‘Taking politics seriously: A prudential justification of political realism’

1. Introduction

Studies into realist political theory have been developed following two main lines of inquiry. First, theorists have devoted much effort and time to clarifying the methodological specificity of realist theorising. The questions animating the first strand of research have been: What are the essential features of a realist approach to political theorising? What is the relationship between political reality and normativity in realist thinking? ¹ Second, theorists have produced sustained defences of the consistency of realism as an approach to political reasoning. The questions animating the second strand of research have been: Is it really possible for realism to avoid moralism? Can political theory qualify as a distinctive kind of normative reasoning? ² And yet the question of how to justify the realist approach has not received the same attention. The systematic inquiries into the metatheoretical basis of realism are still few, and, I argue, they seem unable to offer compelling reasons to a non-realist audience for adopting political realism (Hall 2014; Hall and Sleat 2017; Prinz 2016; Sagar 2016).

Such neglect of justification is not as surprising as it sounds, though. Indeed, it could be explained as a consequence of the anti-foundationalist tendencies of political realism. Notably, for political realists, theories ought to be assessed primarily by looking at political reality and what strikes us as sensible to do, rather than by providing stable foundations and absolute systems of norms (Jubb 2017, 117; Philp 2012). And yet it is critical to provide a compelling justification of political realism. If realists fail to provide a persuasive justification, they will be unable to explain why their method should be favoured over alternative ones.

So, why should we choose political realism as a method for normative political thinking? I offer a justification of realism building upon the complexity of realist thought and taking advantage of the argumentative strategies that realists have moved against political moralism. Specifically, I propose a prudential argument for political realism. In doing so, my primary aim is not interpretive: I do not claim that the argument that I make is the one that the realist thinkers that I consider had in mind. Rather, I try to bring to the surface an argument, more or less implicit in realist writings, that I take to be the most forceful one to employ in defence of political realism.

To do so, I proceed as follows. I first characterise political realism as anti-moralism. I then outline in detail three possible arguments for the realist approach by availing myself of recent inquiries into the metatheoretical basis of realism: The metaethical, the ethical and the prudential arguments. I explain that the prudential argument offers the most solid basis for political realism because it relies on the least controversial premises. Still, I delve into the metaethical and ethical arguments for two reasons: The prudential argument takes advantage of the theses defended by the rival arguments and elaborating the other arguments gives me the chance to show the comparative strengths of the prudential argument.

2. Political realism as anti-moralism

Political realism as a tradition does not constitute a uniform and systematic corpus. Rather, those who define themselves – or have been defined – as realists share some core research interests and theoretical sensibilities (see Galston 2010; McQueen 2017; McQueen 2018). The interpretation of realism that I offer is based on Raymond Geuss’s and Bernard Williams’s writings. This choice is motivated by the fact that Geuss and Williams are usually regarded as the scholars who have played the largest role in shaping the contemporary understanding of political realism (Rossi and Sleat 2014). Hence, my starting point is ecumenical: By studying Geuss and Williams, I seek to propose a sharable justification of political realism, despite the differences that can be traced within this tradition.

By drawing on Geuss’s and Williams’s reflections, I interpret political realism on the basis of what it aims to reject: Political moralism (or political theory conceived of as applied ethics, as Geuss dubs it). Therefore, before proceeding with the outline of the justificatory arguments, we first need to understand what it means to develop an anti-moralist political theory. In Philosophy and Real Politics (2008), Geuss points out:

‘Politics is applied ethics’ in the sense I find objectionable means that we start thinking about the human social world by trying to get what is sometimes called an ‘ideal theory’ of ethics. . . . The view I am rejecting assumes that one can complete the work of ethics first, attaining an ideal theory of how we should act, and then in a second step, one can apply that ideal theory to the action of political agents. (6, 8)

In a similar vein, in ‘Realism and Moralism in Political Theory’ (2005, 1–17), Williams claims:

I start with two rough models of political theory . . . with respect to the relation of morality to political practice. One is an enactment model. The model is that political theory formulates principles, concepts, ideals, and values; and politics (so far as it does what the theory wants) seeks to express these in political action, through persuasion, the use of power, and so forth. . . . Contrast this with a structural model. Here theory lays down moral conditions of co-existence under power, conditions in which power can be justly exercised. . . . My concern here is with what they have in common, that they both represent the priority of the moral over the political. Under the enactment model, politics is (very roughly) the instrument of the moral; under the structural model, morality offers constraints . . . on what politics can rightfully do. In both cases, political theory is something like applied morality. (1–2, emphasis in original)

These two passages exhibit a common understanding of political moralism. For Geuss and Williams, a political theory is moralist when its normative foundations lie outside politics – that is, when the principles it defends are elaborated while abstracting from actual political practices. Indeed, what Geuss and Williams emphasise is that in ethics-first models an understanding of political reality is superfluous for elaborating the normative principles suited to it. In this sense, a political theory must be understood as moralist when political reality plays a subordinate role, and it only starts to matter when the normative principles are applied to political circumstances.

Geuss and Williams believe that such a way of conceiving of political reasoning – namely, deriving political norms from pre-political ones – is seriously flawed. In their opinion, political theorising should be conducted by starting from a serious appraisal of real politics – that is, by paying due attention to politics as a human activity that is governed by distinctive rules, dynamics, values and purposes and tailoring normative reasoning to such distinctiveness (Geuss 2008, 6–18; Williams 2005, chap. 1). Indeed, according to Geuss and Williams, once we pay due attention to the reality of politics and how we engage with it, we realise that political reasoning cannot merely consist of an attempt to adapt moral principles to political practices; rather, political practices should be considered as a source of normativity, not merely a field of application of moral norms. In this sense, realist political theory is so called not because it is a realistic form of political theorising – one that seeks action guidance in actual political circumstances – but rather because it assigns political reality a central normative role. Realists, then, advocate a form of political theory that starts from political reality itself to elaborate the political norms that should regulate our political practices.

