Reflective Equilibrium as a Normative Empirical Model

Transcription

1 Reflective Equilibrium as a Normative Empirical Model Ghislaine J.M.W. van Thiel and Johannes J.M. van Delden University Medical Centre Utrecht, the Netherlands ABSTRACT. People who work and live in a certain moral practice usually possess a specific form of moral wisdom. If we manage to incorporate their moral intuitions in ethical reasoning, we can arrive at judgements and (modest) theories that grasp a moral experience that generally cannot be found outside the said practice. To achieve this goal, we need a legitimate way to balance moral intuitions, ethical principles and general theories. In the present contribution, we describe a version of the model of Reflective Equilibrium, which we call the Normative-Empirical Reflective Equilibrium (NE-RE). RE provides a framework for ethical thinking that includes both (practical) moral experience and normative theory. After an outline of the model, we focus on the role of empirical research and illustrate how empirical data on moral intuitions can add to the comprehensiveness of the set of moral beliefs in NE-RE. Subsequently, we describe how coherence among the elements of an NE-RE can be understood and measured. Finally, we address an important question for any method of moral reasoning: what is the status of the outcome of the reasoning process? This is a matter of justification. We argue for a so-called good reasoning-justified outcome strategy in NE-RE. This strategy is built on a reasoning process in which the reason-giving force of each element is tested and weighed. The thinker has to work towards a coherent view in which only the elements with sufficient justificatory power are retained. If the thinker decides that the elements fit together into a coherent view, a reflective equilibrium is reached. NE-RE is an attempt to describe a model for moral reasoning that can profit from the moral wisdom present among experienced agents through the use of empirical research. KEYWORDS: Reflective equilibrium, moral justification, normative empirical ethics, moral wisdom ETHICAL PERSPECTIVES 17, no. 2(2010): by European Centre for Ethics, K.U.Leuven. All rights reserved. doi: /EP

2 ETHICAL PERSPECTIVES JUNE 2010 I. INTRODUCTION Moral questions and dilemmas in everyday life prompt us to take a normative stance. Sometimes we rely on our moral intuitions and make judgements accordingly, while in other situations we feel the need for more extensive deliberation on a moral case. After a judgement is made, we have to ask: can we justify our moral view vis-à-vis others who may have come to a different judgement? Ethicists have long tried to describe fundamental moral principles from which justified judgements can be derived. However, it seems that general moral principles alone cannot generate the justificatory power that is needed for moral decision making. Progressively developing action-guiding principles and rules requires some kind of interpretation or specification of such principles (Richardson 2000, 304). Currently, most ethicists hold the view that theory and practice should mutually influence each other in the process of searching for reliable moral judgements and theories. Attention to practice has raised questions about the way elements such as moral experience and empirical data can be integrated in ethical reasoning. Our aim is to present a model for moral reasoning that can facilitate this integration. Our starting point is Reflective Equilibrium theory (hereafter RE). In this paper, we present a version of RE modified for the combination of empirical and normative elements. II. REFLECTIVE EQUILIBRIUM THEORY RE was developed by John Rawls for the theoretical purpose of formulating the most appropriate conception of justice (Rawls 1999, xiii). It is a model for justification that revolves around the idea that seeking justified moral principles requires an argumentative process in which general principles and background theories are considered together with a person s considered moral judgements. The expression RE refers to both the process and the result of moral reasoning. 184

