Should nonideal theory rely on ideal theory? Lessons from the Frankfurt School

Abstract

While ideal theory tells us “what a perfectly just society would be like” (Rawls, 1971, p. 8), our current social world is far from perfectly just, and we clearly want to know how to orient ourselves and act in these less than perfectly just, or unjust, circumstances. This is why many political philosophers¹ today agree that what is needed is nonideal theory. There is, however, disagreement on one key issue. While some—most notably Charles W. Mills and Amartya Sen—have argued that nonideal theory does not need to build on ideal theory and should do without it, others have pointed out ways in which ideal theory may still be necessary for doing nonideal theory, and that ideal theory should not be dismissed so easily for the project of nonideal theory. The central question, then, is whether nonideal theory should rely on ideal theory.

In what follows, I would like to take up this question and propose an answer to it. I will do so by considering a more specific question. Taking the Frankfurt School as my point of departure, I would like to examine whether nonideal theory, when guided by ideal theory, can help members of the social world to understand injustices and thus contribute to their self-reflection. Advocates of nonideal theory should be concerned about the ability of nonideal theory to contribute to self-reflection, because whether it can successfully inform collective action to overcome injustice depends on its ability to contribute to self-reflection. In other words, what is at stake is nothing less than the practical relevance of political philosophy. As I will argue, however, reliance on ideal theory renders the ability of nonideal theory to contribute to self-reflection uncertain and therefore a matter of sheer luck. As a result, the ability of nonideal theory to successfully guide action is in constant doubt. I will therefore conclude by suggesting that if political philosophy wants to be practically relevant, nonideal theory should not rely on ideal theory.

My argument will involve four steps. First, I will explain my understanding of ideal and nonideal theory (Section 1), before addressing the debate about the role of ideal theory for nonideal theory (Section 2). I will then turn to the Frankfurt School and outline its basic approach to injustice and social problems in general (Section 3). Finally, I will examine what lessons advocates of nonideal theory can learn from the Frankfurt School, which is where the main action of this paper takes place (Section 4). I will conclude with a summary of the argument.

How best to define “ideal theory” and “nonideal theory” has been the subject of some debate (see, e.g., Mills, 2005; Robeyns, 2008; Stemplowska, 2008; Simmons, 2010; Valentini, 2012). As I understand it, ideal theory is concerned with finding out when the social world, or parts of the social world, are organized in such a way that they can be regarded as fully just. Nonideal theory, on the other hand, is concerned with injustice.

This understanding of ideal and nonideal theory goes back to John Rawls' early work. Rawls introduces ideal theory as the study of “what a perfectly just society would be like” (Rawls, 1971, p. 8). In other words, ideal theory, according to Rawls, attempts to identify the principles of justice that a society would have to realize in order to be considered perfectly or fully just (see also Simmons, 2010, pp. 7–10). It is important to note that in order to identify the principles of perfect or full justice, Rawls introduces the assumption of strict compliance, which states that the members of society act in accordance with the principles of justice proposed by ideal theory.² The function of this assumption is to enable the ideal theorist to develop a clear understanding of what it would mean if certain principles of justice were realized. But it would be misleading to equate ideal theory with theorizing that assumes strict compliance,³ precisely because that would be to miss the whole point of ideal theory, which is to work out when society is perfectly or fully just.⁴

Of nonideal theory, Rawls writes that, unlike ideal theory, it seeks to answer the question “how we are to deal with injustice” (Rawls, 1971, p. 8).⁵ If nonideal theory is understood in this way, then it has two parts. The first part is to provide an evaluation of the present social world—that is, nonideal theory must examine how the social world, or parts of the social world, are organized and identify where injustice exists. The second part is to answer the question of what to do in the face of injustice: nonideal theory must propose collective courses of action that are an adequate response to the injustice in question.⁶

The understanding of ideal and nonideal theory that I have just introduced is in terms of their respective aims. I would like to make two points about this understanding. First, contrary to what one might think, the choice between ideal and nonideal theory is not an either/or matter. Rather, one and the same work of political philosophy can in principle do both ideal and nonideal theory, if it seeks to find out what a perfectly just society would look like and also deals with injustice. Second, and more importantly for the argument to be developed here, although for Rawls ideal and nonideal theory are clearly related in that ideal theory provides “the basis” for nonideal theory (see Rawls, 1971, p. 9) and nonideal theory builds on ideal theory, it should be noted that nothing I have said above implies that they must be related in this way.⁷ That is, it is not part of the understanding of nonideal theory presented here that ideal theory is a necessary precursor to nonideal theory.

