From Moral Principles to Political Judgments: The Case for Pragmatic Idealism

1 Introduction

Political judgments – whether to vote for A or B or to prefer policy x or y – are usually a combination of normative principles or intuitions with an appreciation of empirical facts. For example, I support candidate A (political judgment) because she seems to be the most able and willing (empirical judgment) to fight against poverty (normative consideration). Or I think that kerosene should be taxed (political judgment) because of its high contribution to global warming (empirical judgment), which is unfair to future generations (normative consideration).

When building political judgments aimed at guiding action here and now, one can hardly do otherwise than appreciate what is achievable and what the consequences of a given choice are likely to be given the knowledge we have about the world. ^[1] The interesting philosophical question is whether these kinds of consideration should weigh in the formulation of guiding normative principles themselves or should be considered apart. This is the core of the debate between partisans of idealism ^[2] and realism in political theory.

Three main views can be distinguished in this debate:

1. Idealists believe that abstract moral principles (potentially independent from social facts) are central to normative political theorizing.

2. Non-ideal theorists focus on how best to pursue normative ideals in the world as it is (taking into account relevant social facts and the issue of non-compliance).

3. Realists believe that political normativity should be kept distinct from abstract moral theorizing. In other words, political normativity should not be an application of abstract moral principles to non-ideal circumstances.

Among non-ideal theorists, it is then worth distinguishing between those who believe that there is a place for ideal theorizing in political theory, alongside non-ideal theorizing (Stemplowska 2008; Simmons 2010), and those who insist that ideal theorizing is pointless or misleading (Sen 2009).

At first sight, if a theorist is concerned with guiding political judgments, non-ideal or realist theorizing can seem more attractive, for they deliver normative conclusions that can be (more easily) applied in the existing, far-from-ideal world. I take this willingness to guide concrete political action to be the main reason for rejecting ideal theorizing. The aim of this article, however, is to argue that ideal theorizing might be considered valuable even by theorists moved by such a pragmatic concern (guiding political judgments) and to encourage the combination of ideal and non-ideal theorizing. In so doing, I do not aim to rebut political realism. People who strongly believe that political and moral normativity are of a different nature will not be convinced by the pragmatic idealism that I will advocate. The argument is rather targeted at people who are attracted by political realism or reject idealism (non-ideal theorists of the second category) because of the latter’s lack of action-guidingness.

In the second section, I consider the main charge levelled against political realism – the status quo bias. I argue that this cost of realism will no longer seem worth paying once one realizes that concrete and relevant political judgments can also be reached by starting from ideal principles, and that being committed to abstract ideals does not entail misunderstanding the nature of politics. I thus defend a perspective which can be labelled ‘pragmatic idealism’ and outline a four-step method for proceeding from abstract moral principles to concrete political judgments as a way of combining ideal and non-ideal considerations, taking on board several realist warnings (Section 3). For the sake of illustration, this method is applied to an example drawn from the ethics of migration. Drawing on this example, I then highlight the benefits of proceeding this way rather than disregarding ideal theorizing or adapting norms to reality (Section 4), before making a conclusion.

2 The Critical Value of Ideal Theorizing

In this paper, I use ‘pragmatism’ in the sense of a theoretical attitude concerned with action. The purpose of moral theorizing, from a pragmatic perspective so defined, is ultimately to guide action. In contrast with Elizabeth Anderson’s (2010) approach, for example, which consists in starting from concrete problems or situations of injustice, the pragmatic concern with guiding immediate action, in itself, does not settle the debate between partisans of ideal, non-ideal and realist theorizing. For in order to guide action, you can either think directly about principles applying in the world as it is (realist approach), or theorize based on concrete problems (Andersonian pragmatism), or think about ideal principles and their implications in the world as it is (what I will call ‘pragmatic idealism’). The first two methods are more direct and avoid second-best issues. ^[3] These are two reasons why they might appear more appealing. Hence, the choice for an indirect and more hazardous method must be justified.

The main justification for pragmatic idealism, defined as the combination of high (or even detached) ideals with a willingness to guide action, is to minimize the risks of a conservative bias in the formulation of our moral standards. Their bias in favour of the status quo is probably the main charge that has been levelled against realist approaches to politics, as appears clearly in the work of its critics (Carens 1996; Cohen 1994, 2011), but also its defenders, who have made great efforts to argue either that the status quo is not morally arbitrary (Sangiovanni 2008, p. 161) or that idealism can also be biased in its favour (Rossi 2016, 2019), or that realism need not be conservative (although many of its incarnations are) (Finlayson 2017; Rossi 2019).