3 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL An RE as the result of moral reasoning is a coherent and interconnected set of moral and non-moral beliefs at various reflective levels. The number of beliefs that can be included in the reasoning process of a researcher is necessarily limited, but ideally, in working towards an RE all relevant considerations are addressed. In broad outline, moral reasoning according to RE proceeds in three stages. A person the thinker starts with identifying his or her considered moral judgements or moral intuitions. In theory, the starting point of reasoning can be any of the elements mentioned, but often moral reasoning is triggered by spontaneous and personal interpretation of the facts of a case. The following step is to formulate moral principles that are relevant for the situation under consideration. It is unlikely that these two steps will produce a coherent system of beliefs. Therefore, the thinker will have to respond to the divergence between the principles and his or her considered moral judgements. He or she has to work back and forth between principles and judgements and make adjustments in both their considered moral judgements and their moral principles. This process ends when the thinker accepts a set of principles that coheres with his or her considered moral judgements. The person s beliefs are now said to be in reflective equilibrium. Rawls idea was both welcomed and criticized. One major criticism was and still is the alleged subjectivism inherent in the method. RE would amount to no more than a neat systematisation of the preliminary ideas a thinker has about a given case. In his influential 1979 article, Daniels acknowledges the problem of circularity and suggests that we speak of narrow reflective equilibrium when considered moral judgements and principles are made coherent. He then introduces a third phase of the method, in which the thinker attempts to disrupt the state of narrow reflective equilibrium by considering background theories and alternatives to his or her moral theory (Daniels 1979, 258). Examples of background theories include a theory of personhood or a general social theory. These background theories have to be chosen for their potential to provide 185

4 ETHICAL PERSPECTIVES JUNE 2010 critical input. Again, mutual adjustment of the elements is required until the final result, namely a wide reflective equilibrium, is achieved. Contrary to other approaches in ethics, RE claims no locus of certainty in one of the elements that is brought into the process: there is no privileged type of proposition. The thinker s task is to mould all relevant propositions into a coherent view. The overriding character of the result depends on the strength of argumentation in the reasoning process and on how well it stands up to the consideration of alternatives. The approach of moral theorising, characteristic for RE, liberates ethics researchers from the idea that moral judgments either come from theory or from practice, never from a combination of both. It implies that all considerations and beliefs that are part of the final RE were included for their own justificatory power and amended to fit the context. There are several possible interpretations of RE (Rawls 1999, 43). Many authors besides Daniels have suggested changes to the type and number of considerations that can be included. We offer some examples here by way of illustration. Nielsen suggests that we only allow judgements that are actually agreed upon within a community to take the place of considered moral judgements (1982, ). Beauchamp and Childress propose that we extract considered moral judgements from the common morality and thus include those judgements that all serious moral persons share (2008, 3). Heeger and Van Willigenburg add morally relevant facts as a separate element to be included in our reasoning towards RE (van der Burg and van Willigenburg 1998, 14). Van der Burg adds ideals to the list (1997, 27). RE is the reference theory from which we developed normative-empirical reflective equilibrium (NE-RE; van Delden and van Thiel 1998). Broadly speaking, the strength of an RE depends on three aspects: a) the comprehensiveness of the set of beliefs, b) the ultimate level of coherence among beliefs, and c) the strength of argumentation in the reasoning process. In the present contribution, we will explain how the use of empirical research in NE-RE adds to the comprehensiveness of the result. Moreover, we 186

5 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL offer a practical account of how coherence in RE can be understood and measured. Finally, we address the role of good reasoning for the justificatory power of a moral view that is the result of NE-RE. III. COMPREHENSIVENESS: EMPIRICAL RESEARCH INTO MORAL INTUITIONS We are committed to the idea that the moral experience of agents other than the thinker can enrich ethical reasoning. People who work and live in a certain moral practice have experiences that are generally not found among those outside that practice. These experiences amount to specific moral wisdom, which can be defined as expert-level knowledge and judgement in the fundamental pragmatics of life. This knowledge entails insights into the quintessential aspects of the human condition and human life, including its biological boundaries, cultural conditioning, and intra- as well as interindividual variations (Staudinger and Baltes 1996, 747). It requires the ability to combine knowledge, reflection and life-experience with emotional, social and ethical capacities (Edmondson and Pearce 2007, 233). Moral wisdom is influenced by formative experiences (DePaul 1993, 145-6). A formative experience causes a person to adopt a subtly, or even radically, different moral perspective. It can be provoked by a work of art or literature. However, activities that can lead to a formative experience usually influence a person s moral outlook over time. Thus, people who have a considerable amount of relevant experiences are more likely to have gained moral wisdom than those with less experience. Ethical reflection can benefit from the views of experienced agents because they add to the complexity of reasoning (Woods 1999, 427). For RE, this means that the set of moral beliefs can be enriched when the thinker not only considers his or her own initial reactions to a given case, but also those that are held by others with relevant experience. According to the current understanding of RE, it offers no place for considered moral judgements of other agents than the thinker. The 187

