Positivism and Reasonableness: Authoritarian Leanings in New Atheism’s Thinking

Abstract

Various contemporary phenomena of social regression and authoritarianism are related to religious actors, movements, and beliefs. This text, however, seeks to follow this up with the political–theoretical argumentation that New Atheism has to be understood as a way of thinking which carries illiberal and authoritarian tendencies with it as well. In defence of this position, this article will first reconstruct, with reference to Habermas’s and Rawls’s theory of democracy, elements that must include personal beliefs in order to be considered congruent with democratic values. Subsequently, New Atheism’s conception of rational politics will be presented in order to show in which aspects it contradicts the demands of reasonable convictions. This concerns, in particular, the rejection of reasonable pluralism on the one hand and a non-positivistic view of human beings on the other. As a conclusion, this text supports the proposition that, when speaking of the connection between certain worldviews and today’s illiberalism, New Atheism must also be considered as an unreasonable comprehensive doctrine.

1. Introduction

It is undisputed that many forms of social regression currently observed in liberal democracies—a depletion of the rule of law or the multilateral international order, hostility towards ethnical or sexual minorities, nationalistic hubris, et cetera—are in various ways linked to religious identities, beliefs, and actors.1 The Polish PiS, for instance, supports the narrative that its measures should enforce the cultural values of an equally national and Catholic identity against the influence of a hegemonic cosmopolitan and liberal elite, as, for example, embodied in the European Union or the European Court of Human Rights (see for example Bobowicz and Nowak or Sadlon on this special issue, cf. Köllner 2018). Thus, liberalism on the one hand and religiosity, patriotism, and nationalism on the other serve as expressions of opposing forms of value systems, a supposed constellation of conflict that is referred to by different authoritarian actors and movements, such as in Hungary, Russia, or Turkey.

In this constellation, as an often socio-scientifically underrated player, New Atheism understands itself as both an intellectual enterprise and a social movement that wants to react to the threats of regression posed by the coalitions of religion and reactionary forces. Therefore, New Atheism does not want to be a purely theoretical undertaking, but by embedding itself into the political public sphere tries to exert influence on public opinion. By doing so, it understands itself as an enterprise against the diverse forms of (supposed) “social dogmatism” (Hans Albert) that are practiced by religious actors and believers.2 Following this line of argument, it might seem possible to characterise the different social forms of New Atheism, for example the Brights movement, or public persons such as Richard Dawkins or Daniel Dennett, as part of progressive movements comparable to the Extinction Rebellion or the March for Science. Yet, in my political–theoretical oriented paper, I would like to develop a differentiated assessment of New Atheism’s position in relation to authoritarianism and illiberalism.

An empirical hypothesis at this point, but one that will not be pursued further, is that just like the different types of religious support for authoritarianism and illiberalism (cf. Hennig and Hidalgo on this special issue), New Atheism could also be interpreted as another worldview that opposes the dogmatic ambivalence of modern liberal and pluralistic societies. To be more precise, it can be seen as an illiberal reaction towards religious pluralism in post-secular societies (cf. Gladkich and Pickel 2013, p. 142 or Bechert on this issue, p. 3).3 Insofar as religion does not seem to simply disappear in modern societies, as the classical Weberian secularisation hypothesis assumes, modern post-secular societies provide the basis for atheistic illiberal reactions to pluralism—just as they also provide the basis for religious illiberal reactions to pluralism.

To make the scope of the following clear, the aim is not to analyse empirically whether New Atheism has anywhere—directly or indirectly—led to policies that can be classified as illiberal or authoritarian. Rather, it is to argue on a purely intellectual level that the worldview and political ideas of New Atheism are in contradiction to publicly reasonable ones. If they were in fact to influence policies, there could accordingly be a danger that they would lead to authoritarian decisions. In this sense, it should also be stressed that I do not use the concept of New Atheism as a sociological and historical one insofar as it is meant to denote a concrete social movement in a concrete country or in a concrete time. This could lead one to believe that I am going to address an authoritarian tendency within a clearly identifiable group of people. However, this is not my intention. As an alternative to the concept of New Atheism, one could therefore also speak of a rhetorically militant, political atheism as it is represented by persons who explicitly call themselves New Atheists, but not only by them.

In the first part of my paper, I reconstruct which elements a doctrine must have to count as reasonable and therefore congruent to the values of a democratic state. For this purpose, I will refer to the theories of democracy developed by John Rawls and Jürgen Habermas, which seem particularly appropriate for this for two reasons: on the one hand, both authors deal with the normative requirements of well-functioning democratic systems, and on the other, both also analyse what can be derived from this for a citizen’s communicative actions. Afterwards, I present New Atheism’s self-understanding by analysing more academic and essayistic texts by Hans Albert, Philip Kitcher, and Daniel Dennett, important reference authors of New Atheism.4 Following this, I show in which points the self-understanding of New Atheism as a worldview stands in contrast to the necessary content of reasonable comprehensive doctrines. There are, from my perspective, two elements that a doctrine must contain to be compatible with the values of a democratic society: (a) openness to ambivalence, meaning a differentiation between what is politically right for pluralistic societies and individual notions of truth, and (b) acceptance of an idealistic view of human beings, which identifies them as free agents. Therefore, I conclude that when speaking about religion’s relation to current forms of illiberalism, it is necessary to consider (at least) some forms of New Atheism as an important phenomenon in connection to today’s illiberalism as well.

2. Beyond Agonistic Politics

In the following, I first define reasonableness and explain why it is important for democratic societies, following Rawls’s and Habermas’s political theory (Section 2.1). This step is necessary because I argue that because of New Atheism’s lack of reasonableness, it has a cognitively authoritarian leaning. Then, with reference to Rawls’s conception of reasonable comprehensive doctrines, I will show how reasonable dispositions can be socially realised against the background of a pluralism of worldviews (Section 2.2) and what elements a reasonable comprehensive doctrine has to contain (Section 2.3). This chapter therefore serves to name and to justify the central categories for the following analysis of New Atheism.

