How should we approach uncertain threats of potentially very serious harm? For example, how long should social distancing measures be enforced against the spread of COVID-19? Should we research and develop climate engineering technologies as a measure against climate change harms? Should glyphosate herbicides be banned? Such and similar decisions have potentially far-reaching consequences for the environment or human health; yet they often have to be made under considerable uncertainty, for example, uncertainty about the extent of possible harm, its likelihood, or cause-and-effect relations. Frequently, precautionary principles (PPs) are proposed as an answer to such challenges, telling us that we have to act to prevent harm even if it is uncertain. However, this idea also comes in for criticism as being alarmist, anti-scientific, and in effect doing more harm than good by causing high costs and stifling innovation. The question of how we should deal with uncertain harms is clearly a controversial one. When we seek to address this issue, we are not only faced with the question of whether precautionary principles are justified. More fundamentally, the methodological question arises of how such principles can be justified—what is an adequate method for the justification of a precautionary principle?

One method that is often recommended for justifying principles is reflective equilibrium (RE). Appeals to RE are made in a wide range of philosophical disciplines, from bioethics to political philosophy to logic. Its basic idea—that systematic principles are justified through a process of mutual adjustments to our existing judgments about relevant cases—is readily recited, and it is often seen as a, or even the, method of philosophy (Lewis 1983, Introduction; Scanlon 2003, 149). However, implementations of RE as they can be found in the literature typically either use a sketchy conception of RE, are restricted to simplified cases, focus only on some particular elements of RE, or do not make the application explicit and traceable. This makes it difficult to critically evaluate the method, to assess its potential, and to engage with the criticism directed at it. Moreover, it leaves researchers who want to use the method at a loss: how does one apply reflective equilibrium?

This book provides an explicit and detailed case study for an application of the method of RE. It adopts an elaborate conception of the method and tests whether it can be used to develop and defend a precautionary principle. With respect to RE, I show that there is at least one sufficiently fleshed-out RE conception that can successfully be applied to actual and complex cases. With respect to precautionary principles, the case study demonstrates how a rights-based precautionary principle can be constructed and defended. By focusing in particular on RE as a method for the justification of a precautionary principle, the case study also addresses questions of methodology that so far have been neglected in the debate about precautionary principles. In this way, the book can simultaneously illuminate two different debates.

In this first chapter, I identify desiderata for a case study by giving an overview of the main ideas of reflective equilibrium and surveying existing applications of the method in Sect. 1.1. Section 1.2 introduces the debate about precautionary principles, and explains why this topic was chosen for the case study. Lastly, Sect. 1.3 gives an overview of this book, outlining the chapters in order to show how the identified tasks are addressed.

1.1 Reflective Equilibrium: Main Ideas and Previous Applications

I start by describing the main ideas of reflective equilibrium in Sect. 1.1.1. In Sect. 1.1.2, I give an overview of existing applications of reflective equilibrium, before identifying desiderata for a case study in Sect. 1.1.3.

1.1.1 Main Ideas of Reflective Equilibrium

At the core of reflective equilibrium we have the idea that two sets of elements have to be mutually adjusted with respect to each other: on the one hand, our commitments about a subject matter, and on the other hand, a systematic account of them, for example in the form of principles or a theory. RE thus takes our existing judgments and beliefs as the starting points of theorizing, but without treating them as fundamental fixed-points. It thereby takes into account that our judgments and beliefs are usually not accepted simply at random, but have at least some minimal credibility via the fact that we are committed to them and act on them in our daily practices.