Many political philosophers agree that nonideal theory is important.⁸ Charles W. Mills, for example, argues that political philosophers should try to “guide our actions and make ourselves better people and the world a better place,” and therefore address “the actual workings of injustice” (Mills, 2005, p. 170). Amartya Sen claims that the focus of political philosophers should be on “advancing justice in a society (or in the world)”, that is, on “reducing manifest injustices” (Sen, 2006, p. 217). Ingrid Robeyns writes that “much more work needs to be done on developing nonideal theory” (Robeyns, 2008, p. 362). And Adam Swift agrees that “work that will give us concrete guidance in the particular contexts and circumstances that we actually face” is “indeed a worthwhile enterprise” (Swift, 2008, p. 368).⁹ But so far there is no consensus on whether ideal theory still has a role to play for nonideal theory. Should nonideal theory rely on ideal theory or not? In the following, I would like to take a look at the debate so far.

Looking at the debate so far, one might get the impression that the advocates of engaging in nonideal theory with recourse to ideal theory have the edge over the likes of Mills and Sen. So should nonideal theory rely on ideal theory? I believe that progress on this question can be made by turning to another tradition of social and political thought, namely the Frankfurt School. The Frankfurt School is, of course, internally heterogeneous,¹⁷ but its members share a number of commitments, one of which is of particular interest for my purposes, namely a commitment to the principle of methodological negativism.¹⁸ My aim in this section is simply to explain the content and import of this principle.

MN is evident in Max Horkheimer's talk of the “present distress” (Horkheimer, 2002 [1972/1937], p. 217), or the misery of the present, as the starting point for social criticism. It is also evident in many texts by Theodor W. Adorno, who repeatedly grounds the need for social criticism in experiences of suffering, perhaps most prominently in his collection of aphorisms Minima Moralia (1978 [1951]). Axel Honneth, a contemporary representative of the Frankfurt School, is generally seen as inheriting the commitment to MN from Horkheimer and Adorno, for he is clear that “social discontent … constitut[es] the necessary reference point for empirically justifying critique” (Honneth, 2003, p. 126; see also Honneth, 1994), as does Rahel Jaeggi, who notes that “criticism is possible only where what is criticized, the object of criticism, has succumbed to a crisis of itself” (Jaeggi, 2018 [2014], p. 192). Even Habermas, who is often read as abandoning MN, remains partly committed to it, as his discussion of the thesis of the colonization of the lifeworld shows.¹⁹ What exactly does MN say?

First of all, it is important to be clear about what a social problem is. There are several things to keep in mind. First of all, a social problem is a state of affairs that is undesirable. Social problems include, but are not limited to, injustice.²⁰ Examples of social problems are states of affairs in which individuals cannot really live (Adorno, 1978, p. 15), in which they have difficulty making sense of their lives (Habermas, 1987 [1981], p. 323), in which they are alienated (Jaeggi, 2014 [2005]), or in which they are unjustly denied respect and thus are unable to participate fully in society (Honneth, 2003). Moreover, a social problem is a state of affairs that is socially caused, that is, a state of affairs that arises and persists because of how the social world is configured. Finally, the social problems that social criticism is concerned with are also “social” in the sense of being common or collective. For example, the practices of a small religious group that oppress some of its members are not the kind of social problem that social criticism deals with.²¹

According to MN, however, social criticism does not start directly from social problems. Rather, it starts—or should start—from symptoms of social problems that can be found in the social world. This has three important implications. First, the symptoms of social problems exist prior to and independently of the social critic. Just as someone may have symptoms of a health-related problem regardless of whether a doctor is there to diagnose it, so there may be symptoms of social problems regardless of whether a social critic is there to ascertain whether there is an underlying social problem.²² So the symptoms of social problems are out there independently of the observer. Furthermore, just as the symptoms of health-related problems are often noticed by those affected by them, the symptoms of the social problems that Frankfurt School members write about are often noticed or registered at some level by at least some members of society affected by those problems. Finally, even if such symptoms are registered at some level, the underlying social problems are not obvious. Rather, they require careful diagnosis. This diagnosis will be at once descriptive, tracing the relevant causal relationships, and evaluative or normative, assessing whether there is indeed a problem that, by virtue of being a problem, should be addressed. Making such a diagnosis is what social criticism is about. An important consequence of this is that the symptoms of social problems both provide the impetus for social criticism and act as a constraint on the content of the diagnosis that is provided.