6 ETHICAL PERSPECTIVES JUNE 2010 considered judgements of the individual thinker are those taken up in the reasoning process towards RE. This is necessarily a limited source. The moral wisdom we think is so valuable for ethical reasoning is in many cases out of reach for the ethicist. However, if we incorporate the moral views of experienced agents, we would increase the comprehensiveness and thus the strength of an RE. NE-RE is designed to incorporate empirical data on the moral beliefs of other (groups of) agents. It differs from the original model of RE in two ways: a) the concept of considered moral judgements is replaced by the concept of moral intuitions, and b) empirical research is used to provide data for RE reasoning. Moral Intuitions instead of Considered Moral Judgements When Rawls talked about considered moral judgements (CMJ), he meant the moral judgements that seem clearly to be correct under conditions conducive to making good judgements of the relevant kind (Scanlon 2003, 140). In his earlier work on RE, Rawls described concrete criteria and demands that need to be met by the person holding a CMJ and the circumstances under which he or she came to hold it (1951, 178-9). In later work, however, Rawls does not elaborate on these requirements any further. The concept of CMJ has been criticised for the vagueness of the requirements that need to be met. It seems that the thinker should not only be unaffected by conditions that threaten his or her ability to exercise his or her sense of justice, he or she must also have the will or the intention to reach the correct decision (Rawls 1999, 42). For the practical use of RE, identifying a thinker s or other person s CMJ would require that we rule out the judgements made under non-ideal circumstances. Moreover, the thinker s desire to reach a correct decision has to be verified. These requirements are formulated with the aim to warrant a minimal level of credibility of CMJ. We suggest that the reliability of initial judgements should not be determined before they enter the reasoning process. 188

7 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL This implies that we do not need criteria to warrant the reliability of initial beliefs at the start of reasoning. Instead, we can look for a less problematic concept that describes the initial reactions of people when confronted with a moral case. Moral intuitions are seen as an alternative to CMJ (van der Burg and van Willigenburg 1998, 14). There are several reasons to prefer this concept over CMJ. First, it is appropriate to characterize the beliefs that a person comes to hold without extensive deliberation. The appearance of an intuition is usually rather sudden and not deliberated. Second, the concept of moral intuition offers an account of the type of belief we think is relevant to NE-RE. Moral intuitions are the preliminary interpretations of people that give the holder a sense of the direction in which a judgement about a given case should go (Haidt 2001, 818). There is evidence for the claim that most of our judgements are intuitive and automatic responses to challenges, elicited without awareness of underlying mental processes (Musschenga 2008, 132). It is thus highly likely that the moral judgements of people who work and live in a certain practice, are usually at an intuitive level. It is through these intuitions, therefore, that moral theorising can gain access to detailed information on the moral experience of relevant agents. If we incorporate the intuitions of doctors and nurses, for example, we can work towards an RE that grasps a moral experience that generally cannot be found among people outside the health care practice. Empirical Study in Normative Empirical Reflective Equilibrium (NE-RE) The relevance of empirical research in bioethics is emphasized by the growing incidence of the application of research methods in the social sciences to the direct examination of issues in bioethics (Sugarman 2004, 226). However, the question of the role empirical research is able to play in the process of ethical clarification and decision-making remains at the 189

8 ETHICAL PERSPECTIVES JUNE 2010 forefront of discussions on the subject. In NE-RE, we see several possibilities for valuable empirical input. For the purposes of moral judgement and theory construction, empirical information can be obtained in studies that are guided by non-normative questions. Such empirical data can play a role in so-called reality-revealing questions. These are questions describing the context in which a moral decision has to be made (who, what, when, where, why, how; Borry et al. 2004, 44). Moreover, ethicists have performed empirical studies not only to get the facts right, but also to see how a moral principle works in practice, or as a heuristic tool (Molewijk et al. 2004; Schermer, 2001). In NE-RE, we see a role for empirical research in obtaining relevant moral intuitions. We will clarify and illustrate the aforementioned roles and goals of empirical research in NE-RE, and refer to empirical studies we conducted in the past. Empirical research as prerequisite for serious ethical debate: The first role of empirical research in NE-RE is to provide the thinker with factual information on the state of affairs in a specific practice. This type of empirical research will not answer the ought questions of bioethics, but it is essential in providing sound answers to normative questions. Our study of end-of-life decisions in the care of mentally handicapped people can serve as an example (van Thiel and van Delden, 1997). This study provides empirical information that can be used in moral reasoning on the ethical acceptability of such decisions. For example, to decide whether physicians, as some people fear, are sliding down a slippery slope towards widespread euthanasia, several series of empirical data are needed over time. It is important to note that this information is open to analysis from both moral and non-moral perspectives. Facts about varying hysterectomy rates within different ethnic groups, for example, are important to those who want to discuss the impact of cultural differences on medical decision-making. In NE-RE, this type of empirical data can be brought into the reasoning process as morally relevant information, the role of which is to 190