By requiring that our commitments have to be brought into coherence with a systematic account of them, RE also acknowledges that our judgments might be wrong, or based on biases and prejudices. Trying to fit them into a coherent position with systematic elements can reveal that some of our judgments are not as credible as they seemed on their own. Conversely, it might turn out that statements which are not very credible in isolation can be combined to form a convincing picture.Footnote 1

There are at least two major traditions of RE in philosophy: one builds primarily on the works of Rawls, which is why it is often called a “Rawlsian method”. Rawls (1971) coined the term “reflective equilibrium”, and under this name the method gained importance especially in moral and political philosophy. The other tradition goes back to Goodman (1983), and, originating in questions of theory choice and justification in philosophy of science and logic, sees RE less as a method for a specific task and rather develops a general RE-based epistemology. Nonetheless, there is a lot of overlap in how RE is conceptualized in these two traditions—arguably, Rawlsian conceptions of RE can be seen as particular specifications of the broader RE epistemology. In both traditions, RE is typically understood as an account of justification, which does not necessarily mean that it is truth-conducive (cf. Rawls 1974).Footnote 2

In the Rawlsian tradition (Daniels 1979; Rawls 1971, 2001; Scanlon 2003), RE is often seen as being specifically concerned with the justification of principles of justice that can be accepted by all citizens in a pluralistic society. This made RE a prominent and influential method in political philosophy (e.g., De-Shalit and Wolff 2007). But the potential of the method for moral questions more broadly was quickly recognized, and variants of it can be found in moral epistemology (DePaul 1993; Tersman 1993) as well as in applied ethics (Beauchamp and Childress 2013; Doorn 2010b; Van der Burg and Van Willigenburg 1998b; Van Thiel and Van Delden 2010). Especially in bioethics, a conception of RE that includes empirical data has gained influence as an approach to problem solving in specific cases (de Vries and van Leeuwen 2010).

In the epistemology tradition, the idea to use reflective equilibrium as the basis of a more comprehensive epistemology (Goodman 1951, 1978, 1983; Scheffler 1954) has been most thoroughly explored by Elgin (1996, 2014, 2017). By building on the works of Goodman and Elgin, and connecting them with methodological ideas from Carnap like the method of explication, Brun (2013, 2016, 2020) and Baumberger and Brun (2017, 2021) have developed one of the most elaborate conceptions of reflective equilibrium to date. I describe this RE conception in more detail in Chap. 2, as I adopt it as the basis for testing the applicability of RE in a case study. What is particular about this conception is that it distinguishes between two components of the equilibrium process. On the one hand, we have to bring commitments concerning the subject matter into agreement with a systematic account of these commitments—e.g., principles, laws, or a theory. On the other hand, this agreement between commitments and a system has to be balanced against further demands.

The first sense of “reflective equilibrium” concerns the agreement between commitments and a systematic account. This is achieved through a process of mutually adjusting both sides, with neither commitments nor the systematic account being privileged or safe from revision. This is the idea of “balancing” that is usually associated with reflective equilibrium, and fits with how Goodman characterizes the process of justification in the famous quote from Fact, Fiction, and Forecast:

[Rules] and particular inferences alike are justified by being brought into agreement with each other. A rule is amended if it yields an inference we are unwilling to accept; an inference is rejected if it violates a rule we are unwilling to amend. The process of justification is the delicate one of making mutual adjustments between rules and accepted inferences, and in the agreement achieved lies the only justification needed for either. (Goodman 1983, 64, italics in original)

Concerning the second reading of the metaphor “reflective equilibrium”, (Brun 2020) argues that the agreement between commitments and a system has to be sought between two further forces that drive the process of adjustments in RE. Commitments should not only be in agreement with a system, but also have to respect the initial commitments that we started out with; and the system should not only be able to account for the commitments, but also has to do justice to certain theoretical desiderata which drive the theoretical development in the context at hand (Brun 2020, 26). In other words, the commitments that result after the process of adjustments should still be reasonable in light of the commitments that we started out with. This means that if an initial commitment was revised, we need to be able to give a plausible explanation for this revision. This is also central to Elgin’s conception of reflective equilibrium:

The components of a system in reflective equilibrium must be reasonable in light of one another, and the system as a whole reasonable in light of our initially tenable commitments. (Elgin 1996, 107, italics added by T.R.)

The constraint that commitments should respect initial commitments is balanced on the other side by the demand that we are searching for a systematic account of the subject matter, i.e., principles or a theory that can be used to make justified moral decisions, or to predict events, etc. For this reason, the system should do justice to theoretical virtues like simplicity, fruitfulness, scope, or accuracy (Brun 2020, 25).