To conclude this brief excursus on the Frankfurt School, I would like to consider why one would subscribe to MN. What is the Frankfurt School's rationale for subscribing to MN? As Horkheimer explains, “the self-knowledge of present-day man is … a critical theory of society …, a theory dominated at every turn by a concern for reasonable conditions of life” (Horkheimer, 2002, pp. 198–99; my italics; see also Honneth, 1994, p. 256). As Habermas aptly puts it, the aim is to bring about “self-reflection” (Habermas, 1966 [1965], p. 294; my italics) in the members of society and to guide action on that basis.²³ Thus, by starting from the symptoms of social problems that can be found in the social world, the Frankfurt School seeks to help members of the social world to reflect on and understand the problems they face. By contributing to their self-reflection, Frankfurt School thinkers also hope to help them to address these problems and change the social world for the better.

Against the background of the brief excursus on the Frankfurt School, I would like to return to the question of whether nonideal theory should rely on ideal theory. Are there lessons to be learned from the Frankfurt School on this question? As we saw in the second section, an important argument in favor of taking ideal theory as the starting point for nonideal theory concerns the first part of nonideal theory. The substance of that argument was that nonideal theory should draw upon ideal theory because ideal theory is necessary to identify important but less obvious injustices. Interestingly, as we saw in the last section, the thinkers of the Frankfurt School seem to agree that injustices, or social problems in general, are mostly not there for everyone to see; at the same time, however, they also seem to think that it is possible to identify such injustices or social problems without using ideal theory. In what follows, therefore, I would like to shift the focus somewhat and ask what the implications of relying on ideal theory to identify injustice are for nonideal theory. In particular, inspired by the Frankfurt School and the idea of self-reflection, I want to examine how relying on ideal theory to identify injustices affects the ability of nonideal theory to contribute to the self-reflection of members of the social world, that is, its ability to help members of the social world to understand social problems or, more specifically, the injustices they face. I believe that by answering this specific question, we can also answer the central question of whether nonideal theory should have recourse to ideal theory.

Let me summarize the argument presented in this paper. The central question has been whether nonideal theory should rely on ideal theory. The answer to this question is no: nonideal theory should not rely on ideal theory. My argument was inspired by ideas associated with the Frankfurt School and focused on how relying on ideal theory to identify injustices affects the ability of nonideal theory to contribute to self-reflection, that is, to help members of the social world understand the injustices they face. I have argued that reliance on ideal theory makes the ability of nonideal theory to contribute to self-reflection uncertain and therefore a matter of sheer luck. I have also argued that this should worry nonideal theorists because the ability of nonideal theory to successfully guide action to overcome injustice depends on its ability to contribute to self-reflection. Since the whole point of nonideal theory is to make political philosophy practically relevant, nonideal theory should not rely on ideal theory.

In conclusion, I would like to make two points. First, the argument I have presented here is mainly negative. It is an argument to the effect that nonideal theory should not be done by recourse to ideal theory. While I think it is important to understand why nonideal theory should not rely on ideal theory, I also think that more can be said. In developing my argument, I have not only assumed that it is possible to do nonideal theory without relying on ideal theory, and in particular that it is possible to identify even less obvious injustices or social problems in general without relying on ideal theory, but I have also provided some details about the Frankfurt School's approach to diagnosing injustices or social problems in general. Thus, the work of thinkers associated with the Frankfurt School can be a fruitful resource when developing a positive vision of how nonideal theory could be done without relying on ideal theory to identify injustices or social problems in general. Building on this, it is also worth pointing out that there are interesting parallels between the Frankfurt School and other critical approaches often grouped under the heading of “nonideal theory.” Mills' own approach is one of these (see Mills, 2005, but also Anderson, 2010; Haslanger, 2021). These approaches are all committed to the idea that nonideal theory should start from injustices or social problems in the existing social world, but that identifying such injustices or social problems requires work. They thus share considerable common ground with the Frankfurt School. In light of this, the dialog that has recently begun between the Frankfurt School and these approaches (see, e.g., Haslanger, 2020) is intriguing and promises to show ways of making political philosophy more practically relevant in the future.

There is no conflict of interest to report.