2.1. Why Reasonableness Is Important for Democracy

According to Rawls (1993, pp. 48–54), human beings possess the competence to think and act both rationally and reasonably. Rationality, on the one hand, refers to the capacity to choose and effectively pursue one’s personal life plans in the face of a wide range of possible life goals and life forms. Rational choice and action thus aim at the achieving of a good life. Reasonableness, on the other hand, refers to the ability to follow principles and norms that others can also acknowledge from their point of view, despite opposing preferences. If somebody does not act egocentrically, but in respect to the justified demands of others, then they act reasonably. Reasonableness thus aims at justice, in the sense of a general justifiability of actions or structures.

The question might now arise as to why reasonableness is important for democracy. Why would a mere agonal competition for power and influence not be a normatively appropriate understanding of the democratic process, as it is affirmed in today’s academic discourse by neoliberal, Hayekian approaches, as well as in agonal theories of democracy?5 According to Easton’s (1968, pp. 129–34) classical definition, “political” refers to the authoritatively regulated distribution of material and immaterial goods. Democracies, according to Rawls (1993, p. 4), are furthermore based on the Aristotelian idea that all human beings are in a moral sense free and equal. For a democratic polity or policy, this might lead to a tension between the normatively desired recognition of freedom and the equality of each and every person on the one hand and the intention to authoritatively regulate social life on the other, since political rule and state coercion prima facie imply the restriction of freedom and the asymmetry between the ruler and the ruled. The question therefore is how it is possible to preserve the freedom and equality of every citizen and at the same time order their common coexistence by the help of political rule.

Democratic systems formally attempt to resolve this tension by placing the development of norms, which then guide the exercise of state power, in the hands of those subject to the law, for example by giving them voting rights. In democracies, citizens exercise political rule over themselves so that their freedom and equality are not violated. This is at least a relevant aspect of democratic order that all theories of democracy recognise, including minimalist conceptions (cf. Przeworksi 1999; Kolodny 2014).6

Moreover, with regard to content, it must also be possible that the justification of the norm finds general acceptance—even if only because the norm came about in an open and fair process. Christian List (2011, 286f.), for example, rightly objects to democratic theoretical approaches that merely identify the reaching of an undeliberated compromise supported by the mere majority as normatively sufficient for democratically legitimate decisions. By doing so, these approaches do not make any qualifying demands with regard to the decision’s justification, which might lead to a point where contradictory reasons put forwards by different actors justify the outcome, for example by the different parties that form a governmental coalition. However, this is initially unsatisfactory from an intellectual point of view. As a rule, citizens demand good arguments for policies that restrict their personal freedom. Furthermore, in a non-ideal political system, there are various possibilities for socially powerful actors to politically assert their particular interests at the expense of weak interests. Here, the deliberation of good reasons to justify a policy can provide a remedy by revealing positions that cannot be justified in general in the deliberative process. The public use of reason should consequently guarantee that the norms and policies which henceforth exercise their rule over citizens can also be accepted by them, and thus, in this respect, no illegitimate coercion or even violence is exercised over persons who are actually recognised as free and equal (Habermas 2018, p. 873).

Yet, if one agrees with the principle that democratically legitimate norms must enjoy general acceptance, nothing is said about who is to be committed to this task in a deliberative system.7 Intuitively, officials in the political–administrative system could be identified as the sole addressees of the claim to find generally acceptable justifications for legal norms, for example members of parliaments or constitutional judges. Rawls’s theory of democracy, however, takes the position that citizens have largely the same duties as office holders. The Theory of Justice poses the question of whether ordinary citizens, when they engage in political activities such as casting their vote in an election, should bow to the duty of justice or can simply rationally pursue their interests that are inclined towards their own good (Rawls 1972, 360f.). Indeed, the latter would be short-sighted, given the institutionalised coupling of public opinion and the decision-making processes. Two cases exist for that: first, as a citizen, one frequently selects those persons or parties who directly influence the political decision making for a legislative period. Second, representatives use these reasons to deliberate justified decisions which they draw from a “pool of reasons” (Habermas 1989, p. 28, trans. M.R.) administered by the public. “Democracy […] implies […] an equal share in the coercive political power that citizens exercise over one another by voting and in other ways” (Rawls 1993, 217f.). Consequently, for reasons of institutional design, in democratic systems, there can be no strict differentiation between the duties that affect persons in the political–administrative system and those that affect ordinary citizens.8

Now, sound institutional restrictions or incentives inside the political–administrative system, for example the effective separation of the powers or precautions that prevent the undue influence of socially powerful groups, can enable political actors to make reasonable political decisions based on strategic motives (Habermas 1996, p. 304). Nevertheless, Habermas’s and Rawls’s theory of democracy can be classified in the series of (empirically well-founded) approaches which assume that a certain degree of virtue and public spirit is also necessary for a long-term stable democratic society.9 In order, for example, to transfer learning processes, which have been made at the grassroots level in favour of the emancipation of largely unheard minorities, into political decisions, there must ultimately be a willingness on the part of the wider public and their political representatives to go beyond a purely strategic orientation which merely focuses on their own benefits (cf. Maus 1991, p. 145).10 This, however, cannot be achieved by sound institutional design alone but requires virtuous citizens and politicians to act out of duty.

One important democratic duty is the duty of civility, which relates to the giving and taking of reasons in the deliberative process. Citizens who adhere to the duty of civility do so because they recognise that their political actions, such as their communicative actions in the political public sphere, have an important impact not only on their own individual lives but on the political organisation of the legal community in general. They will therefore—as civility denotes—not recklessly impose their views on others but consider what can gain general acceptance. Citizens who adhere to the duty of civility will, for example, not merely give their egoistic self-interests as reasons for certain policies.

Following a slight revision made in 1997, Rawls (1997, pp. 783–87), cum grano salis, qualified the wide conception as a normatively appropriate formulation of the duty of civility. This allows all those reasons to be given in the decision-making process—cases such as hate speech or incitement of racial hatred not included—of which a person is truly convinced, without immediately filtering them with regard to their general acceptability. Thus, in a first stage, it is quite possible to give reasons for the latter based on generally binding norms that stem from only subjectively recognised comprehensive doctrines. Only in a second stage—“in due course” (ibid., 776)—are citizens called upon to give equivalent reasons that are generally accessible (ibid., 778). In the first stage, for example, it is not a violation of the duty of civility if a devout citizen who rejects a liberal regulation of stem cell research based on religiously founded doubts—“we must not play God”—also makes this known publicly with reference to religious arguments. Only over time should he, according to Rawls’s wide conception, give generally accessible reasons for his position, reasons that could also be shared by fellow citizens of other and no confessed faith.