But how, then, is political reality supposed to play a role in the justification – not merely the application – of political norms? To be sure – as I explain shortly – Geuss and Williams conceive of the appropriate conduct of political theorising in substantively different ways. However, they share one core methodological commitment, whose justification I examine in this article. For Geuss and Williams, the evaluation of political practices and of the norms that should guide them should be conducted in light of an assessment of the value and purpose such practices have, according to the point of view of the actual political agents that participate in them. More precisely, what should be done in political circumstances should reflect what it makes sense to do in those circumstances according to (a critical interpretation of) the system of beliefs of the actual political agents involved. Then, a good share of the final form that a realist theory takes depends on how we interpret this fundamental methodological requirement. In fact, it is possible to develop different realist analyses of political normativity according to how the criterion of acceptability of beliefs and the general methods of elaborating norms are interpreted; and here is where Geuss’s and Williams’s interpretations of political reasoning diverge. Geuss (2016, chaps. 1–2) largely employs a ‘negative’ methodology according to which we ought to orient political action by examining, on the basis of genealogical analysis, which beliefs must be rejected as ideological (and, correspondingly, by following those that pass the test of critical reflection) (see also Prinz 2016; Rossi 2019). Williams (2005, 1–17), instead, adopts a less revisionist, and more constructive, approach: His method can be described as a practice-dependent method in which political practices ought to be reformed in order to fulfil their purpose and the value that their participants attribute to them (see also Jubb 2016; Rossi 2012).

However, for my purposes, it is unnecessary to delve further into the details of the realist methodology. Indeed, I am not interested in arguing in favour of a specific interpretation of the methodology realist political theorists should follow. ³ Rather, I am interested in looking for the reasons that might justify the fundamental methodological insight that moves the realist approach to political theorising. Realists intend to ground political normativity on a critical interpretation of what makes sense to do politically from the point of view of real political actors. Political practices ought to be reformed, or organised, to reflect (a critical interpretation of) what participants regard as worth achieving or safeguarding. Hence, unlike moralism, which defines political norms by disregarding the political circumstances of their application, political realism is a bottom-up procedure that construes the normative analysis from a critical interpretation of actual beliefs and practices. ⁴

Notice, then, that ‘anti-moralism’ does not mean ‘anti-moral’ (Geuss 2008, 1–4). Political realism, properly understood, claims that a political theory ought to start from an understanding of the significance of political practices. Therefore, when political realism is described as a claim about the ‘autonomy of the political’ (Williams 2005, 3), the claim must not be intended as a rejection of ethical thought but as a rejection of moralist methodologies. Ethical thought can contribute to a realist understanding of political normativity precisely because realist political thought focuses on, among other things, what actual agents find politically valuable (Geuss 2016, chap. 2).

Then, why should political theory favour such a bottom-up approach to constructing political normativity? I outline a justification of political realism by taking into account the corpus of Williams’s and Geuss’s works as a whole. Admittedly, mine is just one possible way of grounding political realism, but I argue that it might be the most persuasive one to employ against moralists. Indeed, Geuss and Williams are not entirely systematic writers, and for this reason, multiple interpretations of the grounds of political realism can be consistently offered. Before explaining my interpretation, I introduce two alternative strategies for justifying political realism which have been outlined, more or less explicitly, in the recent literature (Hall 2014; Hall and Sleat 2017; Prinz 2016; Sagar 2016): The metaethical and the ethical arguments. ⁵ The prudential justification takes advantage of both.

3. The metaethical argument: From the limits and sense of ethics

The first argument that can be reconstructed from Williams’s and Geuss’s writings as a basis for political realism revolves around a thesis that they share about ethics. According to this argument, it is realists’ specific understanding of the scope and limits of ethics that grounds their approach to political theory. Geuss and Williams are moral sceptics; still, they believe that some form of ethical discourse is both meaningful and necessary. This first justificatory strategy is based on the substitution of morality with another form of ethical reasoning that, as far as politics is concerned, takes the form of political realism. This is the interpretation favoured by Ed Hall and Matt Sleat, who have recently interpreted political realism as primarily a thesis about the grounds of ethical knowledge (see Hall 2014; Hall and Sleat 2017).

How is it possible that realists are moral sceptics and yet they defend a particular normative approach to politics? Here, words must be read and chosen carefully. Following Williams, I am making very specific use of the concept of morality in this context. ⁶ In Geuss’s and Williams’s works, morality is a form of normative reasoning which aims at defining a system of principles that establishes what ought (or ought not) to be done in a given context (Geuss 2016, 28–33; Williams 2011, 80). The authority of such moral systems typically comes from an independent justification that proves their allegedly objective validity; the justification is usually intended to demonstrate the universal validity of the system of norms. Morality, in this sense, resembles a scientific inquiry: The philosophical endeavour is devoted to defining a consistent system of laws whose justification requires universal recognition (Williams 2011, 193–201). It is easy to see how morality, so conceived, relates to the notion of political moralism that I introduced in the previous section. If morality can define what ought to be done in every circumstance and universally, politics is just one field of application of a moral theory independently justified. According to this conception, there is no point in conceiving of political theorising as requiring a distinctive form of normative reflection. As Hall and Sleat (2017) explain, however, Geuss and Williams harshly reject morality.

But is it possible to abandon morality while saving some form of positive ethical thought? After all, if we reject the idea of an objective and universal normative knowledge, are we not doomed to fall into relativism or nihilism? Geuss and Williams present an alternative view, and their answer is at the root of their defence of political realism – or, better, of a first possible defence of political realism. To understand how they come to defend realism, we first need to understand why, in their opinion, morality is untenable.

Williams and Geuss reject morality for the same reason. They both argue that it is impossible to provide a philosophical justification of a moral system. In fact, according to them, the failure of contemporary moral outlooks is not merely accidental; the whole project of morality ought to be dismissed and rethought. According to Williams and Geuss, at the bottom of contemporary moral theory is a failure to understand the substance of ethical life. Indeed, it is no coincidence that one of the major theoretical influences they share is Nietzsche. ⁷ Following Nietzschean leanings, both Geuss and Williams believe that the project of providing a philosophical justification of morality is doomed to fail because there is no foundation to be discovered. There is no stable basis upon which to ground our moral knowledge: Ethical life is a product of history for which an ultimate foundation – be it rational, transcendental, theological or teleological – cannot be found. So there is no hope of finding an objective point of view, external to our contingent and partial one, to justify our ethical life and prove that what we do, or claim, is ultimately right.