One way moral standards can be biased in favour of the status quo is by being adapted to what seems achievable or feasible in the world as it is. By so doing, we risk legitimizing a state of affairs that might not be immutable. Consider the two following examples.

In debates about the appropriate standard of distributive justice, some might find demanding egalitarian principles unattractive because they are unlikely to be realized or even supported by a majority of citizens. Taking hostility to taxes as a fact, they might be more attracted by a sufficientarian principle requiring only bringing everyone above some threshold of material well-being – a demand well-aligned with a broadly accepted human right to a decent standard of living. Or they might argue that nobody should pay more than half of his or her total income in tax, as recommended by the British Commission on Social Justice, in which Bernard Williams, a well-known realist, took part in 1993–1994 (see Cohen 1994; Barry 2005, pp. 8f.). To be sure, there can be other reasons to be sufficientarian or to argue in favour of a cap on taxation. Yet idealists have good reasons to think that unless these norms are presented as non-ideal or adapted to superable constraints, the upshot of adopting a realist perspective is excessively conservative in that it legitimizes (some degree of) hostility to taxation.

Similarly, some people might be tempted to reject an open-borders ideal because it does not seem realistic. In a world of widespread xenophobia or at least in a world in which many people give more importance to the preservation of their national culture than to transnational solidarity, opening borders, following this perspective, would lead to social disaster. Again, there might be other reasonable reasons to oppose an open-borders regime, but defending a realist approach to this question without reference to more utopian ideals could be seen as legitimizing and ‘naturalizing’ an attitude – national preference – which is politically constructed and thus de-constructible (Axelsen 2013).

Thus, for people who believe that moral norms should not take our moral weaknesses – such as our reluctance to give up part of our gross income or our tendency to favour our in-group – as insuperable, the realist approach might look insufficiently critical.

But why do some realists accept this conservative bias? The main reason is that they see ideal approaches as either irrelevant or dangerous. They can see them as irrelevant because politics is not comparable to a discussion in an ethics seminar. Whatever the fine-grained arguments that philosophers can provide, political agents are moved by other considerations. Politics is about power and conflict, not moral principles (Mouffe 2005; Galston 2010, p. 395). Thus, ideal approaches misunderstand the nature of politics (Sleat 2016; Jubb 2017, p. 115), or so they argue. For the same reason, ideal approaches can also be seen as dangerous (Galston 2010, p. 401; Shapiro 2016), because by abstracting away from real politics, they run the risk of making inappropriate political prescriptions that turn a utopia into a nightmare – for example an ideal of global democracy into global tyranny. On this point, realists have history on their side: utopias have often led to dystopias in practice.

Idealists, however, should not feel their arguments have been rebutted by these criticisms. It might be true that many ideal theorists fail to appreciate political dynamics and therefore make irrelevant prescriptions. But it does not need to be so. An idealist might believe and argue that a just society should be completely different from ours and yet be hopeless about the prospects of realizing this ideal (Estlund 2014), precisely because she recognizes that politics is not responsive (enough) to moral arguments. Yet this raises a challenge for idealists who are not satisfied with a hopeless aspirational theory and who have the pragmatic aim to guide action. They will have to pay particular attention to the reality of politics when formulating political prescriptions based on their moral principles. In light of historical experiences, they should be very cautious, and well aware of second-best issues to minimize the dangerous effects of utopianism. It is the merit of the realist critique to point this out.

Thus, the realist critique is very important for pragmatic idealists. But if there is a way to take this critique on board while avoiding conservatism, as I shall argue, such a solution might be more attractive for those who see the status quo as deeply unjust.

Now, as I mentioned earlier, realist approaches to politics are not necessarily conservative (Finlayson 2017). In particular, when they take the form of ideology critique (Sangiovanni 2008, pp. 161f.; Prinz and Rossi 2017), they have the ability or even the mission to challenge the status quo. What is more, idealist approaches themselves are not immune to status quo biases (Rossi 2016, 2019). They can reflect vested interests or be the product of a conservative ideology. The question, then, is whether one of the two approaches is more likely to be biased than the other.

Rossi (2019) claims that a specific form of realism – as ‘ideology critique’ or as ‘genealogy’ – is more radical and less biased in favour of the status quo than both ideal and non-ideal approaches to normative political theory. The reason is that it sets for itself the mission to scrutinise the sources of ideologies and their genealogy as well as to identify their biases, which an ideal approach does not necessarily do.

