Abstract
The term “social world” is increasingly familiar in philosophy and political theory. Rawls uses it quite often, especially in his later works. But there has been little explicit discussion of the term and the idea of social worlds. My aim in this paper is to show that political philosophers, Rawlsian or not, should think seriously about social worlds and the roles these things play and ought to play in their work. The idea of social worlds can help political philosophers think about what they do in new and fruitful ways and enrich debates about the roles, aims, and methodology of political philosophy.
I begin by analyzing Rawls’s uses of “social world.” I then propose a broadly Rawlsian conception of social worlds as logically possible closed networks of social relations between agents. Next, I put this conception to work, arguing that the idea of navigating the landscape of social worlds can help us better understand the four apparently disparate roles of political philosophy that Rawls presents. Moving beyond Rawls interpretation, I use the idea of social worlds to develop an analogy and distinction between world-oriented and principle-oriented approaches to political philosophy. While principle-oriented approaches grant centrality and importance to engagement with principles of justice, legitimacy, or other political concepts, world-oriented approaches grant centrality and importance to engagement with social worlds. I propose two examples of world-oriented approaches, political philosophy as navigation and political philosophy as world-building, and argue that they are viable and worthy of further consideration.
Introduction
The term “social world” is increasingly familiar in philosophy and political theory. A Google search of it with “philosophy” generates over two million results, as does one with “politics.” It is commonly used in sociology, sometimes as a technical term. One can also find it in many book titles across the social sciences and humanities. 1
Rawls uses “social world” quite often, especially in his later works. But there has been little discussion of Rawls’s use of “social world” and of its use in political philosophy more generally. 2 What he meant by the term is not merely a question of historical interest. Rawls’s work is foundational to analytic political philosophy, and it would be unsurprising if his readers were drawn into his ways of writing and thinking. 3 Indeed, in recent analytic political philosophy, the term “social world” has started to play a larger role. Gaus (2016) appeals to Rawls’s usage of the term when modeling and critiquing ideal theories of justice (39–56, 177–79). 4 The term also appears, undefined, in recent defenses of ideal theory: in Estlund (2020, 4, 114, 134, 212) and in the first sentence of Laurence (2021).
My aim in this paper is to show that political philosophers, Rawlsian or not, should think seriously about social worlds and the roles these things play and ought to play in their work. I contend that the idea of social worlds can help political philosophers think about what they do in new and fruitful ways. Past the fiftieth anniversaries of A Theory of Justice and Political Theory, debates about the aims, methodology, and roles of political philosophy are growing. I suggest that further reflection on the idea of social worlds can enrich these debates.
After beginning by analyzing Rawls’s uses of “social world” , I introduce a conception of social worlds that is Rawlsian in spirit. On this conception, a social world is a logically possible closed network of social relations between two or more agents. Next, I put this conception to work. I argue that the idea of navigating the landscape of social worlds can help us better understand the four apparently disparate roles of political philosophy that Rawls presents. Moving beyond Rawls interpretation, I introduce an analogy and distinction between world-oriented and principle-oriented approaches to political philosophy. While principle-oriented approaches grant centrality and importance to engagement with principles of justice, legitimacy, or other political concepts, world-oriented approaches grant centrality and importance to engagement with social worlds. Last, I propose two examples of world-oriented approaches, political philosophy as navigation and political philosophy as world-building, and argue that they are viable and worthy of further consideration.
Rawls’s Uses of “Social World”
I start with an influential source of “social world” language in political philosophy: the works of Rawls. In the original (1971) and revised (1999a) editions of Theory (TJ), Rawls uses “social world” five times. For example: [The principles of justice] express the result of leaving aside those aspects of the social world that seem arbitrary from a moral point of view. (14 revised) [People] must develop a conception of the social world and of what is just and unjust if the sentiment of justice is to be acquired. (433–34 revised) The conception of justice [including the difference principle] [. . .] seems more likely than its rivals to transform our perspective on the social world and to reconcile us to the dispositions of the natural order and the conditions of human life. (448 revised)
It appears that in these passages, Rawls uses “social world,” preceded by the definite article, simply to refer to society or to the social as opposed to the natural part of the actual world.
In Political Liberalism (2005; PL), Rawls significantly expands his usage of the term. There, he sometimes uses “social world” as in TJ (e.g., 53). But he also uses the indefinite article and speaks of social worlds (plural).
[A] doctrine’s [. . .] adjusting to the general conditions of any normal and human social world [is what] any political view must do. (171) No society can include within itself all forms of life. We may indeed lament the limited space, as it were, of social worlds, and of ours in particular [. . .] (197)
The latter appears when Rawls discusses a view he attributes to Isaiah Berlin, that “there is no social world without loss.” Rawls also frequently refers to the specific social world that is, or contains, a well-ordered society operating in accordance with his conception of justice.
As understood in justice as fairness, reciprocity is a relation between citizens expressed by principles of justice that regulate a social world [. . .] (17) [. . .] we add, so as to present a well-ordered society as a possible social world, the ideas of an overlapping consensus and of a reasonable comprehensive doctrine. (43) The principles of justice must lead to a scheme of basic institutions—a social world—congenial to [the] end [of developing citizens’ moral powers]. (77)
In PL’s original introduction, Rawls uses a similar term, “social possibility.” The possibility of “a reasonably harmonious and stable pluralist society” (xxv) animates his conceptions of liberalism and liberal political philosophy. 5
One can also find “social world” all over Rawls’s work on the history of philosophy. Gaus (2016, 22–23) observes that social worlds are crucial to Rawls’s interpretation of Kant. In “Themes in Kant’s Moral Philosophy” and Lectures on the History of Moral Philosophy (2000; LHMP), Rawls reconstructs Kant’s categorical imperative procedure with reference to social worlds. Rawls characterizes a “perturbed” or “adjusted” 6 social world as the “order of nature” constituted by the laws of nature as we understand them, adding an additional law according to which everyone acts according to whatever practical maxim is being tested. Rawls also improves the test in a way that he takes to be “Kantian in spirit” by requiring comparisons between perturbed social worlds (LHMP: 501–502). “Social world” also appears several times each in Rawls’s lectures on Rousseau’s social contract, Marx’s critique of capitalism (Lectures on the History of Political Philosophy (2007); LHPP), and Hegel’s idea of reconciliation (LHMP). In these contexts the term is used to refer to “our” social world as well as ideal ones.