However, without going into too much detail about the theoretical discourses concerning the concrete content of publicly justified reasons, it can rightly be argued that replacing, to stay with this example, religious reasons with publicly shareable ones is a supererogatory demand: it is questionable as to why religious reasons cannot be generally accepted if they, for example, refer to “buried intuitions” (Habermas 2008a, p. 131) that non-religious citizens might also recognise as practically justified. Lafont (2009) furthermore objected to Rawls’s conception of the duty of civility, stating that it demands too much insofar as, in order to avoid illegitimate coercion, it is sufficient that a claim to political validity cannot be rejected with good reasons, regardless of whether its justification can be generally shared or not.

2.2. Reasonableness and Reasonable Comprehensive Doctrines

A prominent objection to the wide conception of the duty of civility was given by Wolterstorff (1997, pp. 105–9), among others, to the effect that it remains unclear as to why a devout individual should be expected to replace their reasons with publicly accepted ones. On the contrary, one could assume that, for example, authentic religiosity means, on the one hand, that citizens derive their political convictions from their religious views and only from those and, on the other hand, that they cannot truthfully name any other reasons than these religious ones without acting dishonestly. Wolterstorff’s objection must not only be viewed with regard to religious comprehensive doctrines as it also concerns many comprehensive worldviews a person can be convinced of. The predicate of being convinced of something by definition excludes the possibility of simply discarding that of which one is convinced. “If I really thought that there was no more reason for me to believe what I believe than there is for what others believe, then I would not have reasons to believe what I believe” (Christiano 2008, p. 212).11

For Rawls’s political liberalism, this leads to the critical question of whether it is at all possible for many citizens to act in accordance with the wide view of the duty of civility and, if they do, to still sincerely state their truthful reasons or whether they have to disguise themselves for the sake of the demands of public reason, possibly at the cost of enormous cognitive dissonance. Thus, some branches of the critique of Rawls’s Political liberalism, such as that of Wolterstorff (1997, pp. 96–102), lead to the position that Rawls’s theory, contrary to its self-understanding, ultimately rests on metaphysical assumptions that lead to arbitrary argumentations—for example, by qualifying publicly accessible reasons as better than those stemming from subjective comprehensive doctrines. Geuss (2009, p. 41) vividly formulates this reproach against liberal political philosophies in general as follows:

“In Europe up to the eighteenth century the holding of heretical beliefs was one of the public crimes par excellence […], the suppression of heresy was therefore in everyone’s interest. Liberals, however, think that theory (of collective responsibility before God) is wrong and disallow that you are ‘affecting’ your neighbors and fellow citizens in any relevant sense simply by holding a certain belief, even a heretical one. So the liberal can make an effective distinction between public and private in cases like this […] by evaluating the truth or falsity of the theory the agents in question hold about what harms or might harm their interests. The question is then: who does the evaluating? Liberals, of course, think they ought to have the final word […]. In other words, it is the fact that liberals think that the beliefs of religiously minded persons (e.g., that God will hold all responsible for the heresies of any one member of the society) are false that is supposed to count as a reason for thinking that religious people have no grounds for the claim”.

Considering Rawls’s political philosophy, however, such a critique would be too simplistic. He himself recognises the problem raised and summarises the question that moves Wolterstorff’s and Geuss’s line of criticism as follows:

“How is it possible for citizens of faith to be wholehearted members of a democratic society who endorses society’s intrinsic political ideals and values and do not simply acquiesce in the balance of political and social forces? […] How is it possible—or is it—for those of faith, as well as the nonreligious (secular), to endorse a constitutional regime even when their comprehensive doctrines may not prosper under it, and indeed may decline”

(Rawls 1997, 781f.)?

Therefore, I would argue that the criticism of Rawls, as expressed by Wolterstorff, for example, is not fitting. My objection to this criticism is as follows—and this is at the same time an answer to Rawls’s question as to why it is convincing “for those of faith, as well as the nonreligious (secular), to endorse a constitutional regime” (ibid.) by perhaps having to withhold their comprehensive doctrines from deliberation: in a Rawlsian well-ordered society, it is relatively easy for citizens to comply with the requirements of the wide conception of the duty of civility and therefore prima facie distance themselves from their comprehensive doctrines, because this is in turn also a part of their convictions and is justified by their comprehensive doctrine as well (Rawls 1997, 782f.; cf. Jaeggi 2018, p. 11). The wide conception of the duty of civility is not a requirement imposed on them coercively from the outside, but part of their moral convictions.

“In endorsing a constitutional democratic regime, a religious doctrine may say that such are the limits God sets to our liberty; a nonreligious doctrine will express itself otherwise. But in either case, these doctrines formulate in different ways how liberty of conscience and the principle of toleration can cohere with equal justice for all citizens in a reasonable democratic society”

(Rawls ibid.).12

Thus, the differentiation between the duties that follow from what is subjectively rational and the duties that derive from what is reasonable is also to be understood heuristically. For reasonable citizens living in a well-ordered society, Rawls assumes that there is congruence between the requirements of their comprehensive doctrines and the resulting form of life on the one hand and what is reasonable on the other. That which is demanded by, for example, the duty of civility, is at the same time also something that appears, however formulated and justified, as a duty, or at least a recommendation, for acting within the framework of a rational plan of life as well.