9 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL inform the thinker about the facts and circumstances in which a moral judgement has to be reached. Empirical research in the light of a normative question: Another appearance of the empirical in RE is one in which the data is obtained in the light of a normative research question. Elements of moral theory are often incorporated in the empirical studies that yield this type of information. Consequently, the results are linked to moral values. This is the type of research that ethicists usually talk about when they describe empirical ethics research. The research itself can be descriptive, but it cannot be separated from the normative context in which it was collected. The empirical data can have a heuristic function or identify problems associated with moral principles in practice. In NE-RE, this type of empirical information can provide the thinker with morally relevant facts about the effects of particular interventions, aimed at improving how clinicians, for example, attempt to meet an ethical obligation (Sugarman 2004, 226). We will illustrate this type of research by referring to our research into the conduct and opinions of nurses regarding respect for patient autonomy. The moral principle of respect for autonomy is generally accepted as a core-principle in health care ethics. The dominant interpretation of respect for autonomy in health care can be summed up in two categories: informed consent and respect for privacy. We assumed that this principle might induce ethical questions in the context of nursing home care because a significant portion of the population has a diminished capacity for autonomous decision-making. We decided to collect empirical information on the principle and practice of respect for autonomy in the care of nursing home residents. We conducted a study in which nurses were asked for their opinion on this principle and their compliance to the criteria for nursing home care that were derived from documents of the association of Dutch nursing homes. Although the nurses declared en masse that they adhered to the principle of respect for autonomy, we found that they tended nevertheless to fall short 191

10 ETHICAL PERSPECTIVES JUNE 2010 on the implementation of the criteria. Other results revealed that the caregivers encountered not only practical, but also fundamental problems in trying to meet the criteria of respect for autonomy. Of the respondents, 47% answered in the affirmative to the proposition that a caring and attentive attitude towards patients is more important than emphasizing and supporting autonomy (van Thiel and van Delden 1997a, 183). Empirical study of moral intuitions: The third role of empirical research is to enrich the thinker s set of moral intuitions. We argued above that the thinker is only a limited source of moral intuitions because specific moral wisdom is present among experienced agents who work in a given moral practice. DePaul acknowledged this limitation and advised thinkers to expand their moral views by looking at works of art, reading literature and going to the theatre (1993, 144). In addition to these activities, we propose the use of empirical research methods to shed light on the moral intuitions of groups of persons. The moral intuitions that are identified through this type of research can be integrated in the set of beliefs the thinker is considering. The status of these moral intuitions is the same as that of any other element in NE-RE: equal to the others and not immune to revision. The challenge for empirical research into moral intuitions is to employ methods that are suitable for the purpose of identifying intuitions. In previous work, we tested two strategies that may be appropriate for this aim. In a study in nursing homes, we sent questionnaires to one hundred nurses and fifty physicians offering a description of four different views on patient autonomy in the nursing home context. In addition, we designed ten short case descriptions with four options for a response to the case. Each response was based on one of the four interpretations of the principle of respect for patient autonomy and contained a suggestion for a reaction on the issue at stake. The respondents were asked to indicate their preferred view on autonomy and one of the four options with 192