Rawls picks up these threads in Justice as Fairness (2001; JaF). While reappearing in several contexts shared with PL, “social world” is also prominent in the book’s opening section. There (and at the beginning of LHPP), Rawls presents four roles of political philosophy in public culture: the “practical” role of conflict resolution through reason, orientation, reconciliation, and probing the limits of practicable political possibility. In JaF, Rawls uses “social world” when discussing each of the last three.
The members of any civilized society need a conception that enables them to understand themselves as members having a certain political status—in a democracy, that of equal citizenship—and how this status affects their relation to their social world. (2–3) We try to show that the well-ordered society of justice as fairness is indeed possible according to our nature and requirements [of workable political institutions]. This endeavor belongs to political philosophy as reconciliation; for seeing that the conditions of a social world at least allow for that possibility affects our view of the world itself and our attitude toward it [. . .] Our social world might have been different and there is hope for those at another time and place. (37–38, compare 3–4) We view political philosophy as realistically Utopian: that is, as probing the limits of practicable political possibility. Our hope for the future of our society rests on the belief that the social world allows at least a decent political order [. . .] The fact of reasonable pluralism limits what is practicably possible under the conditions of our social world, as opposed to conditions in other historical ages [. . .] (4)
When explaining the difference principle in terms of Kripke’s idea of rigid designators, Rawls also suggests “[t]aking [] cooperative schemes as possible social worlds (let’s say)” (59n26). Yet, we also get an example of a smaller social world. When the strains of commitment seem excessive to us, we might “grow distant from political society and retreat into our own social world” (128). 7 Rawls also entertains the possibility that social worlds might have a shared structure (76). And he again uses “social world” in historical interpretation, now for Mill (119).
The original paper version of “The Law of Peoples” does not contain a single use of “social world,” but the term is all over the book (1999b; LoP). 8 Rawls characterizes both our society and a possible international “Society of Peoples” as social worlds. As in JaF, he connects thinking about social worlds to realistic utopianism, claiming that “our hope for the future of our society rests on the belief that the nature of the social world allows reasonably just constitutional democratic societies existing as members of the Society of Peoples” (5). And he associates “envisioning” social worlds with examining practical (or practicable) possibility (12). In the book’s final section, “Reconciliation to Our Social World,” Rawls connects realistic utopianism to the reconciliatory role of political philosophy. Reconciliation enables one “to see [a social world] as both reasonable and rational” (127).
Was Rawls’s frequent use of “social world” in his later writing inspired by his readings of historical texts? There are some difficulties regarding chronology given that Rawls’s lectures evolved over his career. That said, there are plausible connections to his study of Hegel. The idea of reconciliation and the term “social world” appear together in the original edition of Theory (512). But later, when presenting the roles of political philosophy in LHPP and JaF, Rawls explicitly links reconciliation to Hegel. In 1991, Rawls added lectures on Hegel to his course on moral philosophy and revised them in 1998 (LHMP, xiv and note). Nearly all of the uses of “social world” in those Hegel lectures are in the section on reconciliation (331–36). And Rawls acknowledges at the start of that section that he “learned much from” Hardimon’s (1994) Hegel’s Social Philosophy and “draw[s] from it” (331n). Thus, it is plausible that Hegel, and Hardimon’s interpretation, influenced Rawls’s use of “social world.” 9
We have seen that in Rawls’s later works, he increasingly uses “social world” to refer not only to the social part of the actual world but to a broad set of possibilities, societies, possible societies, basic structures, or something along these lines. The term often reappears in connection to particular ideas, including regulation by principles of justice, value pluralism, and the roles of political philosophy. However, Rawls provides neither a definition nor a general characterization of social worlds. I now turn to this task.
A Rawlsian Conception of Social Worlds
Against Reduction
The quotes I have presented pull us in different directions. I assume that charitably interpreting Rawls requires us to avoid attributing sloppiness or casual polysemy unless we can uncover no reasonable alternatives. So I now consider several possible interpretations of what Rawls means by “social world.” Each takes the form of a quick reduction of social worlds to a familiar philosophical idea. I reject these suggestions, at least for Rawls interpretation. I subsequently present a permissive definition of social worlds that is broadly Rawlsian.
There is an existing paradigm in analytic philosophy for thinking about possible worlds, and there has been much formal work on them in metaphysics, logic, and linguistics. Thus, it is tempting to reduce social worlds to possible worlds with social relations in them. But we should be careful here. Note first that most of the history of political philosophy predated formal work on possible worlds in analytic philosophy. For Rawls to attribute thinking about possible worlds in the modern sense to Kant, Hegel, and other canonical figures would therefore be anachronistic, going against his stated principles for reading texts from the history of philosophy. (Rawls claims in “Some Remarks About My Teaching” that one should try to interpret texts as closely as possible to how the authors themselves saw their work, while explicitly noting any departures from what they say (LHPP, xii–xv).) Moreover, qua political liberal, Rawls would plausibly hesitate to rest his arguments on controversial premises about the metaphysics of possible worlds.