Williams’s Ethics and the Limits of Philosophy (2011) is the work in which this thesis is most fully discussed. There Williams meticulously analyses what he takes to be the most developed attempts to provide a philosophical foundation of our ethical life. He examines Aristotelian foundations (2011, 34–59), Kantian foundations (60–78) and the attempt to objectively ground ethics by seeking to converge on ethical truths (146–72) and he argues that all of them fail. In this way, Williams aims to substantiate the thesis that it is impossible for philosophy to find the ultimate grounding of our ethical life. That foundation does not exist, and the failure of all the major philosophical strategies to meet our need for objective moral answers proves his point (see Hall 2014, 548–51). ⁸ Geuss, albeit in a less systematic fashion, defends the same idea by discussing ethical beliefs in their historical evolution and context. For Geuss, morality is grounded upon illusions which become manifest once we engage in a careful genealogical analysis: What we, here and now, believe to be an ethical truth – eternal, coherent and progressive – is actually the contextual product of specific historical circumstances (see esp. Geuss 2001a; 2001b). In Geuss’s (2005) view, there is no ‘view from nowhere’ (4), and we should deeply recognise this fact if we want to make proper sense of our ethical lives.

What should our reaction be, then, if we agree with Geuss and Williams? Should the realisation that an ultimate foundation for our ethical lives cannot be found leave us in a state of nihilism and despair? This would be a serious mistake. Geuss and Williams believe that realising the contingent nature of our ethical beliefs does not imply the need to immediately reject them, let alone the impossibility of making sense of our ethical lives. Those who think otherwise, they claim, make the fundamental error of again believing that foundationalism is the sole way to ascertain the validity of ethical propositions. In fact, they argue, those who claim that it is impossible to hold any ethical beliefs as truthful because of their contingent nature are trapped in the same deception Williams and Geuss are trying to unmask. Indeed, a similar worry only arises if we are convinced that foundationalism is the sole way to make ethical life meaningful. Those who react with despair are still ‘under the shadow of universalism’ (Williams 2005, 67). The point, rather, is that Geuss and Williams are inviting us to radically change our conception of ethics and ethical life. We do not freely embrace a certain ethical life; we do not simply discover ethical truth and change our lives as soon as we get better information about the truth. Ethics is a necessary part of human identity and relationships. We cannot simply detach from it and consider it a fully independent object of inquiry because it makes us who we are, and hence, we cannot get free of it or examine it from a completely impartial point of view. That is why we ought not to regard the contingent nature of our ethical beliefs as troublesome. Contingent does not mean arbitrary, and once we understand the substance and role of ethics in our lives, contingency ought not to raise suspicions (Sagar 2016, 377). In Williams’s (2006) own words:

Precisely because we are not unencumbered intelligences selecting in principle among all possible outlooks, we can accept that this outlook is ours just because of the history that has made it ours; or, more precisely, has both made us, and made the outlook as something that is ours. We are no less contingently formed than the outlook is, and the formation is significantly the same. We and our outlook are not simply in the same place at the same time. If we really understand this, deeply understand it, we can be free of what is indeed another scientistic illusion, that it is our job as rational agents to search for, or at least move as best we can towards, a system of political and ethical ideas which would be the best from an absolute point of view, a point of view that was free of contingent historical perspective. (193–94)

If this is how we should rethink ethics, our conception of the role of philosophy should change accordingly. Philosophy, as said, cannot help us to construct a moral theory. The role of philosophy cannot be one of justifying ethics. Williams and Geuss conceive of philosophising about ethics in a wholly different manner. Philosophy should help us to better understand, and possibly critically assess, the origin of our beliefs. The fundamental task of philosophy ought to be to enable critical reflection (Geuss 2016, chaps. 1–2). This might take different forms (such as genealogical and interpretive reconstructions, utopian reasoning and artistic endeavours), and it must allow us to reach a truthful account of our beliefs – namely, to allow us to have a better understanding about what our system of beliefs is, which of our beliefs to retain and which ones to revise or reject. ⁹

Politics, like any other practical field, has to be understood along these lines. There cannot be any moralism for the straightforward reason that ethics can never be merely applied ethics: Ethical thinking must always start from an interpretation of actual practices (Hall and Sleat 2017, 281–82). Likewise, political philosophy needs to start from an understanding of people’s ethical beliefs about politics and needs to normatively appraise political practices from such an understanding – provided that the process is guided by critical reflection. Hence, according to this interpretation, political realism is motivated by a more general inquiry into the sense and limits of ethics.

At the beginning of this article, however, I said that my justificatory strategy might prove stronger and more convincing than the rival ones. I deal with this point more directly later, but for now let me point out some of the reasons why this first argument might be regarded as a weak defence of political realism. There are two reasons why the justification just introduced might weaken realism. A first one concerns the kind of justification pursued. This first argument is metaethical: It revolves around a fundamental thesis regarding the nature of ethics and the meaning that ethics might have for us. Convincing as this argument might be, it might nonetheless raise the conundrum that the ultimate justification for realism rests on a metaethical dispute – namely, on an extremely abstract inquiry. This by no means suggests that such a defence is inconsistent. Rather, the point is that it seems to be an unfortunate choice because realism typically requires that we start from real politics, while such a defence remains purely theoretical. ¹⁰ There seems to be much more going on in political realism than a ‘mere’ metaethical reflection; this justificatory strategy cuts off real politics and, with it, some of the themes dearest to the realist literature.