Pace Rossi, I do not see why that form of realism should be considered as incompatible with ideal theory. It seems to me again that the realist critique can be taken on board by idealists – and should be if they want to minimize the risks of biases or conservatism. As we shall see in the next section, ideology critique could be a dimension of the search for adequate fundamental principles.

Yet the issue considered in this paper is not how to formulate fundamental principles; it is about translating these fundamental principles into political judgments. At that stage, ideology critique is not relevant anymore, and when building political judgments, we run a higher risk of conservatism by taking some properties of a political order – such as class struggle, bounded generosity, and moral disagreement – as insuperable or by denying that the status quo is morally arbitrary (Sangiovanni 2008, p. 161). Hence, unless purely normative theorizing is rejected for other reasons, which are beyond the scope of this paper, ^[4] pragmatic idealism (including ideology critique) might appear more attractive for those who care about politically relevant theorizing yet want to avoid any conservative bias.

Should we nonetheless make a distinction between things that can change and others that will not (Galston 2010, pp. 401, 409)? For example, nationalism might be historically situated, constructed, and superable, but some degree of egoism will always inhabit human nature, hence morality should take into account the impossibility of altruism or be aligned with enlightened self-interest.

This sounds to me like an unnecessary pessimistic assumption. It is pessimistic in assuming that humans will always primarily be moved by their own interests. And it is unnecessary in the sense that we do not need to make such an assumption. As long as we keep sight of reality constraints (including limited generosity) when devising practical judgments aimed at guiding direct action, we can hold high moral standards that do not currently seem achievable. The value of doing so is to show the ambition to make humanity better if that proves possible, and by that move to avoid legitimizing current human weaknesses.

Such reasoning applies both to individual morality and to standards of collective organization. Thus, I see no reason to refrain from considering, for example, that a collective organization based on voluntary contributions would be more desirable, from a moral point of view, than one based on market rewards. Such an affirmation does not by itself entail that we should get rid of markets or turn to a communist society from one day to the other. Believing this amounts to misunderstanding the role of moral standards in ideal theorizing. Their aim is not necessarily to guide action in a direct way; they can also offer a critical look at existing institutions. This is, for example, what Joseph Carens (1981) did when imagining a market economy with equal post-tax wages and the changes in our moral motivations that would be required for such scheme to be viable. He was clearly not offering a blueprint for optimal taxation, but he was inviting collective critical reflection on individual motivations in market economies – the kind of reflection that was badly missing in the Commission for Social Justice mentioned earlier. As long as we maintain a clear distinction between what is morally desirable or just and what we ought to do or what the state ought to do ^[5] (Cohen 2011, pp. 225f.; Estlund 2014, p. 121), high moral standards are useful – as critical tools.

These brief considerations are obviously not meant to make a decisive case against realist approaches to politics that are incompatible with idealism. Again, those who reject the distinction between normative and descriptive theorizing or the possibility or desirability of decontextualized normativity will not be convinced. The aim of this section was more modestly to show that for people who are concerned with the action-guidingness and political relevance of political theory, political realism is not the only available option – and it might even appear less desirable than pragmatic idealism given the critical value of high normative standards submitted to some sort of ideology critique (more on which later). Broadly similar arguments have already been made in critiques of political realism. What is lacking in the literature, though, is an explanation of the way ideal approaches can lead to concrete political judgments while incorporating relevant realist warnings.

3 From Morality to Politics

How do we move from high moral standards to concrete and plausible political judgments? By raising this question, I do not want to suggest that normative thinking always goes from abstract principles to concrete judgments. The way many people think – which was systematized by Rawls’ method of reflective equilibrium – is by travelling back and forth between principles and intuitions or (considered) judgments about cases, whatever the starting point. Yet unless we always theorize by starting from concrete cases (which might threaten moral consistency), there will often be instances where we will have to move from moral principles to political judgments.

What is implicit in the question is a distinction not only between principles and judgments, but also between morality and politics. Distinguishing between both realms, however, is consistent with believing that political philosophy involves the application of moral principles to politics – which most realists deny (Williams 2005; Sangiovanni 2008; Galston 2010). My view – shared by others (see Maynard and Worsnip 2018) – is that political judgments are more encompassing than moral ones. They need to take into account not only moral considerations, but also many factual assumptions about agents, markets, and social dynamics. In this regard, political normativity differs from strictly moral normativity, but it nonetheless incorporates guidance from the latter.