The only places I have found “possible world” in Rawls’s major works are (i) his lectures on Leibniz (LHMP, 103–40) 10 ; (ii) when describing the “ethics of creation” in TJ, an approach which he distinguishes from his own (137 revised); and (iii) the footnote from JaF where he connects the difference principle to Kripke’s idea of a rigid designator (59n26). This last quotation is the best evidence that Rawls thinks social worlds are relevantly similar to possible worlds. But his language is tentative: “[t]aking [] cooperative schemes as possible social worlds (let’s say)” (my emphasis). And as I will argue shortly, even linking social worlds to cooperative schemes is not fully consistent with Rawls’s usage. So I do not think this passage is conclusive evidence that, as interpreters of Rawls, we should reduce social worlds to possible worlds.
What about reducing social worlds to situations, partial worlds, or states of affairs with social relations in them? To make one of these moves on Rawls’s behalf would be hasty. If Rawls had these things in mind he could have said so, as all three were commonly discussed in analytic philosophy by the early 1980s. 11 And even if the best metaphysical account of social worlds would treat them as situations, partial worlds, or states of affairs, we should proceed cautiously when interpreting Rawls’s usage given the aims of political liberalism.
The aforementioned quote from PL linking a social world to “a scheme of basic institutions” (77), and various passages where Rawls appears to alternate between uses of “social world” and “society” (e.g., 197), provide some evidence that he had a reduction to societies or basic structures in mind. Throughout his works, Rawls characterizes society as a (fair) system of social cooperation and basic structures in terms of central institutions. Yet as we have seen, Rawls also contrasts political society with “our own” social world to which we can “retreat” (JaF, 128). Although Rawls would plausibly claim that smaller social worlds are still influenced by society and the basic structure, such worlds themselves are neither societies nor basic structures. In addition, we have seen Rawls characterize the things posited to test maxims using Kant’s categorical imperative as perturbed social worlds. These things may turn out to feature the absence of large-scale social cooperation and no basic structure whatsoever: consider lying-promise worlds, no-one-aids-each-other worlds, or states of nature. But Rawls seems happy to use “social world” when referring to these things. He does not indicate that failing the categorical imperative test entails that one was wrong to call them social worlds or that the term ends up referring to nothing at all in such cases.
I agree with Hardimon (1994) that the idea of a social world is “very closely related” to Rawls’s idea of the basic structure of society (16). There are ways political philosophers might decide to restrict the idea of social worlds to sets of institutions, rule-governed social practices, 12 or the like. Yet, as Hardimon goes on to say, “the use of ‘the social world’ [Hardimon recommends] makes it possible to coherently ask whether a given social world forms an interconnected whole and to coherently deny that a particular social world [. . .] forms a system or totality” (16). I think that such an approach is better equipped to handle Rawls’s examples than one that reduces social worlds to societies or basic structures.
What about thought experiments, parts of thought experiments, or models? The first reduction may be tempting for hypothetical social worlds, but I believe it relies on a category mistake. A thought experiment is best understood as a certain type of procedure or cognitive act that may involve an appeal to one or more social worlds. 13 A social world can at most be imagined or contemplated as part of a thought experiment, but like a beaker filled with chemicals, it is not an entire experiment on its own. Moreover, Rawls’s references to the social world(s) we actually inhabit block these reductions. We (hopefully) do not live in a thought experiment, and our own social worlds need not be constituents of thought experiments. Nor do we typically inhabit models.
Gaus’s (2016) conception of social worlds is inspired by Rawls’s use of the term as well as work in social choice theory. But Gaus does not define “social world” as clearly as other technical terms he employs. He emphasizes the arrangement of social worlds across a landscape, treats them as objects usable in modeling, and occasionally suggests that they are possible worlds (96, 108). 14 In Gaus’s model of an “evaluative perspective” or point of view, social worlds are sets of “justice-relevant features”: basic institutions, background conditions, and the like. Predictions of how social worlds might turn out determine their “justice scores” according to the evaluative perspective (42–56). Gaus observes that there can be significant and destabilizing disagreement about the features of particular social worlds (128–31, 162–63) and that shared, public social worlds endorsable by inhabitants with diverse perspectives are particularly attractive (177–79). I do not think Gaus’s operative conception of social worlds is wrong or bad. But we can learn more about Rawls and political philosophy in general by slowing down and asking more basic questions about social worlds than Gaus does for his purposes.
A Definition
Using Rawls as a guide, I propose we define a social world as a logically possible closed network of social relations between two or more agents. This conception accommodates Rawls’s various uses of the term, and I take it to be a Rawlsian one even if it is not his own. 15 Logical possibility entails that social worlds are subject to a very permissive consistency condition. By “agents” I mean beings who are capable of practical thinking and action based on that thinking. This entails an inhabitant condition, and a social world must have at least two such inhabitants. A social relations condition requires that at least two of the agents who inhabit a social world relate to each other interpersonally—by way of communication, conflict, cooperation, and other social relations. A network condition restricts the structure of social worlds so that their features are organized around patterns of behavior. These patterns are intelligible regularities in how a social world operates—including, for example, rule-following, adherence to convention, and mere causal relations. Patterns like these make each social world a cosmos rather than chaos and thus minimally comprehensible. There is also a closure condition. Borrowing a term Rawls uses when introducing his assumption that domestic political societies are isolated, social worlds are “self-contained” (PL, 12). By this, I mean that they have no extraneous content. A social world includes any agents, attitudes, objects, events, etc., that appear in its social relations—and that is all it contains. 16
This characterization of social worlds is very permissive. It follows from this account that we actually live in a social world—perhaps more than one, depending on how we think about individuation and closure. Social worlds may be more or less realistic, feasible, historical, utopian, and/or dystopian. They may be characterized by scarcity or plenty, disagreement or consensus, peace or war. Their inhabitants may have various social technologies: democracy, bureaucracy, laws, market systems, and so on. Social worlds may contain formal and informal social hierarchies and various distributions of wealth, status, and power. And they may contain racism, sexism, classism, ableism, and other forms of oppression. This definition includes a variety of hypothetical worlds from the history of political thought: Plato’s Republic, Augustine’s City of God, Hobbes’s and Locke’s states of nature, 17 Rousseau’s community of equals, Kant’s kingdom of ends, Marx’s perfect communism, Dworkin’s island society, Cohen’s camping trip, and the like. It also includes Rawls’s well-ordered society, while the original position could be insofar as the parties are agents who stand in bona fide social relations. This definition also includes “smaller” worlds, such as families, friendships, and associations, at least if we bracket their broader social context.
Decisively answering questions about the metaphysics of social worlds—their nature, how many there are, how exactly to individuate them, whether they are social objects, etc.—extends beyond my purposes here. Whether Rawls is committed to an uncontroversial and minimal conception of social worlds, analogous to his political conceptions of the person, autonomy, and so on, or to something more robust is also worth considering. It is plausible that different conceptions of social worlds would be useful for different purposes in political philosophy. One might also wonder whether there is a general concept of social worlds that political philosophers or even human beings in general share or could share. 18 These are important topics for further inquiry.
Rawls’s Roles of Political Philosophy
As seen previously, Rawls uses social worlds language when presenting three of his four roles of political philosophy “as part of a society’s public political culture” (JaF, 2). He presents these roles in the opening pages of both JaF and LHPP, having previously discussed two of them in LoP. Systematic discussion of these roles in the literature has begun but is still quite sparse. 19 I will briefly present these roles and then put the idea of social worlds to work, showing how it can give us further insight into how the roles are connected. This also suggests an approach to political philosophy—political philosophy as navigation—that I examine later.
The first role of political philosophy Rawls identifies is what he calls its practical role. (All four might be thought of as practical insofar as they help citizens practically reason or produce practical value.) He emphasizes conflict resolution through an appeal to reason—of the sort accomplished by the doctrine of toleration in the aftermath of the early modern religious wars. When large-scale social disagreements about basic questions seem intractable and threaten social order, political philosophy can help to uncover underlying agreements or maintain “social cooperation on a footing of mutual respect” (JaF, 2).
Orientation is the second role. As agents, we are presented with a wide range of possible aims, ends, and purposes. Political philosophy helps us arrange and evaluate these aims, ends, and purposes in a reasonable, rational, and unified way. It does this for us not just “as individuals or as members of families and associations” but as people who are embedded in societies and granted some sort of political status: for example, as equal citizens (2–3).
Rawls’s third role is reconciliation. He ties this to the Hegelian aim of understanding how our institutions are rational. Insofar as we can think of our institutions as rational, we can thereby accept and affirm them, “calm[ing] our frustration and rage against our society and its history.” But he cautions us that political philosophy can become ideological, contributing to the “defense of an unjust and unworthy status quo” (4). Rawls says we should therefore be careful when doing reconciliatory political philosophy (LHPP) and when invoking the idea of political philosophy as reconciliation (JaF).
The fourth role Rawls identifies is probing the limits of practicable political possibility. He calls this the realistically utopian aspect of political philosophy. Probing the limits of practicable political possibility involves thinking about ideal institutions and relations that could nevertheless exist given “the laws and tendencies of the social world.” Rawls calls this role of political philosophy a “variation” of reconciliation (JaF, 4–5). And as he also claims in LoP, Rawls thinks that recognizing the possibility of an ideal social order can actually help to reconcile us to our own condition. I return to the relationship between reconciliation and probing the limits of practicable political possibility later. 20
Rawls does not deductively argue for his four roles or claim these are the only roles political philosophy plays. 21 What I think he provides, following Floyd (2019, 22–25), is an interpretation or (partial) conception of political philosophy—including but not exhausted by the modern academic discipline. This interpretation emphasizes some features and downplays others. One can ask at least two kinds of questions about the roles of something: what it has actually done and questions about what it should do or its value. I take Rawls to think of his four roles of political philosophy as answers to questions of both kinds.
If I am right to treat the four roles Rawls identifies as an interpretation or conception of political philosophy, a way to assess them is to compare them with other possible interpretations. We might consider other lists. For example, Geuss (2008, 37–55) lists five possible functions of political theory: understanding, evaluation, orientation, conceptual innovation, and critical engagement with ideology. Other interpretations might identify a single primary role or activity: seeking the truth about politics, promoting justice, etc. Risse (2020, chapters 2–3) unifies Rawls’s conception of the role of the political philosopher under the heading of “theory-providing citizen-discussant,” and Geuss’s under the heading of “critic of ideology.” 22 When evaluating various interpretations of political philosophy, criteria we might invoke include historical accuracy, consistency, clarity, explanatory power, truth-conduciveness, moral justifiability, usefulness, simplicity, and unity.
Each of Rawls’s four roles of political philosophy has ample historical precedent and plausibly generates epistemic, practical, and moral value. However, one might worry that Rawls’s list as a whole is not as simple, unified, or useful as possible alternatives. Political philosophy, like most social practices, has emerged over time in a complex way. Perhaps a four-role interpretation does relatively well balancing, say, historical complexity and explanatory power. Yet, for a variety of purposes, such as introducing political philosophy to those unfamiliar with it, teaching students, and distinguishing philosophical subfields, it would be helpful to have an account of the roles that explains how they fit together. Political philosophers, after all, are not the only people to play these roles. Conflict mediators and good political leaders also perform Rawls’s practical role. Activists and policymakers also probe the limits of practicable political possibility. 23 And public culture as a whole orients citizens and reconciles them to their circumstances.
There is a straightforward view of the primary aim of political philosophy that one might think could help Rawls link the roles. This view claims that the aim of political philosophy is or ought to be to discover the truth about justice or other political concepts. 24 Yet, Rawls says nothing about finding the truth when presenting the four roles. This omission makes sense in light of how he frames his later work. In PL, he famously claims that his conception of justice “does without the concept of truth” (94). It would be difficult for Rawls to endorse a truth-oriented characterization of political philosophy given his avowed truth-abstinence.
What about the pursuit of justice or reflection on the concept of justice? Given the scope and influence of Rawls’s theory of justice, one might also have expected him to give justice a central place when presenting the four roles of political philosophy. But in his longer discussion of the four roles in JaF, justice is mentioned only in his main characterization of the fourth role and in his secondary glosses of the second and third roles. And Rawls links the practical role of political philosophy to the problem of order rather than justice. It would have been easy enough for Rawls to say that each role of political philosophy involves developing and communicating reasonable principles of justice. He could also have said that each role promotes justice in the world. Yet he does not employ these natural strategies.
I will now use the idea of social worlds, and the idea of navigating the landscape of social worlds, to construct a unified account of Rawls’s four roles of political philosophy.
25
I begin by glossing each role in terms of social worlds: Conflict resolution through reason. Preserving and improving the social world(s) we inhabit. Orientation. Figuring out which social world(s) we inhabit and our positions in them. Reconciliation. Accepting and affirming the social world(s) we inhabit.
26
Probing the limits of practicable political possibility. Exploring and, in some sense, pursuing better or ideal social worlds.
Understood in this way, each role helps political philosophers and their audiences navigate the landscape of social worlds. (Indeed, “orientation” and “probing the limits of practicable political possibility” independently connote navigation.) Orientation helps us think about our own social world(s) and our positions in them, identifying their characteristics and boundaries. This helps us structure our aims as inhabitants of our social world(s). Probing the limits of practicable political possibility helps us think or “look” outward from our own social world(s) to increasingly utopian alternatives. This may inspire us to try to “move” toward them by changing our social world(s). Reconciliation helps us accept where we are on the landscape by turning our thought inward from alternative social worlds back to our own and encouraging us to make some kind of favorable comparison. Conflict resolution through reason helps us attend to existing conflicts in our social world(s) by examining underlying disagreements and nearby alternatives to make repairs and improvements.
Though I will not examine this here, the idea of navigating the landscape of social worlds might also help us grasp how Rawls’s four roles of political philosophy support one another. Compare actual navigation. For example, a navigator is poorly equipped to pursue or assess alternatives if she does not know where she is. But she also requires some sense of what is nearby to see where she is, make improvements, and so on.
World-Oriented Political Philosophy
I have argued that the idea of social worlds is worthy of further consideration because it is increasingly invoked in political philosophy. And I have suggested that appealing to the idea of social worlds, as well as the idea of navigating the landscape of social worlds, can grant further insight into Rawls’s four roles of political philosophy. I will now show, more generally, that the idea of social worlds can enrich debates about the roles, aims, and methodology of political philosophy. Social worlds can be thought of as primary objects that political philosophers think and talk about. I will discuss the idea of world-oriented approaches to political philosophy and then present two specific ones: political philosophy as navigation and political philosophy as world-building. Short of giving decisive arguments for such approaches, I hope to show that they are viable and worthy of further consideration.
World-oriented approaches are approaches to political philosophy that grant centrality and importance to engagement with social worlds. (One might also speak of world-oriented conceptions or interpretations of political philosophy.) As I characterized social worlds previously, political philosophers who adopt a world-oriented approach would engage with logically possible closed networks of social relations between agents—for example, by thinking, arguing, and writing about them. This engagement would be central and important to their work.
What might a world-oriented approach to political philosophy look like? Consider an analogy. Analytic political philosophers spend much of their time thinking about and working with principles. They often seek to discover or develop true principles of justice, legitimacy, and other political concepts. They present candidate principles in published work to their colleagues and the general public. They compare and contrast principles and argue about them with others who have presented other principles. Some political philosophers may even make the pursuit of principles their single guiding aim. 27 Others may view principles as tools they find or fashion in pursuit of other aims. Similar things might be said about the ways that political philosophers engage with normative theories and arguments. Principle-oriented approaches grant centrality and importance to engagement with normative principles. Something close to a defense of such an approach is found in Cohen (2008, chapter 6; 2011). A principle-oriented approach to political philosophy might, for example, include a guiding aim to discover true, or publicly justifiable, principles of justice.
Political philosophers might make social worlds objects of focus in a similar way. Like principles, social worlds can be discovered or developed, presented, compared, contrasted, and argued about. 28 One might reflect on the justice of social worlds, imagine ideal social worlds, or examine the features of the actual social worlds they inhabit. One might use social worlds as an initial data set or to rank them according to various criteria. Perhaps one might examine and present social worlds alongside principles, orienting their approach around both kinds of objects.
If it is indeed possible to combine world-orientation with principle-orientation in a single approach, how would world-oriented approaches be distinctive? I think Gaus (2016) is correct that “a sharp distinction between a theory focused on principles and one focused on social structures is, ultimately, untenable. All principle-based theories ‘worth our attention’ are sensitive to their social realizations” (22, my emphasis). Yet when examining approaches to political philosophy, distinctions between principles, worlds, social structures, theories, arguments, etc., matter—for the simple reason that they are different kinds of things. Approaches and guiding aims affect what work political philosophers produce and what their inquiry is like. A guiding aim of understanding social worlds would likely yield different methods, ideas, and positions than one of presenting true principles of justice, promoting social equality, or publishing as many papers as possible. Each directs emphasis and attention differently and is likely to foster different priorities, arguments, and ways of reading and writing. Political philosophers’ guiding aims shape questioning and research programs and, when broadly shared, are often taken for granted. When conceptions become dominant, they can also affect how a practice is viewed by outsiders, as well as who chooses to become a practitioner and who remains one. So the distinction between world- and principle-oriented approaches, even if not sharp, is meaningful.
Approaches emphasizing principles of justice, legitimacy, etc., would seem to lead political philosophers in certain directions: toward efficiently packaged central claims, counterexamples, and a division of study organized around these values. Approaches emphasizing social worlds would plausibly encourage detailed investigation of cases and the study of institutions and social patterns real and/or counterfactual. Proof must be in practice, however, and there is much room to debate the virtues and vices of various approaches. And there is room to argue, as James (2013) suggests, that we should let “a thousand flowers bloom” (60). Nevertheless, political philosophers have good reasons to regularly check in with themselves to examine and evaluate their approaches and guiding aims.
I will now discuss how I take world-oriented approaches to relate to several other approaches and proposals. There is plenty of evidence that Rawls’s own approach was principle-oriented, with guiding aims to find reasonable principles of justice and legitimacy and present them to his audience. Even so, in JaF, when Rawls mentions or alludes to normative principles in his comments on each of the four roles of political philosophy, he treats them as a means to fulfilling the roles. Given the amount of time Rawls spent defending his principles of justice in his work, it is tempting to say that he viewed serious engagement with principles as a necessary means for the four roles. Yet understood in such a way, engagement with principles might nevertheless be secondary to navigating the landscape of social worlds.
Sen (2009) explicitly values engagement with objects that he calls social realizations. Sen distinguishes “realization-focused” views from “arrangement-focused” views, inspired by two terms for justice, nyaya and niti, from ancient Indian legal theory (20–21). He prefers realization-focused views, as they conceptualize justice not only in terms of “basic institutions and general rules” but also “what emerges in [. . .] society, including the kind of lives that people can actually lead” (10). I think both arrangements and realizations in Sen’s sense count as social worlds—whether worlds tightly and unrealistically organized by institutions and rules or worlds that are feasible and/or nearby. 29 Indeed, one can have a world-oriented approach that is more or less realistic or more or less focused on ideal theory, depending on which kinds of social worlds one engages with and how one engages with them. And the distinction between world- and principle-oriented approaches is orthogonal to the one between approaches focused on ideal and nonideal theory.
The “political realism” pioneered by Williams (2005) and Geuss (2008) could perhaps be thought of as a set of world-oriented approaches that investigate legitimacy by focusing on individuals and relations in actual social worlds. Something similar could be said about approaches that encourage political philosophers to focus on what human beings in fact do, such as Floyd’s (2017) “normative behaviorism.” I do not think that these approaches to political philosophy are nothing but world-oriented approaches. But the idea of social worlds casts them in a new light and raises new methodological questions. For example: should someone focused on the real world be only world-oriented or also grant principles a significant role? How much attention should political realists or normative behaviorists give to proximate, realistic, but nonactual social worlds, or smaller and larger ones?
World-oriented political philosophy also has some affinities with the work of “practice-dependent theorists” and perhaps also approaches to comparative political thought that center social practices. 30 Erman and Möller (2018) suggest that “the basic practice-dependent idea is that social practices and institutions fundamentally alter relations between people, and, consequently, the first principles of justice appropriate for them” (11). Following Southwood (2011) and Sangiovanni (2016), Erman and Möller reject a narrow, early Rawlsian conception of social practices on which they are necessarily linked to systems of rules (16–17). 31 I do not intend to decisively adjudicate between world- and practice-oriented approaches. Yet even if one rejects a rule-based account of social practices, world-oriented approaches appear to have a significant advantage: social worlds can be larger in scope than social practices. A single social world (for example, a state) might contain inhabitants who are engaged in multiple social practices (baseball, voting, etc.). The interactions between practices might have dynamics that are normatively significant. But social worlds can also be smaller. Thus, they are a flexible object for the methodological orientation of political philosophers.
We find ourselves in what Floyd (2022) calls a “methodological moment.” Debates about the methodology, aims, and roles of political philosophy have increased in prominence. Various political philosophers have called for more public engagement. Nussbaum (2016, 203–209) argues that philosophers ought to present their views in ways that are publicly accessible. Laurence (2021, ch. 4) argues that it is crucially important for theorists of justice to keep in mind potential agents of change in their societies. Moody-Adams (2022) suggests viewing participants in social movements as contributors to political philosophy. de Shalit (2020) encourages political philosophers to pursue a method of “public reflective equilibrium,” including discussion and debate with their fellow citizens. And “grounded normative theory,” as described by Ackerly et al. (2021), often involves substantial engagement with subjects of empirical study from a standpoint of intellectual humility. These various “public” approaches could be thought of as world-oriented approaches that involve direct interaction between theorists, audiences, and subjects of study in shared social worlds.
Two World-Oriented Approaches
Political Philosophy as Navigation
I will now explore two world-oriented approaches in further detail. I take these to be viable and interesting approaches to political philosophy, with historical precedent and expected value, even if, all things considered, they are not best or most comprehensive. First is political philosophy as navigation. It includes a single guiding aim: to help one’s audience 32 navigate the landscape of social worlds. If what I argued earlier is on the right track, it is a Rawlsian one in a broad sense.
Given this guiding aim, the rest of the approach takes the form of trying to contribute to navigation. This involves helping the people in one’s audience figure out where they are on the landscape, where they might want to go, and better and worse destinations and paths. A political philosopher with this approach would think a lot about social worlds: their audience’s own worlds and many others. Examining principles, developing theories, and presenting arguments are compatible with this approach. However, someone with this approach would engage with these things for the sake of aiding with navigation. And they would not think of principles, theories, arguments, or even knowledge of social worlds as the ultimate target of their political-philosophical work.
Talk about navigating the landscape of social worlds is of course metaphorical. Navigation metaphors—the ship of state, climbing mountains 33 —have played significant roles in the history of political philosophy. One significant virtue of the navigation metaphor is that it simultaneously captures both theoretical and practical aspects of political philosophy. A position held by many who care about political philosophy is that it ought to be both theoretical (aimed at truth or something like it) and practical (aimed at action-guidance and the promotion of value). And helping someone navigate requires you to contribute to their thought as well as their action. Truth can be sought and highly valued consistent with a world-oriented approach—true beliefs plausibly help the climber or the sea captain navigate—but it is not everything. One also needs other things when crossing land or sea: a physical map for reference, a list of attractions and dead ends, some glasses or sunglasses, and enough inspiration and nourishment to keep going. It is unlikely that the audience of political philosophy can do without the analogs of these things—images, goals, heuristics, motivation, hope—sustained only by the presentation of truths and their own true beliefs.
Adopting political philosophy as navigation as an approach would not logically or rationally commit its possessor to many substantial positions on first-order questions of political philosophy. But one’s approach is not neutral in its effects on the theories and other research outputs one develops. There is a natural but contingent link between political philosophy as navigation and nonideal theory. A political philosopher with the approach of political philosophy as navigation would want to help people navigate, and actual people already inhabit some social worlds. So it would make sense if that political philosopher gave more attention to actual, realistic, and feasible social worlds. Now many ideal theorists care a lot about helping with navigation. They might say that the best way to navigate is to chart a course for ideal social worlds or that political philosophy ought to be focused on ideal worlds in some other way. Yet the navigation metaphor does have some limits. Helping hypothetical agents navigate among hypothetical worlds does not seem like a desirable guiding aim or use of one’s time, even for a philosopher.
We also have reason to think that adopting political philosophy as navigation would discourage the proliferation of theories that defend normative principles as necessarily true or true in all possible (social?) worlds. It is difficult to see why political philosophers focused on helping people navigate the landscape of social worlds would desire, require, or even be in a position to find necessarily true principles rather than ones good enough to help people navigate where they are. Similarly, political philosophy as navigation seems less likely to prop up views that appeal to a single master principle of utility, liberty, or equality based on its necessity or rationality, although such views might do quite well based on their simplicity, uniformity, or moral accuracy. In contrast, political philosophy as navigation plausibly would be more likely to produce work that is descriptively adequate, action-guiding, persuasive to its audience, and practically efficacious. And what counts as work that achieves the guiding aim of political philosophy as navigation would likely vary based on the circumstances of its authors and readers—since what counts as help with navigation varies based on where one happens to be.
Political Philosophy as World-Building
Engagement with social worlds often takes the form of thinking about actual ones or those that have been presented in previous discourse. But world-oriented approaches to political philosophy could also emphasize the activity of world-building: creating and presenting novel social worlds. Consider a quite different and un-Rawlsian world-oriented approach to political philosophy, political philosophy as world-building, with a guiding aim of creating and presenting novel social worlds. Most would agree that creativity can be useful to the normative theorist, fueling thought experiments and pumping intuitions. But the value of utopian and, more broadly, unrealistic thinking is not exhausted by its direct contributions to theorizing about justice and other political concepts. There are many things that theories, propositions, and rational arguments cannot do very well. 34 Displaying nonactual social worlds of various kinds can capture attention, generate discussion, motivate political action, and much more.
Some recent defenses of ideal theory gesture at these other sources of value. Estlund (2020) and Laurence (2021) both argue that there is value in presenting and examining possibilities that are not known to be feasible. However, one can consistently endorse this claim alongside the view that unrealistic thinking should not directly inform theorizing about justice. This is a plausible perspective: one that rejects ideal theory of the ordinary sort but endorses other utopian contributions to public culture, perhaps even by political philosophers. In any case, distinguishing world-building from theorizing can help political philosophers locate the value of unrealistic thinking without taking a specific stance on its relationship to normative theorizing.
Thinking of the political philosopher as, in part, a world-builder can also help us demystify the historical connection between utopian literature and utopian political philosophy. More’s Utopia inspired an entire genre of utopian writings: Bacon’s New Atlantis, Campanella’s City of the Sun, and many others. These works present ideal societies and are both philosophical and literary. But despite their philosophical influence and importance, these works are rarely treated as canonical or given much attention in analytic political philosophy. 35 Thinking of political philosophers as world-builders as well as theorists can allow us to understand utopian texts in deep, dynamic, and more charitable ways than we might if our main purpose was to mine them for arguments. Rawls himself places his work in this tradition by calling his political philosophy “realistically utopian.” He presents a theory of justice and also presents a picture of a well-ordered liberal society. The well-ordered society does not appear ex nihilo—it is built from familiar ideas and idealizations of actual societies—but it is nevertheless Rawls’s own creation. It is not only a model of justice, nor a mere tool for his thought experiments, but an offering to his audience that can serve many functions and be taken up in many ways. The term “world-building” typically brings to mind science fiction, filmmaking, and role-playing games like Dungeons and Dragons. Perhaps it should also make us think of political philosophy.
One might worry that world-building threatens to unleash the imagination from the rightful rule of reason, leading political philosophers and their audiences away from the truth. Unrealistic thinking can, of course, carry us to silly or dangerous places. The proper course of action, in my view, is for political philosophers to self-consciously use their imaginations, 36 whether in theorizing or in world-building. If one is hesitant, there is another, more realistic way that political philosophy as world-building could be understood. It could be understood as a creative process of building and rebuilding actual social worlds. 37
Conclusion
In this paper, I have shown that the idea of social worlds is worthy of careful reflection in political philosophy. I have tracked Rawls’s use of “social world” and presented a conception of social worlds as logically possible closed networks of social relations between agents. I have argued that the idea of social worlds casts Rawls’s four roles of political philosophy in a new light. And I have presented a family of world-oriented approaches to political philosophy, including political philosophy as navigation and political philosophy as world-building. While I have not conclusively defended any such approach, I hope to have shown them to be worthy of further consideration.
I end with some questions. What does “social world” refer to in political philosophy, and what do we want it to refer to? What are the metaphysics of social worlds? How can thinking about social worlds improve interpretation of historical works? How should political philosophers think and talk about social worlds? And how might thinking about social worlds alter conceptions of and approaches to political philosophy?
There are many paths one might follow. Political liberals may prefer a thin conception of social worlds to one dependent on deeper metaphysical assumptions. Those who want to bring social choice theory to bear on political philosophy may prefer a conception on which social worlds are flexible formal constructs. Some may desire a reduction of social worlds to possible worlds to align with trends in metaphysics and the philosophy of language. Social ontologists may treat social worlds as social objects: some built in space and time, others in thought and language. Comparative political thinkers might count social worlds associated with practices, cultures, or traditions among their central objects of comparison. And to counteract anachronistic interpretation, historians of political thought might develop multiple conceptions of social worlds.
In any case, thinking carefully and comprehensively about social worlds will give political philosophers a better sense of what they do and what they can be.
Footnotes
Acknowledgements
For helpful feedback, thank you to Andrew Barton, Valentina Cantori, Jasmine Gunkel, Drew Hogan, Mahmoud Jalloh, Robin Jeshion, S.A. Lloyd, Jonathan Quong, Collis Tahzib, Stephanie Van Fossen, Robert H. Wallace, Karolina Wisniewska, and the referees and editors of this journal. Thank you also to an audience at the 2021 “After Justice” Rawls conference hosted by the University of Bucharest.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was completed with the support of a USC Dornsife/Graduate School Ph.D. Fellowship.
1.
See MacLean et al. (2021, 233–35). Examples in philosophy include Goldman (1999), Searle (2011), and Gilbert (2013). For early discussions of the term in sociology, see Strauss (1978) and Unruh (1980). 
uses “social world” (“soziale Welt”) to refer to “the totality of legitimately regulated interpersonal relations” (120). I do not take a stance here on how my proposed conception relates to Habermas’s ideas, including “lifeworld” and “system.”
2.
Exceptions that helped inspire this paper include Gaus (2011, 2016) and Gaus and Hankins (2017). Things have changed since wrote: “The idea of a social world, and moving from one to the other is, I admit, somewhat unusual in contemporary liberal political philosophy” (32).
4.
Compare Wiens (2012, 2015); .
5.
Rawls also presents his arguments in PL as a possibility proof of “a just and stable society of free and equal citizens” amid reasonable disagreement (4, 47).
6.
“Perturbed” in “Themes”; “adjusted” in LHMP.
8.
Though LoP (1999) was published before JaF (2001), the editor’s foreword to the latter indicates that “by 1989 the manuscript had evolved into something close to its current form” (xii). Thank you to a reviewer for guidance.
9.
On Rawls and Hegel, see Schwarzenbach (1991); Bercuson (2014, ch. 2); and Gledhill (2020). also notes the Rawls/Hardimon/Hegel link (1179n4).
10.
“Possible worlds” are a central element of Leibniz’s philosophy. So it is unsurprising that Rawls would use “possible world” in presenting Leibniz’s views.
11.
12.
As a reviewer has noted, one might follow a different path and interpret usage of “social world” based on his rule-based account of social practices in “Two Concepts of Rules” (20–46). Another path might focus on Rawls’s idea of a “social union,” although that involves shared ends absent from some of the social world cases (TJ, 456–64 revised).
13.
14.
Gaus favorably presents arguments from Elster’s (1978) and , 210–14) “possible world analyses” of ideal theory.
15.
Rawls might oppose such a permissive conception of social worlds due to his commitment to realistic utopianism. However, I think that given his usage, it would make more sense for him to say that unrealistic social worlds do not matter for political philosophers or matter as much as other social worlds. This is different from saying they are not social worlds at all.
16.
Outside of LoP, Rawls treats political societies as closed (TJ, 7 revised; JaF, 13).
17.
18.
See Rawls (PL: 14n15).
19.
See McKean (2017), Floyd (2019, 22–23), Risse (2020, ch. 3), and Laurence (2021). Compare .
21.
Rawls writes (emphases added): “We begin by distinguishing four roles that political philosophy may have” (JaF, 1), and “I see four roles that political philosophy may play” (LHPP, 10).
22.
25.
27.
I think of guiding aims as those aims most central to an agent’s practical reasoning in specific domains or generally: for example, a doctor’s aims to promote health and to not harm their patients.
28.
29.
32.
Rawls views the audience or addressees of political philosophy as citizens in general (LHPP, 1). Compare Young (1990, 13); , ch. 4).
35.
See Sargent (2010); .
37.
One might also think of worldmaking and world-building interchangeably. One might follow Goodman’s (1978) conception of worldmaking as making true “versions.” Or, one might adopt a conception of worldmaking on which it reshapes the actual world, following Getachew (2019) and .