Even though this conception of RE is well elaborated (see Chap. 2 for a fuller discussion of it), it might still leave the practitioner who wants to apply the method at a loss: I know that I should start from my initial commitments, but how do I get access to them and how do I select them? How can I determine which theoretical virtues are relevant, and how they should be interpreted? What does it mean for the system and the commitments to be in agreement, and how should I resolve conflicts between the two?

Recently, Walden (2013) has defended reflective equilibrium exactly on the grounds that it does not in itself give a definite answer to these and similar questions. According to Walden, RE is anti-essentialist in the sense that it denies that we can say much about our epistemic inputs, methods, and standards in advance of inquiry. On these grounds, he defends RE against opponents who attack the method based on specific ways to define it, e.g., who equate commitments with intuitions and attack RE on the grounds that intuitions are not a plausible input. This defense fits well with Elgin’s view that our standards and goals are higher-order commitments which can, in principle, be revised as well (see, e.g., Elgin 1996, 99, 104; 2014, 247–248)—but it doesn’t seem to do much for the agent who is searching for a method to apply.

That RE does not say something more essential about the nature of inquiry led Foley (1993) to reject the idea of RE as unhelpful:

It tells you essentially this: take into account all the data that you think to be relevant and then reflect on the data, solving conflicts in the way that you judge best. On the other hand, it does not tell you what kinds of data are relevant, nor does it tell you what is the best way to resolve conflicts among the data. It leaves you to muck about on these questions as best as you can. (Foley 1993, 128)

However, such worries have not deterred researchers from invoking the method of RE for their work; and Walden acknowledges that for specific projects, the method of RE can be given a more definite character (Walden 2013, Fn. 15). But, as the next section shows, while we can gain valuable insights from previous applications, they fall short of providing the basis for a systematic assessment of how RE can be applied as a method.

1.1.2 Applications of Reflective Equilibrium

Given the claim that reflective equilibrium is an important—maybe even the method—for justification, the lack of explicit, well-documented applications is surprising. Beauchamp and Childress (2013) note several unresolved problems about the method of RE, one of them being that there is a striking discrepancy between claims to be using the method versus actual uses of RE:

First, ambiguity often surrounds the precise aim of the method. It might be used in reflecting on communal policies, constructing a moral philosophy, or strengthening an individual’s set of moral beliefs. The focus might be on judgments, on policies, on cases, or on finding moral truth. Second, it is not entirely clear how to know when our effort to achieve reflective equilibrium is going well, or how to know when we have succeeded. Explicit uses of the method (by contrast to claims to be using it) are difficult to find in the ethics literature. Most discussions are heavily theoretical and distant from contexts of practice. [...] Third, the wide-ranging objectives of even a weak wide reflective equilibrium are at minimum intimidating and may be unattainable ideals of both comprehensiveness and coherence. (Beauchamp and Childress 2013, 410, italics added by T.R.)

A look at the literature confirms this assessment: very different purposes are associated with applying RE, and there is a lack of explicit applications. I will say more on these two points in the following. The third point, whether or not the method is too demanding to be applicable, cannot be assessed until the other two points have been addressed.

Aims of Applying RE

The objectives that people pursue when stating that they are applying RE differ widely, but can be roughly grouped under four main categories.

The first group sees RE as a method for the justification of (often normative) conceptions, principles, or theories. The most prominent example is probably Rawls (1971), who aims at a theory of justice. Another example is Swanton (1992), who sees the purpose of RE in the justification of a value-conception, in her case that of “freedom”. Similarly, De-Shalit and Wolff (2007) want to use RE to give an account of disadvantage that, as part of an egalitarian theory, can be applied to actual society in order to diminish disadvantage. And (Van Thiel and Van Delden 2001) aim to use RE to give an account of the principle of respect for autonomy in health care.

In the second group, RE is seen as an approach to decision-making with respect to a particular (moral) problem. For example, Kushner et al. (1991) refer to it as a “systematic approach to the process of working through an ethical dilemma”. In this understanding, RE is a method of decision making that can be used to adjust existing principles and judgments with respect to a particular problem in order to come to a justified decision—in the sense of seeking an answer to the question “What is morally right to do in a particular case?” (e.g., Rutgers 1998, who discusses this for the context of veterinary ethics).