A second reason why this first argument might be weak concerns the argument’s very structure. Even if it is possible to reconstruct the path from (a form of) scepticism to realism by making use of Williams’s later writings and Geuss’s scattered suggestions, some doubts regarding the tenability of the project might arise. As Paul Sagar (2016, 378–79) notes, the analysis so far might not be completely convincing ¹¹ : A radical sceptic might object that, given the unavailability of stable ground for our ethical beliefs, it is hard to retain some belief as truthful; that is, it is hard for some belief to pass the test of reflection, and not to be regarded as the fruit of a persistent self-deception. The radical sceptic will have either to find some other way to defend realism ¹² or to reject the realist method. ¹³

4. The ethical argument: From the dangers of moralism

As I said, setting aside the metaethical argument for the moment, there are two further strategies one could pursue to defend political realism. The second argument could be dubbed the ethical argument – where ethics is understood in the realist sense that I briefly discussed above – because it focuses on showing that some of political moralism’s implications are unacceptable. While the metaethical argument debated the methods of inquiry that normative political theories ought to follow, the ethical argument revolves entirely around showing the practical dangers towards which political moralism leads – dangers that, the argument goes, ought to be eschewed and that realism is able to prevent.

A major part of Geuss’s and Williams’s writings is dedicated to reflections about the dangers posed by moralist thinking in political circumstances. This is a well-known feature of realist writings, which have often been described as overtly destructive because they mainly focus on highlighting the shortcomings of moralism instead of constructing a full-fledged alternative (see Galston 2010; Hurka 2009). This line of interpretation dominates a recent article by Janosch Prinz, which – by providing the first careful analysis of Geuss’s political methodology – offers valuable insights into the justification of political realism (Prinz 2016). As Prinz (2016, 4–9) correctly points out, Geuss’s central interest is not so much to engage in an abstract dispute aimed at establishing the best theoretical framework for guiding political action as it is to denounce, and possibly contain, the perils of moralism. Even though Prinz does not deal with Williams’s writings in his article, an analogous case could be made regarding Williams’s political theory. A major part of Williams’s political writings is deconstructive too: Williams (2005, 62–74, 145–53) most of the time deals with denouncing the absurd – in his view – implications of moralism, and he does so by making manifest the consequences of such a theoretical approach.

This crucial aspect of their thought remains mostly neglected if we interpret political realism as a primarily metaethical stance. The metaethical route, though consistent, seems to lose sight of some of the distinguishing features of realist thought. Instead, if we focus on the deconstructive dimension of political realism, it is possible to build a negative justification in its favour entirely based on the shortcomings of moralism.

How does this argument unfold? The reason why realists regard moralism as an approach to political reasoning that leads to unacceptable outcomes is its theoretical structure. The problem with moralist thinking is not, in fact, confined to a single (or some) specific kind(s) of moral theories; rather, the dangers fostered by moralism are due to the type of reasoning it supports. Indeed, the troublesome feature Geuss and Williams identify in moralism resides in its insensitivity to political reality. As we saw at the beginning of this article, moralist theories conceive of political reasoning as a form of reflection in which normative principles independently justified – typically moral principles – are applied to political circumstances. It follows that political reality – once the normative principles have been identified – cannot affect the justification of the normative principles defended by the political theory. Rather, in a moralist theory, the political circumstances of application of the theory only affect the feasibility of the normative principles; that is, real politics should be shaped according to the normative principles identified, insofar as it is possible to do so. Importantly, this means that realism must not be confused with a form of consequentialism. Consequentialist theories too must be regarded as moralist theories. The normative principles which a consequentialist theory defends require checking and weighting the real-world consequences of actions, but this requirement, per se, does not affect in any sense the validity of the principles applied, whose justification is thus insensitive to political reality. ¹⁴ The problem Williams and Geuss see in moralism does not strictly consist in its inability to consider the consequences of political actions. The trouble resides in the irrelevance to which political reality is condemned in moralist paradigms – in the stark separation between political theory and political reality when it comes to justifying the normative principles that should be applied in political circumstances.

Indeed, according to Geuss and Williams, the stark separation between political normativity and political reality that characterises moralist theories might lead to the justification of political principles which foster conflict, the misuse of violence, instability and oppression. As realists explain, this is because – given the nature and features of political reality – preventing such harms can only be achieved on the basis of a context-dependent normative analysis of the political circumstances under consideration. In other words, if those harms are to be prevented, the political context must play a justificatory role with respect to the principles that apply to it. To illustrate the point, let us examine some examples which serve as schema for the several cases Williams and Geuss discuss.

In ‘Humanitarianism and the Right to Intervene’, by examining the case of humanitarian interventions, Williams (2005, chap. 12) discusses what it would mean to apply moralist theories to politics. What would happen, he wonders, if we applied the everyday Principle of Rescue ¹⁵ – according to which we ought to assist people we happen to find in peril and whom we have the means to assist – to international relationships? This would be a paradigmatic example of moralist reasoning employed in political circumstances since political normativity would derive from normative principles justified for interpersonal relationships – that is, prior to, and independently from, politics. As Williams explains, the international case involves several complexities that make the Principle of Rescue inapplicable in political circumstances unless we accept its destabilising consequences. Indeed, in international relations, in deciding whether to intervene in another country to aid its population, several additional questions come into the picture, among which: Is the intervention welcomed by the government which would receive the aid? Would the intervention destabilise international political relationships? Which state is supposed to intervene? Is the intervention approved of by the population of the country which is providing help? If the realities of politics are considered carefully, it seems clear that the Principle of Rescue, if applied to politics, would severely undermine stability and trigger collateral conflicts. To escape such potential harmful consequences, it would be necessary to analyse the actual politics and the conditions upon which decent international relationships could be preserved. Hence, it looks dangerous, and therefore inappropriate, to apply principles designed to bracket political reality (in this case, designed for interpersonal relations) to actual political circumstances. A more politically savvy approach would require a contextual normative assessment of the normative principles that should be followed in political circumstances.

Similar considerations emerge also in Williams’s discussion of human rights as political institutions. As Williams (2005, chap. 6) is keen to emphasise, the charge of violating human rights is ‘the most serious of political accusations’ (72). The accusation of violating human rights has the force to justify enacting severe measures – even the use of power and violence – to eradicate the source of the violation. For this reason, Williams (2005) argues, defining a right as a human right is a choice that ought to be taken very carefully; it ought to be a matter of political good sense (72–74). This means that the definition itself, not the mere application, of what can be regarded (or not regarded) as a human right must be political in character – namely, compatible with the possibility of ordered international relationships. Indeed, if human rights are conceptualised without regard to the actual political circumstances of their application, the risk will be either to induce severe political conflicts or to turn rights into impotent rhetorical devices, given their incompatibility with the possibility of maintaining ordered political relationships (64).

Along these lines, Geuss (2010) warns against the potential harms of the insensitivity to political reality that characterises moralist theories when he says, ‘The moralizing approach tends to assign potentially infinite value to the struggle against certain […] visible forms of evil. If the task at hand is of infinite importance, some fabrication of evidence, suborning of or intimidation of civil servants, infliction of “collateral damage” on innocent populations, etc. are forgivable offences’ (33; see also 2016, 35–38). Here, Geuss is pointing out that, when political ideals are designed abstractly and then applied to political circumstances, moralism might indeed exacerbate conflict and violence. Indeed, if such shortcomings are to be avoided, political reality should have a role in shaping political theory, not the other way around. The attempt to mould political reality according to a normative paradigm independently justified leads to a dangerous form of political theorising which excuses, or even encourages, the use of force.

Someone might think that such examples only work when uncharitable characterisations of moralist theories are considered. If less naive theories were taken into account – the rebuttal would go – such shortcomings would easily be prevented. After all, the objector might claim, who would directly apply moral principles to political circumstances without making some apt amendments? However, such a criticism would misinterpret Geuss’s and Williams’s point. The point of their observations is that no matter the complexity of the theories considered (or of the amendments applied to them), unless the normative principles are subjected to a context-dependent analysis, the above-mentioned dangers will always constitute possible consequences of moralist theories. Indeed, the conditions upon which violence, conflict and disorder can be prevented can only be assessed by looking at the political context of application – at the real political relationships in place between the considered political actors – not abstractly. An adaptation of moralist theories to politics will not suffice. ¹⁶

There’s more. Besides fostering conflict, violence and disorder, moralist theories are accused by realists of being potentially oppressive normative structures. Such a criticism is most clearly articulated by Geuss, who explains that moralist theories, by separating the justification of political normativity from the analysis of actual political dynamics, are theoretical tools that can reinforce existing power structures or manipulate consensus. Geuss (2008) takes as an example A Theory of Justice, by John Rawls (1999), which he harshly criticises by pointing out, ‘One is immediately struck by the complete absence in it of any discussion of […] the basic issues of politics. The topic of “power”, in particular, is simply one he never explicitly discusses at all’ (90). The trouble with this mischaracterisation of politics, at least according to Geuss, is that the lack of a discussion about power and power dynamics makes Rawls unable to see some of the gravest problems which affect politics. Consequently, his theory not only offers no guide to counteracting these problems but actually reinforces them. ¹⁷ According to Geuss, Rawls is above all unable to recognise the fact of ideological power – the idea that the interests of some people might be systematically manipulated and oppressed – because the dominant systems of thought do not contain the conceptual elements the oppressed need to realise their disadvantage. ¹⁸ As a proof of such a shortcoming, Geuss mentions Rawls’s choice of constructing his theory upon the device of the veil of ignorance: The idea that principles of justice ought to be identified by agents unaware of any information about their concrete conditions of life in society, such as gender, class or job. In this way, Rawls crystallises ideologies because, by completely removing from sight concrete power dynamics, his theory does not contemplate any device able to tackle them (Geuss 2008, 86–89). In sum, by diverting our attention from reality to theory, moralist theories hinder existing power dynamics. This is why Geuss (2010, 42) recalls that moralist theories can be a powerful tool in the hands of those who hold political power. This is something Williams is well aware of too. Indeed, while discussing the criteria that should be employed to assess the legitimacy of political institutions, Williams specifies that those criteria need to be compatible with the Critical Theory principle, which states that ‘the acceptance of a justification does not count if the acceptance itself is produced by the coercive power which is supposedly being justified’ (2005, 6). By emphasising the role of the Critical Theory principle in the assessment of legitimacy, Williams, then, is pointing out that political power can indeed manipulate subjects’ beliefs and that this is a peril that should be avoided as much as possible.

The features that I have exposed so far must not be taken as features that every moralist theory shares. In fact, some of the alleged shortcomings pull in opposite directions: At times, moralist theories can justify disruptive political actions (as with the case of humanitarian interventions); at other times, they can foster an uncritical preservation of the status quo (as with the case of moralist theories which are consistent with the dominant normative ideology). The examples discussed were meant to give a sense of the possible shortcomings that Geuss and Williams envisage in moralism as a form of political thinking.

However, there is a common trait that unites all of them. Since moralism isolates the justification of political normativity from political reality, it is unable to clearly see what happens in the real world and to adequately address some of the most serious political troubles we face in actual circumstances, such as the containment of conflict, disorder, violence and oppression.

Political realism, then, presents itself as the remedy to the distortions of moralism. By making a bottom-up interpretation of the beliefs of actual political actors its source of normativity, political realism manages to eschew the practical dangers portrayed above. The normative relevance attributed to what makes sense to do from the point of view of actual political agents manages to ensure the conditions of a political order that subjects can find acceptable and – as far as possible – not coercively imposed, nor a source of conflictual relationships. And the critical interpretation of subjects’ beliefs makes possible an active examination of ideologies and power structures. In this way, realism manages to heal the dangerous divide between political theory and political reality created by moralism.

Is this line of defence of political realism convincing? A first doubt is that such realist considerations commit a fundamental error since they collapse theory onto reality, and yet, these are two separate domains: Theoretical inquiries are merely concerned with understanding normativity; they do not have the concrete consequences for the real world that realists fear. This is a point which Geuss and Williams would harshly object to. For Geuss (2016, 20) and Williams (2005, 72), there is a fundamental connection between theory and reality: Political theories are forms of political actions because they can affect the real world. Therefore, it makes perfect sense to reflect on their consequences.

Notwithstanding the first doubt, the major objection that might be raised against this (ethical) second line of defence is that something seems to be missing. As has been commonly said in opposition to realism, criticism by itself cannot work (Hurka 2009). If moralists are right about political reasoning, realists might well find the implications of the moralist theories unacceptable, but they should nonetheless respect those implications as true accounts of what ought to be done. Moralists, in other words, could be unmoved by the ethical sort of realist critique. If realists want their reflections to be convincing, they must advance some alternative positive idea of how to think about politics. Some theorists, in fact, went as far as to suggest that there are implicit assumptions in realists’ arguments that motivate their unease with moralism (see Erman and Möller 2015; Larmore 2013).

Maybe this second line of defence could be saved from the accusation of being inconclusive by making explicit the covert assumptions that allegedly ground realists’ rejection of moralism. Following this route, there seem to be two sorts of implicit assumptions that might support the realist case: A metatheoretical one and a principled one. The trouble is that neither of them seems convincing. In the first case, the dissatisfaction with moralism could be grounded on an implicit prior understanding of the nature of ethical knowledge. As we have seen, this is the route followed by Hall and Sleat. The reason why moralism ends up looking unacceptable in the cases considered above is that it contravenes what makes sense from the point of view of the agents involved, which is the only valid source of ethical knowledge. However, if that is the reason, the second sort of argument would collapse on the first sort, and I have already pointed out some of the latter’s potential weaknesses. Alternatively, instead of a prior metaethical background, there could be some principled reason why Geuss and Williams deem unacceptable the implications of moralist theories. For instance, they could believe that the first task of politics is to avoid oppression in all of its forms (both on a normative and on a practical level) and maintain that moralism fails to satisfy this fundamental principle, given the implications mentioned above. ¹⁹ But then realism would seem to fall into a form of tacit moralism since it could be charged with assuming a principle prior to politics – that is, a fact-insensitive assumption. Overall, this second line of argument in favour of realism seems to be doomed to fail since either its results are inconclusive or it relies on implicit theoretical assumptions that weaken it.

But is this a necessary conclusion? The realist has another way out, which I believe is the most interpretively accurate. The realist could deny that every critique must be accompanied by a constructive alternative. Indeed, the whole point of Geuss’s and Williams’s reflections is to argue for the priority of reality over theory in ethical reasoning. It is then entirely consistent that they make their point by simply showing political reality – that is, by exhibiting why moralism does not make sense to us as a guide for practical life. ²⁰ They do not need an alternative account of what valid ethical reasoning would consist in because this would be the scientistic way of thinking they are arguing against. Hence, there is a way to make the second line of defence viable. However, this line of defence comes at a price: It cannot convince the moralist of the preferability of political realism. Exactly because the realist defends his position by embracing criteria of validity which the moralist does not share, the realist argument cannot strike a chord with the moralist. The moralist and the realist would be like preachers of different faiths; convincing the moralist would be more a matter of conversion than of rational persuasion.

This is why I believe that the third argument in defence of political realism, which I am about to introduce, has to be regarded as better. As I explain shortly, the third argument has tools to convince the moralist of the preferability of political realism, and it also can overcome the weaknesses of the first (metaethical) defence.

5. The prudential argument: From responsibility in political action

As I briefly pointed out at the beginning of this article, the third argument is meant to overcome the weaknesses of the first two; however, to achieve this goal, it makes use of both of them. This means that, in order to justify realism, the third argument takes into account a wider set of realist topics because it refers to both its metaethical and ethical dimensions, while the first two (as I showed) can be constructed independently. For this reason, I take the third argument to be stronger also from an interpretive point of view.

I dub this final argument the prudential argument because it is centred around the idea that endorsing political realism as a practical guide for political action would be prudent. ‘Prudent’ does not here mean ‘convenient’ in a derogatory sense; rather, it means ‘responsible’, as in a justification which follows from taking politics seriously.

The prudential argument is divided into three steps. The first two steps come from the arguments previously introduced. To begin with, from the metaethical argument, we can deduce that there is deep uncertainty about the conditions of validity of our ethical judgements and, accordingly, about how to conduct normative thinking. In fact, we can loosen the conclusions of the first argument and make safe use of it without having to deal with its metaethical core. To recognise, as I do, that there is epistemic uncertainty about ethical claims is sufficient to recognise that realists raise good arguments against the possibility of acquiring true moral knowledge. The doubts raised against moral knowledge are certainly worth considering – so much so that the disagreement about these issues is still wide and is far from being resolved. ²¹ Hence, my first step consists in noting that Geuss and Williams raise some meaningful doubts about whether we can grasp some truth about ethical thinking and consequently about political thinking too.

The second step takes advantage of the ethical argument. From that argument, which comes from the ‘negative’ realist literature, we learned that normative theories are performative (i.e. they might have some impact on the real political world) and that the structure of moralist thought is in some specific sense worrisome. As we have just seen, moralist theories can lead to a variety of concrete damages, including that they might foster an inappropriate use of force, incite disruptive actions or frustrate our capacity to critically question our practices and choices (thereby hiding or perpetuating concrete injustices). As explained, all these shortcomings of moralism are due to its logical structure: Since moralism does not attribute normative relevance to political reality, it makes political reasoning insensitive to it. By pointing this out, as stated above, I do not mean to suggest that all moralist theories foster the sorts of consequences I mentioned. Rather, I more modestly mean to say that these shortcomings are potential consequences inherent to their very structure. So my second step merely consists in recalling Geuss’s and Williams’s suggestions about the potential practical dangers of moralism. But am I endorsing some sort of pre-political value judgement at this stage by defining such consequences as dangerous? This is a doubt I above considered as one possible objection to the ethical argument. Yet my answer is negative: For the purposes of the argument, I merely need to assume that we commonly see the above consequences as grave. Hence, mine is a contingent interpretive claim, not an evaluative judgement. Some of the implications of moralism, in fact, appear to threaten what we hold dearest in our everyday lives. In a realist perspective, the ‘securing of order, protection, safety, trust, and the conditions of cooperation’ (Williams 2005, 3) – which moralism appears to threaten – does not just describe the purposes political practices are meant to fulfil; rather it reflects some of the fundamental interests of human beings. As Williams points out, to remedy such threats represents the ‘first political question’ which is ‘“first” because solving it is the condition of solving, indeed posing, any others. It is not (unhappily) first in the sense that once solved, it never has to be solved again’ (Williams 2005, 3, emphasis added). That is, politics is the condition of possibility for the pursuit of any other good, so the fulfilment of our political interests has a special priority from a human point of view. So, I consider such interests as a commonsensical ‘platitude’ about politics – to use Williams’s words – and therefore, I take moralism to be regarded as dangerous from the point of view of actual political agents. ²²

We come now to the third and final step of the argument. Given the previous two steps, we can argue for rejecting moralism as a guide for political action, and adopting realism instead, for prudential reasons. This means that realism can be defended as the less risky form of practical reasoning in political circumstances. The choice of political realism as our guide for political practices best safeguards our political interests because it ensures the best cost/benefit balance among them. The merit of this general prudential claim emerges vividly when we consider moralist theories with the sorts of political consequences discussed in the ethical argument. In such cases, by choosing whether to favour moralist reforms or a realist normative guidance, we face a choice about what kind of interests to prioritise in the political domain: Either we might defend moralism because we believe (or hope) we found the truth about how to conduct politics and we are disposed to act according to that truth and accept the consequences of doing so or we might endorse realism because we believe that preserving our political interests and protecting them from the dangers of moralism ought to take precedence. Now, given the first two steps of the argument, it seems sensible to conclude that political realism is the most prudent approach. In fact, on balance, given the persistent disagreement about the ability of moralism (and, relatedly, of the diverse theories it gives rise to) to find the truth about political normativity, and given its perceived potential shortcomings, selecting moralism as a practical guide for political circumstances appears to be an excessively risky choice: Moralism causes severe harm, while the gain it promises is uncertain. So, in such cases, following moralism as a practical guide looks unwise since the benefits it promises are overridden by the serious costs incurred. Moralism, according to this line of argument, is a gamble which is hardly worthwhile. Given such considerations, prudential reasoning justifies assigning priority to preserving political interests over implementing a conception of moral truth in political circumstances. Such reasoning is reflected in the well-known realist motto that ‘politics should be prior to morality’, meaning that the creation of a stable and ordered political community is what we should first and foremost seek to achieve politically. Moreover, this means that actual politics should shape political normativity and define the space moral considerations can gain in political circumstances. Indeed – as we have already seen when discussing the ethical argument – given the persistence of moral disagreement, the conditions for the creation of a stable political community can only be found in actual politics, by looking at what actual political agents might find politically acceptable, not by implementing a moral outlook in the given political context. Realism, then, offers a way to protect the fundamental political interests of actual agents by grounding normativity on an interpretation and reflection about what makes sense to do politically from the point of view of actual political agents. ²³ On balance, prudential reasoning would conclude that, in this kind of circumstance, it would be politically wise to follow a realist methodology.

However, it might be replied, certainly not every moralist theory needs to have the sort of practical implications portrayed above. So can the prudential argument generalise its conclusion? After all, a possible objection might go, we could simply contrast the implications of moralist paradigms with the real world and assess whether it is wise to follow the moralist prescriptions. Yet – the realist would reply – this is precisely the sort of reasoning that makes the prudential argument a forceful general justification of the realist approach. Indeed, the prudential argument is precisely constructed around the idea that moralist prescriptions could, and should, be weighed against what actual agents find valuable. But the prudential argument also suggests that whenever the conclusions of moralist and realist analyses conflict, we should safeguard the interests of actual political actors, given the fragile status of moralist theories. Therefore, once we accept that normative prescriptions ought to be subjected to prudential analysis, either moralism becomes redundant (because it reaches the same conclusions as realism) or it becomes unwise (because it conflicts with what realism would counsel). ²⁴

Can this third justificatory strategy convince the moralist? Given its prudential nature, I believe that this third argument possesses the tools to persuade the moralist about the superiority of realism as a practical guide in political circumstances. In fact, even if the moralist regarded the pursuit of truth as a highly valuable goal, its uncertain status and the risks involved could give her a reason to recognise that realism is wiser – at least for the time being and on a practical level.

Still, the prudential argument might look puzzling and fragile for a number of reasons. Let me unfold it by addressing some of the doubts that might be raised against it. A first group of objections can be directed against the very idea of a prudential justification. For these critiques, the very project of prudentially justifying political realism is troublesome. Such doubt can take three different forms. The first one consists in claiming that a prudential justification is by definition unsound. Since the task of the theorist is to seek a better understanding of normative political theory, evaluating the practical costs of the theories is irrelevant, or, more correctly, focusing on the costs of theories might undermine the theorist’s research. In other words, a prudential justification commits a category error: It focuses on practical advantages, while the aim of a justification should be to identify what method delivers the best normative understanding of politics.

However, a realist would not accept this sort of objection. In order to appropriately understand the prudential argument, a fundamental point about realism must be understood. The reason Geuss and Williams would not regard a prudential justification as a category error is that, in their view, theorising qualifies as a political action and, crucially, every political action must be conducted responsibly. I have already mentioned that Geuss and Williams regard theories as performative constructs – that is, as products that can impact the real world. By being so characterised, however, theories must be subjected to the sort of evaluations that every political action must go through. Among those evaluations, judgements of responsibility are a fundamental but largely underappreciated component of realist thought. By speaking of responsibility in political action, Geuss and Williams refer to the traditional Weberian notion that political actions ought to be analysed and assessed also in light of their expected consequences (Geuss 2010, 30; Williams 2005, 12 and 72). To act responsibly in political circumstances means, in this sense, to carefully examine the concrete impact of the intended action and to weigh the costs and the risks of a course of action. Indeed, Max Weber’s (2004, 83–94) appeal for responsibility in political action is meant to remind us that we act in a real world and that political choices might have a severe impact on it; that is, it is a call to take politics seriously – the very same call, as I explained in section 1, that animates realist political theory. ²⁵ Taking politics seriously requires turning our attention from political theory to political reality – appreciating, and taking seriously into account, the fact that political actions have consequences for actual people and, therefore, must be assessed by carefully weighting the risks they might engender. Hence, political theorising, which constitutes an active intervention in the real world, ought to be evaluated with similar standards. It is, then, perfectly legitimate to assess theories in light of the costs of realising them. Realist analyses, as we have seen, are a warning against the concrete dangers of moralism. We could even go as far as to say that the category error is committed by those who unrealistically separate reality and theory when the question of which guidance to pursue in political circumstances is at hand.

Let us now turn to the second possible objection to employing prudential reasons as justifications. According to the objection, the prudential argument would encourage people to develop self-deceptive beliefs because it would justify acting as if realism were the approach which truly reflects what ethical life demands of us, while in fact, this approach is just the most convenient. The prudential justification, in other words, would go against the very foundation of the realist project, which demands a truthful understanding of our political world. However, the argument that I am proposing here merely claims that we should take the realist approach as a practical guide; the norms we endorse ought to be conceived of by following a realist methodology. The prudential argument explicitly declares that adopting political realism follows from assessing the risks of moralism and does not exclude the pursuit of parallel inquiries – metaethical and ethical – into what political practices truly demand of us.

A final objection to the prudential justification could be made about its contingent nature. Like every prudential argument, this kind of justification is merely contingent: It does not prove that realism is the form of practical reasoning that ought to govern our political life; it merely shows us that, given the present circumstances, it happens to be convenient. This is a fragile achievement which could be lost as soon as circumstances change. The realist would reply that this is certainly true, as prudential justification is necessarily contingent. However, the realist would deny that this weakens the argument, for two reasons. First, the justification is less contingent than it might appear: The facts mentioned in support of it are contingent but persistent. The facts upon which the prudential justification is built are the deep disagreement about the conditions of validity of metaethical and ethical theories, the practical consequences that moralism might give rise to and a shared interest in preserving some fundamental political values (such as protection, order, trust, stability and the conditions of cooperation). All of these facts are stubborn political facts which will hardly change in the foreseeable future. Second, it is not even clear why arguments that are grounded on contingent facts ought to be regarded with suspicion. In fact, this is exactly the point that the realist aims to make by arguing that we ought not to trace a sharp line of separation between theory and reality.

Leaving aside the possible objections to the idea of prudential justification, let us consider a final doubt regarding the argument, this one concerning the prudential calculus itself. Someone might object that the prudential assessment is wrongly conducted because the pursuit of truth is infinitely more valuable than any other goal. Therefore – the objection would go – the prudential analysis would not allow us to exclude moralism, since taking the risk of moralism would always constitute the best choice, given the infinite value of the potential gain. There are two answers the realist might give. First, she might, once again, recall that the prudential argument does not prevent us from searching for the truth about metaethical or ethical issues; political realism is just the best practical approach. Second, the realist would note that the prudential calculus aims to take into account the interests of the participants; hence, it does not matter whether the theorist would perform the calculus in a different manner. The relevant point of view is the enlarged one of those who take part in political practices. This is because, as I explained, the prudential argument originates from the necessity of responsibly considering what happens in the real world when moralism is adopted, and this requires abandoning the restricted point of view of the theorist in order to endorse the enlarged one of political participants. As Williams (2005) emphasises, political realism ‘speaks to humanity’ (59).

6. Conclusion: Going realist in order to take politics seriously

I want to close my article by saying a few more words about my argument and its relevance. As I specified in the introduction, my central goal was to outline a prudential justification for adopting the realist method in normative political theory. As I explained throughout, though, the prudential justification is just one possible justification of the realist method. I argued that there are at least two other arguments – what I dubbed the metaethical and the ethical arguments – which might back the realist case for adopting a bottom-up methodology. However, I claimed that the prudential argument is the only justificatory strategy which a moralist might find compelling. As I tried to show, the prudential argument has such potential because it employs means unfamiliar to the moralist: Whereas attempted theoretical justifications – the ones pursued in the metaethical and the ethical arguments – prove unpersuasive, a prudential justification has some chance at persuading because it maintains that the realist method is the wisest practical choice.

Is this the sole advantage of the prudential argument? I do not believe so. I regard the prudential argument as being also the most loyal to the spirit of realist political thinking. Indeed, I take it to be a fruit of the fundamental insight of political realism: That politics ought to be taken seriously, where taking politics seriously means placing at the core of normative reflection a focus on comprehending the reality of politics (Geuss 2008, 9–18; Williams 2005, 12–14). But we can read such a requirement in two different – though deeply connected – ways, according to the importance we assign to politics and reality. On the one hand, this requirement can be read to mean that political thinking ought to be structured around a deep understanding of political phenomena. On the other hand, this same requirement can be read as suggesting that political theorists ought to be aware that what they are reflecting about is, indeed, a part of reality. ²⁶ On this second reading, political theory ought to be conducted in light of the fact that politics is not merely a theoretical object of study; it is the world we inhabit, in flesh and blood. For political realism, once we come to a serious understanding of the reality of politics, we come to recognise that normative theorising and judgement ought to be conducted by seriously inquiring into the concrete implications that theories have and by assessing which consequences we are ready to accept. Political theories cannot merely satisfy a requirement of theoretical consistency; they need to be evaluated by paying attention to their expected functioning of real political practices and to the comparative costs and benefits of assuming them as practical guides. The prudential argument, being structured around a call for responsibility in political action, follows from the need to take politics seriously: It urges us to turn our gaze from the theoretical machinery to reflect on what would be the wisest practical strategy. Hence, even though all three arguments start from analyses of political practices, it is with the prudential argument that the realist invitation (and challenge) to take politics seriously becomes pivotal.