I would now like to offer a method suited to political normativity so conceived. If we want to maintain high moral standards detached from reality constraints and yet reach political judgments, we need to proceed through several steps and keep their separation in mind even though, in practice, we might be tempted to lump them together. Before going through these steps, note that political judgments can bear on candidates to elections or directly on policies. When judging candidates, some personality traits like fair-mindedness, sincerity, or leadership certainly matter (see Dovi 2007), but we should also take into account the values they affirm and their favoured policies. Because competing candidates can affirm broadly similar values, it is often their policy proposals that will be the salient and interesting criterion of differentiation. Hence, I will focus here on judgments about particular policies. ^[6]

To illustrate the method with a concrete application, let us imagine a strong cosmopolitan person reflecting upon the right immigration policy to defend in her political context – an affluent society facing immigration pressure and currently very selective about who is allowed in. One could define strong cosmopolitanism as the combination of a) the rejection of any special moral obligations to compatriots and b) the recognition of the morally arbitrary character of national borders. From this perspective, there is at least a prima facie preference for a world without borders (Carens 2013). ^[7] Selecting the best immigration policy, however, requires more careful reflection. Opening borders might be too straightforward an application of the moral principle.

4 Clarifying Commitments and Disagreements

Once these four steps have been followed carefully, one can formulate a political judgment inspired by a moral principle or principles. The judgment, however, does not exactly derive from the principles. People could disagree at each step, for different reasons, and reach a very different judgment while starting from the same moral basis.

One reason is that empirical evidence is often contested and can result in different and conflicting predictive models. Additionally, those who believe in value pluralism must accept the possibility that different people sharing the same values will weigh them differently and disagree at the policy level for that reason. However, one might expect that two people starting from the same moral premises, ordering values in a similar way, and facing the same empirical evidence should reach identical political judgments.

Yet another possible source of reasonable disagreement is that the choice of a given path towards an ideal involves uncertain strategic decisions. The socialist movement, for example, has always been divided on the best political strategy: reforming or resisting capitalism. What would have happened if social democrats had refused the welfare state compromise, which had the effect of depoliticizing the working class (see Przeworski 1985)? These kinds of questions can be settled neither by philosophical speculation, nor by empirical evidence. Thus, two people starting from the same moral premises, ordering values in a similar way, and facing the same empirical evidence might still disagree about strategy and reach different political judgments. Nevertheless, one important benefit of this four-step method is to make it easier to locate political-normative disagreements, which matters for both academic debates and political deliberations.

The example from the ethics of migration highlights the fact that political disagreements can have an important variety of sources: opposed conceptions of justice; different ranking of values; competing assumptions about economic constraints, social dynamics, or electoral possibilities. Yet what frequently happens in political and academic debates is that people think they are disagreeing although they are not addressing the same question. For example, strong cosmopolitans usually refuse to consider selective immigration as just. This, however, does not mean that they would object to some form of it in practice. They will either disagree or agree with more pragmatic recommendations about immigration depending on whether the question is ‘What would justice require, ideally?’ or ‘What should we do now?’

Similarly, carefully distinguishing fundamental principles from rules of regulation, following Cohen’s (2008) suggestion, reduces the likeliness of seeing fundamental principles attacked on the basis of their alleged practical implications (as in Anderson 1999). Fundamental principles are distinct from political prescriptions. ^[15] We need to take different steps to arrive at political judgments, and along the way other considerations might move us further from what could seem to be the direct implication of a principle (such as open borders for strong cosmopolitanism). Hence, carefully distinguishing the different steps of a normative reasoning is important to clarifying commitments and disagreements, and thereby to reducing misunderstandings and fruitless debates.

Another conclusion is that the reasoning that brings a person to defend a given policy is an important matter, both philosophically and politically. Consider once again the example of differentiated social rights for immigrants. The same policy could be defended by people on the far right and people on the far left (although the former would probably want to reduce the number of immigrants as well as restrict their rights). The difference between the two positions, on this issue, might not be the content of the policy, but its justification. On one side, the justification is that compatriots matter more, morally, than foreigners, who are therefore not entitled to the same rights and opportunities; on the other side, it is that electoral and budgetary constraints create a trade-off whose least undesirable solution might be to temporarily reduce immigrant access to social protection.