To provide an illustrative sketch, the duty of civility requires that religious citizens do not use state power to politically enforce their religious beliefs. As their faith cannot possibly be generally shared in a pluralistic society, such a practice would be unreasonable. This restraint—if the concept is at all apt here—does not, however, come in the sense of “Jefferson’s compromise” (Rorty 1999, p. 169), externally imposing a modus vivendi restriction on the believing citizens, because their community of faith is simply not powerful enough to impose their comprehensive doctrines on all others. Rather, as members of a well-ordered society, they are also truly convinced from an internal point of view that, in accordance with their belief, it would be wrong to use state power to impose their religious convictions on their fellow citizens. Thus, they can also consider such an action ethically wrong from the point of view of their religious doctrine because, for example, it might violate the God-given dignity of human beings to disesteem the individual freedom of faith, conscience, and creed.13

What does this mean for the relation between the duty of civility and the duties or instructions that are derived from reasonable comprehensive doctrines? For a reasonable citizen, it will not be possible in a particular situation to neatly discern their reasons for action. What the duty of civility demands overlaps, however this may look in detail, with the duties or creeds that involve leading a good life. For Rawls, a good conduct of life is always the right conduct of life, even though the good, unlike, for example, that which is prominently in Kierkegaard’s (1959, 182f.) philosophy of existence, does not merge into the right alone.14

2.3. Elements of Reasonable Comprehensive Doctrines

Thus, while reasonable comprehensive doctrines are important for individuals as they provide them with a holistic explanation of the outer and social world and point the way to a good, fulfilled life, they are important for democratic societies insofar as they are a condition of the possibility for stability in pluralistic democracies. So, what are the elements that reasonable comprehensive doctrines must include in order to fulfil this function? What do reasonable citizens have to be convinced of in their own way in order to act out of democratic duty? From the current state of research, which in part goes beyond Rawls, eight elements can be identified that a reasonable comprehensive doctrine must contain.15

• In the sense of the duty of civility presented above, reasonable comprehensive doctrines must encourage their bearers to give generally acceptable reasons in the political public sphere when it comes to deliberating legal norms, which are then enforced by the state. Rawls (1997, p. 773; 2003, p. 55) suggests that reasonable comprehensive doctrines encourage those who are convinced of them to give reasons that are generally shareable, which, as already indicated, is a supererogatory demand. Rather, for the purpose of avoiding illegitimate coercion, it is sufficient that reasons advanced in deliberation cannot reasonably be rejected (Scanlon 1998, p. 191; Lafont 2009).
• It has to be added that reasonable doctrines will only advocate generally acceptable norms in all politically relevant questions (Rawls 1997, p. 777). They will not, for example, justify a political claim by saying that it is generally acceptable if and only if it is for their own good. Yet, in the case of another political question, if it is perhaps not for their good, they will suspend the principle of public justification.
• Moreover, citizens with reasonable comprehensive doctrines do not simply happen to give generally acceptable claims of validity, but they likewise only apply such justification procedures that could gain intersubjective acceptance (Schmidt 2008, p. 100). Consequently, not only must the political output be generally acceptable in content if a worldview is to be evaluated in terms of its reasonableness, but also the procedure with which it chooses to reach a concrete justification. Reasonable religious citizens, for example, will not recognise the right of other believers to obey their conscience just because they simply have no political power to impose their faith on them. Rather, they will recognise that the justification of the right to freedom of conscience must also be such that it can be accepted by others.
• This, in turn, is based on the fact that reasonable citizens recognise their fellow citizens as free and equal beings, whose autonomy must be respected (Rawls 1997, p. 778)—“tolerance is a primary virtue”, as Judith Shklar (1964, p. 5) puts it.
• This is also based on the fact that in modern, democratic societies, there will simply be an irrevocable pluralism of different world views. In the words of Fraenkel ([1964] 2011), pluralism is a structural element of modern societies that can only be abolished by the use of coercion or violence. Rawls (1989, p. 235; 1993, pp. 54–58) explains this characteristic of modern societies by reminding us that the anthropologically constant burdens of judgement, the idola tribus, always make most of our judgements conditional on being subjective or even flawed as the cognitive faculty of humans is limited in principle. For example, as is well known in epistemology, no unambiguous inductive conclusions can be drawn from observations; rather, they are always dependent on factors such as theoretical premises. Now, moreover, modern societies are characterised by a high degree of functional differentiation, which goes hand in hand with the differentiation of the possible experiential worlds depending on sociological factors such as occupation, gender, place of residence, milieu, et cetera. This diversification of the possible experiential worlds favours the fact that the already operative burdens of judgement are thus given a much larger contact surface. At the micro-level, those who live in modern societies therefore form differently coloured worldviews depending on their social position. One could say that Rawls is in a way approaching the well-known Marxian hypothesis, according to which social being determines consciousness. The result of this is that modern sociality, mediated by the mechanism of functional differentiation, will on the macro-level develop an irrevocable pluralism of reasonable yet possibly divergent comprehensive doctrines, whereby different worldviews cannot necessarily be attributed to cognitive failures, but rather to different social positions.
• From the acceptance of the burdens of judgement follows the preparedness to learn from others (Benhabib 1992, p. 98; Apel 1993, p. 510).
• From this also comes, at the same time, the preparedness to accept only such truth claims in deliberation—if it is necessary to deal with them—that can be justified beyond different worldviews, which means truth claims made by common sense or science, as far as they are widely undisputed and comprehensible (Rawls 1993, p. 224).
• It follows from what has been said that, as far as the basic institutional order of a society is concerned, reasonable doctrines accept a system of certain fundamental human and citizen rights that cannot be rejected. These are the classical liberal rights to negative liberty and the republican rights of participation, but also social, ecological, and technological rights, which make the realisation of the former possible in the first place (Habermas 1996, 122f.; Rawls 1997, p. 774).

In society, three goals can be achieved by means of the common adoption of reasonable points of view: (i) respect is shown towards everyone as a free and equal person; (ii) as the society’s basic structure and its policies are not based on a mere power struggle between different interests but on good reasons, it is stable; and (iii) the principles and rules guiding social life can be intellectually justified, and thus, they have an epistemic quality.

3. New Atheism as an Unreasonable Comprehensive Doctrine

A look at history, but also at current events, immediately shows how religious convictions can manifest socially as both reasonable and unreasonable comprehensive doctrines. Political religions are placed on a continuum from non-violent resistance to oppression on the one hand to religious terrorism and war on the other. In this respect, it would be a one-sided approach to regard them as unreasonable in principle.16

With New Atheism, a new doctrine is entering the public stage of modern societies in the so-called West, represented by authors such as Daniel Dennett, Richard Dawkins, Michael Schmidt-Salomon or Philipp Möller; parties such as the Partei der Humanisten in Germany; organisations including the Giordano Bruno or Hans Albert Foundation; and social movements such as the Brights. New Atheism is characterised by the fact that it is not only a private worldview—“I don’t believe that God exists”—but also a decidedly social and political project that wants to exclude religion from the public sphere (Kitcher 2008, 13f.; cf. Gladkich and Pickel 2013; Fehige 2019; Bechert 2021, p. 3). Instead of religious or other reasons being interpreted as obscure, only those reasons that can be judged according to scientific standards should be able to claim political validity, whereby a scientistic understanding of scientific standards is prejudiced here. If this is done, so the promise goes, authoritarian or illiberal political decisions can be circumvented.

