverse concept, and therefore we cannot claim that religion writ whole is responsible for violence. Instead, he argues that we need to look at the conditions under which religion might turn violent (Cavanaugh Reference Cavanaugh2009: 8), and posits several characteristics that distinguish religious conflict (Cavanaugh Reference Cavanaugh2009: 141–3). Much like Armstrong, he finds that attributing violence to religion is unwarranted, largely due to the difficulty in separating out religious motives from secular ones – but concludes instead that this does not absolve religion.
THE DIFFICULTIES WITH DOCTRINE
These works all illustrate the ‘difficulty with doctrine’. Armstrong’s arguments in particular do not allow us to assess properly any potential causal relationship between religion and violence, because they are unfalsifiable. This problem manifests itself conceptually, empirically, and logically.
Conceptual
First, both Armstrong and Cavanaugh agree that there is no one simple or universal definition of religion. Armstrong (Reference Armstrong2014: 123) argues that ‘a religious tradition is never a single, unchanging essence that compels people to act in a uniform way. It is a template that can be modified and altered radically to serve a variety of ends.’ Similarly, Cavanaugh (Reference Cavanaugh2009: 3–4) states that ‘there is no transhistorical and transcultural essence of religion … attempts to separate religious violence from secular violence are incoherent. What counts as religious or secular in any given context is a function of different configurations of power.’ For her part, Armstrong goes a step further, and repeatedly argues that religions are not only different from each other and highly diverse, but that they are multifarious, containing both benevolent and malevolent interpretations.
Yet despite these claims that there is no general definition of religion, Armstrong constantly generalizes about ‘religion’ across time, space, and doctrine. Here, insofar as religions share any doctrinal characteristics in her account, it is that they all promote peaceful and loving communities – limiting their capacity to incite violence a priori. Conceptually, this is having it both ways. If there is no universal definition of religion, then how can we conclude it is not responsible for violence (or anything else)? Why, for example, is the ‘violence of secularism’ (Armstrong Reference Armstrong2014: 317) a given, but religion is malleable and therefore not complicit? As Cavanaugh points out, it is impossible ‘to separate religious from economic and political motives in such a way that religious motives are innocent of violence’ (Cavanaugh Reference Cavanaugh2009: 5), especially if some religious traditions (such as Islam) themselves do not separate out the political from the religious.
Armstrong’s argument thus hinges on distinctions she herself disavows. On the one hand, she repeatedly claims that religion and politics are historically inseparable; on the other, she declares the former innocent of charges and the latter responsible for violence. But if religion cannot be separated from secular life, then we cannot have truly ‘secular’ conflict, nor can we eliminate the possibility that the inherent ‘structural violence’ of the state is religious in origin.
Further, despite arguing that there are multiple competing interpretations, Armstrong also asserts that a ‘proper’ understanding of doctrine in Judaism or Islam, for example, would not permit certain violent practices. Yet if multiple interpretations of doctrine exist, and Armstrong gives us no standards for evaluating them, then why is hers the ‘proper’ interpretation? Why conclude, for example, that the motives of Jewish militants were ‘far from being inspired by their religious tradition, the militants’ conviction violated core teachings of Rabbinic Judaism’ (Armstrong Reference Armstrong2014: 369)? She further argues that ‘when Muslims attack churches and synagogues today, they are not driven to do so by Islam. The Qur’an commands Muslims to respect “the people of the book”’ (Armstrong Reference Armstrong2014: 396). That may be true, but since religion is multifarious and open to interpretation, why should we accept Armstrong’s particular reading here? Why would an interpretation that leads some to kill infidels be any less valid? Perhaps most bizarrely, the 9/11 terrorists and their collaborators are said not to be that familiar with Islam, many of them having recently converted (Armstrong Reference Armstrong2014: 381). Atta and his co-conspirators used a document that ‘atomizes the mission’ and broke it down into a schedule and a sequence. This violated the Islamic precept of tawhid (making one) and integrating their activities and thoughts, making their entire enterprise unIslamic in Armstrong’s view. In Armstrong’s reading, using a calendar is a non-Islamic activity: and even though there are no ‘correct’ interpretations, we are to accept this one as definitive.
These assertions of what constitutes a ‘proper’ interpretation of doctrine deny that the actors involved have a clear understanding of their own agency. For example, Armstrong argues that members of Hamas ‘travel on the path of Allah’, but are not motivated by religion (Armstrong Reference Armstrong2014: 364). The author’s interpretation takes precedence over the stated aims of the agents directly involved. This begs the question of what makes her right in either her interpretations of doctrine or her causal explanations. The resulting, and troubling, consequence is the paternalism inherent in claiming a ‘proper’ interpretation of another’s religion: the claim is here of a superior claim to knowledge of another’s belief, an authority asserted but not so readily granted. Yet there are no logical, historical, or theological grounds to conclude that Armstrong’s is the ‘proper’ reading.
Thus, Armstrong’s account has difficulty with both identification (distinguishing religion and religious motives clearly and distinctly) and interpretation (establishing whether something was done in accordance with doctrine or not, whether it was the ‘proper’ orthodox interpretation or blasphemy). And here Armstrong offers us little conceptual clarity, leaving herself open to charges of cherry-picking. But this works both ways: if we cannot make a distinction, then we cannot absolve or exonerate religion, any more than we can find it culpable. If religions are open to interpretation, then the radical views of the fanatic or the zealot carry as much weight as that of ‘moderate’, liberal academics.
Empirical
Secondly, there is no a priori standard by which we could find religion responsible for violence. Armstrong does not specify the conditions under which we could consider violence religious or caused by religion. Inevitably, acts done in the name of religion in Armstrong’s accounting are individual aberrations, blasphemy or misinterpretation of an inherently peaceful religious tradition. As a result, her argument is not empirically falsifiable: there is no standard or criteria for evidence that would prove her account wrong.
Here, Cavanaugh offers a standard for distinguishing the role of religion; for him, a conflict is religious if: (1) combatants oppose each other on the basis of religious difference; (2) the primary cause is religious rather than political, economic, or social; (3) religious causes must be analytically separable from these other potential causes; and (4) the rise of the modern state can be eliminated as a cause (Cavanaugh Reference Cavanaugh2009: 141–2). Yet even here, problems arise: firstly, one of his standards for distinguishing a religious cause is that religion must be a cause. This is less than helpful empirically. Secondly, in its application, this standard becomes problematic: Cavanaugh criticizes Martin Appleby’s account of the religious participation and legitimization of Yugoslav civil wars as not providing a ‘serious warrant for attributing violence to a sui generis impulse called religion’ (Cavanaugh Reference Cavanaugh2009: 48). But why? The combatants often identified themselves by their religious affiliation (and more importantly, targeted at least one group on that basis), and religious actors repeatedly motivated and justified the violence (see Perica Reference Perica2002). What constitutes a serious warrant? Finally, and critically, religion may very well be a causal factor in conflict, even when it is not the only factor, yet in both Armstrong and Cavanaugh’s accounts, for religion to act as a cause it must be the only reason for the rise of conflict or other phenomena.
Armstrong does not provide a standard. Instead, violent impulses are inevitably ascribed to secular actors, or religious actors gone astray. Yet given what a powerful force religion is, it would be truly unusual if it was never strongly implicated in violence. Yet Armstrong does not provide a standard by which we would know. Without a clear standard for what would constitute a religious cause, Armstrong’s arguments and characterizations become unfalsifiable.
Instead, she falls back on the indistinguishability of religious, social, and political causes, especially prior to 1700. Even so, how can we conclude that medieval anti-Semitism or the Spanish Inquisition were not caused by religious considerations? Nor is it the case that there was no distinction made between religious and secular forces; as Malcolm (2015) points out, there were lengthy debates from the Middle Ages onwards about the difference between Church and State, human and divine law or theology and philosophy.
Logical Status
Armstrong argues that she is attempting to falsify the claim that religion implies violence. The problem is that the logical structure of her argument precludes this conclusion. The statement ‘religion implies violence’ is logically true if: (1) we observe both religion and violence; (2) we see no religion and yet we see violence (violence has other causes); and (3) we see no religion and no violence. The statement is only falsified, and Armstrong is only correct, if (4) we see religion but no violence. Yet that last condition does not obtain, in both Armstrong and Cavanaugh’s accounts. There are no logical grounds on which to conclude that religion does not imply violence.
Instead, Armstrong pursues a different line of argumentation and emphasizes how much conflict has been non-religious. She demonstrates there is plenty of secular, non-religious conflict, such as the First World War, the Second World War, the horrors of Stalin or Pol Pot, of Nazi and Communist ideologies. There are two problems here: first, we are again suddenly able to distinguish which conflict is secular or religious – a position Armstrong earlier disavowed. Armstrong’s argument here only serves to emphasize a peculiar inconsistency: the impact of religion cannot be considered apart from the state, but the converse does not hold. For all the attempts to show how religion and the secular state ‘coinhere’, Armstrong’s argument rests on the ability to separate out the two and point out that it is the secular, not the sacred, that is responsible for violence, oppression, inequality, and so on.
Secondly, and fundamentally, no matter how much ‘secular’ conflict exists, this does not absolve religion. The presence of secular conflict does not make religion inherently peaceful. This is a bit like saying that since exposure to radiation causes cancer, smoking does not. Further, religion may not be sufficient for conflict in general, but it may be necessary in particular cases: some smokers do not get cancer, but some cancer patients only got cancer because they smoked.
In the end, it is not that Armstrong’s argument is necessarily wrong; it is that we cannot know whether it is, based on her concepts, empirical evidence, and the logic of her argument. As a result, while Armstrong’s examination of various religious traditions is both fascinating and deeply sympathetic to the intricacies and subtleties of human religious experience, it is difficult to conclude where and how religion is responsible for violence, conflict, and other political outcomes.
MOVING FORWARD
How then, can we retain the sensitivity to the diverse and changing nature of religious experience and yet more clearly distinguish the role of religion in a given context? If our goal is a more falsifiable explanation, then we cannot simply rely on asserting one interpretation over another, conceding that religion is undefinable, or arguing that politics and religion are indistinguishable. Instead, we can: (1) pay close attention to the religious rhetoric and declarations of political actors; (2) disaggregate the broad category of religion into constituent aspects, and examine these; and (3) examine how the constituent aspects of religion interact with specific disaggregated aspects of politics: how churches form coalitions with each other and with political parties, for example, or how some doctrinal claims resonate broadly in political campaigns while others do not. We can then provide standards of evidence, construct counterfactuals, and specify causal mechanisms that allow the reader not only to gain confidence in the account being advanced but also to weigh it against other accounts. These suggestions are incomplete – but they may make the study of religious influence more tractable and more falsifiable.
First, when political actors use religious rhetoric to justify their actions, we should heed the actors’ own stated motivations. Rather than insisting on multiple interpretations but then asserting our own as the ‘proper’ one, we can rely on what actors see and state as their religious motivations. Here, as Melford Spiro (Reference Spiro1966: 112) argued, ‘the most obvious basis for religious behavior is the one which any religious actor tells us about when we ask him.’ Rather than imputing ‘proper’ or improper motivations, an actor’s self-stated motivation should suffice to call an act religiously motivated. Such statements may be instrumental, rather than sincere – but they are not said without a reason. The instrumental use of religious rhetoric nonetheless presupposes a willing audience, and an expectation that such claims will resonate. If adherents use religious rhetoric instrumentally and it resonates, how is that not the use of religion? Who are we to say what its proper interpretation is? Stated religious motivations are unlikely to be cheap talk. Actors speaking in a religious idiom do so for a reason – and by taking them at their word, we accord them the agency that paternalistic insistence on ‘proper’ interpretations denies.
Second, rather than attempting a universal definition of religion, we should strive to examine those aspects of religion that may be causally relevant in a given context. One option is to disaggregate religious groups into doctrine, organizational forms, adherent profiles, and so on.Footnote 1 We can then develop more specific (and falsifiable) propositions about how doctrine might affect the choice of institutions, or how hierarchical (episcopal) religious organizations mobilize as opposed to more locally autonomous (presbyterian) groups, for example. We would then be better situated to appreciate the potential power of these aspects of religion and then see whether these stipulated relationships hold.
To focus on one of these aspects, doctrine consists of the set of beliefs and indicated practices espoused by a given religious denomination. If doctrine is the body of principles and teachings that describe both the tenets of the religion and the practices it advocates, then it is the basis for religious rhetoric: how actors use doctrine to motivate, justify, and propagate their goals. Doctrine thus structures both thought and practice: it provides the basis and the legitimation for religious rhetoric.
Why and how does doctrine matter? Here, a key assumption is that ‘to grant causal status to doctrines forces recognition that the most fundamental aspect of any religion is its conception of the supernatural’ (Stark and Finke Reference Stark and Finke2000: 34). Critically, doctrine is unfalsifiable: no worldly evidence can overcome its claims. Since it is a belief system that cannot be disconfirmed, the claims of religion on politics can be absolute and irrefutable (but also open to symbolic, as opposed to material, concessions: see Atran et al. 2007). Different interpretations might be possible, but denial is not. As a result, religious rhetoric robs opposition of its arguments, making it a very powerful tool for elites, political and otherwise.
Further, unlike other ideologies, doctrine provides divine sanction (and otherworldly rewards) for human action, making it a particularly compelling set of justifications and motivations for human behaviour. As a result, doctrinal arguments and religious rhetoric lower barriers to certain behaviours (including violent actions) – all the more so if the doctrinal interpretive tradition prizes the ‘pristine purity’ of its original articulation.Footnote 2 If robbing someone of their humanity is a step closer to making killing them possible, imagine what it means to have someone declared a blasphemer, heretic, or an unbeliever. Such individuals are not just different or less human; they also abnegate and deny the most sacred of values, and therefore become easy targets. In the Thirty Years War (1618–48), even if there were coalitions between Protestant and Catholic princes (as Cavanaugh (Reference Cavanaugh2009) notes), Armstrong concedes that ‘in the minds of the participants, however, these wars were certainly experienced as life-and-death struggle between Protestants and Catholics. Religious sentiments helped soldiers and generals to distance themselves from the enemy, blot out all sense of a shared humanity, and infuse the cruel struggle with a moral fervor that made it not only palatable but noble’ (Armstrong Reference Armstrong2014: 249–50). Thus, when Armstrong (Reference Armstrong2014: 220) notes that ‘when Pope Urban summoned the First Crusade, he had usurped the kings’ prerogative in his bid for papal supremacy’, the question is not whether or not he was seeking power. Rather, why were his arguments so convincing? Why did subsequent crusades, led by Barbarossa, Philip II and Richard I, also rely on these religious arguments?
Finally, religious doctrine serves as a powerful template and a source of common knowledge (Chwe 1994): a focal point and a framing device, a common understanding of both what others know and how they understand certain policy problems or challenges. That doctrine changes over time, or is open to multiple interpretations, does not mean it is causally impotent. The political views of the religious can change, with the Progressivist views of American evangelicals on Prohibition, populism, and abolition shifting to their contemporary conservatism and emphases on abortion, reproductive politics, and family. Yet, even as interpretations changed, the power of religious justification did not, sanctifying the goals and edifying the adherents. Doctrine may very well be open to multiple and changing interpretations, but those interpretations can be powerful determinants of political acts and rules in a given context.
The templates and common understandings generated by doctrine can be powerful focal points. Doctrine can influence both the choice of institutions and the long-term outcomes that result. Robert Woodberry (Reference Woodberry2011) argues that Protestant churches and missionaries promoted mass literacy, printing, and education, in an emphasis on the ability to read and to access the word of God directly. These in turn spurred social mobility, opportunities for women, and long-term health improvements, among other favourable outcomes. In the US, ‘from the body of Puritan thought come three elements that proved especially important for subsequent American political practice: covenant theology, the emphasis on the total depravity of humankind, and the concept of the chosen people’ (Wald and Calhoun-Brown Reference Wald and Calhoun-Brown2014: 43). These doctrinal tenets led to the idea of government as covenant and conditional allegiance; checks and balances as a system of governance under the assumption that human beings could not be trusted with power; and the concept of a civil religion, a common code subscribed to by the country’s religion that imparts a sacred character to the nation (Wald and Calhoun-Brown Reference Wald and Calhoun-Brown2014: 49, 54), respectively. In a broader articulation of this thesis, Protestantism facilitated democracy: the doctrine encourages individual conscience, with the Bible as key authority rather than priests or religious authorities, and ‘tend[s] towards separation and independence from ancient church structures and traditions as well as political authorities’ (Woodberry and Shah Reference Woodberry and Shah2004: 48). More recently, elements of Catholic doctrinal teaching such as subsidiarity and the primacy of transnational identities have informed the organization and development of institutions, including the European Union (whose founders, among them Robert Schuman and Konrad Adenauer, were all committed Roman Catholics).
Dominant religious doctrines shaped would-be state institutions even before the state was founded. For example, Lutheran, Calvinist, and Catholic doctrine each views the source of poverty differently and sees its moral import in distinctive ways. Accordingly, each Church imposed different regimes of poverty assistance in early modern Europe: integration, punitive work, and exclusion, respectively (Kahl Reference Kahl2005, Reference Kahl2014). Here, the attention to doctrine resolves several outstanding puzzles in the study of the welfare state: for example, why the US and UK hold individuals responsible for their own poverty, while Scandinavian countries and Germany view it as a social responsibility. Conversely, in several countries where traditional class- and party-based accounts of the welfare state would lead us to predict sparse provisions we see instead a strong, progressive role for the state in family policy. The underlying reason was Church–State conflict in the nineteenth century, won by the secular liberal forces of the nation state (Morgan 2009).
Nor do doctrinal differences preclude inter-denominational coalitions on given issues. For examples we can turn to the Evangelical–Catholic pro-life coalition of the 1990s in the US, or the conflict between the autocrat/Islamicist disavowal of human rights as ‘Western imperialism’ and the Catholic Church, which opposed this assault on what it saw as universal human values at the Vienna World Conference on Human Rights in June 1993 (Weigel 1999: 26). Yet doctrine also acts as a constraint on the ambitions of such coalitions: the agreement between evangelicals and conservative Catholics on the need to eliminate abortion in the late twentieth century did not lead to similar consensus on reproductive technologies, as we will see in the next section. Doctrinal commitments meant that these coalitions could only pursue some goals together and had to avoid others.
Instead, doctrine can interact with institution-building to produce historically contingent but very robust social and political outcomes. Primarily Roman Catholic countries, from Poland to Italy to France, show very different patterns of Catholic Church behaviour and influence (Grzymala-Busse Reference Grzymala-Busse2015). The doctrinal goals of the Church are formally identical, but the political demands differ, as do the policy outcomes. In another example, the markedly higher perception and approval of inequality between men and women among Muslims is not a function of scripture alone. But fiqh, or Islamic jurisprudence, is the product of the early centuries of Islam, when gender inequality was taken for granted. Subsequent Islamic jurisprudence continued to value and privilege these early classical texts. A legal tradition and juridical institutions ‘rooted in religion itself’ (Fish Reference Fish2011: 211) thus produced conservative attitudes to gender roles and the inferior legal status of women. As Fish notes, Christianity does not have this kind of jurisprudence but Hassidic/Ultraorthodox Jews do, with similar results for sexual segregation, sexual purity laws, and disavowal of legal or political equality between men and women. Conversely, even the most categorical of commandments meet political reality: for example, ‘one of the central commitments of Islam is to a divine monopoly of the making of the law’ (Cook Reference Cook2014: 270). But even here, it is easier to declare a monopoly than to enforce it. Islamic jurists found ways around it, ‘not, of course, by arrogating to themselves God’s exclusive right to legislate but by suitably reinterpreting the closed canon of his legislation’ (Cook Reference Cook2014: 279; see also Kuran Reference Kuran2011).Footnote 3
This brings us to a final point about reconsidering the impact of religion: rather than asking whether or not religion matters, we should look at how specific aspects of religious doctrine or organization interact with other factors. Rather than blurring the lines between religion and politics, we should examine instead how specific doctrines, organizational structures or adherent profiles interact with political institutions, regime types, or politician ambitions. Interaction is not the same as causal irrelevance: if aspects of religion interact with another factor, their impact is conditional, but may still be critical. Doctrine interacts with secular structures and pressures to produce distinct configurations of influence, institutions, and policy. The rise of fundamentalism, for example, is both a backlash against secular political developments and innovations and a reformulation of religious orthodoxy. The over-representation of Islam in civil wars is not simply the result of intrinsic qualities of Islam, but of the way in which elites use it to mobilize and to outbid each other, and the susceptibility of societies to such outbidding (Toft Reference Toft2007). The Taliban, Hamas, and Hezbollah, meanwhile, are such effective forces of violence due not only to their religious rhetoric, but to their resilient and cohesive structure, focused as it is on enhancing team effort, preventing free-riding and defection, and obtaining credible signals of commitment from its members (Berman Reference Berman2009). Thus doctrine is critical; but it does not act alone.
As a result, doctrine may not be sufficient for some political action, but it may very well be necessary. If it matters so much to its adherents that they are willing to exterminate each other over small differences in doctrine (see the Sunni vs. Shia conflict, or the fate of Uniates in Eastern Europe, caught between Catholicism and Orthodoxy),Footnote 4 then we ought to take into account the ways in which doctrinal claims can be unusually compelling to their intended audience – and unusually attractive to their would-be promoters. Failing to do so blinds us to the differences in institutional configurations that result, to the very high costs paid by those who travel only a short distance beyond their original identity, and to the peculiar coalitions (and lack thereof) that result when religion enters politics, and vice versa. Specifying such interactions also leads to more careful analyses of when and how doctrine matters: and just as importantly, when it does not.
EXPLAINING DIFFERENCES IN POLICY
A focus on doctrine as an aspect of religion, and on the ways in which it interacts with institutions and electoral considerations, helps to explain several empirical puzzles. For example, why is it that embryonic stem cell research is restricted in the US, but in vitro fertilization (IVF) is not, even though both involve the manipulation and destruction of embryos? Here, doctrine and religious coalitions interacted to produce distinct outcomes in the 2000s. A conservative religious coalition between Evangelicals and Roman Catholics worked hand in hand with Republican Party representatives to restrict embryonic stem cell research as another way of ‘protecting the unborn’. Yet precisely because doctrines clashed, IVF remained unregulated at the federal level, despite the theological opposition of a key member of the coalition, the Roman Catholic Church.
In contrast to the European cases, even as embryonic stem cell research and IVF have in common the creation and the subsequent destruction of embryos, they remain politically distinct issues in the US. After its introduction in the US, IVF became an accepted technology – while stem cell research became linked to cloning and to human experimentation and remained far more controversial. Congress has refused to fund embryo research for fear of encouraging abortions, while leaving IVF and stem cell research in the unregulated private sector and under the authority of states.
Both doctrine and electoral calculations dictate a more muted stance on IVF than on abortion or even stem cell research. First, Catholic stances on IVF do not reflect a broader consensus among conservative Christians. The Catholic Church has unequivocally condemned in vitro fertilization – in addition to leading to the destruction of ‘spare’ embryos, the procedure is viewed as immoral in that it separates the act of intercourse from procreation. However, this last point is unique to Catholic doctrine and has not resonated more broadly among other denominations. The Southern Baptist Convention has stopped short of condemning IVF, arguing instead that parents are responsible for unused embryos left behind (Allen Reference Allen2013). Most evangelical churches have either not espoused a stance on IVF, or hold conflicting views on the procedures. Since there is no one religion that represents national moral interests in the US, a broader inter-denominational agreement is critical, but there was no theological basis for one here. As one critic pointed out, embryonic stem cell research and IVF both destroy embryos, but the opponents of stem cell research do not pursue the banning of IVF: ‘they’re not doing that because it’s politically suicidal and self-contradictory: fertility treatment, after all, is just a latter-day means to be fruitful and multiply; what could be more “pro-life”?’ (Rosenberg Reference Rosenberg2001).
As a result, political mobilization by religious organizations against IVF in the US is nearly non-existent. According to their proponents, even ‘personhood’ amendments that would eliminate abortion and end stem cell research nonetheless allowed IVF.Footnote 5 Without consistent pressure from religious groups or their political allies, the federal response is to leave the matter largely unregulated; unfunded by the federal government, most of the research and procedures have been developed in the private sector.
Contemporary interpretations of both Catholic and Protestant doctrine agreed on stem cell research, however. Here, the Lutheran Church–Missouri Synod and the Southern Baptist Convention joined the Roman Catholic Church in opposing such research (Wertz Reference Wertz2002: 676). These religious organizations cast embryonic stem cell research as equivalent to abortion, and thus, to murder. Critics ranging from the Christian Coalition of America to the Traditional Values Coalition to the National Conference of Bishops spoke of the need to preserve the rights of the unborn and human dignity by banning embryonic stem cell research. Southern Baptists argued that ‘The Bible teaches that human beings are made in the image and likeness of God (Gen. 1:27; 9:6) and protectable human life begins at fertilization’ (Southern Baptist Convention 1999).
Reframed as protecting the unborn, opposition to embryonic stem cell research resonated with religious views and found numerous allies within the Republican Party (Banchoff Reference Banchoff2012: 3 and 231). Federal regulations had already prevented the use of federal funds for experimentation on human embryos since 1974. The moratorium was extended indefinitely in 1989 by the Department of Health and Human Services Secretary Louis Sullivan, who accepted arguments from theologically conservative members of the Human Fetal Tissue Transplantation Research Panel that such research would increase abortions by legitimating the use of foetal tissue from abortion.Footnote 6 When the Clinton administration authorized the use of federal funds for research on left-over embryos in 1995 (White House 1994), Congress intervened to pass the Dickey Amendment, prohibiting the Department of Health and Human Services from funding any research involving specifically created research embryos or the destruction of any embryos.
The situation changed with the isolation of human embryonic stem cells in 1998. The Clinton administration decided that these cells could be harvested without directly causing the destruction of embryos – and thus the Dickey Amendment no longer applied. It was left to President George W. Bush in 2001 to authorize the use of federal funds for embryonic stem cell research. He was put under heavy pressure by the Catholic Church and pro-life organizations, which equated stem cell research with abortion and demanded its end. Bush received numerous letters from the US Conference of Bishops, and met privately with Pope John Paul II, who further urged him to end the funding. Bush’s public address in 2001, explaining his decision to allow federal funding for research with existing embryonic lines, explicitly referred to the ‘culture of life’, a concept espoused by the Pope.Footnote 7 By limiting research to existing lines, Bush avoided the ‘life and death decision’ altogether.
The rhetorical connection between embryonic stem cell research and abortion was a double-edged sword: the religious pragmatists approved Bush’s decision, but the purists saw the limits as inadequate. Pat Robertson praised the Bush decision as ‘firmly protecting the rights of the unborn’, while other groups, including the Conference of Bishops, decried the decision as continuing the ‘killing’ (Goodstein Reference Goodstein2001). Evangelicals and Catholics could be found on both sides, yet even the purists were not willing to oppose the president further, hoping for future legislative action. They were not disappointed: in 2006 and 2007 Bush further vetoed legislation to expand stem cell research to donated embryos left over from IVF treatment.
In short, both embryonic stem cell research and in vitro fertilization result in the destruction of embryos that abortion opponents find so objectionable. Yet the response to these two issues differed greatly from each other – and from abortion, where a religious-partisan coalition saw enormous electoral gains to be made. Further, religious organizations took very different tacks addressing the two issues. IVF was largely left alone, given the diverging religious views and political costs (and perhaps its high clinical profits). Stem cell research, in contrast, was likened by its opponents to abortion; once this rhetorical move was made, groups opposing abortion would pounce to eliminate it.
EXPLAINING VIOLENCE?
If religion is such an important aspect of historical human experience, if it is inextricably bound up in the very fabric of humanity, then it is likely to be a powerful force in other domains, violent or not. After all, if our love for our children, friendship and loyalty, desire for peace, order, and stability, can all lead to violence, why should our attempt to commune with the divine be exempt? Yet many scholars, including Armstrong or John Esposito, portray Islam and other denominations as intrinsically peaceful religions (Armstrong 2014; Esposito 2010; Esposito and Mogahed 2009). As Steven Fish (Reference Fish2011: 6) notes, ‘to such scholars, misunderstanding on the part of non-Muslims is to blame for any negative impressions … ever resistant to prejudice and intent upon overcoming ethnocentrism, these writers assume that Muslims cannot be, say, more prone to violence, inequality, or authoritarianism than non-Muslims are.’
Armstrong’s arguments here counter a long line of thought, from Montesquieu’s Spirit of the Laws to Samuel Huntington’s ‘bloody borders’ of Islam, which views religion as intrinsically violent. More recent empirical analyses have found a peculiar embrace between violence and Islam. Steven Fish (Reference Fish2011: 151–2) finds that Muslims and Christians are equally likely to engage in large-scale political violence, but Muslims are far more likely to engage in terrorism. In his words, ‘terrorism is not a uniquely Muslim disease, but its perpetrators in recent times are disproportionately Islamists’, with over three-fifths of terrorist incidents, and nearly 70 per cent of all deaths perpetrated by Muslims (Fish Reference Fish2011: 153). Similarly, Monica Duffy Toft (Reference Toft2007: 115–16) notes that 81 per cent of religious civil wars involve self-avowedly Islamic motives and actors, and 26 per cent of all civil wars. These civil wars, moreover, tend to be more intractable, lasting longer and more likely to recur.Footnote 8
What is the role of religious doctrine here? Religions are not intrinsically peaceful (or violent) but their doctrines can be taken up with enormous zeal to justify violence and to recruit further adherents. Here, taking actors’ rhetoric seriously, examining doctrine as an aspect of religion, and analysing how they interact with political opportunities allows us to pinpoint more carefully how religion may be implicated in violence. One example is Da’esh (the militant extremist group, also known as the Islamic State of Iraq and Syria, or ISIS). As Graeme Wood (Reference Wood2015) pointed out in an article in the Atlantic Monthly magazine on the doctrinal bases of Da’esh, bad governance, the reliance on oil, and shifting social mores all matter in the rise of Da’esh, ‘but focusing on them to the exclusion of ideology reflects another kind of Western bias: that if religious ideology doesn’t matter much in Washington or Berlin, surely it must be equally irrelevant in Raqqa or Mosul. When a masked executioner says Allahu Akbar while beheading an apostate, sometimes he’s doing so for religious reasons.’ Wood goes on to quote a prominent scholar of Islam, Bernard Haykel, that while many people want to absolve Islam and call it a religion of peace, Islam is ‘what Muslims do, and how they interpret their texts’.
The article generated enormous controversy: several of Wood’s critics, much like Karen Armstrong, repeatedly pointed out that nothing in Islam predetermined the rise or spread of Da’esh (Jenkins Reference Jenkins2015a, Reference Jenkins2015b; Moghul Reference Moghul2015). We may grant that immediately but, as we noted earlier, that kind of critique is incoherent. It confuses two distinct statements about necessary and sufficient conditions: (1) Da’esh in its particular form and with its specific tactics could not arise without the doctrine of Islam; and (2) nothing in Islam necessarily predetermines Da’esh, or inexorably leads to vicious conflict. Wood and Haykel argue the former – and their critics argue the latter, claiming either that Islam does not necessitate Da’esh-style tactics, or that multiple textual interpretations of the Qur’an exist.Footnote 9 But the question is not whether Islam or any other religion inevitably produces violent extremists; it is whether they could exist, take the form and use the tactics they do, and succeed without the religion and the powerful doctrinal ideology it offers. In the case of Da’esh, crucifixions, beheadings, and slavery are all textually faithful tactics used by Da’esh to establish its desired caliphate. The caliphate itself, a stated goal of Da’esh, is a fusion of religious and political authority that transcends the nation state and necessitates territorial control. It is a specific product of the interaction of Islamic doctrine and political history. As a result, it is not surprising that the restoration of the caliphate as a goal resonates much more deeply than a continuation of Muslim political leadership itself, and leaders from Abu Musab al-Zarqawi (the founder of Da’esh) to Ayman al-Zawahiri (the leader of al-Qaeda since 2011) have sought its restoration (Cook Reference Cook2014: 325).
Nor is the argument that religions contain multiple interpretations the decisive one here. We may grant that Da’esh cherry-picks the Qur’an to justify and to motivate its actions – and that its interpretations are not representative of either the potential universe of interpretations or the actual practice and beliefs of its fellow Muslims. For the same reasons, the totalitarian ideology of Da’esh is not the only one compatible with Islam: the gamut runs from pietists, such as many Salafis, who reject political engagement, to Muslim fascists (in 1930s Egypt), Muslim democrats in Turkey and Tunisia, and even Muslim leftists (Cook Reference Cook2014: 335–6). But that does not undermine the assertion that Da’esh or other organizations can be (and are) motivated by their religion, and their particular interpretation of doctrine, without which they would neither hold the norms nor exercise the tactics that they do.
Political violence in the name of religion is not uniquely Islamic; those who wish for a ‘Protestant Reformation’ for Islam should recall the bloody violence that accompanied the Reformation in Europe (Cook Reference Cook2014: 458). Religious struggles not only divide societies but also undermine the capacity of the state to control violence, often with bloody results. For example, ‘by causing religious wars, massacres, and persecutions, the “Protestant Reformation” was a major cause of the witch-hunts’ (Stark Reference Stark2003: 250, emphasis in original). Other episodes, such as the Crusades, which Armstrong claims were simply the violent aspects of a secular struggle of power, were nonetheless justified in a religious language – a testament to the power of doctrine as both exhortation and incentive. And here it is precisely because doctrine can justify action as divine will, promise rewards beyond earthly compare, and demarcate the boundaries of the vilified Other that it is such a potent force in fomenting violence.
In short, religious doctrine is rarely a sufficient condition for violence, policy pressure or a set of normative commitments. But it may very well be necessary: without their particular interpretation of religious doctrine, neither the pro-life activists in the US nor the Islamic militants in Syria could articulate their views in the way they do, and with the intensity that they do. And here, rather than asserting that their interpretation of doctrine is ‘incorrect’ or somehow violates the given religion’s intrinsic qualities, we would do well to listen to these actors themselves and take the power of their religious rhetoric as seriously as they (and their intended audiences) do.
CONCLUSION
In the end, Armstrong’s claims come down to ‘religion does a lot of different things. To claim that it has a single, unchanging, and inherently violent essence is not accurate. Identical religious beliefs and practices have inspired diametrically opposed courses of action’ (Armstrong Reference Armstrong2014: 393). But to acknowledge the multifarious and changing nature of religious doctrine, and its many different interpretations, is simply the first step. It is precisely because religion does inspire different courses of action that we need to take seriously the stated motivations and justifications of the actors involved, the doctrinal bases that they are using, and the compelling nature of religious arguments. It is not that religion inevitably produces violence; it is that some violence (and other political outcomes) cannot exist without the doctrinal templates and rhetorical justifications of religion.