11 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL respect to each case description. We explicitly asked them to choose the option that best reflected their normative ideas, and not that which most adequately described everyday practice. In this way, we tried to obtain empirical information on the moral intuitions of caregivers regarding respect for the autonomy of nursing home residents (van Thiel and van Delden 2001). More recently, we employed a qualitative empirical study to explore the moral intuitions of the general public in relation to the so-called Ashley case. Ashley is a young and severely disabled American girl. When she was 6 years old, she was given medical treatment to stop her growth. In addition, a hysterectomy was performed to avoid problems with menstruation and the growth of Ashley s breasts was arrested. For the purpose of assessing the way Ashley had been treated, we performed a qualitative study in order to access the moral experience of other people. We decided to search for intuitions among the postings on internet Message Boards that were opened for discussion on this case. We extracted 438 pieces of text from a total of 3004 postings. Three criteria guided the selection: a) the text comprised an interpretation of the case, b) the writer expressed a judgement in terms of right, wrong, acceptable, praiseworthy etc., and c) the text was not a lengthy and well-thought out discussion of the case. We hoped that the application of these criteria would allow us to extract moral intuitions from the contributions. The empirical research presented here shows that data of both quantitative and qualitative nature can be relevant for RE. Moreover, a variety of methods of inquiry can be used. The thinker should decide which empirical input is relevant and how it can be obtained on a case-by-case basis The thinker who wants to produce a reflective equilibrium has to consider empirical elements together with normative principles and background theories. In this process, the thinker aims for coherence among all relevant considerations. 193

12 ETHICAL PERSPECTIVES JUNE 2010 IV. THE SEARCH FOR COHERENCE There is no clear description of what makes a set of beliefs coincide with a reflective equilibrium. Rawls speaks of the mutual support of many considerations, of everything fitting into one coherent view (1971, 21). For the practical use of RE we need a specification of what coherence is and how it can be measured. The Concept of Coherence Coherence is valued because it can be seen as a property of an information set that boosts our confidence that its content is true (Bovens and Hartmann 2003, 601). However, the nature of coherence and the way people should evaluate their beliefs with respect to coherence is poorly described (Beauchamp and Childress 2008, 387; DeGrazia 2003, 221; Rauprich 2008, 52). Intuitively, coherence is a matter of how well a body of beliefs hangs together : how well its components produce an organized, tightly structured system of beliefs, rather than a haphazard collection or a set of conflicting subsystems (BonJour 1985, 93-5). Rawls does not offer us a lead regarding the exact meaning of coherence. Others have suggested several options. We outline some of the more comprehensive views: Daniels speaks of RE when the best fit or the strongest mutual support among the components is achieved (Daniels 1996). Things fit well when there are at least no inconsistencies. However, on any reasonable coherentist account, coherence is more than mere logical consistency. BonJour illustrates this point with two sets of propositions, A and B. A contains this chair is brown electrons are negatively charged today is Thursday. B contains all ravens are black this bird is a raven this bird is black 194

13 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL Clearly both sets of propositions are free of contradiction. But in the case of A, this consistency results from the fact that its components are almost entirely unrelated to each other. Although they are not in conflict, they also fail to be positively related in any significant way (BonJour 1985, 96). What is needed for meaningful coherence is substantial mutual support between the elements of a set of beliefs (Rauprich 2008, 66). To specify coherence as substantial mutual support, we can use the notions comprehensiveness and interconnectedness (Sayre-McCord 1996, 166). Comprehensiveness represents the aim of RE reasoning to consider as many relevant beliefs as possible. The notion of interconnectedness can be specified by referring to inference relations (BonJour 1985, 96). These are the positive connections between beliefs that constitute coherence. Each individual belief can be involved in many different inferential relations and the degree to which this is so represents a determinant of coherence. Measuring Coherence The investigation of the inferential connections between elements is one way to measure coherence in a specific set of beliefs. Thagard points out four types of coherence, which are useful for a thinker in NE-RE in deciding whether his or her set of beliefs can qualify as a reflective equilibrium (1998, ). In summary: a) Explanatory coherence: this is the coherence between observation and understanding. The importance of this type of coherence lies in the fact that some normative principles are tied to empirical claims. Guarini gives the example of capital punishment. The argument supporting a general principle that capital punishment is acceptable may depend on its presumed deterrent effect. But whether capital punishment actually has a deterrent effect is a largely empirical question (2007, 29). b) Deductive coherence is the coherence between principles and judgements. There is a positive connection if a principle and a particular judgement are both likely to be either accepted or rejected. Positive connections between principles and judgements are constituent parts of coherence. 195

