This item is only available as the following downloads:
Running Head: MANIPULATING METAETHICAL BELIEF S DOES FIRMNESS OF CONVICTION CAUSE OBJECTIVISM ABOUT A MORAL ISSUE? BY DAVID TURON A Thesis Submitted to the Division of Social Sciences New College of Florida in partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Steven Graham Sarasota, Florida January, 2013
MANIPULATING METAETHICAL BELIEFS ii TABLE OF CONTENTS Abstract ................................ ................................ ................................ ............................. iv Introduction ................................ ................................ ................................ ........................ 1 Morality and Moral Judgment ................................ ................................ .......................... 1 What is morality? ................................ ................................ ................................ ......... 1 Why study moral judgment? ................................ ................................ ........................ 2 Moral Dilemmas ................................ ................................ ................................ ............... 4 Definition and commonly used dilemmas. ................................ ................................ ... 4 The use of moral dilemmas in empirical psychology. ................................ .................. 5 The influence of external factors on judgments of moral dilemmas. ........................... 7 Moral Objectivism ................................ ................................ ................................ .......... 13 Why study judgment about moral objectivism? ................................ ........................ 13 Previous findings regarding judgments of moral objectivism. ................................ ... 14 The Present Study ................................ ................................ ................................ ........... 21 Method ................................ ................................ ................................ .............................. 2 3 Participants ................................ ................................ ................................ ..................... 23 Materials and Procedure ................................ ................................ ................................ 2 3 Results ................................ ................................ ................................ ............................... 2 6 Discussion ................................ ................................ ................................ .......................... 3 1 Why Did the Priming Manipulation Fail? ................................ ................................ ...... 33 Unintended priming effects. ................................ ................................ ....................... 3 3 Dilemma specific priming susceptibility. ................................ ................................ .. 3 4 Limitations of the Present Study ................................ ................................ .................... 3 8 Population pool and sampling methods. ................................ ................................ ..... 3 8 Construct validity. ................................ ................................ ................................ ...... 3 9 Directions for Future Research ................................ ................................ ....................... 41 References ................................ ................................ ................................ ......................... 4 3 Footnotes ................................ ................................ ................................ ........................... 4 6
MANIPULATING METAETHICAL BELIEFS ii i Appendix A : Supplemental Theoretical Discussion Regarding the Implications of Empirical Research for Meta Ethics ................................ ................................ .............. 4 8 How should we approach moral semantics? ................................ ................................ 4 9 Two Kinds of Outlooks in Moral Semantics ................................ ................................ 50 The monistic outlook. ................................ ................................ ................................ 50 The pluralistic outlook. ................................ ................................ ............................... 5 2 Problems for the pluralistic outlook. ................................ ................................ .......... 5 6 The Meta Ethical Theoretical Virtues ................................ ................................ ............ 60 Explanatory relevance. ................................ ................................ ............................... 6 2 Value for normative inquiry.. ................................ ................................ ..................... 6 3 Metaphysical tenab ility. ................................ ................................ ............................. 6 4 Psychological tenability.. ................................ ................................ ............................ 6 5 Summary and Additional Philosophical Issues ................................ .............................. 6 7 Appendix A References ................................ ................................ ................................ 70 Appendix A Footnotes ................................ ................................ ................................ ... 71 Appendix B : What is Meant by Objectivity ? ................................ ................................ .. 7 6 Appendix B References ................................ ................................ ................................ .. 7 9 Appendix B Footnotes ................................ ................................ ................................ .... 80 Appendix C : Materials ................................ ................................ ................................ .... 81 Section 1 Version A: Save Lives Condition ................................ ................................ ... 81 Section 1 Version B: Do Not Kill Condition ................................ ................................ 82 Section 2: PANAS ................................ ................................ ................................ .......... 8 3 Section 3: Moral Dilemma ................................ ................................ ............................. 8 4
MANIPULATING METAETHICAL BELIEFS iv DOES FIRMNESS OF CONVICTION CAUSE OBJECTIVISM ABOUT A MORAL ISSUE? David Turon New College of Florida 2013 ABSTRACT Atrocities are often seen as paramount cases of immorality events that involved some of the most surely immoral actions out of any imaginable. From the certainty about the immorality of those actions it seems that perceived objectivity the feature of being immoral regardless of any particular sentiment or sensibility follows. Is it actually the case, h owever, that firmness of conviction about the moral status of an issue causes the intuition that the conviction is objectively true? The present study was designed to address this question. Forty participants were primed to endorse one or another rule befo re being presented with a moral dilemma and prompted to make a judgment both about the dilemma and about the objectivity of that judgment The priming manipulation was not successful in sway ing responses to the moral dilemma and thus no causal inferences c ould be determined Theoretical explanations and future directions for research are discussed Thesis Adviser: Steven Graham Division of Social Sciences
MANIPULATING METAETHICAL BELIEFS 1 Introduction Morality and Moral Judgment What is morality? Throughout human history, philosophers have sought to answer t he question, In Western culture, the rigorous inquiry into the nature of morality goes at least as far back as Socrates The human practice of making moral judgments of course began much earlier than inquiries into its nature by philosophers. This might lead one to ask more grandiose questions: d id humans always make moral judgments? Do non human animals such as apes make moral judgments? How did the practice originate? These questions are certainly worthy of careful consideration, and indeed many have taken them up and expended much effort in working toward an answer. Although these questions are tangential they reveal a core a mbiguity that perv ades all inquiry about morality : what precisely is meant by the term moral judgment or the moral domain (the domain of action or judgment pertaining to morality)? T he definition of moral may be laden with particular attitudes about the c ontent of morality ; seeing some factor as morally relevant or irrelevant varies between different systems of moral attitudes. For example, one might define the moral domain as that which concerns the regulation of the infliction of harm on others. Though t he moral domain has often been thought to essentially involve the regulation of harm inflicted on others (Turiel, 1983) there has been an effort to expand the definition of the moral domain to include other kinds of attitudes, such as the moral revulsion toward harmless but disgusting acts Shweder (1990 ) approach, introduced a threefold division of moral
MANIPULATING METAETHICAL BELIEFS 2 codes. This division involved the formation of moral categories based on the kinds of reasons availa ble to justify moral judgments. Moral Foundations Theory, proposed by Haidt and Joseph ( 2004) expanded the number of categories to five harm/care, justice/fairness, authority/subversion, loyalty, and sanctity/purity. This approach was largely motivated by the recognition of the pitfalls of overly narrow definition s of morality; it would be a grave error to deny definition of morality that just so happens to align with the conception of morality within the culture belonging to the researchers Despite the difficulties in specifying how the moral domain may be distinguished from other domains of human behavior, the failure to do so would seem, ipso facto, to cause difficulties in distinguishing moral judgment from other forms of judgment. Although traditional assumptions regarding the distinction between morally relevant situations and morally irrelevant situations have come into question, it may be fair to retain a few modest assumptions regarding this distinction whilst remaining neutral with respect to the aforementioned concerns with making assumptions about what counts as morally relevant. Judgments regarding the permissibility of killing other humans h ave canonically been treated as moral judgments. 1 Judgments about whether something is just the right thing to do or whether an act is seen to tend to evoke guilt in people who commit the act canonically have been treated as judgments of moral rightness an d wrongness, respectively. It is quite modest to assume that each of these canonical examples count as morally relevant especially given the widespread assumption that they are morally relevant throughout the most recent literature in moral psychology.
MANIPULATING METAETHICAL BELIEFS 3 W hy study moral judgment? Much work in social psychology has focused on moral (and immoral) action rather than judgment. For example, studies such as ( 1963 ) famous experiment on obedience to authority as well as the host of literature on prosocial behavior shed light on the factors that shape the likelihood of acting in a way that is conventionally considered morally good or bad. 2 The empirical study of moral judgment serves distinct purposes from the study of moral action Whereas the study of moral action provides information pertaining to the processes by which people decide to follow moral injunctions, the study of moral judgment provides information pertaining to the processes by which people distinguish between morall y acceptable and unacceptable behavior. The study of moral judgment might thus be said to shed light on what the content of morality is, as understood by people. Although t his question has been the subject of moral philosophy for mi llennia t he study of the psychology o f moral judgment is distinct as an area of inquiry from moral philosophy. Moral philosophy is often divided into two categories : substantive ethics and meta ethics. Substantive ethics involves argumentation and reasoni ng about what ought to be done (through the formulation of moral theories and principles). Substantive ethics, therefore, is about the content of morality. Meta ethics by contrast involves philosophical argumentation about the nature of morality. Although both meta ethics and the M eta ethicists have sought to shed light on moral semantics and linguistics, moral metaphysics, moral epistemology, definitions of morality, formalizations of moral logic
MANIPULATING METAETHICAL BELIEFS 4 and the nature of psychological states belonging to the moral domain. All of these meta ethical tasks stand much to gain from the evidence gene rated from the findings of empirical psychologists and may potentially also be help ful for psychologists in the process of design ing studies. However, they primarily involve the development of the conceptual underpinnings of ethics or morality via the use of reasoned arguments and thus are not properly empirical endeavors. The psychology of moral judgment, in contrast to both meta ethics and substantive ethics the actual conception of morality as employed by humans There are many kinds of questions pertaining to the psychology of moral judgment (e.g., the development of moral judgment, the evolutionary or cultural origins of moral codes and the psychological processes involved in the formation of moral judgments ) Studies on the psychology of moral judgment are typically carried out by presenting some moral quandary to participants and asking the participants to report whether a particular action is morally right or wrong. Such mor al quandaries are often (but not always) presented in the form of a hypothetical moral dilemma. Moral Dilemmas Definition and commonly used moral dilemmas. A moral dilemma is a scenario in which moral considerations come into conflict with other moral con siderations. Some moral dilemmas instruct the readers to imagine themselves in the scenario depicted, whereas other moral dilemmas present some series of events that do not involve the participant. Moral dilemmas also sometimes present a particular choice of action for participants to judge but sometimes instead ask the
MANIPULATING METAETHICAL BELIEFS 5 participants to decide which choice of action they prescribe for the scenario. The excerpts below are some examples of moral dilemmas that are often utilized by psychologists Each of these dilemmas involve the choice between letting many people die and causing the death of a fewer number of people in their place. Despite this similarity between these dilemmas, people tend to be more willing to sacrifice some people to save a larger number in the first but not the second two dilemmas. The trolley dil emma (taken from Bartels, 2008). In the path of a runaway tr ain car are five railway workmen who will surely be killed unless you, a bystander, do something. If you flip a switch, the train will be diverted onto another track, where it will kill a single railway workman. In this situation, would you flip the switch ? The footbridge dil emma (taken from Bartels, 2008). In the path of a runa way train car are five railway workmen who will surely be kille d unless you, a bystander, do somethin g. You are standing on a pedes trian walkway that arches over the tracks next to a large stranger. Your body would be too light to stop the train, but if you push the stranger onto the tracks, killing him, his large bod y will stop the train. In this situation, would you push him? The surgery dilemma (taken from Bartels, 2008) You are a surgeon with a number of patients. Five of them need organ transplants. Each of them needs a different organ or they will surely die. Y ou have another patient who is healthy and would be an ideal organ donor for the others. If
MANIPULATING METAETHICAL BELIEFS 6 you transplant his organs (against his will) into the bodies of the other patients, they will live but he will die. In this situation, would you perform the transpl ant? The use of moral dilemmas in empirical psychology. Much of the work in moral psychology since Lawrence Kohlberg has involved the use of moral dilemmas in the form of hypothetical vignettes that participants are asked to judge. There are numerous advantages and disadvantages to this approach. Using hypothetical moral dilemmas is useful because it provides a way to study judgments about some of the most intense ly morall y troubling situations situations that by their very nature cannot be staged for the purpose of research either for ethical or practical reasons It is also useful because it allows researchers to present issues in a controlled, uniform way in order to t est for subtle influences, such as the influences of wording or framing effects that could not as readily be controlled in an analogous way in a staged enactment of a morally troubling situation. Since moral judgment, as distinct from moral actio ns, can be expressed by people through written language, researchers may measure moral judgments straightforwardly simply by asking participants what they believe about some vignette Unfortunately this methodological approach is not without its drawbacks A sking a participant to report what she or he would do in a hypothetical situation requires her or him to project her or himself into the situation and predict how she or he would act. actually align with what those people would do in such situations. Additionally, one might be concerned with the ecological validity of the usage of hypothetical moral dilemmas as tools to study moral judgment; it is unclear whether the presentation of a m oral dilemma
MANIPULATING METAETHICAL BELIEFS 7 in the form of text on a page or screen yields different moral judgments than the presentation of a moral dilemma in concretely instantiated forms. Some (e.g., Navarette, McDonald, Mott, & Asher, 2012 ) have attempted to develop a method that a voids this stipulated drawback by simulating moral quandaries in more visceral ways than through text on a page or screen such as through a constructed imme rsive virtual reality program Unfortunately this alternative method is far more difficult to utili ze than the traditional method of presenting text to participants and asking them to imagine the story represented in the text. A certain degree of m odest y in conclusions drawn from resea rch may be sufficient to avoid the skeptical pull of the se aforementioned concerns Systematic co variance between moral judgment and various factors within the context of circling numbers after reading text that portrays a hypothetical scenario may be quite informative abo ut the nature of moral judgment ; the fin dings of such studies may be interesting in their own right, even if they do just pertain to making moral judgments of this removed textual form. Although concerns about ecological validity may prevent researchers from generalizing their findings to all c ontexts in which moral judgment occurs, they do not invalidate the conclusion that such findings reveal some features of some moral judgments. The influence of external factors on judgments of moral dilemmas. Moral attitudes are not constant or static. It moral attitudes may shift throughout life. For example, as one ages one may come to avow the permissibility of capital punishment by recognizing the value of the reasons in favor of that view. However, a number of studies b y social psychologists have revealed
MANIPULATING METAETHICAL BELIEFS 8 that moral attitudes may even be transiently influenced by various situational factors created by experimenters (e.g., priming ) In making judgments regarding moral dilemmas, people do not always simply recollect genera l moral principles that are stored from past endorsements, but at least sometimes choose to endorse general principles on the basis of situational factors. Unrelated emotions appear to influence some moral judgments (particularly, moral judgments that are more emotionally charged) Valdesolo and DeSteno (2006) studied how positive affective stat es affect judgment of moral dilemmas. They first randomly assigned p articipants to observe a stimulus to induce positive (in one condition) or to induce neutral affect (in the other condition) and then immediately presented all of the participants with both the footbridge dilemma and the trolley dilemma. Participants tended to report being in a more positive affective state wh en p resented with the positive clip. The participants who were presented with a positive stimulus, but not the neutral stimulus, tended to advocate a utilitarian response (advocating the sacrifice of one in order to save five) to the footbridge dilemma by a factor of 3.8. No such effect occurred for responses to the trolley dilemma Thus it was concluded that people tend to endorse utilitarian judgments more often whe n levels of affect are higher. Valdesolo and DeSteno explained this difference between the two dilemmas with the claim that the footbridge dilemma is more emotionally charged than the trolley dilemma. E xternal factors even when people explicitly deny those factors as relevant to their moral judgment Uhlmann, Pizarro, Tannenbaum, et al. (2009) f ound that the endorsement of moral principles is at least sometimes politically motivated, even if people are not aware of this bias
MANIPULATING METAETHICAL BELIEFS 9 identification related to how they would respond to the race of a person in a hypothetical moral dilemma. They asked participants whether it would be justified to sacrifice one person to save 100 people The y collected data on self reported political orientation and treated that as a participant variable as well as randomly assigned the participants to one of two conditions. In one condition, the name of the one person being sacrificed was and in the other condition, the name was For the former condition, the 100 people who woul d die if the one person was not sacrificed consisted of the New York Philharmonic and for the latter condition, the 100 people consisted of the Harl em Jazz Orchestra. Liberals ( specifically defined as 1 standard deviation below the mean self reported political orientation) were less willing to endorse the sacrifice if the sacrificed individual was perceived as a member of a minority race but liberals assigned to the other condition ( i.e., those who read the version of the dilemma in which the sacrificed individual was intended not to be perceived as a member of a minority race) and conservatives in either condition were more likely (to the same degree) to endorse the sacrifice of the one person to save the 100 people In the ir next s eparate study, they presented participants with both of the previously used dilemmas, but asked the participants considered when deciding when to sacrifice one person to save 100 people The y found the same pattern of resu lts in this version of the task as the previous version but all participants regardless of political orientation responded to the second dilemma in a way consistent with the moral principle used to decide the first dilemma, even if the racial identities of those in the dilemmas were changed. Every participant endorsed the view that race should not be a factor in such judgments. Therefore, self reported liberals
MANIPULATING METAETHICAL BELIEFS 10 tended to shift the moral principle deemed apt for the situation just in order to avoid sacrifi cing a member of a racial minority despite the fact that they did not report race as a relevant factor in determining the right choice in moral dilemmas. The same principle was tested for political conservatives as well. They presented p articipants with a scenario in which either American troops attacked Iraqi insurgents or Iraqi insurgents attacked American troops. In either case, the attacking party decided that there might be unwanted deaths of civilians, but that collateral damage would be necessary to bring about a greater good. Only c onservat ives ( defined by more than 1 standard deviation away from the mean political orientation) tended to be significantly more permissi ve of the military attack when the American troops were the attackers (all others judged the attack similarly between the two conditions) These results suggest that at least among those who are politically opinionated, general moral principles are endorsed not only on the basis of previous episodes of endorsement, but also on the basi s of politically relevant information involved in particular situations in which the principle must be used, even when that information is not explicitly reported as information that should be taken into account in determining the adequate principle for th e situation It is unclear, however, whether this phenomenon can be generalized to all moral dilemmas. The motivated cognition of moral judgments elicited by priming effects is not limited to cases in which motives arising from political orientation are in volved One moral judgment Eibach, Libb y & Ehrlinger (2009) performed a study to determine whether being harmless but offensive acts (a woman undergoing plastic surgery to affix animalistic horns to her skull,
MANIPULATING METAETHICAL BELIEFS 11 a congenital dwarf voluntarily participating in dwarf throwing contests, and a man watching videos of animals copulating to become sexually aroused) They randomly assigned p articipants to answer a number of demographic questions either before or after the primary task judging the moral permissibility of the harmless but offen sive acts. Among the demographic questions one asked participants whe ther or not the par ticipant was a parent The p articipants were also asked to judge a control situation that involved (harmful) property rights violations. When not primed with their parental role (when the partici pants answered the demographic questions at the end), parents did not differ from non parents in their judgment of the harmless but offensive acts However, parents showed significantly more disapproval of these acts when primed to think about t heir roles as parents before judging the moral permissibility of the actions There were no significant diff erences between parents and non parents in judgment of the control question about property rights violations in either of the two conditions. Since the non parent s who answered the demographic questions before the primary task did not tend to judge the actions differently from thos e who answered the demographic questions last, the difference in moral judgment cannot be attributed solely to answering d emogra phic social role (in this case, parenthood) can influence moral judgments (in this case, by judging offensive acts more severely). Covertly p riming certain rules c an also influence the judgments of moral dilemmas that involve those rules that were primed Broeders, van den Bos, Muller et al. (2011) conducted three studies involving the effects of priming rules in different ways on
MANIPULATING METAETHICAL BELIEFS 12 judgments of the trolley and footbr idge dilemmas. In all of the studies, the y randomly assigned participants to be exposed to some form of prim ing involving either the rule save lives or the rule do not kill (The second study included a third control condition in which no rule was primed.) They also randomly assigned p articipants in each of the studies to respond either to the trolley dilemma or the footbridge dilemma following the priming manipulation In the first study, the priming manipulation involved a story about a UN mandate which incorporated the rule being primed in the relevant condition. In the second study, the priming manipulation involved the presentation (via a puzzle on a computer) of symbols representing the goals of saving lives or refraining from killing. The symbols for save lives included a red cross, an ambulance, and a lifebuoy, whereas the symbols for do not kill included a peace symbol, the Ten Commandments, and the symbol of the Du tch organization Tegen Zinloos Geweld (whi ch translates to against senseless violence ). The third study involved a subliminal (parafoveal) form of priming during an ostensible hand eye coordination task; they instructed participants to fixate on the center of a computer screen and respond to arrow s quickly, but flashed images of the symbols used in the second study in the corners of the c omputer screen for a very brief period (20 ms) followed by a masking image. Intriguingly, each of the studies yielded the same pattern of responses to the moral di lemmas. Although average moral judgments did not significantly vary across the priming conditions for the trolley dilemma, responses to the footbridge dilemma tended to shift in the direction consistent with the rule being primed; for those assigned to res pond to the footbridge dilemma, the average judgment for those primed with the rule save lives tended more toward the direction of judging it morally right to push a man off of a footbridge in order to stop the trolley, and the
MANIPULATING METAETHICAL BELIEFS 13 average judgment for those p rimed with the rule do not kill tended more toward the opposite direction. These findings suggest that simply being primed with a certain rule can shift moral judgments, but the extent of efficacy of such a form of priming is limited to certain moral dilem mas. Although there are other important studies pertaining to moral judgment these aforementioned studies are the most relevant to the rationale and the procedures utilized in the present study It should be stressed that although these studies have demon strated that moral judgments or attitudes may vary across situational contexts, situational influences do not constitute the entirety of the variability of moral intuitions. Individual differences with regard to overall worldviews, socially learned attitud es, private reflection, and the like may turn out to be indispensible in accounting for why people have the moral views that they do. However, these studies nonetheless demonstrate that these other potential factors relating to individual differences do no t account for all of the variability in moral views either. Moral judgments are determined in part by the particular rules and values accessible in recent memory, the activation of particular schemas, and misattributed affective states. Each of these situa tional influences on moral intuitions therefore can be utilized as tools by psychologists to understand the causal relationship between moral judgments and other psychological phenomena. Moral Objectivism Why study j udgment about moral objectivism ? By em pirically determining the implicit meta ethical beliefs of people and factors that shape these implicit beliefs, moral psychologists may refine their understanding of intuitions about the objecti vity of moral truths. This is valuable to philosophers who
MANIPULATING METAETHICAL BELIEFS 14 ar gue in favor of certain meta ethical views on the basis of intuitions about the nature of moral truths (e.g., the intuition that moral assertions purport to suppose that there are objective moral facts ) Empirical research on these intuitions can help to revea l their nature and causes ( For additional detailed discussion of the relation between the empirical investigation of judgments of moral objectivism and meta ethics, see Appendix A For additional discussion about the use of the term moral objectivism see Appendix B. ) Previous findings regarding judgments of moral objectivism To what extent and why do people find moral issues to be objective truth apt matters? Recently, both psychologists and philosophers have conducted empirical studies concerning the objective status of moral statements. In one of a series of studies on folk beliefs in the status of morality, Shaun Nichols (2004) presented participants with four different kinds of scenarios involvi ng disagreement between two individuals. The scenarios involved disagreements over whether the earth is flat, whether it is okay to drink soup straight from a bowl, whether it it is okay to hit people if one feels like it. For each scenario, the participants were asked to choose whether (and later indicate why) they agreed with either one side of the disagreement, or whether there is no fact of the matter about the issue. Nicho ls found that participants were most likely to agree that there is a fact of the matter about the flatness of the earth, followed by whether it is okay to hit someone, followed by whether it is rink soup from a bowl. Although Nichols found that the majority of participants agree d with the claim that there
MANIPULATING METAETHICAL BELIEFS 15 is a fact of the matter about whether it is okay to hit people if one feels like it, he did find that a some participants (17 out of 46 total) did not agree with that claim, but nonetheless agree d with the claim that there is a fact of the matter about whether the earth is flat. This finding was taken to entail that there is not universal agreement in folk beliefs in the objectivity of moral trut hs. objectivism with respect to a variety of issues. In their first study, Goodwin and Darley sessments of various statements. They as beliefs about the correctness of the statements and their views on whether there is an objective fact of the matter about the statements. Goodwin and Darley divided the statements into four domains for the purpose of interpret ing the results (these divisions were not labeled in the materials presented to the participants): scientific/factual, ethical, conventional, and aesthetic. To assess the the statement in question was an attitude or opinion as opposed to a truth or falsehood, and whether their beliefs about the statement and the beliefs of someone who disagreed with them could both be correct. Following this section of the study, the y inspe cted the responses and selected five statements in which the participant indicated either strong agreement or disagreement with a statement (the set always included two ethical statements, one conventional statement, one statement of taste, and one factual statement). For each of these statements, the participants were asked to write out the reasons they suspected for why others had disagreed with their views. The participants were then asked to specify the grounding for moral statements. The y were presente d with
MANIPULATING METAETHICAL BELIEFS 16 four groundings they a every good person on earth, regardless o a society could not survive without its citizens holding these beliefs, and their truth is self evident ) and were asked to mark each of these groundings that they agree justifies moral statements in general. Additionally, the participants were asked to indicate whether the existence of a supreme being is necessary for there to be right or wrong acts. They f ound that al though participants tended to converge in agreement with regard to the moral rightness or wrongness of many of the ethical issues, there were considerable differences (e.g., ranging from 2 % to 61 %) in whether the participants tended to regard the issue as true There were substantial differences between the overall mean ratings of belief in objectivism with regard to each domain of statements; scientific/factual statements were regarded as the most objective, followed by ethical statements, fo llowed by conventional statements, and finally followed by aesthetic statements. Goodwin and Darley found that the more groundings any participant indicated as grounding morality the more likely the participant endorsed objectivism about the ethical state ments. Finally, they found that strength of agreement regarding an ethical statement was positively correlated with (and predicted 26 % of the variance of) objectivism about ethical statements. In a follow up study, Goodwin and Darley (2008) replicated th e procedure of their first study with the exception of a modification to the first measure of moral objectivism. Instead of aski ng whether a statement was true false or an opinion or attitude they asked the participants whether there can be a correct an swer as to whether the statement is true or not. The y found similar but not completely identical results to t he
MANIPULATING METAETHICAL BELIEFS 17 results from the first study; in the modified version of their study, objectivism did not differ between factual and et hical statements. As a follow up to these previous studies, Goodwin and Darley (2012) conducted several related studies to determine the factors related to the endorsement of moral objectivism T hey were particularly interested in the relationship between the valence of moral beliefs ( i.e., whether they are judgments of moral commendation or moral condemnation ), perceived consensus about the beliefs, and the tolerance of disagreement over the issue with beliefs about the objectivity of the moral issue. The participants first read a number of moral and non moral scenarios. Some of the moral scenarios consisted of positive judgments whereas others consisted of negative judgments ( i.e., something to be judged as morally unacceptable or something to be judged as morally praiseworthy or beneficial). In the first study, the participants indicated their judgment of each scenario, the extent to which they thought there was a correct answer regarding the issue, and an estimate of the percentage of United States citize ns who they thought would agree or disagree. The y told the participants that there was another participant who disagreed with each of the answers indicated previously The y then asked the participants whether the other person must be mistaken. The y also as ked the participants how comfortable they would be to have th is disagreeing person as a room mate. The objectivity of issues involving negative moral judgments were significantly higher than ratings of objectivity for issues involvi ng positive moral judgments. Both the alleged disagreeing per son as a room mate were strongly positively correlated with judgments of objectivity of the issue.
MANIPULATING METAETHICAL BELIEFS 18 In thei r second study, Goodwin and Darley sought to reveal the extent to which information about the consensus of moral judgments ratings of tolerance for alternative views regarding those particular issues as well as their ratings of obj ectivism regarding those issues. They randomly presented the participants with one of two versions of a list of moral and non moral dilemmas, each of which indicated (bogus) percentages of people who agreed with a particular judgment of each dilemma. The v ersions differed in the purported consensus on the moral issues. They found a small but significant tendency for participants, regardless of whether they agreed or disagreed with the majority, to indicate higher ratings of moral objectivism for issues that they were led to believe were held by the majority. However, conclusion that people generally tend to be moral o bjectivists has been challenged by other findings Sarkissian, Park, Tien, et al. (2011) sought to determine whether peopl e tend to be less likely to endorse moral objectivism when moral disagreement is construed as occurring between members of distant cultures ( with very different values ) They presented each participant with two extreme moral transgressions and were told to imagine a classmate who judged the actions to be morally wrong. To test the hypothesis that construing moral disagreement as cross cultural reduces the endorsement of moral objectivism, they randomly assigned the participants to one of three conditions F or each of the conditions, the participants were asked to rate level of agreement with the statement that either their classmate or someone who disagrees with their classmate about the moral wrongness of each of the transgressions must be wrong. In one con dition, the person disagreeing with the classmate was jus t called Sam In the second condition, the y stated that the person
MANIPULATING METAETHICAL BELIEFS 19 disagreeing with the classmate was a member of an Amazonian tribe called the Mamalons which was described as a warrior culture with very different values than its neighboring cultures. In the last condition, the y asked the participants to imagine an extra terrestrial disagreeing with the classmate. The extra terrestrial was described as coming from a culture whose values were totally different from human values They found that participants tended in general to agree less with moral objectivism for disagreements between more distant cultures ( M ean ratings of agreement with the claim at least one must be wrong [with regard to the moral assessment] Sam condition, significantly less for the Amazonian tribe condition, and significantly less in the extra terrestrial condition than in the Amazonian tribe condition.) They then replicated this procedure with introduc tory philosophy students in Singapore and found the same pattern. In a third study, they conducted the procedure again with American college students, but instead of presenting one of the conditions to each participant, they presented each of the versions to every participant. This revised procedure produced the same pattern of results. In a fourth study, Sarkissian et al. replicated the within subjects design in their third study, but introduced a new between subjects variable. The participants (American c ollege students) were randomly assigned to respond to either a version of the moral transgressions in which the transgressor was American, or in which the transgressor was Algerian Although this added condition in which the identity of the transgressor wa s varied had no effect for the Amazonian tribes member and extra terrestrial versions of the vignette, there was a significant interaction for the same culture condition; when the hypothetical judges of the moral transgression were both part of the
MANIPULATING METAETHICAL BELIEFS 20 same cu lture as the participant, the participants tended to agree more with the statement that one of the judges must be wrong. In their final two studies, Sarkissian et al. (2011) investigated whether this same phenomenon would occur with factual beliefs (e.g., the belief that pasta grows on trees) as well as moral beliefs. In the fifth study, they randomly assigned participants (again, for both studies five and six, American college students) to either the moral or the non moral condition. In both conditions, th e y asked the participants to judge whether, in a disagreement over the issue between a classmate and an Amazonian tribes member, one disagreeing party must be wrong. The participants were much more likely to say that one of them must be wrong for the factu al issue than for the moral issue. Their final study replicated this method with the addition that half of the participants were presented with a version that asked whether either the classmate or the Mamilon must have an incorrect belief, and the other ha lf were presented with a version that asked whether at least one of them might have good reasons for their beliefs. The participants tended to agree more with the claim that one of the disagreeing parties had to be unjustified in having the moral beliefs t hat they did than they did with the claim the same claim about non moral beliefs. Sarkissian et al. (2011) claim that the results of all of these studies suggest that people tend to be meta ethical relativists after all; just as whether January is a winter month or a summer month is relative to the hemispheres of the earth whether an action is morally right or wrong is relative to the values of the culture to which one belongs. However, there are points of contention studies demonstrate that people in general are meta ethical relativists. A lthough they did replicate the procedure with students from Singapore, all of the studies were conducted
MANIPULATING METAETHICAL BELIEFS 21 with samples of college students from psychology or philosophy classes. It i s possible that other populations may respond differently to these procedures. The Present Study By determining the malleability of views regarding moral objectivism as well as the factors that causally influence views regarding moral objectivism, both psy chologists and philosophers stand to gain more knowledge about moral attitudes. (For further discussion regarding the philosophical implications of this particular empirical question, see Appendix A.) It was hypothesized that firmness of conviction in a mo ral issue causally facilitates the tendency to endorse objectivism about the moral issue. Goodwin and Darley (2008) found a connection between the firmness of conviction in a moral issue and the tendency to endorse objectivism about the moral issue, and so I intended to test whether this connection was a causal connection. To test this hypothesis, I utilized a priming method in order to experimentally manipulate the firmness of conviction in a moral issue. l issue are causally influenced by the firmness of their moral judgments pertaining to that particular issue should have an the objectivity of that particular moral issue. Broeders et al. (2011) fo und that firmness of judgment of moral dilemmas can be susceptible to influences from priming rules that are relevant to the moral dilemma Replicating o ne of their method s for priming rules I attem pted to experimentally regarding a moral issue in order to determine whether objectivism about a moral issue co varies with varying degrees of conviction
MANIPULATING METAETHICAL BELIEFS 22 Although only the exposure to priming could be directly experimentally manipulated, there are reasons to suppose that the hypothesized outcome of the study the hypothesized outcome in which both firmness of moral conviction and moral objectivism vary (in the same way) between the priming conditions a nd are positively correlated with each other would constitute evidence that firmness of moral conviction causally influence views regarding moral objectivism. Prior to the introduction of additional assumptions, the inference from the hypo thesized outcome that the firmness of moral judgment causally influenced views regarding moral objectivism would be no more valid of an inference than two competing explanations the inference that views regarding moral objectivism causally influenced the firmness of moral judgment and the inference that some third variable (either the independent variable itself or some variable that was causally influenced by the independent variable) causally influenced both firmness of moral conviction and views regard ing moral objectivism. However, additional assumptions cast doubt on the plausibility of these alternative causal explanations. Since the experimental manipulation involved rule priming, the immediate effects of the independent variable must have be en rel ated in some way to the particular content (rules) involved in the priming procedure. Those primed rules were directly related to the two competing principle s involved in the moral dilemma, and assuming that the firmness of conviction of moral judgments of dilemmas are connected to the perceived moral weight of each of the competing principles, the firmness of moral conviction regarding the dilemma must therefore have been connected to the primed content. However, the primed content (the rules) did not dire ctly relate to the matter of moral objectivism. For
MANIPULATING METAETHICAL BELIEFS 23 this reason, it is implausible that if the hypothesized outcome occurred, the priming procedure had a direct causal influence on the judgment of moral objectivism about the issue. For this reason, it is a lso plausible that the priming procedure had a direct causal influence on the firmness of the moral judgment of the dilemma. Thus, although the experimental setup did not necessitate that the hypothesized outcome would entail that firmness of moral convict ion causally influences the propensity to advocate moral objectivism about a moral issue, such an explanation would best fit that outcome given the relative implausibility of alternative explanations. Method Participants Forty participants were recruited f or the study at a small liberal arts college in southern Florida. All participants were at least 18 years of age, as indicated by their informed consent to participate in the study. In the interest of minimizing the time expenditure from the participants, demographic information such as age, gender, and ethnicity were not collected because there was not ample reason to have anticipated these factors to bear any systematic relationships to the phenomenon in question. The participants were recruited through a n email sent to the student forum at the college. The recruitment email stated that the study was being conducted for a senior thesis and requested volunteers to complete a few straightforward and brief tasks. Materials and Procedure P articipants were randomly assigned to read and respond to one of two possible versions of the first portion of the materials (See Appendix C for a copy of the materials used.) They were deceived about the true purpose of the study in order to minimize the
MANIPULATING METAETHICAL BELIEFS 24 effects of demand characteristics. This portion of the experiment was intended to prime the participants with a particular rule but the participants were not given this information until the debriefing I remained blind to the condition of each participan t. As consistent with Broeders et al. (2011), the first section consisted of a paragraph about a UN peacekeeping mission in the Congo in which a general issued a particular mandate for the mission. The mandate varied betw een the two conditions; the Save Li ves condition contained the mandate to save lives, and the Do Not Kill condition contained the mandate to refrain from killing. These mandates were intended to prime the participants with rules relevant to a moral dilemma within the section to follow. All other features of this section were identical between the two conditions. After the presentation of the background scenario involving the UN peacekeeping mission the participants were instructed to answer two open response questions. In the first question the participants were instructed to think about and write down how they would implement the mandate if they were in the about and write down how the mandate applied in their daily lives. The second section of this first portion of the study consisted of a filler mood assessment task which was presented ostensibly as one of the primary measures of interest for the study. Upon completing the first portion of the study the participants were then given a packet containing the materials for the second portion of the study The participants were led to believe that this packet was intended to investigate a separate research question from the first portion of the study. I t old the participants that the purpose of this section was to determine the relationship between the time spent reading and responding to the questions in the packet and the content of the responses to the questions in the packet.
MANIPULATING METAETHICAL BELIEFS 25 Upon beginning, I acted to give the impression that the participant was being timed ( I pressed buttons on a cell phone both at the time the participant started looking at the packet and at the time t he participant indicated that s he or he completed the packet). Unbeknownst to the p articipants, the time spent completing the packet was not measured at all. This section of the study was the same across both conditions and consisted of the surgery dilemma along with a number of questions pertaining to it. The questions following the dil emma consisted of four 8 point forced choice Likert style questions about the following: what the participant thought was the morally right course of action in the dilemma, what the participant thought he or she would do if put in the situation presented i n the dilemma, whether both the participant and a hypothetical person who believed a contradictory view about the dilemma could possibly be correct in their moral assessments, and finally whether or not this hypothetical person could have good reasons for he r or his moral assessment. The objectivi sm measure was determined by the answer to the third question following the dilemma. Upon completing the study, the participants were debriefed and thanked for their participation before finally being awarded compensation for their time spent participating in the st udy. The debriefing included a series of questions meant to probe for suspicion about the true purpose of the study. The participants were first asked about their general impressions of the study, next asked if they believed that there was any connection b etween the first packet and the second packet (and if so, what connection in particular, and the time at which they formulated this hypothesis about this connection), and finally if they suspected any deception about th e true purpose of the study. For any of those
MANIPULATING METAETHICAL BELIEFS 26 participants who indicated that they did suspect deception about the true purpose of the study, I asked him or her to explain whether the deception he or she suspected was due to a general suspicion about psychology studies or about something part icular to the present study as well as whether they had any particular ideas about how they were being deceived and the truth that was being concealed by such deception. The participants were classified into three groups pertaining to suspicion about the s tudy. The first group consisted of participants who did not suspect any deception or only had a general suspicion of deception without a particular idea of how they were being deceived, the second group consisted of participants who suspected that the firs t packet was intended to influence responses on the second packet in some way without particular ideas about what kind of influence was intended, and the third group consisted of participants who suspected that the first packet was meant as a priming manip ulation to influence the responses on the moral dilemma in the second packet. Following these questions, I fully explained the true purpose of the study as well as the reason why the deception was required for the study to be carried out. Results Although there were a total of 40 people who participated in the study, two participants (one in each condition) were excluded from the analysis because, as revealed by the probe for suspicion, they accurately determined the true purpose of the procedures while com pleting the study. Thus, out of the participants whose responses were not removed from the data prior to the analysis, 19 were assigned to each condition. I read the written responses for the first portion of the study to ensure that all of the participant s completed that task as instructed. Further, I found that six participants in the Save Lives
MANIPULATING METAETHICAL BELIEFS 27 condition referenced the principle of refraining from killing in their open responses to this task, but no participants in the Do Not Kill condition referenced th e principle of saving lives in their open responses. In the sections to follow, each analysis was because the reference to the principle from the opposite condition in the ir open response questions indicated that the manipula tion may have failed to prime those participants as intended Additionally, the statistical analyses were performed both on the actual scale values for the questions and for the degree of conviction, wh ich was determined by calculating the distance of each of scales For each of the four questions pertaining to the moral dilemma, the possible values ranged from 1 to 8. The participants tended to indicate a reluctance to sacrifice the healthy individual in the moral dilemma to save the five dying individuals ( = 2.03 SD = 1.13 ). 36 of the 38 participants circled a number in the range corresponding to the choice to refrain from sacrificing the healthy individual. Participants assigned to the Save Lives condition tended to indicate a similar level of reluctance to sacrifice the indi vidual ( = 2.16 SD = 1.01 ) than participants assigned to the Do Not Kill condition ( = 1.89 SD = 1.24 ). An independent two tailed t test revealed that neither this difference was statistically significant ( t = .72 p = .48 ) nor was the difference in degree of conviction regarding this question ( t = 1.35 p = .19 ) The difference in responses to this question between the two conditions, excluding the six participants in the Save Lives condition who mentioned the principle of refraining from killing, w as also not significant ( t = .24 p = .81 ) nor was the difference in degree of conviction between the two conditions ( t = .6, p = .56) Thus, it
MANIPULATING METAETHICAL BELIEFS 28 cannot be concluded that the priming manipulation had any impact on the projected willingness to sacrifice the healthy individual in the moral dilemma The participants tended to favor the claim that their projected choice of action regarding the moral dilemma was the right or best choice, but not to an extreme degree ( = 5.5 SD = 1.87 ). Participants assigned t o the Save Lives condition tended to indicate a similar level of confidence in favor of their projected choice regarding the moral dilemma being the right or best choice ( = 5.58 SD = 1.71 ) than participants assigned to the Do Not Kill condition ( = 5 .42 SD = 2.06 ). An independent two tailed t test revealed that neither this difference was statistically significant ( t = .26, p = .8), nor was the difference in degree of conviction regarding this question ( t = 1.22, p = .23). The difference in responses to this question between the two conditions, excluding the six participants in the Save Lives condition who mentioned the principle of refraining from killing, was also not significant ( t = .85, p = .41), nor was the difference in degree of conviction bet ween the two conditions ( t = .57, p = .57). Thus, it cannot be concluded that the priming manipulation had any impact on the firmness of conviction in the rightness of the projected choices of action regarding the moral dilemma. The participants tended to favor with the claim that a contradictory view about the right choice of action regarding the moral dilemma need not in principle preclude the correctness of their view about the right choice ( = 2.87 SD = 1.8 ). Participants assigned to the Save Lives c ondition indicated a very similar level of agreement with this claim ( = 2.95 SD = 1.84 ) than participants assigned to the Do Not Kill condition ( = 2.79 SD = 1.81 ). An independent two tailed t test revealed that neither this difference was
MANIPULATING METAETHICAL BELIEFS 29 statistically significant ( t = .27, p = .79), nor was the difference in degree of conviction regarding this question ( t = .31, p = .76). The difference in responses to this question between the two conditions, excluding the six participants in the Save Liv es condition who mentioned the principle of refraining from killing, was also not significant ( t = 1, p = .33), nor was the difference in degree of conviction between the two conditions ( t = .94, p = .36). Thus, it cannot be concluded that the priming mani pulation had any impact on the judgment of objectivism about the issue involved in the moral dilemma. The participants tended to agree with the claim that someone might have good reasons to favor a contradictory view about the right course of action regard ing the moral dilemma ( = 6.71 SD = 1.56 ). Participants assigned to the Save Lives condition indicated a similar agreement with this claim ( = 6.79 SD = 1.58 ) than participants assigned to the Do Not Kill condition ( = 6.63 SD = 1.57 ). An independ ent two tailed t test revealed that neither this difference was statistically significant ( t = .31, p = .76), nor was the difference in degree of conviction regarding this question ( t = .57, p = .58). The difference in responses to this question between th e two conditions, excluding the six participants in the Save Lives condition who mentioned the principle of refraining from killing, was also not significant ( t = .29, p = .78), nor was the difference in degree of conviction between the two conditions ( t = .01, p = .99). Thus, it cannot be concluded that the priming manipulation had any impact on the recognition of the potential for good reasons favoring a contradictory view about the right course of action regarding the moral dilemma Although there were no significant differences in the questions regarding the moral dilemma between the two priming conditions, there were several correlations
MANIPULATING METAETHICAL BELIEFS 30 between the responses to some of these questions that did reach levels of statistical significan ce. The willingness to sacrifice the healthy individual in the moral dilemma was moderately negatively correlated with the degree of judgment that their projected choice of action regarding the moral dilemma was the right or best choice ( r = .38 p = .02 ) ; the participants who indicated a higher level of willingness to sacrifice the healthy individ ual tended to indicate a lower conviction that their projected choice was the right or best choice. The willingness to sacrifice the healthy individual in the mo ral dilemma was almost significantly positively correlated with the level of agreement with the claim that someone might have good reasons to favor a contradictory view about the right course of action regarding the moral dilemma ( r = .30 p = .07 ), but wa s not significantly correlated with the measure of objectivism regarding the issue in the moral dilemma ( r = .16 p = .34 dilemma was the right or best choice, however, was moderately positively correlated (almost statistically significantly so) with the measure of objectivism regarding the issue in the moral dilemma ( r = .316 p = .053 projected choice of action regarding th e moral dilemma was the right or best choice was strongly negatively correlated with the level of agreement with the claim that someone might have good reasons to favor a contradictory view about the right course of action regarding the moral dilemma ( r = .52 p < .001 ). The level of agreement with the claim that someone might have good reasons to favor a contradictory view about the right course of action regarding the moral dilemma was also moderately negatively correlated with the measure of objectivism regarding the issue in the moral dilemma ( r = .40 p = .01 ).
MANIPULATING METAETHICAL BELIEFS 31 Although the assessment of self reported mood was primarily intended as a filler task (which was presented prior to the moral dilemma but after the priming manipulation) there were some correlations between the level of self reported moods and the questions pertaining to the moral dilemma that did reach levels of statistical significance. The willingness to sacrifice the healthy individual in the moral dilemma was modera tely negatively associated with the self reported degree of being in an interested mood ( r = .35, p = .03) and was moderately positively associated with the self reported degree of being in a distressed mood ( r = .36, p = .03). The measure of objectivism regarding the moral issue was moderately positively associated with both the self reported degree of excitement ( r = .37, p = .02) and the self reported degree of being jittery ( r = .34, p = .035). The level of agreement with the claim that someone might h ave good reasons to favor a contradictory view about the right course of action regarding the moral dilemma was moderately positively correlated with the self reported degree of feeling proud ( r = .38, p = .02). Additionally, a two tailed independent t tes t revealed that the self reported degree of attentiveness differed significantly between the two conditions ( t = 2.13, p = .04); participants assigned to the Save Lives condition tended to report being less attentive ( = 2.68, SD = .95) than did particip ants assigned to the Do Not Kill condition ( = 3.42, SD = 1.17). No other self reported mood differed to a statistically significant extent between the two conditions. Discussion Although the data did not support the hypothesis that firmness of conviction about a moral issue causally influences the endorsement of objectivism about that issue, it did not demonstrate that no such causal influence exists Nonetheless, the connection
MANIPULATING METAETHICAL BELIEFS 32 between the firmness of conviction about the surg ery dilemma and objectivistic intuitions about it was quite minor, and thus even if the firmness of conviction was a causal source of variance for objectivistic intuitions about the issue, it may only serve as a small fraction of the causal source of varia nce in objectivistic intuitions about the issue (10 % at most, assuming that the effect size of the present study is generalizable) The data from the present study indicated that reports of being more excited or jittery (slightly) better predicted the deg ree to which one endorsed higher levels of objectivism about the issue than did the firmness of moral conviction about the issue. Therefore under the assumption that the findings regarding the tested population can be generalized to a larger population, a nd that the tendency to endorse objectivism about the surgery dilemma is not in some way different from the assessment of the objectivity of solutions to other kinds of moral quandaries the causal explanation for the presence or lack of objectivistic intu itions about the solutions to moral quandaries primarily involve s factors other than However, since Goodwin and Darley (2008) found a stronger degree of association (R 2 = .26) between firmness of moral convicti on and objectivism about the same moral issue, the relationship between firmness of moral conviction and objectivism about the same moral issue may vary between kinds of moral issues. Further research can help to elucidate this question. The other connect ion s between the four questions regardi ng the moral dilemma were not as theoretical ly important as the aforementioned connections but are at least worthy of mention. About 16 % of the variance in the responses to the objectivism measure was accounted for by the degree to which one agreed with the claim that there
MANIPULATING METAETHICAL BELIEFS 33 own view. Further, the degree to which participants thought that one might have good reasons favoring an alte rnative moral view was moderately negatively associated with the strength of conviction in the moral issue. Thus, the reported perception of a lack of good reasons for an action better predicted objectivism about the negative moral value of the action than did reported firmness of conviction about the issue. It is possible that this association between objectivism and the perception of good reasons may be unique to moral issues; S arkissian et al. ( 2011) found that participants tended to agree more with the claim that others could have good reasons to believe in a factual error than have good reasons to have a disagreeable moral belief. Perhaps the association between having good re asons and objectivism differs between different domains ( i.e., factual and moral domains) of issues. Unfortunately since only one issue was presented in the present study, future research is necessary to assess this hypothesis. Why did the Priming Manipula tion Fail? Unintended priming effects. A number of participants, during the debriefing, voluntarily took the initiative to share their reasons for selecting their choice for the surgery moral dilemma. Many of these explanations included reference to and emphasis on a violation of consent of the sacrificed patient in the dilemma. Although no measure in the present study can indicate the extent of consent related considerations responses, it would at least be feasible to suppose that these kinds of considerations can participants in both conditions were re quired to read over an informed consent form immediately prior to beginning the tasks in the study. Given that the mere presence of the
MANIPULATING METAETHICAL BELIEFS 34 informed consent form implicitly conveyed certain rules, i.e., requesting individual consent before even subjecting peop le to tasks in a psychological study, it would not be implausible to suspect that all of the participants were primed with consent related rules in a similar way that the participants were intended to be primed with harm related rules and life saving relat ed rules. Insofar as such stipulated priming did occur, doubt may be cast on the efficacy of the priming manipulation in the Save Lives condition; although participants may have been primed to focus on considerations that involved saving live s, the partici pants may also have been primed to focus on considerations involving doing good (by saving lives) only in the context of rules pertaining to respect for autonomy ( i.e., consent). Priming rules related t o respect for autonomy may have diminished or even eli minated the ability to construe subsequently primed rules in ways that would be contradictory to rules pertaining to the respect for the autonomy of persons. This issue could be ameliorated by adding filler tasks that would allow enough time to elapse betw een the presentation of the consent form and the tasks containing the priming manipulations. Dilemma specific priming susceptibility. If problems with the methodology used in the present study may be assumed not to have been the culprit for the failure o f the priming manipulation to have an impact on moral judgments, the most plausible explanation for why priming particular rules did not affect judgment on the moral dilemma is just that the process of making a moral judgment may vary between different kin ds of situations or dilem mas. T he psychological process involved in solving some dilemmas may be more susceptible to the influence of a primed moral rule than others. This explanation derives its plausibility from the fact that
MANIPULATING METAETHICAL BELIEFS 35 the footbridge dilemma, but not the trolley dilemma, was influenced by various kinds of rule priming in the studies conducted by Broeders et al. (2011). Broeders et al. (2011) suggest ed that for judgments of a moral dilemma to be substantially influenced by extraneous information, t here must be ambiguity and feelings of uncertainty present in the dilemma They do not specify whether such ambiguity refers to moral ambiguity or ambiguity in the imagined situation itself and read either way, the accuracy of this suggestion is not clear. The use of the notion of extraneous information implies an assumed generalization from the kind of priming employed in the procedures of their studies and the present study to other forms of extraneous influences. Although the study performed by Valdesolo and DeSteno (2006) corroborates this generalization (since the ir affect induction only influenced judgments of the footbridge dilemma and not the trolley dilemma), further evidence is required to ascertain its val idity. Further, it is a currently open empirical matter whether the differences between the dilemmas in which rule priming has an effect on moral judgment arise from differences in the ambiguity present in the situations conveyed in the dilemmas. Given tha t such rule priming has only been empirically tested thus far on the footbridge dilemma, the trolley dilemma, and the surgery dilemma, and only reliably influenced judgment on the footbridge dilemma, it is not clear that the only features that the latter t wo dilemmas lacked and the former dilemma possessed was a greater sense of ambiguity. If a sense of ambiguity is meant to refer to uncertainty about the effects of a choice within the scenario, t here is the same degree of ambiguity present in the surgery d ilemma as the re is in the footbridge dilemma. J ust as one might feel uncertain that pushing a large man in front of a trolley may not adequately stop the trolley ( despite the explicit specification within the vignette to
MANIPULATING METAETHICAL BELIEFS 36 assume that pushing the man will gu arantee that the trolley will stop ) one might feel that the organ transplants may fail to save the dying patients in the surgery dilemma (despite similar specifications to the contrary) Alternatively, if Broeders et al. (2011) meant that there is a highe r sense of moral ambiguity (viz., uncertainty about the right course of action) that suffices for susceptibility to rule priming, then people must at least tend to exhibit more certainty about the right course of action in the surgery dilemma than in the f ootbridge dilemma. Although a the means to complete a quantitative comparison ty regarding the two dilemmas are not available, it was the case that the participants of the present study were closer to the midpoint than the end point on the certainty scale for the rightness of their choices regarding the dilemma; dilemma was only approximately one point away from the midpoint on the 8 point scal e. There are alternative explanation s for the lack of susceptibility to rule priming on the surgery dilemma P eople may have specific conceptions of the restrictions and obligations that are inherent to occupying the role of a doctor. A doctor is supposed to be someone whom patients can trust someone who owes to her or his patients honesty, respect, and above all the interest in every being. The Hippocratic Oath, after all, obliges doctors to do no harm. Further, people in Western society (and other contemporary societies) may feel entitled, in virtue of being in the patient role, to be consented (if possible, given that one is not in imminent danger of fatality or unable to consent ) before even being treated by a doctor. It is thus unaccep table, and even punishable, to perform any procedure without obtaining prior consent, except in special circumstances (circumstances that do not obtain in the surgery dilemma) Additionally, i f
MANIPULATING METAETHICAL BELIEFS 37 it were acceptable for doctors to force unwitting patients to sacrifice themselves, then hospital This would taint the reputation of the hospital as a safe place and the doctor as trustworthy. The difference between the footbridg e and surgery dilemmas, according to this explanation, stems from the presence of the particularity of the societal role s (embedded with important duties and entitlements) involved in the surgery dilemma Certain principles that dictate what kind of behavi or is prohibited become highly salient just by construing the situation in terms of the societal roles of doctor and patient. This explanation could be empirically tested by somehow construing the footbridge dilemma such that the person doing the pushing i s a doctor and both the large man being pushed to stop the trolley and the people on the tracks being saved are the s (Such an adaptation may require additional modifications.) If the effects of rule priming found in Broeders et al. (2011) were still exhibited in this alternative version of the footbridge dilemma, then this explanation would be falsified. Of course, this explanation accounts only for the lack of susceptibility to rule priming in the surgery dilemma and not the trolley dilem ma since the base case situation portrayed in the trolley dilemma does not include any particular societal roles. There may simply be a distinct explanation for the lack of susceptibility to rule priming in the trolley dilemma (perhaps, for example, Broede 2011 explanation) Additionally, if this hypothesized explanation regarding the present findings turned out to be correct, it may be possible that a different form of priming would have an influence on moral judgments of the surgery dilemma. P erhaps, f or example, if participants were primed to enter an abstract or concrete mindset (assuming that these mindsets are related to the extent to
MANIPULATING METAETHICAL BELIEFS 38 which people consider their duties within a particular societal role) they may be less likely to take the societal role of being a doctor into consideration in making the ir moral assessment. Limitations of the Present Study Population pool and sampling methods. There are a number of potential issues with the sampling methodology involved in the present study As was mentioned earlier, it may be the case that American undergraduate students are distinct in important ways from the majority of other human adults. 3 Although the particular phenomenon of interest the causal relationship between the strength of mo ral conviction about an issue and intuitions favoring moral objectivism about the issue may best be understood as a feature of moral convictions and obj ectivist intuitions in general, it is still possible that there are differences between different populations with regard to this relationship. I f intuitions favoring moral objectivism are related to epistemic attitudes or other worldviews, such as religio us worldviews, that are not as widely held by American college students as they are for the majority of other people objectivism about a moral issue may not be generalized to a larger population There may be further differences between students at small liberal arts colleges and students at major universities or more conventional educational institutions. For these reasons, it would be quite problematic to assume that the findings of t he present study reveal universal features of human moral psychology. Although there is not clearly evidence available to substantiate this concern, it is perhaps equally unclear that there is evidence to justify neglect of it.
MANIPULATING METAETHICAL BELIEFS 39 Additionally, the recruitme nt procedure may not have been successful in testing a sample that was representative of the population of college tested Because the recruitment was accomplished via an online student forum, only students who choose to view the forum (and chose to view t he particular recruitment post on the forum) could have been recruited. Further, since participation was initiated voluntarily by the participants, the participants sampled in the study may have had a distinct motivation to make contributions toward empiri cal research, which could potentially be associated with a number of other distinctive characteristics such as highly charitable dispositions or positive attitudes about empirical research (which may further be associated with particular worldviews and epi stemic attitudes) A positive attitude about empirical research might be related to characteristic ways of responding to questionnaire items; for example, having a positive attitude about empirical research might be associated with certain attitudes about facts and values (e.g., the attitude that only facts can be objective, and that moral values are not factual) These possibilities though currently only speculative, present additional concerns for research on moral objectivism. However, they could be exp licitly addressed through the use of questionnaires and alternative recruitment procedures Construct validity. There are problems with the validity of the objectivism measure. The objectivism measure, which asked participants whether it is possible in pr inciple for both their own view and a contradictory view to be correct, was adapte d from Goodwin and Darley (2008, 2012). Since the measure used in this study was connected to the operational definition of moral objectivism as the view that two conflicting moral views cannot both
MANIPULATING METAETHICAL BELIEFS 40 be correct, any concerns with the operational definition entail concerns with the measure of moral objectivism about an issue. It is not entirely clear such an operational definition of moral objectivism adequately picks out what p hilosophers, or perhaps even people in general mean by moral objectivism. It is possible to believe that a moral principle is objectively true, yet contain qualifications regarding its context of application embedded with in it. For example, one may coherently believe that it is objectively true that eating pork is morally wrong for morally wrong to eat pork purported ly objective way ( i.e., the rule is true regardless of any belief, sentiment, or other psychological feature of a moral inquirer). A non Jewish person would be correct to say that eating pork is morally permissible (assuming that the practice of eating mea t in general is not morally unacceptable on other grounds ), whereas a Jew ish person would be correct to say that eating pork is morally impermissible. Although a linguistic analysis might reveal that the Gentile and Jew are meaning to say different things through their moral assessments of ish and non Jewish people to be correct about their moral assessments of eating pork answered affirmatively by participants in psychol ogical studies despite that the same ish and non Jewish people to be correct a bout the moral assessment that i f one is Jewish, one ought not to eat pork where the Jew ish person agrees and the non Jewish person (and thereby believe that there is an objective moral issue at stake). Although it is not clear that moral views regarding the surgery dilemma are analogous to the moral views in
MANIPULATING METAETHICAL BELIEFS 41 this example regarding the c onsumption of pork the example nonetheless raises doubt about the validity of the operational definition of moral objectivism that was utilized here and in previous research. Since the operational definition of moral objectivism utilized here seems to div erge in some cases from the more commonly utilized and accepted notion of moral objectivism, there may be an operational definition of moral objectivism that Directions for Future Res earch As mentioned previ ously, there are several empirical questions that would help determine the correct explanation for the efficacy of rule priming on the footbridge dilemma but not the surgery and trolley dilemmas. In order to determine whether a sense of ambiguity is responsible for the susceptibility to influence by extraneous factors, people could simpl y be asked to report their sense of ambiguity both moral ambiguity and the ambiguity of the situation itself of the footbridge, trolley, and surgery dilemmas, and that the reported ambiguity could thus be directly compared. Further, to test the hypothe sis that the societal roles in the surgery dilemma fix the conception of acceptable kinds of behaviors, those societal roles could be transposed onto the roles occupied by the individuals in the footbridge dilemma as it is presented to participants followi ng a priming manipulation Future research on rule priming in general might involve the use of other moral dilemmas as well as other rules that are primed in order to influence moral judgment of those dilemmas. F uture research regarding the belief in the objectivity of moral truths may benefit by the utilization of a revised operational definition of moral objectivism and corresponding measures to that revised operational definition that more closely track
MANIPULATING METAETHICAL BELIEFS 42 what philosophers and lay people mean by moral obj ectivism. With this revised measure of moral objectivism, it would be philosophically illuminating to test the hypothesis that belief in the objectivity of issues in the moral domain bears the unique feature of being negatively associated with the belief t hat there are no good reasons justifying an alternative, contradictory moral view. Finally, since the firmness of conviction in a moral issue is not, based on the present findings, associated with objectivism about that moral issue, the same sort of experi mental method could be adapted to test the causal influence of further hypothesized factors on moral objectivism. Such factors may include for example, beliefs about the source of moral truths (e.g., a divine source )
MANIPULATING METAETHICAL BELIEFS 43 References Bartels, D. M. (2008). Principled moral sentiment and the flexibility of moral judgment and decision making. Cognition 108 381 417. doi:10.1016/j.cognition.2008.03.001 Broeders, R., van den Bos, K., Mller, P. a., & Ham, J. (2011). Should I save or shoul d I not kill? How people solve moral dilemmas depends on which rule is most accessible. Journal of Experimental Social Psychology 47 923 934. Elsevier Inc. doi:10.1016/j.jesp.2011.03.018 Eibach, R. P., Libby, L. K., & Ehrlinger, J. (2009). Priming family values: How being a parent affects moral evaluations of harmless but offensive acts. Journal of Experimental Social Psychology 45 1160 1163. doi:10.1016/j.jesp.2009.06.017 Goodwin, G. P., & Darley, J. M. (2008). The psychology of meta ethics: exploring objectivism. Cognition 106 1339 66. doi:10.1016/j.cognition.2007.06.007 Goodwin, G. P., & Darley, J. M. (2012). Why are some moral beliefs perceived to be more objective than others? Journal of Experimental Social Psychology 48 250 256. Elsevier Inc. doi:10.1016/j.jesp.2011.08.006 Haidt, J., & Joseph, C. (2004). Intuitive ethics: How innately prepared intuitions generate culturally variable virtues. Daedalus, 133 55 66. Ham, J., & van den Bos, K. (2010). On Unconscious Morality: The Effects of Uncons cious Thinking on Moral Decision Making. Social Cognition 28 74 83. doi:10.1521/soco.2010.28.1.74
MANIPULATING METAETHICAL BELIEFS 44 Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? The Behavioral and Brain Sciences 33 61 83. doi:10.1017/S0140525X099 9152X Kohlberg, Lawrence (1958). The Development of Modes of Thinkin g and Choices in Years 10 to 16 Unpublished doctoral d issertation, University of Chicago Illinois, USA Milgram, S. (1963). Behavioral study of obedience. Journal of Abnormal and Social Psychology 67, 371 378. Navarrete, C. D., McDonald, M. M., Mott, M. L., & Asher, B. (2012). Virtual morality: emotion and action in a simulated three Emotion 12 364 70. doi:10.1037/a0025561 Nichols, S. (2004). After object ivity: an empirical study of moral judgment. Philosophical Psychology 17 3 26. doi:10.1080/0951508042000202354 Sarkissian, H., Park, J., Tien, D., Wright, J. C., & Knobe, J. (2011). Folk Moral Relativism. Mind & Language 26 482 505. doi:10.1111/j.1468 0017.2011.01428.x Shweder, R. (1990) In defense of moral realism: R eply to G abennesch. Child Development, 61 2060 2067 Turiel, E. (1983). The Development of Social Knowledge: Morality and Convention. Cambridge, England: Cambridge University Press.
MANIPULATING METAETHICAL BELIEFS 45 Uhlmann, E. L., Pizarro, D. A., Tannenbaum, D., & Ditto, P. H. (2009). The motivated use of moral principles. Judgment and Decision Making 4 479 491. Valdesolo, P., & DeSteno, D. (2006). Manipulations of e motional context shape moral judgment. Psychological science 17 476 7. doi:10.1111/j.1467 9280.2006.01731.x
MANIPULATING METAETHICAL BELIEFS 46 Footnotes 1 Although such judgments might also be treated as legally relev ant or conventionally relevant, their status in relation to legal or conventional concerns would not preclude their status as morally relevant. 2 It should be noted that whether these behaviors actually are good or bad is irrelevant to the validity of the research. It is just typically assumed that helping others is a morally good thing, all else equal, and delivering fatal shocks to a person is ty pically a morally bad thing, even when an authority issues an order to do so. 3 See Henrich, Heine, and Norenzayan (2010) for a thorough critique of the usage of populations consisting exclusively of American college students.
MANIPULATING METAETHICAL BELIEFS 47 Figure 1 Mean responses to the four que stions about the moral dilemma separated by condition Error bars attached to each column denote the SEM for the column. Higher scores represent greater willingness to sacrifice the healthy patient in the dilemma, obje ctivism about the moral dilemma, and the possibility that there are good reasons for an alternative, contradictory view regarding the moral dilemma, respectively for each question from the left side to the right side of the horizontal axis. 1 2 3 4 5 6 7 8 Willingness to Sacrifice Patient Certainty that Choice was Right Objectivism Measure Possibility of Good Reasons for Alternative Means Across Conditions for the Primary Moral Judgments Save Lives Condition Do Not Kill Condition
MANIPULATING METAETHICAL BELIEFS 48 Appendix A: Su pplemental Theoretical Discussion Regarding the Implications of Empirical Research for Meta Ethics How precisely does empirical research on folk views regarding moral objectivism inform meta ethical questions? More specifically, how does the present resear ch question the question of whether the firmness of conviction in a moral issue exerts a causal influence over the tendency to endorse a form of objectivism about that moral issue inform meta ethics? The latter, more specific question shall be the prim ary question of concern, as the answer to that question provides an example that suffices as part of an answer to the former question. The short er response to this question of interest would be something like the following: evidence about the causes of be lief in moral objectivism can elucidate the reasons why some find moral objectivism intuitive. By revealing the reasons for the existence of an intuition, philosophers thereby stand to gain a clearer means to apprehend and evaluate the intuition in questio n. Thus empirical research aimed at the question the present study was aimed at has a limited application to philosophical questions concerning the actual objectivity of moral truths ; it does not have implications for any philosophical arguments in favor o f moral objectivism that do not rely on and utilize the assumption that moral objectivism is just intuitive. However, arguments that do take as a premise that moral objectivism is intuitively correct are subject to critique on the basis of empirical findings pertaining to the causes of those intuitions. If, for example, the causes of the intuition do not correspond to sound, justifying reasons, then we have reason to repudiate the intuition rather than build philosophical views from it
MANIPULATING METAETHICAL BELIEFS 49 The longer response to the question regarding the relevance of the present research on meta ethics involves its implications for moral semantics. It shall be argued that empirical research uncovering the causes of features of moral attitudes, such as the feature of objectivism, provides evidence that can be used to support or rebut views about the relation between moral attitudes and descriptive beliefs (for example, the view that moral attitudes are a species of descriptive belief). Before spelling out exactly how the present research informs moral semantics in this way, some philosophical groundwork on moral semantics in general is in order. How S hould W e A pproach M oral S emantics? Meta ethics consis ts of a cluster of philosophical questions pertaining to the status and nature of ethics or morals. Its internal diversity results from the diversity of aspects of the status and nature of ethics or morals (e.g., the epistemological nature, the metaphysica l status, the semantics of moral discourse, etc.). Meta ethicists are concerned, for example, not just with what the ought means in the way it is used within sentences to descriptive utterances or directive utterances, but also with other projects, such as the determination of the actual features that make moral verdicts true or false (e.g., moral reality, facts, properties, etc.). The former might be labeled moral seman tics or pragmatics whereas the latter might be labeled moral metaphysics. The present argument is concerned specifically with the former category moral semantics or pragmatics (henceforth just referred to as moral semantics for simplicity) 1 which one c an understand as the philosophical project whose purpose is to explain the features of language employed in moral discourse ( i.e., its meaning).
MANIPULATING METAETHICAL BELIEFS 50 The intention of the present section is to clear the path and sort out useful ways for which moral semanticist s may postulate and decide between competing accounts; I do not mean here to critically engage with any particular meta ethical accounts, but rather to discuss how we should understand moral semantics In the sub section to follow, I distinguish between tw o different kinds of outlooks towards semantic accounts monistic outlooks and pluralistic outlooks that moral semanticists can undertake. I offer reasons for meta ethicists to prefer a pluralistic outlook over a monistic outlook. In the sub section fol lowing that, building from this endorsed way of approaching moral semantics, I sketch out some criteria with which moral semanticists can use to evaluate particular accounts of moral semantics, and evidence resulting from empirical research concerning the causes of intuitions regarding moral objectivism finds a place among those criteria. Two K inds of O utlooks in M oral S emantics The monistic outlook. A project in moral semantics is intended to explain or elucidate the meaning of moral language by working out an account of moral discourse 2 There are numerous possible outlooks or approaches one may take on the interpretation of such accounts. 3 Monistic outlooks involve the interpretation of the account as describing a single, privileged kind of meaning of all genuine moral utterances. If an interpretation of this sort is assumed any utterance whose speaker intends to mean something different from the correct theory of meaning of genuine moral claims cannot be a genuine moral utterance. Any speaker who thinks otherwise is simply mistaken about the kind of utterance she or he is making. If, for example, it were true that all genuine moral
MANIPULATING METAETHICAL BELIEFS 51 utterances involve the ascription of non natural moral properties, Jones would simply be m by eat that food with your hands (as opposed, say, t o a directive statement pertaining to mere conventions ) and the correct semantic analysis of the utterance dictates that the meaning of the utterance did not involve any implicit ascriptions of non natural moral properties, then Jones would be mistaken in believing that the utterance was a genuinely moral one T he features of the sentence would just indicate that the sentence could not count as a use of genuinely moral language Thus monistic outlooks are imperious in two ways: regardless of the intentions or beliefs of speakers, they may constrain the meanings of utterances and they may also constrain the kinds of utterances that are genuinely moral in status. A monistic outlook on accounts of moral semantics can thus be characterized as the endorsement of the following propositions. (I 1 ): All and only genuine moral language belongs to the discourse type M 4 (I 2 ): All and only utterances of discourse type M have some meaning of kind A with respect to the content of the utterance. (I 3 ): For any utterance, M The first proposition accounts for the unity of the semantic account; moral language is taken to be a unified explanandum picked out by M Further, the second proposition adds that all uttera nces fitting into discourse type M can be understood in relation to some kind of meaning or translation, referred to as A whatever it is that
MANIPULATING METAETHICAL BELIEFS 52 utterances in M must all be about. If the semantic account is illuminating, the contents of the A term will be c ontents that we grasp with clarity. The third proposition provides a n absolute means to identify moral language via linguistic features of the utterance itself, and to distinguish moral language from non moral language. Together, these three propositions c onstitute the characteristic features of the monistic outlook on semantic accounts. As an example a monistic interpretation of some nondescript form of realism would take discourse type M to involve some particular form of d escription or fact stating acti vity. An utterance constituting this kind of discourse is understood to mean certain things about the (moral) facts; the reduction to a kind of fact statement relating to moral facts accounts for the A term in I 2 Finally, this realist account might identi fy particular features of sentences that, according to the account involve some kind of reference to a specific class of facts ( i.e., the class of moral facts, whatever that may be) such as the right sort of syntactical structure in conjunction with the u se of certain words like should or right or even the presence of certain kinds of cues (such as a tone of indignation), as the means to distinguish moral discourse from non moral discourse. The pluralistic outlook. In contrast to monistic outlooks or interpretations of accounts, what I call the pluralistic outlook of an account of moral semantics is only purport ed to explain one meaning of some moral utterances (albeit, an important one given that it significantly illuminates our understanding of moral discourse). 5 A pluralistic interpretation of a semantic accou nt may be taken, at minimum, as the replace ment of the universal quantifier in (I 1 ) with an existential quantifier. Thus s uch a kind of outlook is at minimum characterized by the implicit endors ement of either of the following proposition s
MANIPULATING METAETHICAL BELIEFS 53 (C 1 a ): Some genuine moral language belongs to the discourse type M (C 1b ): Genuine moral language could belong to the discourse type M The two versions of C 1 can be distinguished by the presence of a modal operator which C 1b introduces. Although C 1b offers a more precise characterization of the pluralistic outlook to be discussed in some of the following sections, the crucial, more basic move from monism to pl uralism is captured by C 1a (Henceforth, reference to C 1 is intended as reference to either C 1a or C 1b .) A semantic account, under a pluralist interpretation, is intended to describe the me aning of moral utterances that, while a coherent account may not b e the only valid or privileged account of the meaning 6 C 1 does not inherently contradict a form of monism regarding moral semantics, but the endorsement of C 1 over I 1 offers at least several advantages. C 1 allows for the possibility of inter individual va riations in the semantics of moral discourse. Such inter individual variations might result from cultural, developmental, philosophical, personal, and motivational differences between users of moral language. Additionally, C 1 allows for the possibility tha t moral talk is a patchwork of diverse, sui generis kinds of communicative activities, even intra individually. D iscourse about certain acts being heinous and unacceptable for reasons related to care and respect for other persons might be fundamentally dis tinct from discourse about what kinds of character traits constitute ideals to strive towards. 7 Jones might for example, engage in both kinds of discourse, yet engage in each in such a way that neither kind of discourse could be reduced to the other. This way of understanding accounts of moral semantics also better aligns with what Helen Longino calls the theoretical virtue of ontological heterogeneity (Longino, 1995).
MANIPULATING METAETHICAL BELIEFS 54 mo ral talk, pluralistic accounts allow for the possibility that this heterogeneity is important in some sense. One might question just as a matter of fact whether such heterogeneity with regard to moral discourse actually exists. Perhaps, one might contend, there are some underlying features of all moral language, such as the endorsement or internalization of particular kinds of values pertaining to conduct. I am not convinced, however, that this contention provides reason enough to suppose that there is a su bstantial commonality between all forms of moral discourse; even if all moral systems involve some kind of internalization of values, there does not appear to be a consensus about which kind of values are distinctively moral or non moral. 8 Different political orientations evidently correspond to different views about what kinds of values (e.g., care for others, respect for sanctity, etc.) count as morally relevant as opposed to just conventionally relevant (Graham, Haidt, and Nosek, 2009). Different cultural worldviews involve the employment of different sources of moral value, such as the concept of honor in cultures of honor (Nisbett and Cohen, 1996), as well as moral concepts particular to a religious system such as sin w ithin Christian t raditions and dharma in certain Indian religious traditions. I take these examples as evidence that points toward the view that there are no particular kinds of values that are essential to morality, but ultimately I must defer a more complete examination of the issue. 9 Since a clear verdict on this issue cannot sufficiently be generated without a great deal of focus, the best I can do here is to point toward these sources of evidence about the diversity of foundational concepts of moral value, and leave it as an advantage for pluralism about moral semantics that it can better
MANIPULATING METAETHICAL BELIEFS 55 accommodate irreducible heterogeneity regarding moral concepts, values, or language than can monism about moral semantics. In addition to the ability to accommodate heterogeneity in th e explanandum, the adoption of C 1 as a replacement for I 1 moral discourse greatly reduces the vulnerability of the account to counter examples to the semantic account ( i.e., instances of paradigmatically moral talk that defy the scheme proposed by an account of moral semantics); whereas the previous advantage concerned the sui generis plurality of kinds of moral concepts, this advantage is concerned with problem cases for any given semantic account of moral dis course. Although it is obvious that mere dialectical safeguarding in no way suggests some greater proximity to the truth of any matter (prominent examples of this point include unfalsifiable scientific theories), it does offer a different kind of advantage It might turn out that no single, unified semantic account of moral talk can address all counter examples. Given the fact that the current state of moral semantics (and meta ethics more generally) does not approximate a consensus, it seems at least plaus ible to suggest that no present single, unified semantic account of moral talk addresses all counter examples that proponents of a competing account might present. The specific advantage of adopting C 1 is that there need not be worry that this state will n ever change. This advantage just results from the fact that C 1 allows for a kind of pluralism about the issue. Another advantage for the pluralistic outlook is that it has the ability to accommodate the possibility that the meanings of utterances meta ethi cists are interested in are determined (at least in part) by the intentions of their speakers. The view that meanings of utterances are determined by the intentions of the sp eaker comes notably
MANIPULATING METAETHICAL BELIEFS 56 from H. P. speaker or (Gr ice, 1968). As an example illustrating this view of meaning, if Smith believes that morally wrong acts derive their wrongness from (part of) the meaning certain acts are morally wrong just is that those acts are fo rbidden by God, insofar as that is what Smith intends to convey to her audience in making claims about morally wrong acts If one refrains from affirming a monistic interpretation of moral semantics, then one may intentions; the meaning implicit supposition that acts are morally wrong in virtue of b ei ng forbidden by God This might best be characterized as the denial of I 3 and the endorsement of an alternative proposition C 3 (C 3 ): For any utterance, if its speaker has intentions belonging to kind R in making the utterance, the utterance is meant to b elong to discourse type M 10 Many may find themselves skeptical about broad claims regarding the reduction meaning. However, such worries 3 All that matter s here is simply that speaker meaning can (and often does) capture what we are after when we inquire about the meaning of an utterance; an account of the meaning of moral language, insofar as it is intended to elucidate the phenomenon of human moralizing r ather than linguistic traditions themselves, may very well just be intended to track speaker meaning. Problems for the pluralistic outlook. There are potential issues pertaining to language that follow from the adoption of C 3 By understanding the meaning
MANIPULATING METAETHICAL BELIEFS 57 the Gricean notion of speaker meaning, meaning becomes entangled with the speaker, 11 If the meaning of an utterance is shaped by some intentions of the speaker in making the utterance, and those intentions are cast within the conceptual framework with which the speaker is operating, then the meaning of an utterance will be partly determined by the conceptual framework which speakers bring to bear in communicating linguistically. 12 While, in the previous example, Smith may believe that acts derive their moral wrongness from being sinful and from being God or sinfulness. This appears, then, to result in an inconsistent triad of propositions. (1) What is meant by a moral utterance is shaped by the implicit beliefs involved (2) Two speakers can be in agreement about particular moral claims, despite having different contradictory background beliefs about the concept of moral value. (3) Speakers can only be in agreement about a particular claim if they both avow the same claim. This triad might be taken as evidence against the view that the meaning of any moral claim involves implicit suppositions regarding the status of moral claims in general (that is, some may find it most plausible to reject the first proposition in the triad because it is the weakest and most controver sial ). As an example illustrating this solution to the triad harbor skepticism about the very concept of sin
MANIPULATING METAETHICAL BELIEFS 58 disagree that war is sinful because I do no t believe in the idea of sin at all, but I think we as a prima facie result of this example that the meaning of moral wrongness can be isolated from the various beliefs about the concept of moral wrongness held by someone engaging in moralizing. However, alternative interpretations of this example do not require the repudiation of (1). It seems quite plausible that the use of the term morally wrong by a speaker varies a s a function of the socio linguistic context of the utterance One sort of response vindicating (1) might involve the idea that the richness of implicit suppositions built into the meaning of moral wrongness and other moral concepts can vary ( intra individ ually) between contexts of utterance in an accordion like manner; speakers might adjust the meaning of moral wrongness such that only some core concept of moral wrongness is shared with an interlocutor who might have vastly different beliefs about the natu re of moral wrongness. Since those vastly different beliefs about the nature of moral wrongness simply are not relevant to the particular context of moral discourse, it might be practical and reasonable to deflate the meaning of moral wrongness such that w hat is left is some common core of the meaning which the moral wrongness share. Alternatively, a defender of (1) could make the argument that (3) admits of an ambiguity regarding the criterion for determining the sameness of a claim. Sameness might be established by some threshold of similarity between the content of claims rather tha n perfect identity. Thus Jones and Smith could truly agree that war is morally wrong in Smith
MANIPULATING METAETHICAL BELIEFS 59 idiolect than it does in Jones A defender of (1) could also deny (2) that speakers are actually in agreement and claim instead that the speakers are mistaken in believing that they are in agreement. This interpretation might further involve making the distinction between full and part ial agreement, reading (3) as true of full but not partial agreement, and finally denying (2) insofar as agreement is read as full agreement. I offer these replies in defense of (1) not to support a decisive argument that (1) is true, but rather as evidence at least to support the tenability of (1 ). T his issue requires more attention and analysis before the truth of (1) can be more thoroughly assessed. Given that these approaches to resolving this inconsistent triad that do not involve the repudiation of (1) are potentially valid this section shal l proceed under the assumption that (1) is tenable. In addition to this concern about moral agreement, o ne might argue against the Gricean approach in C 3 that there just are linguistic conventions for use of the term moral wrongness determined by the Eng lish language, and therefore there is a true or correct account of the meaning of moral wrongness (and other moral terms) because such an account would perfectly describe the conventions of the English language. This point might be taken to suggest that a monistic account of moral semantics is the most natural approach. I offer two responses of varying degrees of modesty. The more modest (and more pertinent) response is just that one can deviate from and prescribe revisions to a linguistic convention, and thus it does not follow from this point about the conventions of the English language that pluralistic outlooks toward moral semantics are incoherent or even philosophically unilluminating. The more contentious response is that there may no t be a single linguistic convention for the use of the term moral wrongness (and other
MANIPULATING METAETHICAL BELIEFS 6 0 moral terms) according to the English language, even holding the context of utterance constant T his more contentious response invol ves a grand claim pertaining to langu age, thus a proper assessment cannot be accomplished here. However, if one finds plausibility in this latter contention, the monistic outlook would become quite unattractive, if not utterly untenable. Given that nothing claimed here absolutely refutes the propositions constitutive of the monistic outlook, I have nonetheless presented some reasons to suppose that the pluralistic way of interpreting accounts of moral semantics has a number of advantages It is, as argued here, coherent to hold either a monist ic or pluralistic outlook on moral semantics, but potentially more advantageous to advocate the latter over the former. The M eta E thical T heoretical V irtues The discussion hitherto, insofar as it has concerned the theoretical advantages of competing outlo oks regarding accounts of moral semantics, has been concerned with the theoretical value of meta meta ethical views ( i.e., the particular outlook taken toward the project of producing a semantic account of moral discourse). In this section, I shall sketch what I refer to as some meta ethical theoretical virtues (pertaining specifically to projects in moral semantics rather than other areas of inquiry within meta ethics); 13 I shall enumerate the ways regarding just how (and why) one could select some accounts of moral semantics over others. Insofar as one must utilize some means to determine the correct answer to any question, monistic outlooks on moral semantics require some criterion, either explicit or implicit, for determining the purportedly correct account of moral semantics. This requirement for a criterion, however, holds equally for pluralistic outlooks of moral semantics. Although the move from monistic to pluralisti c outlooks of
MANIPULATING METAETHICAL BELIEFS 61 moral semantics involves the repudiation of the assumption that there is a single, correct semantic account for all instances of genuinely moral talk, an advocate of a pluralistic outlook may still distinguish between better and worse semanti c accounts of the meaning of moral language. The criterion for distinguishing between better and worse accounts is, I shall argue, not a single criterion after all, but a collection of multiple criteria. Although the kind of approach I take in explicating these multiple criteria for the apprehension of meta ethical accounts is not strictly incompatible with a monistic outlook, it fits much more naturally with a pluralistic outlook. To suppose that there is a single, correct account of what all genuine mora l terms mean requires that the criteria of apprehension are only relevant to the epistemic justification for believing the account to be true, and not constitutive in any way of the (fixed) truth or correctness of the account. On the other hand, although e ven within a pluralistic outlook of moral semantics, there might be some fixed truth about what any given speaker means in making a moral utterance on some actual occasion, the pluralistic outlook opens the possibility for meta ethicists to develop their a ccounts as prescriptive vis vis some new, modified way of engaging in moral discourse. 14 A pluralistic approach to moral semantics opens the possibility that some accounts of moral semantics (or, perhaps, some components of these accounts) are prescriptiv e or normative rather than wholly descriptive. 15 Moreover, insofar as the pluralistic outlook allows for the presence of normative components within a semantic account of moral language, the theoretical value of an account of moral semantics might be constr ued as constitutive of its correctness. Thus the explication of specifically meta ethical theoretical virtues is quite valuable for engaging in such projects inasmuch as it
MANIPULATING METAETHICAL BELIEFS 62 clarifies the means to justify such prescriptions. The discussion to follow is inte nded to provide an initial outline of some theoretical virtues that are central to semantic accounts qua theories of the meaning of moral language (other more general theoretical virtues may apply as well). Explanatory relevance. Since the purpose of acco unts of moral semantics is to elucidate or explain the meaning of moral language, accounts of moral semantics must attain a certain degree of relevance to the appropriate domain (moral discourse). As one may have reason to harbor a skeptical attitude towar d the existence of any universal features of moral discourse, it may be more accurate to refer to domains of moral discourse rather than a single domain, and thus an account of moral semantics is explanatorily relevant inasmuch as it is relevant to whichev er of these domains it is intended to elucidate. Further, although this dimension of theoretical evaluation might appear at first glance to be bivalent in form (that is, it may appear that semantic accounts either just fail or succeed in being explanatoril y relevant), since the explanandum (a kind of moral discourse) bears many features (e.g., affect induction, aptness for embedding in inferential reasoning, truth aptness, etc.), the explanation (an account of moral semantics) will be relevant to the explan andum to the extent to which the discourse type it describes includes these features; there are just as many possible degrees of explanatory relevance as there are disparate but characteristic features of moral discourse. Accounts of moral semantics bear ing normative components, though indeed intended to introduce some way of engaging in moral talk with new features, may still retain some partial connection to existing practices of moral discourse. Insofar as
MANIPULATING METAETHICAL BELIEFS 63 normatively laden accounts retain such a conne ction, they are apt for evaluation with respect to the theoretical virtue of explanatory relevance in just the same way that purely non normative accounts are. The normative components of such accounts of moral semantics, though, are not subject to evaluat ion on the basis of explanatory relevance (perhaps because it is not their function to explain anything at all). The normative components are apt for evaluation on the basis of the other theoretical virtues, however. Value for normative inquiry Semantic accounts, just like any kind of theoretical account, are intended to have implications that do something for us; there are reasons that motivate the fundamental i s (morally) particular kind of linguistic activity for the sake of cataloguing human activities, the greater reason perhaps stems from the connection to substantive ethical inquiry (questions pertaining to what one should do). Moral semantics provides import for substantive ethics because there are questions pertaining to substantive ethics whose answers will be shaped by (our understanding of) the meaning and purport of moral claim s. This connection to substantive ethics becomes especially salient in matters pertaining to both intra and inter cultural moral dispute; if we accept some kind of meta ethical relativism, then our understanding of the need and potential for a resolution of inter cultural moral dispute may shift accordingly. The extent of justification of moral beliefs, which may rightly be categorized as an issue pertaining to moral epistemology, will also be profoundly influenced by the meaning and purport of moral claim s. Moreover, we want our practice of moralizing to be vindicated if possible. We want to
MANIPULATING METAETHICAL BELIEFS 64 have good (justifying) reasons to believe that things like mass murder are morally unacceptable, and to be capable of providing these reasons to dissuade others, not j ust on acts. That want gives rise to another goal the goal to have a clear apprehension of the issue, even if it fails to turn out the way we originally wanted it to tu rn out. An account of moral semantics that helps bring us closer to this goal, even if it does so only through influencing an intermediary domain of inquiry such as moral epistemology, bears some degree of value for normative inquiry. Metaphysical tenabili ty Accounts of moral semantics almost always involve some kind of relationship to metaphysics. Although it is conceivable for some account of semantics to be compatible with many metaphysical assumptions, many analyses of moral language tend to have a cl ose connection to various metaphysical commitments. For example, moral realism of any kind posits some sort of moral reality natural or non natural and any kind of posited reality constitutes a metaphysical supposition. Some accounts of moral semantics of course, might entail a greater number of metaphysical commitments than others. The mere burden of such metaphysical commitments does not by itself necessarily provide a reason to favor the account any less. However, given that those metaphysical commi tments are essential to the account, a rejection of those commitments on independent grounds (for example, a rejection of the tenability of moral properties or facts) does ipso facto provide reason to devalue the account of moral semantics. Of course, any tacit commitments to false beliefs entailed by a semantic account do not provide evidence that the semantic account in question fails to describe what actual
MANIPULATING METAETHICAL BELIEFS 65 people are meaning when they talk morally; such considerations are only pertinent to the evaluatio n of normative components of semantic accounts. The way to proceed from such a situation in which metaphysical commitments essential to the account have been rejected on independent grounds is not at all clear, and will depend ultimately on the status of c ompeting accounts of moral semantics. If the overall virtue of the account exceeds or remains equivalent to the virtue of competing accounts, then one may have reason not to reject the account entirely. However, if the metaphysical commitments essential to the account are deeply troubled, one might be motivated in response to construct a similar account that avoids the problematic metaphysical commitments, such as a form of fictionalism in which the importance of the literal metaphysical implications of the meaning of a claim are downplayed. Psychological tenability Just as accounts of moral semantics may require certain metaphysical commitments, accounts of moral semantics may essentially require a certain explanatory story about language and its relatio nship to psychological processes. For the sake of otherwise) to psychological states or processes, and the nature of the psychological processes purportedly connected t psychological commitments For example, e xpressivism in meta ethics can be roughly characterized as the view that moral utterances lack descriptive content but instead are expressions of particular psychological states such as affective states. One might take expressivism, therefore, as the view that there is some kind of expression relation between (non descriptive content bearing) psychological states or processes and utterances that
MANIPULATING METAETHICAL BELIEFS 66 explains the meaning of moral talk. Thi s very notion of an expression relation is just the commitments. ipso facto reasons to take issue commitments are ipso facto reasons to take issue with the account. However, both the philosophy of mind and empirical psychology together constitute the appropriate area of inquiry for determining t may just be experimentally determinable facts of the matter about the psychological processes involved in the performance of moral speech acts and related relevant psychological processes. These facts might help us develop our understanding of a moral judgment qua a psychological process, as well as the relation between moral judgment and moral language. Further, there may be determinable psychological facts of the matter about why moral judgments have the phenomenological features that they do (as well, of course, as to the extent of the universality of the phenomenological features in question). universal or objecti ve in nature. One possible ( realist or cognitivist ) explanation for this introspection is that moral beliefs are just the same kind of psychological entities as factual beliefs. Experimental methodology is clearly capable of being utilized to assess this question and determine why ( i.e., the underlying ca uses accounting for how) moral beliefs might seem to be universal or objective in nature. The answer to such a question may or may not turn out to demonstrate that moral beliefs are the same kind of thing as factual beliefs. 16
MANIPULATING METAETHICAL BELIEFS 67 Just as challenges to the me taphysical commitments of accounts of moral semantics need not necessarily provide reason enough to reject the accounts outright, altering the account in some minimal way, s uch as adopting a form of fictionalism. The picture here, again, is not one in which problems with an account force us to consider absolute rejection as the only alternative to absolute acceptance. Accounts of moral semantics, so long as one accepts some f orm of a pluralist outlook, can be evaluated with respect to overall virtue relative to competing accounts. Loosening the imperious grip of a monistic outlook on moral semantics allows for a certain kind of flexibility; it grants the possibility that an ac count of moral semantics needs only apply to some form of moral discourse, and may not offer the final and absolutely correct understanding of moral language as some people actually use it. This pluralistic move allows meta ethical accounts of moral langua ge to be thought of, in part, as a way that we could and even ought to engage in moral discourse even if we do not actually do so already. Unlike with metaphysical commitments, the psychological commitments of an account of moral semantics only need to be tenable for a normative theory insofar as those commitments are psychologically possible and do not describe the actual psychological explanation that accurately represents most ordinary cases of moralizing; whereas we cannot simply make a particular metap hysical system true by willing it, it is possible for us to decide to engage in certain practices corresponding to a psychological story involved in some proposed meta ethical view. 17 Summary and A dditional P hilosophical Issues
MANIPULATING METAETHICAL BELIEFS 68 Although the approach I have advanced (a pluralist outlook) does not preclude the possibility of a certain kind of de facto monism about moral semantics (viz. the possibility that a single semantic account can in fact adequately explain the meaning of every moral utterance), the plur alist outlook boasts several advantages over the monist outlook heterogeneity (and the evidence against the actual homogeneity of moral concepts), it adds stability to semantic accounts by reducing their vulnerability to counter examples, and it better accommodates the understanding of meaning as (Gricean) speaker meaning (which may best serve the meta language). Nevertheless, a more detailed analysis of a number of issues may impact the assessment of the value of monist and pluralist outlooks. Such issues would at least include analyses of meaning and philosophy of language more generally as well as an analysis on the fact of th e matter regarding the sui generis heterogeneity of moral concepts. In addition to the assessment of monist and pluralist outlooks on moral semantics, I have presented and discussed several theoretical virtues pertaining to meta ethics explanatory releva nce, value for normative inquiry, metaphysical tenability, and psychological tenability. The outlook one takes toward moral semantics ( i.e., a monistic or pluralistic outlook) shapes how one understand the relation between these theoretical virtues and the correctness of any given semantic account of moral discourse. There may be additional important meta ethical theoretical virtues worthy of discussion. Most crucial to the present research, however, is the virtue of psychological tenability. In short, empi rical research pertaining to the area of the present study has import for meta ethics in virtue of providing evidence about the similarities and dissimilarities between moral
MANIPULATING METAETHICAL BELIEFS 69 attitudes and non moral attitudes, which has implications for the general psychological framework with which we may utilize to distinguish between moral and non moral attitudes. F urther, such a framework has implications for the virtue of competing accounts of moral semantics which under a pluralistic outlook of moral semantics partially constitutes a ny If in particular, it turned out that objectivism about moral issues was caused by very different underlying factors than objectivism about factual or conventional issues, there may be reason to su spect that there are divergent psychological processes at work in the cognition of moral issues and factual or conventional issues. Although the present research has neither sufficiently established nor refuted this hypothetical feature of moral attitudes, future empirical research in a similar vein may shed a substantial amount of light on the matter.
MANIPULATING METAETHICAL BELIEFS 70 Appendix A References Graham, J., Haidt, J., & Nosek, B. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of person ality and social psychology 96 1029 10 46. doi:10.1037/a0015141 ence meaning, and word meaning. Foundations of Language 4 225 242. Haidt, J. (2001). The e motional d og and i ts r ational t s ocial i ntuiti onist a pproach to moral judgment. Psychological Review 108 814 834. doi:10.1037//0033 295X. Haidt, J., & Joseph, C. (2004). Intuitive ethics: How innately prepared intuitions generate culturally variable virtues. Daedalus, 133, 55 66. Haslanger, S. (2005). What are we talking about? The semantics and politics of social kinds. Hypatia 20, 10 26. Longino, H. E. (1995). Gender, politics, and the theoretical virtues. Synthese 104, 383 397. Nisbett, R. & Cohen, D. (1996). Culture of Honor: The Psych ology of Violence in the South Boulder, CO, US: Westview Press. Sinnott Armstrong, W. (2007). Moral Skepticisms USA: Oxford University Press.
MANIPULATING METAETHICAL BELIEFS 71 Appendix A Footnotes 1 As Walter Sinnott Armstro ng has pointed out, the name moral linguistics may be best suited to describe this overall area of inquiry as many topics of consideration taken up in it pertain to analyses of the pragmatics of moral language (Sinnott Armstrong, 2007). I shall use the ter m moral semantics to encompass what might be better labeled moral linguistics because despite the potential relevance of pragmatics in such analyses, meta ethical accounts of moral or value language must always help answer the semantic question that is, they must shed some light on the semantics of words like should or ought 2 Meta ethicists, of course, may also be interested in other explananda besides those relating just to morality; one might be more interested in value in general, and thus be more in terested in developing the semantics of value or normativity In this section I restrict the discussion to explananda relating to morality, but nearly everything claimed in this section about moral discourse may also be applied with ease to meta ethical projects pertaining to value or normativity more generally. 3 It may be noted that the various approaches in developing accounts of moral semantics reflect features of meta meta ethical s tances, viz. stances about how to understand the purport of meta ethical projects. The features of these stances are apt for philosophical evaluation in much the same way that meta ethical accounts are. 4 A discourse type is defined here as a set of commun icative traditions known by some community that are used to transmit information pertaining to a particular domain. We might think of a discourse type as a family of related linguistic tools that together perform a particular function. For example, we might characterize technical vocabulary
MANIPULATING METAETHICAL BELIEFS 72 in some area of cellular biology as a discourse type. Each item of vocabulary has a specialized meaning or function for talking about cellular processes or cellular structures. What unites the whole set of vocabulary as a discourse type is just that it is used for talking about cellular biology, and we might construe some of the vocabulary within that discourse type as constituting an even more specific, different discourse type. All that unifies the parts of a discou rse type is that the constituent linguistic tools are all used for one particular goal or purpose. 5 The terminology here may require some explanation. What makes this kind of actual endorsement of a multip licity of semantic accounts of moral discourse, but rather that under this outlook, it is possible to endorse a multiplicity of semantic accounts. 6 Even proponents of pluralistic kinds of accounts, however, could hold something like monistic or imperious attitudes about their accounts; they might believe that their proposed account ought to be accepted universally because it is better in some way than all of its alternatives, and thus attribute privilege to discourse type M over other posited discourse typ es intended to account for moral language. 7 In this example, I am casually referring to the distinction between deontological moral terms and aretaic moral terms, but the division between sui generis kinds of moral discourse might turn out better to be di vided rather differently, such as through some alignment with the foundations in Moral Foundations Theory (Haidt and Joseph, 2004), or perhaps even between agent centered and non agent centered moral imperatives. I leave this as an open question for future investigation.
MANIPULATING METAETHICAL BELIEFS 73 8 Furthermore, different kinds of values reflect different concepts and thus different forms of discourse, as shall be discussed in greater depth in the passages to follow. 9 Such an examination would include a discussion regarding the assu mption that these various moral concepts cannot be reduced to one ultimate fundamental value. Philosophical inquiry regarding value monism and pluralism (not to be confused with monism and pluralism about moral semantics) is of course directly relevant to this question, as are certain empirical findings pertaining to moral reasoning, such as the phenomenon of moral dumbfounding (see Haidt, 2001). 10 We might add that any discourse type is constituted by particular linguistic i ntentions in making an utterance. If, say, my speaking jokingly is (partly) understandable as my intention to say something that the audience would find amusing, and we might suppose that there is some discourse type say comedic discourse whose constit uent linguistic tools include the same kind of tools whose function is to produce amusement in interlocutors, then my intentions in speech match up to some linguistic tools within this comedic discourse type, and therefore can be said to fit this discourse type. This aligns with the aforementioned characterization of discourse types as being unified or classified on the basis of their purpose or function. 11 This Gricean notion of meaning, however, does not entail that meaning is only 12 We mig ht suppose that a conceptual framework is reducible to some set of concepts, schemas, beliefs, or attitudes. A precise reduction will designate exactly which of these potential constructs to choose from. I do not intend to presuppose the truth of any
MANIPULATING METAETHICAL BELIEFS 74 one p articular account about the nature and constitution of conceptual frameworks but rather intend to leave the issue open, but assume henceforth that conceptual frameworks involve certain cognitive constructs that I will call implicit beliefs 13 It should be noted that the notion of a theoretical virtue here need not be inflated with philosophical ba ggage associated with the term virtue All that is required is a certain minimal recognition of each virtue as a domain of evaluation of a theoretical account. 14 It would seem most natural here to take pluralism about moral semantics as the endorsement specifically of C 1b rather than C 1a for if a proposed way of engaging in moral discourse is meant as prescriptive rather than as descriptive of our existing practi ces, then it may not be the case that some instances of moral language belong to the proposed discourse type, whereas it would still be the case that moral language could belong to the proposed discourse type. 15 The distinction between normative and non no rmative meta ethical projects is ameliorative approaches and conceptual approaches, respectively (Haslanger, 2005). Just as Haslanger notes regarding conceptual and ameliorative approaches, meta ethical projects are not wholly normative or non normative, but the two kinds of projects represent distinct goals. 16 Such investigation may also provoke us to scrutinize the entire folk psychol ogical concept of belief, which naturally would have profound implications for our understanding of what we call factual beliefs and moral beliefs
MANIPULATING METAETHICAL BELIEFS 75 17 Of course, since introspection by no means grants us transparent access to the psychological story that re presents what is going on when we act, we might be fooling ourselves about what we are really doing.
MANIPULATING METAETHICAL BELIEFS 76 Appendix B: What is meant by objectivity? Although the definitions of o bjectivity and subjectivity are the subjects of intense philosophical analysis another if it relies less on p. 5). Moral objectivism, as the term is used here, is understood as the view that there are moral trut hs that do not rely on the makeup or position of any individual for which the moral content holds true (and further, that moral content is not perspective dependent simply on the basis of being moral content). Moral objectivism can vary with regard to its width of scope. One might hold a broad kind of moral objectivism that applies to all issues categorized as moral. This might be understood as a general meta ethical theory about the status of moral discourse and values. Alternatively, one might hold a narr ow kind of moral objectivism that pertains to each moral issue on a particular basis; one might simply hold objectivism about a particular moral issue or some group of moral issues. For example, although one might disavow the claim that there is an objective truth of the matter about the moral rightness of acting kindly toward strangers, one might nevertheless avow that there is an objective truth of the matter about the moral wrongness moral domain, or are morally relevant. Moral objectivism about the truth of some moral content M entails that contradictory judgme nts of the moral content M cannot all be true or correct. 1 The view that two or more contradictory judgments of the same moral content M can in principle possibly all be correct because their correctness relies on the makeup
MANIPULATING METAETHICAL BELIEFS 77 or position of the subject maki ng the judgment shall be labeled as moral subjectivism about the moral content M. 2 Affirming the view that two or more contradictory judgments about some moral content M can all possibly be true or correct does not imply moral subjectivism about the moral content M, though, unless the reason for holding this view about these judgments of the moral content M involves some particular commitment to a view regarding the truth maker for the moral content M viz. the view that moral content M is made true or false by some features of the owner of a belief involving the o accept that it is impossible in principle for two contradictory evaluative judgments about the same moral content M to b e correct or true implies a moral objectivism about the issue involving moral content M by virtue of modus tollens Although it is coherent for a moral subjectivist to deny that contradictory evaluative judgments of the same moral contents in fact are all correct or true, it is not coherent for a moral relativist to deny that contradictory evaluative judgments of the same moral contents cannot possibly all be correct because it is possible for any moral judgment to be correct given the appropriate makeup or position of the moral subject making the judgment. However, the meaning of possibly in this context might vary in important ways. If a relativist read possibly in this case to involve the various potential makeups or positions only within a particular cul tural framework, then a relativist might have good reason to deny that the contradictory views could all be correct. Possibly is intended to be read in this context in a broader sense. It should be noted that these notions of moral objectivism and moral r elativism are meta ethical in nature and not normative in nature. A normative form of moral
MANIPULATING METAETHICAL BELIEFS 78 relativism dictates what one ought to do in facing moral disagreement, viz. act tolerantly or respectfully toward the proponent of the conflicting moral judgment (s ee Quintelier and Fessler, 2011 for a conceptual analysis and review of empirical work done on the psychology of normative moral relativism). Meta ethical relativism and normative moral relativism may relate to each other in important ways; the relationshi p between meta ethical relativism and normative moral relativism would provide a useful subject matter for future studies in moral psychology.
MANIPULATING METAETHICAL BELIEFS 79 Appendix B References Nagel, T. (1986). The View from Nowhere New York, NY: Oxford University Press. Quintelie r, K., & Fessler, D. (2011). Varying versions of moral relativism: The philosophy and psychology of normative relativism. Biology and Philosophy 27, 95 113. doi:10.1007/s10539 011 9270 6
MANIPULATING METAETHICAL BELIEFS 80 Appendix B Footnotes 1 This feature of moral objectivism follows from the assumption that the law of non contradiction applies to all objective truths or facts. It might be wor thwhile to empirically determine whether lay individuals actually do universally accept the law of non contradiction. 2 This account may turn out to be philosophically untenable insomuch as this non objective feature of a subject matter implies that ostens ibly contradictory judgments about the same moral contents are actually about different contents, and thus are not strictly speaking contradictory.
MANIPULATING METAETHICAL BELIEFS 81 Appendix C : Materials Section 1 V ersion A : Save Lives C ondition assigned to read. During missions abroad soldiers have to apply to certain rules. This is called a mandate. During a confl ict in 1994 in Congo, Africa, the mandate of the UN peace keeping force present in the country was to use armed violence to end the conflict. As a result of several warnings beforehand and during the conflict by the UN commander in chief on the spot, the C anadian lieutenant general Romo Dallaire, UN soldiers were allowed to conduct offensive actions. They were both allowed to use their weapons to protect the civilian population, as well as when they themselves were attacked. The mandate was based on the pr Now, try to imagine what it would have been like to be in the position of lieutenant general Romo Dallaire. What are some of the specific ways that you would try to ensure that the UN soldiers followed the mandate based on the princ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ _____________________ __________________ _________________________________ ________________________________________________________________________ for you in your daily life? _______________ _________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ _______________________________________ ________________ _________________ ________________________________________________________________________
MANIPULATING METAETHICAL BELIEFS 82 Section 1 V ersion B : Do Not Kill C ondition assigned to read. During mis sions abroad soldiers have to apply to certain rules. This is called a mandate. During a conflict in 1994 in Congo, Africa, the mandate of the UN peace keeping force present in the country was to refrain from using any armed violence to end the conflict. D espite several warnings beforehand and during the conflict by the UN commander in chief on the spot, the Canadian lieutenant general Romo Dallaire, UN soldiers were not allowed to conduct any offensive actions. They were only allowed to use their weapons when they themselves were attacked. The mandate was Now, try to imagine what it would have been like to be in the position of lieutenant general Romo Dallaire. What are some of the specific ways that you would try to ensure ________________________________________________________________________ ________________________________________________________________________ ________________________ ________________________________________________ _______________________________________ _________________________________ ________________________________________________________________________ Now think about the principle of the mandate. What does the mean for you in your daily life? ________________________________________________________________________ ________________________________________________________________________ _______________________________________________________ _________________ _______________________________________ _________________________________ ________________________________________________________________________
MANIPULATING METAETHICAL BELIEFS 83 Section 2: PANAS For the following section, you will be presented with a number of words that describe different feelings and emotions. Read each item and then list the number from the scale below next to each word. Please write the number in each blank that indicates to w hat extent you feel this way right now, that is, at the present moment. 1 2 3 4 5 Very Slightly or Not at All A Little Moderately Quit e a Bit Extremely _____________ 1. Interested _____________ 11. Irritable _____________ 2. Distressed _____________ 12. Alert _____________ 3. Excited _____________ 13. Ashamed _____________ 4. Upset _____________ 14. Inspired _______ ______ 5. Strong _____________ 15. Nervous _____________ 6. Guilty _____________ 16. Determined _____________ 7. Scared _____________ 17. Attentive _____________ 8. Hostile _____________ 18. Jittery _____________ 9. Enthusiastic _____________ 19. Activ e _____________ 10. Proud _____________ 20. Afraid
MANIPULATING METAETHICAL BELIEFS 84 Section 3: Moral Dilemma Scenario: You are a surgeon with a number of patients. Five of them need organ transplants. Each of them needs a different organ or they will surely die. You have another patient who is healthy and would will die. 1. In this situation, would you perform this transplant? Please read statements A and B and then circle the number that best matches what you believe. A. Yes, I would perform the transplant. B. No, I would not perform the transplant. 1 2 3 4 5 6 7 8 Certain of B Certain of A 2. Do you think that your hypothetical choice is the right or best choice to make in this sce nario? A. Yes, I believe that my choice selected in 1. is the right or best choice to make. B. No, I do not believe that my choice selected in 1. is the right or best choice to make. 1 2 3 4 5 6 7 8 Certain of B Certain of A 3. Suppose someone else participating in this study holds an opposing view on this issue that directly contradicts your own view. What do you think about this conflict in views? Please read statements A and B and then circle the number that best matches what you believe. A. In principle, at least one of us must be mistaken; we cannot both be right. B. In principle, it is a possibility for neither of us to be mistaken; we could both be right. 1 2 3 4 5 6 7 8 Certain of B Certain of A
MANIPULATING METAETHICAL BELIEFS 85 4. Again, imagine this same person who holds a view that directly contradicts your own. Do you think that it is possible that this person has good reasons to support her or his view? A. This person might have reasons supporting her or his view that I see as goo d reasons. B. This person could not have reasons supporting her or his view that I see as good reasons. 1 2 3 4 5 6 7 8 Certa in of B Certain of A