In the following, the argumentation of New Atheism will first be presented with reference to the arguments developed by Hans Albert and Philip Kitcher (Section 3.1). Both authors were chosen due to the fact that their reasoning is finely worked out, although their style is sometimes nevertheless essayistic. It will then be argued that New Atheism can be classified as a comprehensive doctrine in the sense of Rawls’s theory of democracy. However, it lacks essential elements to qualify as a reasonable comprehensive doctrine (Section 3.2). This result is significant insofar as it leads to the conclusion that New Atheism’s self-description as a project of enlightenment and defence of liberal values is erroneous; rather, it exhibits a cognitively authoritarian leaning in New Atheism. Finally, the more moderate argument put forward by Daniel Dennett will be discussed, which states that religion should not be excluded from the public sphere because it is “evil” in principle, but rather because inclusion serves as a dangerous door opener for religious fanatism and the like (Section 3.3). This, however, is an untenable “perversion argument” subsequent to Albert O. Hirschman’s Rhetoric of reaction.

3.1. Outlining New Atheism

In its political argumentation, New Atheism operates on the premise of the epistemic superiority of a certain idea of what scientific knowledge is. In this respect, it is necessary to first briefly address some epistemological considerations (a) before the political arguments that follow (b).

(a) To illustrate the underlying epistemological position of New Atheism, it helps to have a look at Hans Albert’s development of the so-called Münchhausen trilemma. With this trilemma, Albert, a companion of Karl Popper, intended to underpin critical rationalism argumentatively by presenting it as the only intellectually honest way out of the justification problems of foundationalism.

Foundationalism assumes that justified knowledge x is possessed if at the same time there are foundations f(x) that qualify x as epistemically justified. If, for example, one sees a book, one would then be able to say that the knowledge of the existence of the book is justified, if, on demand, foundations such as glasses were given. “I am epistemically justified in assuming that there is a book in front of me, because I wear glasses that let me see it.”

However, the question arises as to how one can be certain that one’s own knowledge is really true, for might not the respective foundation f be defective or unsuitable for arriving at true knowledge? What justifies the foundation through which something else is justified? In the attempt to answer this question within the framework of foundationalism, one becomes entangled, according to Albert (1975, pp. 29–34), in the Münchhausen trilemma.

(I) First, the answer could be that the foundation f, in the given example of the glasses, is justified because they not only help one to see the book f(x), but also other objects, such as a pen next to the book f(y), a paper sheet f(z), et cetera—they even help their owners to orientate themselves in the outer world. However, the first horn of the trilemma is that this attempt at justifying f does not justify the foundation itself, but only lists further cases of supposedly successful recognition with the help of the foundation f in question. Unless a justification of f itself is given, all these further cases could all be simple coincidences of factually true knowledge, meaning it is uncertain whether they are really true, or, in the other case, merely illusions.

(II) Based on this, one could try to systematically check the epistemic adequacy of the foundation f by, for example, taking off the glasses and checking if they are alright. This strategy, however, would not be convincing either, because in doing so, one would merely replace one foundation f, the glasses, with another one f₂(f), the eyes—and if there is doubt about their epistemic justification, then one could perhaps ask an oculist who would simply count as a third foundation f₃(f₂(f)) that might be false and so on. Trying to justify knowledge in this way would therefore lead to an infinite regress.

(III) The third horn of the trilemma consists in “dogmatically” abandoning the attempts at justifying the foundation (Albert 1973, p. 31; 1975, p. 97; 2000, p. 15). Accordingly, one simply believes in the functionality of one’s own foundations and could possibly support them with reasons as in (I) and (II). However, this ultimately suspends the attempt to justify the foundation in a generally comprehensible way.

According to Albert (1975, p. 29, trans. M.R.), from the trilemma’s insolubility follows that all “knowledge”, which justifies itself in a foundationalist manner, is “self-fabricated and thus worthless for apprehending reality”. For the epistemically justified evaluation of any body of knowledge, a change to the critical–rationalistic paradigm is recommended, as is exemplary in today’s natural sciences and empirical, quantitative social sciences.17 According to this, our knowledge of the world is only an accumulation of (for the time being) not falsified hypotheses about what is actually the case. In order to avoid “dogmatisation”, these hypotheses should offer the possibility of being “critically examined” by trial and error (ibid., 35). If, however, a hypothesis such as “God exists” cannot be examined in an intersubjectively comprehensible way because, to stay with the example, the existence of God might not reveal itself equally to everybody, then such a hypothesis must be rejected as perhaps true but nevertheless epistemically unjustified.

(b) Up to this point, Albert’s argumentation has been concerned with the predicate of epistemic normativity, the assumptions of which can be said to be epistemically justified. Following this, Albert (1979, p. 17; 2000, pp. 144, 215–17; 2011, p. 94) builds a bridge from epistemic to political normativity by identifying certain factually given worldviews, especially monotheistic religions, which, as illustrated by the third horn of the trilemma, evade critical examination of their foundations “dogmatically”, for example by referring to “theologically sanctioned texts” (Albert 1975, p. 160, trans. M.R.). They might, following Albert, accept certain assumptions derived from critical investigation, such as the one that is undertaken by science. However, they cut off foundations, which are relevant for their world view, from this process of “critical examination”.18 This is therefore not only an intellectual problem, but also a democratic one, because:

“the social aspect of dogmatism includes, above all, institutional arrangements created by groups with dogmatic belief systems […] to immunise their members against the influence of dissenting views and dangerous ideas and information”.

(ibid., 97, trans. M.R.)

Albert’s argument is sociological: epistemic “dogmatisation” can only persist through institutionally enabled social dogmatisation.19 One option regarding this institutionally founded stabilisation of “dogmatic” convictions consists of making political claims on the bases of “supposedly revealed truths” (ibid. 160, trans. M.R.) in order to endow them with the power of legal norms. If this succeeds, positive law then serves to consolidate or at least privilege certain “dogmatic” world views and ways of life.