14 ETHICAL PERSPECTIVES JUNE 2010 c) Deliberative coherence is the coherence between actions and goals. The positive deliberative connection is when an action facilitates a goal. Incompatibility of an action with a goal is a negative constraint between elements (Guarini 2007, 30). d) Analogical coherence. Arguing for moral principles and judgements analogically implies supporting a conclusion in one case by comparing it to a similar case whose moral status is more obvious. Analogical arguments are rarely convincing on their own, but they can contribute to the overall coherence of a view. The strength of an RE depends largely on the level of coherence of the result. RE thinking should be directed at coherence throughout the process of reasoning. The claim that coherence requires more than internal consistency demonstrates the need for a sophisticated approach. In the process of reasoning, the thinker has to strive for a set of beliefs that is not only consistent, but also comprehensive. This aspect of coherence fits well with the overall goal in RE to consider the broadest set of moral and non-moral beliefs. Our version of the model (NE-RE) adds to the comprehensiveness of the set of beliefs through empirical investigation of moral intuitions of other agents than the thinker him or herself. Interconnectedness refers to the relations among beliefs that can either be strong or loose, as in BonJour s example of propositions. To measure coherence in a set of beliefs, a thinker can investigate the connections between several types of elements of the set. Thagard s description of four types of coherence is useful for this task. A thinker in RE can measure the positive and negative connections between a) observation and explanation, b) principles and judgements, c) actions and goals and d) cases. The more and the stronger the inferential connections, the higher the level of coherence. There is no clear cut-off point for an RE. However, a thinker can evaluate the inference relations between his or her beliefs. Some will have many positive connections, and others will only have a few. For a reflective equilibrium, the beliefs that are situated at the heart of the system 196

15 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL should be positively connected to each other. Moreover, the requirement of comprehensiveness means that even though a small set of beliefs may be free of inconsistencies, the thinker has to make an effort to keep all beliefs he or she deems relevant aboard the reflective equilibrium. Thus, the thinker should not readily accept a small but consistent set of beliefs. Instead, he or she has to have good reasons to dismiss beliefs. In a transparent reasoning process, the thinker can justify his or her choices and be criticized for it by others. V. JUSTIFICATION: CREDIBILITY AND GOOD REASONING Suppose a person has considered a comprehensive set of beliefs at different levels of reflection, tested and adjusted these beliefs and came up with a moral view based on a consistent and interconnected whole. We may conclude that this person rightly claims that his or her beliefs are in reflective equilibrium. However, the important question as to whether the resulting perspective on a given moral case is justified remains unanswered. This is a matter of moral justification. A familiar objection against RE is the no-credibility objection (DePaul 1993, 23-27). The argument states that if the justification of the result of an RE process is based on the coherence of a set of beliefs, each of these beliefs has to be reliable enough to guide the process of reasoning. Moral intuitions are not sufficiently trustworthy to fulfil this function because they are open to error. To counter the no-credibility objection, some authors suggest the stringent selection of initial judgements at the start of the reasoning process in order to prevent the bad ones from getting in (Singer 2005). Unfortunately, this strategy always results in a dramatic reduction of the number of initial judgments that are allowed in RE. This conflicts with the ideal that we consider the broadest set of beliefs in RE. Moreover, the demand for credible moral intuitions might be contrary to the dynamic 197