Philosophically, the difference is clear: the two positions derive from opposite understandings of the scope of justice. And as Christopher Bertram (2019, p. 7) puts it, ‘it distorts our moral understanding, specifically, our understanding of justice, to represent a pragmatic compromise with unjust attitudes [the unwillingness to pay more taxes] as exemplifying what justice itself demands.’ Thus, the advantage of a pragmatic-idealist approach is that it makes the distinction easier: global freedom of movement combined with equal rights would be more just, but under specific circumstances, a more pragmatic compromise might be justified.

Politically, distinguishing the two positions matters as well, for only the strong cosmopolitan position engenders the moral discomfort that will motivate attempts to overcome or reduce the reality constraints on the pursuit of the ideal. The strong cosmopolitan defending differentiated rights will not stick to this policy proposal. She will also likely argue in favour of higher taxes, a higher budget for social protection, reduced international tax competition, global redistributions or other policies aimed at reducing the sources of immigration, a better social and economic integration of immigrants, etc. ^[16] This is why it is important to make our fundamental moral beliefs stand out clearly, and why it matters to keep high normative ideals: to motivate the relentless efforts needed to improve our world despite reality constraints. Incidentally, maintaining a connection between concrete political engagement and high normative ideals might be especially welcome in a context of policy convergence between left and right if we want to maintain a specific ideological identity and distinguish it from political opponents (see Cohen 1994; Barry 2005).

Finally, clearly picturing the different steps involved in political-normative thinking also highlights the hypothetical nature of political judgments. They are provisional conclusions based on moral premises and a series of empirical assumptions. If some of these empirical assumptions change, we should reconsider the judgment. This hypothetical status, it seems to me, appears less clearly when normative standards are adapted to reality constraints as is the case with realist approaches.

5 Conclusion

In this article, I have argued against adapting our moral standards to what is perceived as feasible or what is likely to bring about desirable consequences under existing circumstances. The value of high moral standards is to make the moral flaws of our world appear more clearly and to critically assess existing institutions and attitudes. High standards allow us to avoid legitimizing existing injustices and not to fall in the trap of adaptive preferences (Cohen 1991).

I have also argued that one mission of political philosophy or theory should be to help build political judgments, guiding concrete actions in the world as it is. Therefore, it is not enough to defend abstract – and possibly hopeless – principles of justice. We need to be able to turn these into concrete political judgments by relying on social sciences and with an appropriate dose of realism about politics. The method appropriate to political-normative thinking that I advocate to this end comprises four steps:

1. clarification and ordering of our guiding moral principles and values

2. anticipation of the effects of competing existing policies

3. consideration of new alternative policies and their effects

4. considerations of political achievability

Although most theorists are likely to proceed through all these steps, be it in a disordered way, there is a widespread tendency in realist theorizing to mix some of these or to alter the guiding moral principles in light of practical considerations. I believe that this is a mistake ^[17] and that two main gains can result from using the method of reasoning I advocate instead.

First, as I have argued in the third section, it is likely to clarify (and possibly reduce) philosophical and political disagreements by making it easier to identify whether people disagree on policies, fundamental principles, or intermediate norms such as rules of regulation.

Second, this method permits a more critical role for political philosophy. It does not have to choose between practical relevance and criticism or between pragmatism and idealism. It is possible and desirable for a theory to maintain high moral standards pushing humanity upwards and yet to play a useful social role in guiding political action in the world as it is – provided that moral standards are carefully scrutinized from an ideological analysis perspective and then articulated with empirical considerations along the lines suggested above. The level of ideality or realism of our discourses will then often depend on the audience we are addressing and the aims of our communicative act (Carens 2013, pp. 288f.; Valentini 2012, p. 660).

Thus, if the main values of political realism are its political relevance and its respect for the particular nature of politics, the cost of realism in terms of status quo bias is not worth paying, because, as I have tried to show, idealism is not incompatible with relevance and does not forbid taking into account the peculiar ‘nature’ of the political (without assuming it to be immutable) as a game usually based on conflict, power, deception, and low moral motivation. Ideal principles can abstract away from this reality, but political judgments based on these principles will have to take this ‘nature’ into account if they are to be relevant.

Using this method also involves costs, to be sure. First, we certainly lose time in laying out our reasoning and explaining how we arrived at a specific political judgment while starting from a very remote principle. It seems to me however that all theorists will accept this cost. Second, and more importantly, we lose the immediate perception of a connection between the normative principles we recognize and the policies we advocate, possibly creating an unpleasant degree of cognitive dissonance (as is well illustrated by the immigration case discussed in Section 3). Yet the noblest way of reducing this dissonance is to bring the world closer to our ideal, and not the other way around.