The third group sees RE as a framework for structuring discussions and interpersonal decision-making. For example, Doorn (2010a) sees RE as a framework that can enable decision-making in a pluralist context with different stakeholders, and Schroten (1998) proposes it as a framework for discussions in ethic committees. Similarly, Brandstedt and Brännmark (2020) suggest a version of RE that is intended as “a tool for public reasoning about practical problems which aims to facilitate shared solutions”.

Another idea is to use RE to analyze and reconstruct debates, which for example Hahn (2000) demonstrates with respect to the foundational crisis of mathematics. Brun (2014) suggests that RE can be used for modelling and justifying the various steps involved in argument reconstructions.

The different ideas about how RE can be used have even prompted some to speak of different types of reflective equilibrium methods (Van der Burg and Van Willigenburg 1998a, 12). This diversity also extends to the structure of the method, i.e., what people think is required in order to apply RE, and is connected with different ideas about the agent(s) involved in the process (for an overview of different ways to conceptualize the social aspects of RE, see Baderin 2017).

In this book, I will investigate RE as a method that aims at justifying principles, theories, or similar. This clearly fits best with the goal of justifying a precautionary principle. Additionally, I will conceptualize it as a method that can be applied by a single epistemic agent, and not, e.g., as a deliberative group project. Both decisions will give the case study a more stringent focus while laying the groundwork for further work on different ways to spell out RE as a method.

How Has RE Been Applied so Far?

Consistent with the observation made by Beauchamp and Childress (2013, 410), I could find no example where the actual process of reflective equilibrium, i.e., systematically adjusting commitments and system with respect to each other (as well as with respect to relevant background theories and theoretical virtues), has been done in a way that is both explicit and comprehensive. Either the process is not well-documented—we might grant it to Rawls (1971) that he applied RE, but his arguments are not documented in a way that allows for a clear identification of the steps and elements of the RE process and how they were adjusted under the influence of which considerations (the same holds for De-Shalit and Wolff 2007). Or the process has only been rudimentary started and has not been followed through. Or it is missing altogether, and the application stops after describing part of the input such as a set of commitments or some candidates for background theories. Especially in more empirically oriented studies, the focus is typically on how to obtain inputs for the RE process: for example, by sending out questionnaires (Van Thiel and Van Delden 2001), through conducting semi-structured interviews and analyzing those later (Ebbesen and Pedersen 2007), or by forming one’s commitments based on empirical information about the moral experience of others (de Vries and van Leeuwen 2010).

In the cases where the application does not stop after describing (parts of) the setup, typically all of the (few) adjustments take place on the side of the commitments, without confronting them with a systematic account in the form of principles or a theory. One of the few exceptions are Van Thiel and Van Delden (2001), who confront several candidate principles with commitments of practitioners that were identified through questionnaires. They reject those candidates as inadequate and introduce their own proposal, but this new candidate does not again get confronted with the commitments, and remains only a rough sketch.

The studies that made use of structured discussions describe some adjustments of commitments (Doorn 2010b; Van de Poel and Zwart 2009), but these adjustments seem rather to be about learning effects and how participants of a discussion adjust their position under influence of additional input. They thus cannot clearly be attributed to an application of reflective equilibrium.