16 ETHICAL PERSPECTIVES JUNE 2010 character of the method of RE. A significant role of moral intuitions is to fuel the thinking process of the thinker. Selection at the start of reasoning can hamper this function of intuitions. Another approach to the no-credibility objection is to move away from the discussion concerning the characteristics of moral intuitions and focus on the argumentative process. This strategy is employed by DePaul (1993). We have developed his ideas further and argue for the good reasoning-justified outcome strategy for moral justification in RE (van Thiel and van Delden 2009) Good Reasoning Justified Outcome Strategy Following this strategy, the thinker starts by identifying the broadest set of relevant moral intuitions. Typically, the moral views of people other than the thinker him or herself are taken up to enrich the initial set of moral intuitions. Qualitative empirical work designed to obtain these intuitions is usually necessary. In the subsequent process of moral reasoning, moral intuitions, principles and theories can gain or lose justificatory power. RE provides a model in which they are examined together, adjusted, accepted or expelled. The guiding principle of examination is the level of coherence among different beliefs. The thinker will try to achieve coherence by mutual adjustment of beliefs. In the end, only the beliefs with sufficient justificatory power (derived from their coherence with other elements) are part of RE. The moral intuitions in this RE can be considered to have sufficient credibility, because they were tested and confirmed in the reasoning process towards RE. In the good reasoning-justified outcome strategy, much importance is bestowed upon the argumentative process. To achieve an equilibrium that has strong justificatory power, the reasoning process and the accompanying attitude of the thinker should meet several criteria. First, clarity concerning facts and arguments adds to the justificatory power of RE, because 198

17 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL the normative force of the outcome of RE depends in part on the strength of the reasons that have featured in the process (Holm 2008, 13). A transparent mode of reasoning is therefore important. Second, the thinker should avoid getting caught up in his or her own intuitions by taking on an attitude of openness. This requires in the first instance that he or she is aware of potential biases and motivated to correct them. Second, the thinker should deliberately try to seek alternative ways to interpret the moral aspects of a case. Having an open mind with regard to the moral intuitions of other people is an element of this attitude. Finally, a thinker in RE is interested in reasonable judgements. He or she is motivated to find and retain considerations that others cannot reasonably reject. This implies that a reasonable thinker is sensitive to the perspective of other agents who are similarly motivated to find good reasons for actions (Scanlon 1998, 171). In this way, reasonableness adds to the justificatory power of an RE. VI. CONCLUSION The normative empirical version of the RE model enables a thinker to incorporate moral views of others into his or her reasoning process. We defend the idea that this is valuable, because it is likely that the moral experiences of people who are involved in a given moral practice differ from those of people outside the practice. The moral experience of people is expressed in their moral intuitions. Therefore, we took on the challenge of describing a version of RE in which (empirical) information on moral intuitions of (groups of) people is taken up. We called this NE-RE (the Normative Empirical Reflective Equilibrium). The role of moral intuitions in moral reasoning is the subject of a longstanding debate in ethics. Many authors raise significant concerns with regard to the reliability of moral intuitions. Contrary to many responses to this problem, we move away from the search for criteria to 199

18 ETHICAL PERSPECTIVES JUNE 2010 guarantee sufficient credibility from the outset of NE-RE reasoning. Instead, we follow DePaul in his focus on the reasoning process towards RE. We describe characteristics of good reasoning to outline a process in which the reason-giving force of considerations (empirical or normative) is determined. Each element can gain or lose justificatory power in this process. If the thinker decides that the elements fit together into a coherent view, a reflective equilibrium is reached. The level of justificatory power of the elements may vary, but the resulting RE can be defended against others. Therefore, we think that NE-RE is a methodologically sound account of a model for moral reasoning that can profit from the moral wisdom that is present among experienced agents. WORKS CITED Beauchamp, Tom L. and James F. Childress Principles of Biomedical Ethics. 6 th revised edition. Oxford: Oxford University Press. BonJour, Lawrence The Structure of Empirical Knowledge. Cambridge, MA: Harvard University Press. Borry, Pascal, Paul Schotsmans and Kris Dierickx What is the Role of Empirical Research in Bioethical Reflection and Decision-Making? An Ethical Analysis. Medicine, Health Care and Philosophy 7: Bovens, Luc and Stephan Hartmann Solving the Riddle of Coherence. Mind 112: Daniels, Norman Wide Reflective Equilibrium and Theory Acceptance in Ethics. The Journal of Philosophy 76: Daniels, Norman Justice and Justification. Cambridge, Cambridge University Press. DeGrazia, David Common Morality, Coherence and the Principles of Biomedical Ethics. Kennedy Institute of Ethics Journal 13: DePaul, Michael R Balance and Refinement. Beyond Coherence Methods of Moral Inquiry. London: Routledge. Guarini, Marcello Computation, Coherence, and Ethical Reasoning. Minds & Machines 17: Haidt, Jonathan The Emotional Dog and its Rational tail: a Social Intuitionist Approach to Moral Judgement. Psychological Review 108: Holm, Soren Background paper on Article 14 of the Universal Declaration on Bioethics and Human Rights from a philosophical perspective. Report of the Meet- 200