This quite daring hypothesis may at first sound abstract, but it gains plausibility when one thinks of, for example, efforts by religious fundamentalists to prevent legislative initiatives such as the establishment of same-sex marriage by referring to “theologically sanctioned texts”. This is highly problematic in terms of democratic theory because, as mentioned, democratic rule, although it is rule accompanied by the option of state sanction, does not violate the freedom and equality of all those subject to the law in that, on the one hand, the citizens are the authors of the law, and, on the other, they jointly seek for reasons that can be accepted by all affected. Prima facie, the “supposedly revealed truths” of a specific part of the population do not meet this qualification because they are based on personal beliefs.20

According to Albert, what reasons can be appropriate for the deliberation of democratically legitimate norms? In order to answer this question, one must look back to the solution of the Münchhausen trilemma, according to which only those reasons that are not justified with the help of any believed foundations but can be critically examined via trial and error and (provisionally) withstand this control can be intersubjectively accessible.21 It follows from this that the reasons put forward in the democratic formation of a political will must also be able to be examined via trial and error to fulfil the principle of public justification. That which passes this test can then be used as a generally accessible resource for justifying democratically legitimate policies (Albert 1975, pp. 173–182).

In this sense, Philip Kitcher (2008, p. 13) states vividly:

“The commitment to making religious allegiance a private matter must be maintained, and, in consequence, religious considerations must not be allowed to enter public debate. Campaigning for greater public understanding of science can readily be supported by appeal to democratic principles: insistence on using scientific standards as constitutive of genuine knowledge would support the dissemination of correct information, thus enabling citizens to identify their genuine interests. For, after all, the possibility that democratic involvement will promote freedom seems to depend on giving the citizenry the best possible factual basis for forming their judgement.”

The only rational potential that ultimately seems to remain for religion in the perspective of New Atheism is to point to individual and social pathologies (Albert 1975, 174f.). One could think of the assumption that religiosity is reactivated where there is social or cultural precarisation (cf. Kitcher 2008, p. 14). Accordingly, the public appearance of religion can serve as an indicator for social problems, but without ascribing to it an intrinsic value. Religion is, to use a famous quotation from Marx ([1844] 1978, p. 54), “the sigh of the oppressed creature”:22

“In my secularist view, those standards [for public reasoning] can only come from insistence on the priority of science, but that insistence must make room for the religious impulses and concerns that militant secularists currently ignore. Secularism must be humane, recognizing the needs for community, for social support, for ways of exploring why human lives matter”.

(Kitcher 2008, p. 15)

3.2. Why New Atheism Is an Unreasonable Comprehensive Doctrine

New Atheism’s epistemological critique of religion proceeds in such a way as to reject religious faith (at least in its monotheistic form) as intellectually dishonest, albeit on the basis of supposedly unsatisfactory evidence.23 Only through a so-called “dogmatic” circumvention of the critical examination of faith could monotheism still exist in its common form. This critique of religion becomes relevant in political theory because epistemic “dogmatisation” also makes use of political instruments. However, in liberal, enlightened societies, it is concluded, only evidence-based reasons can be suitable for the democratic decision-making process.

Intuitively, such a position may be convincing at first: what religious fanatics believe to have been revealed may hardly be suitable for a generally acceptable and therefore reasonable decision. However, according to my following refutation, New Atheism, similar to religious dogmatism, is also not suitable for naming generally acceptable reasons and is thus in itself dogmatic and authoritarian. The reason for this is that, thinking of the burdens of judgement, different epistemic points of view exist which can lead to (provisionally) equally justified, yet contradictory, truth claims. Yet, if, in modern societies, there are different, equally well-founded assumptions about what is the case, then, in the event of reasonable disagreement, reference to one epistemic point of view cannot be sufficient to generate general acceptance (a). Furthermore, there are good reasons to assume that New Atheism leads to counterintuitive normative consequences that contradict democratic existential presuppositions and practices (b). Accordingly, if one considers certain preconditions of the democratic mode of cohabitation, one must inevitably refrain from a positivistic worldview such as that which New Atheism proposes.

(a) New Atheism identifies what is the case positivistically as that which can be observed. By doing so, New Atheism does not make a methodological point, but an ontological one: it presupposes that intellectually justified knowledge has to be observable because the world simply consists of natural entities (cf. Audi 1992, 52f.; 2006, pp. 88–91).24 It is true that we might seem to perceive immaterial phenomena, for example mental ones, but all this is supposed to be due to physical factors such as biochemical processes. Therefore, the existence of an immaterial being called God, a human soul, and so on is to be rejected in the sense that it is not empirically provable.

However, why is such a standpoint as this unreasonable? Among other things, it is unreasonable because it does not acknowledge the fact of reasonable pluralism, as mentioned in point 5 in Section 2.3. To argue in favour of this, I first refer to Robert Audi’s discussion of ontological positivism.25 Following Audi (1992, p. 53; 2006, 93f.), ontological positivism operates with the principle of global determinism. The principle of global determinism states that if everything that exists, even seemingly immaterial phenomena, supervenes on material processes, then ultimately everything that is and will be follows the determined laws of cause and effect. The idea of intentional action, however, is based on the premise of mental causation, which contradicts the principle of global determinism. Audi argues that we cannot know whether there is such a variable as mental causation or simply only global determinism. For the question as to which worldview is better justified, a worldview with mental causation or a positivistic worldview with global determinism, there remains only the comparison of the explanatory power of both ways to look at the world. Therefore, the epistemological question thus shifts to whether we can explain more phenomena in our outer world by assuming that everything is globally determined or by assuming non-material mechanisms such as mental causation. Audi concludes that, when looking at our world, the explanatory power of the principle of global determinism on the one hand and mental causation on the other is quite the same. We can think of phenomena such as free choice that cannot be described in terms of global determinism but are not seldom, for example, in our everyday experience or in qualitative social research.

Benjamin Libet (2005, 562f.), whose famous experiments are often wrongly cited as evidence for the untenability of the idea of mental causation, takes a similar position on the explanatory competence of this principle and the problematic status of an overemphasis on global determinism by saying:

“The assumption that a deterministic nature of the physically observable world (to the extent that may be true) can account for subjective conscious functions and event, is a speculative belief, not a scientifically proven proposition. Nondeterminism, the view that conscious will may, at times, exert effects not in accord with known physical laws, is of course an unproven speculative belief. […] Given the speculative nature of both determinist and nondeterminist theories, why not adopt the view that we do have free will (until some real contradictory evidence may appear, if it ever does)? Such a view would at least allow us to proceed in a way that accepts and accommodates our own deep feelings that we do have free will.”

Thomas Nagel (2012, p. 4) even goes beyond Audi and Libet in the sense that he not only opposes ontological positivism with at least the same explanatory power of a non-materialistic worldview, but even explicitly speaks of its “failure” in explaining the world:

“What is lacking, to my knowledge, is a credible argument that the story has a nonnegligible probability of being true. […] My skepticism is not based on religious belief, or on a belief in any definite alternative. It is just a belief that the available scientific evidence […] does not in this matter rationally require us to subordinate the incredulity of common sense”.

(Nagel ibid., 6f.)

From this it can be concluded that, bearing in mind the epistemic justification of the principle of mental causation, one can assume at least the possibility of the existence of non-material entities, including a divine person, a soul, and so on, as can be seen in the common forms of monotheism. As the explanatory power of positivist descriptions of the world is not greater than that of those explanations which work with the assumption that in certain aspects of the outer world there are entities and phenomena which cannot be reduced to the purely material, one cannot defend a general rejection of the epistemic justification of the belief in God—although this does not, of course, say ex positivo how an intellectually justified belief must look.

What does this mean for democratically organised coexistence in modern, pluralistic societies? It is evident that there are good reasons to reject a positivistic worldview as the only possible one. Rather, we must assume that different non-materialistic, reasonable, and comprehensive doctrines exist due to the burdens of judgement. Therefore, it is unreasonable to stipulate that only purely evidence-based reason should count in deliberation. This would simply unjustifiably favour a specific doctrine (cf. Grätzel 2015; Bassiouni 2020, p. 67).26

This is the reason why Rawls (1997, pp. 775–80) underlines that publicly acceptable reasons are not secular reasons—at least not secular reasons in the way New Atheism would define them: “We must distinguish public reason from what is sometimes referred to as secular reason and secular values. These are not the same as public reason. For I define secular reason as reasoning in terms of comprehensive nonreligious doctrines. […] A view often expressed is that while religious reasons and sectarian doctrines should not be invoked to justify legislation in a democratic society, sound secular arguments may be. But what is a secular argument? Some think of any argument that is reflective and critical, publicly intelligible and rational, as a secular argument; and they discuss various such arguments for considering, say, homosexual relations unworthy or degrading. Of course, some of these arguments may be reflective and rational secular ones (as so defined). Nevertheless, a central feature of political liberalism is that it views all such arguments the same way it views religious ones, and therefore these secular philosophical doctrines do not provide public reasons. Secular concepts and reasoning of this kind belong to first philosophy and moral doctrine, and fall outside of the domain of the political”.

Of course, sometimes deliberation has to deal with better founded truth claims, and evidential reasons might be better suited for deciding what we should do in order to avoid illegitimate coercion, for example in questions about climate policy. However, this alone does not constitute the deliberative process. Not only evidential questions are of political importance, but also questions of what is the ethically and morally right thing to do or how to cope with different interests (cf. Habermas 1996, p. 319). In this sense, New Atheism’s picture of what a normatively sound process of will-formation should look like is not just authoritarian in that a doctrine is unfoundedly preferred. Rather, New Atheism also commits a pragmatic “constriction of political discourse” (Habermas 1994, p. 4) by not sufficiently considering issues such as moral ones.

Habermas (2008a, 140f.) nicely describes this one-sidedness as follows: “Scientism often misleads us into blurring the boundary between natural scientific knowledge which is relevant for understanding ourselves and our place in nature as a whole, on the one hand, and a synthetic naturalistic worldview constructed on this basis, on the other. This form of radical naturalism devalues all types of statements that cannot be traced back to empirical observations, statements of laws, or causal explanations, hence moral, legal, and evaluative statements no less than religious ones.”

(b) Up to this point, the objection against ontologically understood positivism has been epistemological in nature: as other paradigms of explaining the world also have a similarly high explanatory competence, it would be epistemically unjustified to accept only a purely positivistic doctrine as reasonable. However, following Audi’s critique of positivism, one can, in my view, raise another objection to the understanding of politics as implied by New Atheism. My thesis is that the worldview as represented by New Atheism ultimately contradicts certain existential presuppositions of the democratic constitutional state and is therefore also unreasonable.

I would like to make this comprehensible with the help of the method of the wide reflective equilibrium (cf. Rawls 1999, 18f., pp. 42–45; Fishkin 1984, 117f.; Schmidt 2012, 148f.). The wide reflective equilibrium is characterised by the fact that, in its application, not only is the coherence of considered judgements to be examined depending on unquestionably valid principles, but also attempts are made to bring together the principles themselves, as well as other elements through variation and rejection. These further elements include, for example, argumentation figures such as thought experiments or, as will be important in the following, positive theoretical assumptions.

Following Audi’s discussion of mental causation, my objection to New Atheism is that the democratic constitutional state is necessarily dependent on the assumption of the existence of mental causation. Conversely, a democratic constitutional state cannot be justified in purely positivistic terms. For my argumentation, I would like to refer to two constitutive factors from the practice of democratic life: one being the principle of public justification (1), as mentioned in point 1 in Section 2.3, and the other being the attribution of responsibility (2), accompanied by the option of legally regulated sanctions. These two principles, I argue, are dependent on the premise of a freedom of will, which may be conditional and limited, but is nevertheless existent.

(1) The principle of public justification postulates that citizens must be cognitively capable of transforming their political demands in the light of the potentially justified claims of their fellow citizens, as well as acting according to these good reasons given by others (cf. Elster 1992, 111f.). If, however, one was to assume that political preferences and the resulting actions were merely the product of biochemical causal mechanisms, then neither the deliberation of good reasons nor the claim to act according to them would make sense (cf. Frankfurt 1969). This applies because the search for legitimate justifications for a certain policy or polity, which may also lead the individual to refrain from egocentric claims, depends on the ability to act against inner resistance in the light of better reasons (cf. Kaiser 2004).27

(2) Closely linked to the assumption of the attribution of free will as a necessary condition for deliberation and rational action is the possibility to attribute responsibility, which becomes democratically relevant in at least two ways. On the one hand, it is significant to be able to hold political actors such as members of parliament, but also citizens in the political sphere, accountable for their decisions. On the other hand, it is important to sanction misconduct. Regardless of whether one refers to the principle of retribution, of prevention, or a synthesis of both for the justification of sanctions (cf. Hart 1968b, p. 26), many authors in philosophy (of law) (cf. Hart 1968a; Günther 2000, p. 478; Apel 2002, p. 16; List 2014) take the position that a sanction can only be legitimate if the person acting has a specific responsibility for their actions that goes beyond a mere deterministic “neuronal programming” (Merkel 2008, p. 104, trans. M.R.). Only if conduct can be identified as an action—even if only a negligent action—does an attribution of responsibility and guilt appear to be convincing and justified (Habermas 2002, p. 100; 2008a, p. 141; 2019b, pp. 298–319; Buchheim 2004, p. 164; Zunke 2008, pp. 181–89).

However, actions in this sense cannot exist in the context of an ontologically positivistic worldview since here all human behaviour ultimately supervenes on biochemically determined mechanisms. Accordingly, what people do can only be described as an effect or a reaction to a certain stimulus, but not as an action. Therefore, it marks a normatively significant difference “whether one sees one’s counterpart as a determined bio-machine or—at least in principle—as a freely acting subject” (Müller and Schmidt 2015, p. 11, trans. M.R.).28

(∑) This consideration of the principles of public justification, responsibility, and free will can be illuminated in the light of the wide reflective equilibrium as a critique of the congruence between New Atheism and democratic values as follows: if the principle of public justification and the attribution of responsibility are assumed to be constitutive elements of democratic practices, the existence of free will and therefore of mental causation has to be assumed as well. The assumption of mental causation, in turn, is, as Audi shows, incompatible with ontological positivism. As New Atheism’s conception of politics is based on an ontologically positivistic worldview, it contradicts the values implicit in everyday democratic practice.

Michael Sandel (2007, p. 25) sums up this position well: “One aspect of our humanity […] is our capacity to act freely, for ourselves, by our own efforts, and to consider ourselves responsible—worth of praise or blame—for the things we do and for the way we are”.

To put it briefly, either the logical defence of a democratic organisation of society would require the abandonment of ontological positivism, or the upholding of ontological positivism would require the abandonment of those premises that are constitutive for democracy.

3.3. “Protective Coloration for Their Fanatical Coreligionists”? New Atheism’s Rhetoric of Reaction

A more sociological thesis that can be counted as part of New Atheism’s worldview is that even a deliberatively qualified inclusion of religiously based demands is to be omitted because it would be a door opener for all other, possibly troublesome, religious claims. Daniel Dennett (2006, p. 299), for example, expresses this hypothesis, vividly following Sam Harris:

“by their good works they provide protective coloration for their fanatical coreligionists, who quietly condemn their open-mindedness and willingness to change while reaping the benefits of the good public relations they thereby obtain.”

Now, on the one hand, this seems to be an empirical question that should not simply be answered generally by a philosophy book. On the other hand, it is nevertheless possible to reject this hypothesis with the help of Albert O. Hirschman’s political theory. A stance such as Dennett’s is therefore not convincing because it follows the argumentative logic of what Hirschman (1991, p. 11) terms a perversity thesis.

Arguments within the framework of the perversity thesis claim that attitudes or actions that might be supportable at first glance are to be rejected because they are necessarily accompanied by an unintended, opposite effect. Yet, the logical problem with arguments that follow the line of a perversity thesis is that they do not give sufficient reasons for this supposed general mechanism (Hirschman 1991, p. 35). For example, if A recommends policy l to achieve purpose x, B could formulate a counterargument in the sense of a perversity thesis, if B claims that x is a legitimate purpose, but rejecting l because l would lead to the opposite of x. In fact, what B might want to achieve is to prohibit l and, accordingly, x, but knows that this position is not generally acceptable.29 Thus, for example, welfare state policies being rejected with the neoliberal warning of an entry into a communist economy of scarcity or, to take another example from the connection of religion and politics, the banning of same-sex marriage as a means to avoid marriages between adults and minors can be identified as perversity arguments.

Dennett’s idea that the deliberately controlled inclusion of religious reasons in public discourse will function as a “protective coloration for fanatical coreligionists” works as a perversity argumentation, because there is no striking reason why the inclusion of emancipative, democratic, yet somehow religious, arguments should lead to the intrusion of normatively deficient religious reasons. Why should a critical public not be able to separate the first from the second?

A historical retrospective further illustrates the empirical untenability. Logically, following Dennett, the criticism of racial segregation by the vastly religion-influenced US-American civil rights movement would have to be rejected as democratically problematic because even if its concerns were justified, its religious framing would bear the danger of opening the discourse to religious fanatics. Additionally, the same imperative of restraint would have to apply equally to the religiously inspired resistance fighters under National Socialism or the Latin American liberation theologians who speak out against exploitation.

4. Conclusions

The presented considerations lead to a potentially surprising conclusion, which is that New Atheism can be identified as a comprehensive doctrine because it formulates a comprehensive idea of the outer, social, and inner world of its supporters. It is an unreasonable comprehensive doctrine, although its rhetoric might signal something different. On the one hand, the epistemologically unfounded focus on purely evidential reasons for deliberation, with the simultaneously prescribed exclusion of every other reason, cannot be reasonably accepted by all other citizens of a democratic society. Even if this requirement for exclusion is postulated with regard to securing democratic values, it is authoritarian. On the other hand, reasonable comprehensive doctrines are supposed to have a stabilising effect in the sense that they overlap in a consensus regarding democratic values. However, it is obvious that New Atheism’s reference to ontological positivism is logically incongruent with the values of a democratic constitutional state, and thus, it cannot overlap with reasonable comprehensive doctrines. While religions are in themselves highly varied and must therefore be examined in terms of reasonableness case by case, New Atheism can be rejected as unreasonable in general.