We also find some demonstrative applications of the method by authors who discuss and develop an account of RE (DePaul 2011; Sinnott-Armstrong 2006), but these are typically very restricted toy examples. For example, Sinnott-Armstrong (2006) demonstrates how narrow RE might work by presenting us with a number of cases and judgments corresponding to them, e.g., “The Watching Case: A baby crawls into a pool of water. A bystander can save the baby’s life easily at little cost. Otherwise the baby will drown” and “The Watching Judgment: It is morally wrong for the bystander to let the baby drown”, and describes how an agent might generalize moral principles from such judgments, that then get confronted with other cases where they run into problems, e.g. the “Doctrine of Doing and Allowing” runs into problems when confronted with the “Trolley case” and “Trolley judgment” (that it is not morally wrong to pull the lever and direct a runaway trolley from five onto one). Even though adjustments are shown, and even though Sinnott-Armstrong makes clear that it is also possible to adjust commitments, the “equilibration” rather seems to proceed as some sort of “counterexample-philosophy”. There is no search for a system of principles that fits with a broad set of initial commitments, and instead the adjustments proceed by searching again and again for counterexamples which are invariably treated as a reason to alter the set of principles. Also, while Sinnott-Armstrong continues to discuss additional conditions and constraints that should be added to RE in order to achieve “wide coherence”, e.g., second-order beliefs about the reliability of first-order beliefs, bringing in other people’s beliefs, meta-ethical beliefs, metaphysical beliefs, etc., he does not bring this to bear in an (exemplary) application.

While systematic descriptions of an RE process are therefore absent, for descriptions and assessments of RE states matters are indeed even worse. There are some examples in which it is compared how participants of discussions adjusted their positions (Doorn 2010b; Van de Poel and Zwart 2009), but only Doorn (2010b) tries to assess the coherence of judgments, principles, and background at least informally. Van Thiel and Van Delden (2001) do compare theory candidates with commitments, but not systematically, and although they do make a new theory proposal, they do not assess how coherent the position would be with that system, and what adjustments might still be necessary. Explicit criteria for assessing positions and identifying weaknesses are thus still an important desideratum for applications of RE. This and other desiderata are listed in the next section, where I summarize the most pressing challenges for applying reflective equilibrium as a method.

1.1.3 Desiderata for a Case Study

As the above section shows, previous applications of RE fail to sufficiently exemplify how the method can be applied. This is not to discredit their academic merit. Most of them were concerned with concrete and specific challenges and problems, and merely wanted to use RE as a method to tackle these challenges. However, the deficiencies of existing applications make it difficult to assess what role RE can play as a method. This makes a detailed case study necessary in order to test whether it is possible to find instances of its elements in actual applications, and whether its criteria can be used in insightful ways to guide a process of adjustments, and to assess whether a position is justified.

In particular, we can identify the following challenges for applying reflective equilibrium as a method:

  • The aims and purpose of applying the method are unclear;

  • It is unclear what kinds of input, and whose inputs, to select—are the relevant commitments intuitions, (considered) judgments, or beliefs? Should they be of philosophers, practitioners, or the broader public?;

  • It is unclear what it entails to apply the method, i.e., what its steps are and how to complete them, and how to decide about adjustments;

  • It is unclear how to measure progress and to assess whether or not one did actually reach a state of reflective equilibrium;

  • On the one hand, there are worries that, because of its coherentist and holistic character, the method might be too demanding to be actually applicable;

  • On the other hand, there are worries that, because it has no fixed standards, RE is too vacuous to provide any useful guidelines or real constraints, and thus falls short of being a method.

Clearly, the last two worries—whether RE is either too demanding or too vacuous to be applied as a method—cannot be assessed until the other problems have been solved and a real attempt has been made to apply RE. To provide a real test for RE as a method, this application should be to an actual, complex problem, instead of working with simplified toy examples. Thus, we can identify the following desiderata for a case study on the applicability of RE as a method:

  1. 1.

    Reflective equilibrium (RE) should be spelled out as explicitly and precisely as possible, including: (i) its theoretical and methodological foundations, (ii) the aim of the method, and (iii) how RE can be specified for particular applications.

  2. 2.

    RE should be applied to an actual, complex problem in order to provide a real test case.

  3. 3.

    The initial position of the process of adjustments should be explicitly described, e.g., the input and the specified criteria.

  4. 4.

    The application should be traceable step-by-step, describing what was adjusted when and with respect to what, and how the adjustment in question can be defended with the RE criteria.

  5. 5.

    The application should include a detailed comparison of the initial and resulting position, including assessing the resulting position with respect to RE criteria and whether or not a state of RE was reached.

  6. 6.

    The application and its results should be evaluated and critically discussed in order to learn from it for the use of RE as a method.

A case study that aims to meet these desiderata will allow us to assess the benefits and challenges of applying RE. If RE can be successfully applied, the case study will also help to provide guidelines for further applications. For example, it is an open question whether the process of adjustments that is part of RE should be understood as a method that one can follow step-by-step when trying to justify something. Is reflective equilibrium best understood as a constructive method, i.e., a method that describes a procedure that one can follow in order to make epistemic progress? Or is it better understood as spelling out requirements for justification that we can use for the reconstructive appraisal of epistemic positions? By testing RE in a constructive application with the goal of formulating and justifying a precautionary principle, this book aims to provide important insights on these questions.

In the next section, I introduce the topic of precautionary principles, which was selected as the topic of the case study in order to meet desideratum (2). Section 1.3 then gives an overview about the whole book, describing how and where the desiderata (1)–(6) are addressed.

1.2 Precautionary Principles as a Test Case for Applying Reflective Equilibrium

The question of how a precautionary principle (PP) can be justified is an ideal test case for the applicability of RE. It provides an actual, complex problem: the basic idea behind PPs is that we have to take action to prevent harm even if we are uncertain about its likelihood or extent. It is often summarized as “better safe than sorry”. However, it is controversial how this general idea of precaution should be spelled out as a principle.

PPs emerged first in regulatory contexts, arguably as an answer to ineffectual policies of the past which failed to protect the environment and/or human health in important respects (Harremoës et al. 2001). While talk of “the” PP in the singular is common, formulations and interpretations vary so greatly that, if only to be on the safe side, it is appropriate to use the plural form and speak of precautionary principles (cf. Hartzell-Nichols 2013).

Precautionary principles are often seen as a supplement or even as an alternative to traditional approaches to risk regulation, like quantitative risk assessment and cost-benefit analysis. Because the latter require knowledge about the relationship between options and outcomes as well as the probabilities of the various possible outcomes, they condemn us to idleness in cases where we lack that sort of knowledge, if we consider them the only maintainable approach to risk management. Thereby, so the criticism goes, they inhibit immediate measures which would be necessary in order to avoid serious damage. PPs, in contrast, demand action even if this knowledge is not or only partially available.

However, precautionary principles also have been criticized for various reasons, such as being too vague to be action-guiding, paralyzing the decision-process through conflicting recommendations, or being anti-scientific and promoting a culture of fear (for an overview of the criticisms see, e.g., Randall 2011; Sandin et al. 2002).

The challenge of how to spell out the basic idea of precaution in the form of a systematic precautionary principle is thus a suitable test case for the applicability of RE: first, it presents us with a practical problem in which we, as imperfect epistemic agents, have to find a justified principle as a basis for our decisions. Second, there are already a number of existing proposals for PPs to draw on. Third, PPs are often confronted with objections like that they lead to incoherent recommendations. Resolving such inconsistencies seems exactly what RE is designed for. Additionally, it is a topic of practical relevance in the face of, e.g., current threats arising from global climate change.

1.3 Overview of this Book

This book tests a specific conception of reflective equilibrium (RE) in a case study on an actual, complex problem: how can a precautionary principle (PP) be justified? In doing so, it addresses the desiderata (1)–(6) identified in Sect. 1.1.3.

To clarify the tasks ahead, it is useful to make the distinction between epistemology, methodology, and method. Although RE is often referred to as either a method or a methodology, it is rarely explained what is meant by these terms—and sometimes they seem to be used almost interchangeably.Footnote 3 To further complicate things, RE is also often called an epistemology, or account of justification (e.g., Tersman 1993). Thus, in order to obtain a clearer picture of what it means to apply RE as a method, it is important to distinguish these three categories.

For the purpose of this book, I propose the following distinctions (cf. Ackerly and True 2013): firstly, methods are concrete tools and techniques of research, in the sense of a set of instructions or a specification of steps which should be followed to achieve a given (sub-)objective (see also Caws 1967, 339; McPherson 2015, 653). Secondly, methodology is a theory and analysis of how research should proceed. That is, it describes the general research strategy by answering questions such as which method, or combination of methods, is adequate for pursuing a specific research question. As Ackerly and True (2013, 137) put it, a methodology does not prescribe particular methods no matter the question, but is better understood as a framework which guides decisions at various stages during the research process. And thirdly, epistemology is concerned with a theory and analysis of what has epistemic value, e.g., under what conditions a belief qualifies as knowledge.

Consequently, if we want to apply RE as a method for the justification of a principle such as a precautionary principle, we need, firstly, to be clear about its epistemological foundations, e.g., what it means for a principle to be justified. Secondly, we need to develop a methodology that guides the specification and application of the method with respect to a specific objective, and in light of the underlying epistemology.

Chapter 2 describes the epistemological foundation of RE as a method of justification. It describes the conditions under which an epistemic position is justified through being in a state of reflective equilibrium. The chapter introduces the different elements that are part of RE—commitments, a system, background elements, and theoretical virtues—as well as the relations between them. As explained above in Sect. 1.1.1, I adopt a conception of RE from the works of Elgin, Brun, and Baumberger. While this conception is well elaborated, it is concerned primarily with the epistemological analysis of RE as an account of justification. As such, it does not tell us how the elements and criteria of RE should be specified for a particular project. For example, it tells us that we need a system, like a theory or a set of principles, and that this system should do justice to theoretical virtues. But questions such as which theoretical virtues are relevant for the development of a particular system is not inherent to RE as a general account of justification.

We thus need to develop a methodology that allows us to spell out RE as a specific method for a specific justificatory project, which is the main task of Chap. 3. This chapter develops guidelines for specifying RE as a method, and discusses methodological issues that need to be resolved. For the purpose of the present book, this is done with respect to two important restrictions, which will help to further sharpen the method: firstly, the method should be applicable by a single epistemic agent, and not require group deliberation or similar group processes. Secondly, I am concerned with reflective equilibrium as a method that aims at justifying principles, theories, or similar; and not, for example, as a method for finding a justified consensus, or as a decision framework. Chapters 2 and 3 thus contribute to desideratum (1), to be as explicit and precise as possible about the theoretical and methodological foundations of RE, and to show how RE can be specified as a method with respect to a particular aim.

Chapter 4 gives a survey about different interpretations of precautionary principles, proposed justifications for them, as well as objections and possible rejoinders. This contributes to desideratum (2), that RE should be applied to an actual, complex problem. The results from Chap. 4 also provide the basis to identify the input of the RE process, i.e., candidates for commitments, principles, and background elements of the RE application. This input, and the reasons for selecting it, is described in Chap. 5, which addresses desideratum (3), an explicit description of the initial position from which the process of adjustments starts. As part of Chap. 5, the method of RE is also specified with respect to the particular project of justifying a precautionary principle, which further contributes to desideratum (1).

Chapters 6–8 then present the actual case study. To cover as many aspects of applying RE as possible, the application is divided into three parts: Chap. 6 demonstrates how RE can guide systematization and theory development, Chap. 7 describes step-by-step how commitments and principles are adjusted in alternation, and Chap. 8 shows how a specific position can be spelled out and appraised with the RE criteria. Chapters 6–8 thus address desideratum (4), a traceable step-by-step application that makes it clear when adjustments are made and how they are defended. Chapter 8, in which the equilibration process comes to a preliminary end point, also addresses desideratum (5) by assessing the resulting position and discussing whether a state of reflective equilibrium was reached. With respect to precautionary principles, the application yields a rights-based precautionary principle.

Intermediate results are discussed after each chapter of the application. Chapter 9 summarizes them and discusses what we can learn from the case study for the applicability of reflective equilibrium as a method. It thereby addresses desideratum (6). As the case study shows, RE can be specified in a way that makes it applicable while being neither too demanding nor too permissive. In the case study, RE does put real constraints on the justification process, while allowing us to structure the process of adjustments and facilitating the search for further relevant considerations. However, the application in this book is also very complex, partly due to the fact that, as a case study, it has a strong focus on how RE can be applied. Not everyone might want to engage with the method at this level of detail. Thus, Chap. 9 also discusses how one can make use of the methodological benefits from RE without having to spell out every aspect in as much detail as is done in Chaps. 6–8.