19 VAN THIEL AND VAN DELDEN REFLECTIVE EQUILIBRIUM AS A NORMATIVE EMPIRICAL MODEL ing of the working group of IBC on social responsibility and health (Annex V). Unesco. Molewijk, Albert C, Anne Stiggelbout, Wilma Otten, Heleen Dupuis and Job Kievit Empirical Data and Moral Theory. A Plea for Integrated Empirical Ethics. Medicine, Health Care and Philosophy 7: Musschenga, Albert W Moral Judgement and Moral Reasoning: a Critique of Jonathan Haidt. In The Contingent Nature of Life. Bioethics and the Limits of Human Existence. Edited by Marcus Düwell et al., Dordrecht: Springer. Nielsen, Kai On Needing a Moral Theory. Metaphilosophy 13: Richardson, Henry S Specifying, Balancing and Interpreting Bioethical Principles. J Med Philos 25: Rauprich, Oliver Common Morality: Comment on Beauchamp and Childress. Theoretical Medicine and Bioethics 29: Rawls, John Outline of a Decision Procedure for Ethics. Philosophical Review 60: Rawls, John A Theory of Justice. Revised edition Cambridge, MA: Harvard University Press. Sayre-McCord, Geoffrey Coherentist Epistemology and Moral Theory. In Moral Knowledge? New Readings in Moral Epistemology. Edited by Walter Sinnott-Armstrong and Mark Timmons, New York: Oxford University Press. Scanlon, Thomas M What We Owe to Each Other. Belknap: Harvard University Press. Scanlon, Thomas M Rawls on Justification. In The Cambridge Companion to Rawls. Edited by Samuel Freeman, Cambridge: Cambridge University Press. Schermer, Maartje H.N The Different Faces of Autonomy: a Study on Patient Autonomy in Ethical Theory and Hospital Practice. PhD diss., University of Amsterdam. Singer, Peter Ethics and Intuitions. Journal of Ethics 9: Sugarman, Jeremy The Future of Empirical Research in Bioethics. Journal of Law Medicine & Ethics 32: Thagard, Paul Ethical Coherence. Philosophical Psychology 11: van Delden, Johannes J.M. and Ghislaine J.M.W. van Thiel Reflective Equilibrium as a Normative-Empirical Model in Bioethics. In Reflective Equilibrium. Edited by Wibren van der Burg and Theo van Willigenburg, Dordrecht: Kluwer. van der Burg, Wibren The Importance of Ideals. The Journal of Value Inquiry 31: van der Burg, Wibren and Theo van Willigenburg, eds Reflective Equilibrium. Dordrecht: Kluwer. van Thiel, Ghislaine J.M.W. and Johannes J.M. van Delden, Keimpe de Haan and Alex K. Huibers Retrospective Study of Doctors End of Life Decisions in Caring for Mentally Handicapped People in Institutions in the Netherlands. British Medical Journal 315:

20 ETHICAL PERSPECTIVES JUNE 2010 van Thiel, Ghislaine J.M.W. and Johannes J.M. van Delden. 1997a. Dealing with Patient Autonomy in Dutch Nursing Homes. Health Care in Later Life 2: van Thiel, Ghislaine J.M.W. and Johannes J.M. van Delden The Principle of Respect for Autonomy in the Care of Nursing Home Residents. Nursing Ethics 8: van Thiel Ghislaine, J.M.W. and Johannes J.M. van Delden The Justificatory Power of Moral Experience. Journal of Medical Ethics 35: van Thiel Ghislaine, J.M.W Moral Wisdom in the Balance. Reflective Equilibrium as a Normative Empirical Model for Bioethics. PhD diss. Utrecht University. van Willigenburg, Theo Inside the Ethical Expert. Kampen: Kok Pharos. Woods, Martin A Nursing Ethic: the Moral Voice of Experienced Nurses. Nursing Ethics 6: