Self - Regulation and

In this paper we explore a self-regulatory perspective on the self-evaluative moral emotions, shame and guilt. Broadly conceived, self-regulation distinguishes between two types of motivation: approach/activation and avoidance/inhibition. We use this distinction to conceptually understand the socialization dimensions (parental restrictiveness versus nurturance), associated emotions (anxiety versus empathy), and forms of morality (proscriptive versus prescriptive) that serve as precursors to each self-evaluative moral emotion. We then examine the components of shame and guilt experiences in greater detail and conclude with more general implications of a self-regulatory perspective on moral emotions. keywords: moral emotions, self-regulation, shame, guilt, parenting styles Self-Regulation and Moral Emotions 3 Tracing the Self-Regulatory Bases of Moral Emotions Until recently, morality was thought to entail moral reasoning that emotions only served to undermine (Kohlberg, 1984; Piaget, 1951). Emotions similarly were regarded as “hot” feelings to be contrasted against “cold” cognitions, and emotion and morality were thus discrete areas of investigation in psychology. Interestingly, as psychologists increasingly began to understand emotions as inherently cognitive (e.g., Lazarus, 1991; Russell, 2003), morality too began to be viewed as inextricably tied to emotions (e.g., Haidt, 2001, 2007; Rozin, Lowery, Imada, & Haidt, 1999). We now view moral experience as comprising not only a list of rules and ideals, but also strong emotions that serve as moral regulators of our self-evaluations and actions. Self-regulation describes purposive (but often not self-controlled) processes motivated by some internal object or goal (Carver & Scheier, 2008). It is a perspective that can incorporate both emotions and cognitions and suggests a “syndrome” (Averill, 1982) of connected appraisals, phenomenal experiences, and action tendencies. As such, self-regulation may aid in linking emotional experiences and moral rules, and in particular the interplay between the two. Classic social psychological theory has understood morality as “an internalized representation of normative standards” that are embedded in and derived from social interaction (Baldwin, 1906; Mead, 1934; Newcomb, 1950). Our experiences of moral emotions represent the ethics, ideals, and obligations inherent in being a member of society; and these are believed to be binding on the self (see Feldman, 2006). To understand the nature of this “binding,” we need to recognize the centrality of the community to the individual—our nature as social beings and our need to belong Self-Regulation and Moral Emotions 4 (Baumeister & Leary, 1995; Mead, 1934). The development of self-regulation in the moral context fundamentally depends on our interactions as social beings and our need to maintain social relationships. In this paper we explore a self-regulatory perspective on morality and its relevance for the primary self-evaluative moral emotions, shame and guilt. We turn first to self-regulation as broadly conceived in psychology. In past research, we have applied the two basic regulatory orientations--approach/activation and avoidance/inhibition--to the moral domain (Janoff-Bulman, Sheikh, & Hepp, 2008) and have used this distinction to differentiate shame versus guilt experiences (Sheikh & Janoff-Bulman, 2010). In addition to elaborating on the self-regulatory bases of shame and guilt, the purpose of this paper is to provide an understanding of the development of moral emotions through parental socialization. We propose that the central dimensions of parenting (i.e., parental restrictiveness versus nurturance) are reflections of approach and avoidance regulation and promote two different moral regulatory systems that in turn underlie shame and guilt. In the final sections of the paper, we examine these moral emotions in greater detail and explore their implications for emotion and morality. Self-Regulation: Approach (Activation) versus Avoidance (Inhibition) Findings from diverse fields of psychology support a central distinction between two self-regulatory orientations: an approach system based in behavioral activation and an avoidance system based in behavioral inhibition (for reviews, see Carver & Scheier, 2008, and Gable, Reis, & Elliot, 2003). Thus Carver and Scheier’s (1990, 1998, 2008; see also, Carver, 2006) model of self-regulation conceptualizes behavior as movement to or from one of two referents; movement towards a goal, a desired end-state, is regulated Self-Regulation and Moral Emotions 5 by a discrepancy reducing feedback loop, whereas movement away from an “anti-goal,” an undesired end-state, is regulated by a discrepancy enlarging feedback loop. The approach system is responsible for the activation of movement towards positive goals, and the avoidance system leads to the inhibition of movement towards goals that may lead to negative or painful outcomes. Carver and Scheier’s model reflects the influence of Gray (1972, 1982, 1990), who posits two regulatory systems, namely the Behavioral Activation System (BAS) and the Behavioral Inhibition System (BIS), based on principles of conditioning and distinctions between positive and negative reinforcers, rewards and punishments, and appetitive and aversive stimuli. The BAS, the appetitive motivational system, is sensitive to reward and promotes the activation of behavior. The BIS, the aversive motivational system, is sensitive to punishment and promotes suppression, or inhibition, of behavior. Gray’s conceptual nervous system highlights the activation and inhibitory behavioral tendencies and appetitive and aversive motivations that are derived from positive and negative reinforcers, respectively. The distinction between the BIS and BAS systems is supported by studies of their neural substrates (for a summary, see Carver & Scheier, 2008; also see Carver & Harmon-Jones, 2009). Davidson and colleagues (e.g., Davidson, 1995, 1998; Davidson, Ekman, Saron, Senulis, & Friesen, 1990; Sutton & Davidson, 1997) have demonstrated that relatively greater left frontal activity is associated with the BAS and relatively greater right frontal activity is associated with the BIS. Earlier studies manipulating affect found that pleasant film clips eliciting positive feelings, pleasant tastes, and monetary incentive all increased left-sided anterior activation. Unpleasant film clips Self-Regulation and Moral Emotions 6 inducing fear and disgust, unpleasant tastes, threat of monetary loss, and preparation for giving a speech (among “social phobics” only) increased right-sided anterior activation (Davidson et al., 1990; Sutton & Davidson, 1997). Additionally, Carver and White (1994) have also elaborated upon the BIS and BAS as personality dimensions and constructed the BAS/BIS Scales to assess chronic dispositions for these self-regulatory orientations. Overall, approach and avoidance regulation differ in both focus and action tendency. Approach regulation focuses on a positive end-state or goal and involves activation. Avoidance regulation, in contrast, focuses on a negative end-state or “antigoal” and involves inhibition. Two Forms of Moral Regulation We recently applied the distinction between approach and avoidance (i.e., behavioral activation versus inhibition) to morality (Janoff-Bulman et al., 2009). From this perspective, moral regulation involves the development and operation of two regulatory systems, in which morality and immorality form the goal and the “anti-goal,” respectively, and activation and inhibition define distinct behavioral orientations. Specifically, the prescriptive system regulates the approach of positive, moral end-states comprising objects and goals that are desired and rewarding, whereas the proscriptive system regulates the avoidance of negative, immoral end-states comprising objects and goals that are undesired and punishing. The behaviors motivated by the proscriptive and prescriptive moral systems are linguistically represented, respectively, by “I should,” signifying the activation of behavior (e.g., “I should help others”), versus “I should not,” the inhibition of behavior (e.g., “I should not harm others”). Morality in the proscriptive system is particularly Self-Regulation and Moral Emotions 7 focused on restraint and curbing of negative desires, or temptations. This is the domain, for example, of many of the popularized “seven deadly sins,” which involve excesses or indulgences that presumably call for self-control. The moral injunctions here are often second-order desires, those that are derived from reflecting upon and evaluating our firstorder immoral temptations. Proscriptive immorality then is characterized by giving in to those temptations, those motivations to engage in undesirable acts. Prescriptive morality, in contrast, concerns the activation of desirable behaviors. These include not only prosocial actions, involving acts of benevolence, charity, and generosity, but also industriousness, self-reliance, and hard-work, all of which involve the activation of behavior evaluated as good (see Janoff-Bulman et al., 2008). Prescriptive immorality is characterized by the absence of a desire or motivation to engage in “right” conduct. Moral injunctions may call upon second-order desires to overcome preferences for inaction, inertia or apathy; more often, however, they may actually reflect first-order desires: one may want to volunteer or be inclined to help a friend, in the absence of countervailing desires (Janoff-Bulman et al., 2009). Research on the two regulatory systems has demonstrated a moral asymmetry. Proscriptive moral regulation is the harsher of the two systems; it is condemnatory and strict, whereas prescriptive moral regulation is commendatory and less strict (JanoffBulman et al., 2009). In studies involving approach-avoidance priming, linguistic representations associated with approach and avoidance, and associations with individual difference measures of approach-avoidance sensitivity (i.e., BIS/BIS scales) we found support for the links between avoidance and proscriptive morality, and between approach and prescriptive morality. Further, proscriptive morality was regarded was Self-Regulation and Moral Emotions 8 regarded as more mandatory, whereas prescriptive morality was perceived as more discretionary; and greater blame was attributed for proscriptive immorality than prescriptive immorality, but greater moral credit was attributed for prescriptive morality than proscriptive morality (Janoff-Bulman et al., 2009). Support for distinguishing between the two regulatory systems is also evident in research by Kochanska and colleagues (Aksan & Kochanska, 2005; Kochanska, 2002; Kochanska, Coy, & Murray, 2001) on the development of morality in young children. These researchers distinguished between “do’s” and “don’ts”: do’s involved activating and sustaining an activity (e.g., toy cleanups), and don’ts involved prohibitions and suppressing behaviors (e.g., not playing with a forbidden attractive toy). In their longitudinal research with children at 14, 22, 33, and 45 months, they found that do’s were more challenging for children and there was greater compliance with don’ts (i.e., prohibitions). In addition, compliance was stable across ages within context (do’s and don’ts), but not across the two contexts (Aksan & Kochanska, 2005; Kochanska, 2002; Kochanska et al, 2001). Kochanska et al. (2001) conclude that their data provide “impressive evidence of substantial differences” between do’s and don’ts in early selfregulation. Morality and immorality are internalized standards representing rewarding endstates to approach (“shoulds”) or punishing end-states to avoid (“should nots”); as Carver and Scheier (2008) note, these standards are often dynamic and range in level of abstraction. They are symbolic representations, derived from the “categories made salient by the individual’s particular sociocultural and historic context and from the models, images, and symbols provided by the media and by the individual’s immediate Self-Regulation and Moral Emotions 9 social experiences” (Markus and Nurius, 1986, p. 954); they include moral imagery, tenets of religion, and also desired and undesired images of the self (Epstein, 1973; Gergen, 1972; Markus and Wurf, 1987; Mead, 1934; Ogilvie, 1987; Schlenker, 1985). Further, the conduct associated with these end-states varies. The liberal tradition (e.g., Kohlberg, 1981, 1984; Turiel, 1983, 2002) has conceived of morality as based on people’s understandings of “rights, justice, fairness, and the welfare of people” (Turiel, 2006, p. 10). More recently, moral psychology has expanded the domain of moral standards and action to include those pertaining to sexual behavior, work ethic, cleanliness, and other instances of more personal conduct (e.g., Haidt, 2001, 2007; Schweder, Much, Mahapatra, & Park, 1997). As these standards of conduct are internalized during the socialization process, the child enters the moral world and is able to navigate accordingly. Moral Development: A Self-Regulatory Approach Moral socialization is the process whereby an individual internalizes norms of right and wrong from relational experiences with others. The parental-figure, or “the carrier of culture” (Sullivan, 1953, p.35), is the first vehicle through which transmission of standards of conduct occurs; the parent represents the intersection of the child and the community during his or her formative years. The child’s moral development is a transition from external regulation by the parents, as highlighted by Kohlberg’s (1971) preconventional morality, to internal regulation by the child (see also, Aksan & Kochanska, 2005; Kochanska, 1993; Kochanska et al., 2001). Parent-child interactions not only influence which standards the child will internalize, but also the nature of the child’s self-regulatory and self-evaluative capabilities (Kagan, 1984). Self-Regulation and Moral Emotions 10 Moral socialization begins around the age of two, when the child starts to “make inferences about symbolic meanings” (Kagan and Lamb, 1987, p. 10; Piaget, 1951; Case, 1985; Werner & Kaplan, 1963). Kagan and Lamb (1987) provide an example outside of the moral domain in which a child experiences distress after failing at a self-imposed task, building a toy tower. The distress alludes to some recognition of the discrepancy between the child’s own performance at building the toy tower and the representation of a more perfect one. The representation of the perfect toy tower functions as the evaluative standard, guiding behavior towards achieving that particular outcome. The socialization of morality is analogous: through relations with others, especially parental figures, children internalize moral standards from which they then regulate their behavior and evaluate themselves. The relevant goal and anti-goal then becomes attaining morality and avoiding immorality, with self-evaluative and emotional consequences for their successes and failures. Parental Restrictiveness and Nurturance Dimensions The child’s moral regulatory abilities arise from the interaction between socializing agents and the child, making both the behavior of the socializing agents and the responses of the child important. Two parenting dimensions central to socialization research are parental restrictiveness and nurturance (e.g., Baumrind, 1966; 1967; similar also to Sullivan’s [1953] security versus satisfaction and Higgins’s [1987] security versus nurturance needs). Interestingly, each dimension seems to have a regulatory function that directly parallels either avoidance or approach motivation, or more specifically, proscriptive or prescriptive regulation. We argue that parental restrictiveness socializes a primary focus on immorality and (self-oriented) anxiety, whereas nurturance socializes a Self-Regulation and Moral Emotions 11 primary focus on morality and (other-oriented) empathy (see Figure 1). While anxiety is considered to be a distressing state elicited by a perceived threat, empathy is generally used in two manners: one denotes a cognitive ability to hold the perspective of another and the other denotes an affective concern or sorrow from perceiving someone else’s distress or suffering (e.g., Batson and Shaw, 1991); our use of empathy is consistent with the latter perspective and is similar to Eisenberg and Miller’s (1987) “sympathy.” Parental restrictiveness refers to the amount of parental monitoring and the rigidity of limits set for the child, whereas nurturance refers to the amount of affective warmth expressed by the parents in parent-child interactions. Parental restrictiveness is mainly displayed by the threat or use of punishment and psychological and/or physical control, whereas nurturance includes displaying affective praise, encouragement, and physical affection. If viewed through a moral regulation lens, restrictiveness and nurturance may serve as environmental precursors for the development of moral emotions, and shame and guilt in particular. Parental Restrictiveness Given that the moral function of proscriptions is to restrict immoral behavior, restrictive parenting promotes the socialization of prohibitions and the experience of anxiety in the child (see Figure 1). In these interactions the parent is communicating to the child that engaging in the proscribed behavior will incur negative, punishing outcomes. The focus of restrictive parenting is on obedience and the inhibition of behavior (the “don’ts”), which is likely to produce anxiety because of the threat of punishment. Similar to Gray’s (Gray, 1982; Gray & McNaughton, 1996) and Carver’s (e.g., Carver & White, 1994) assertion that anxiety is an avoidance motivator (see also, Self-Regulation and Moral Emotions 12 Higgins, 1987, 1997), in moral regulation anxiety guides what the child should not do in order to avoid punishing outcomes. In their longitudinal research with young children, Kochanska et al. (2001) found that fearfulness was positively associated with children’s success at refraining from prohibited behaviors (don’ts), but was not at all associated with measures of success in the domain of praiseworthy behaviors (do’s). Certainly, the socialization of “should nots” often successfully keeps the child from engaging in harmful, even potentially dangerous, behaviors; however, a predominance of restrictive, punitive parenting in which relations are primarily threatening and lack nurturance may communicate to the child that she or he is inclined to be immoral and engage in immoral conduct, this being the reason for the parent’s highly punitive orientation. As a result, the child is likely to be highly proscriptivelyoriented, vigilant in avoiding immorality, and particularly prone to anxiety. Past researchers have indeed found a positive association between parental restrictiveness and adolescent anxiety--as well as loneliness, low academic achievement, low self-esteem, low self-reliance, and high self-derogation (for a review, see Barber & Harmon, 2002). Because adherence to proscriptions is mandatory and violations incur punishment, a highly proscriptively-oriented child is likely to be self-oriented, with concerns turned inward, and attention focused specifically on the attenuation of his or her anxiety. As a self-oriented state, anxiety disrupts the process of feeling concern for others, and thus past research has found that children of punitive parents exhibit lower levels of empathy and prosocial behavior (for a review, see Eisenberg, Spinrad, & Sadovsky, 2006; e.g., Krevans & Gibbs, 1996; Eisenberg, Fabes, Schaller, Miller, Carlo, et al., 1991). Further, in her influential research, Baumrind (1966; 1967; see also e.g., Maccoby & Martin, Self-Regulation and Moral Emotions 13 1983) crosses the restrictiveness and nurturance dimensions to distinguish between authoritative, authoritarian, permissive, and neglectful parenting styles. The authoritarian parent is high on restrictiveness and low on nurturance, relying on punitive force, a focus on threats, and obedience to socialize the child. Although adolescents of authoritarian parents were least likely to show “problem” behaviors such as drug use, they were less competent and prosocial than those of authoritative or democratic parents and also exhibited high levels of anxiety and distress (Baumrind, 1991; see also, Lamborn, Mounts, Steinberg, & Dorbbusch, 1991; Steinberg, Mounts, Lamborn, & Dorbbusch, 1991). This corresponds to expectations from a dual self-regulatory perspective: parents high on restrictiveness emphasize proscriptions rather than prescriptions. Their children are apt to manifest high levels of anxiety, a strong focus on prohibitions (e.g., avoiding drug use), and low levels of activation-based moral conduct (e.g., prosociality and industriousness). Parental Nurturance Rather than the threat of negative, punishing outcomes, parental nurturance focuses on positive, affectively rewarding outcomes that the child can achieve. The child cares for the well-being of those who are nurturing to him or her, and this empathy regulates what the child should do to maintain these warm, loving relationships. Nurturing parenting socializes prescriptions, and the presence of empathy makes prescriptive regulation largely “other-focused,” where one’s concern is turned outwards (see Figure 1). Even when they include behaviors one should do for oneself (e.g., selfreliance), they nonetheless entail the positive expectations of others. For example, Grusec, Goodnow, and Cohen (1996) have noted a relation between industriousness and Self-Regulation and Moral Emotions 14 empathy: the authors have found that establishing standards to achieve, such as routine household chores, increases children’s empathic concern and prosocial helping. These findings make sense if both behaviors are viewed as prescriptively-oriented. A predominance of nurturing interactions communicates to the child that she or he is inclined to be moral and engage in moral conduct. From the prevalence of warm, nurturing interactions arises a valued, loved sense of oneself and the ability to care for others, especially parental figures; here, a strong prescriptive orientation and empathic responding is socialized. Psychologists have long maintained that early, secure attachment to a parental figure is an important antecedent to the development of empathic concern (e.g., Bowlby, 1969; Mussen & Eisenberg-Berg, 1977; Sullivan, 1953). Studies on attachment theory have shown that secure attachment and maternal responsiveness predict children’s empathic responding (e.g., Kochanska, Forman, and Coy, 1999). The importance of secure attachment, however, is the “abundance of love and nurturance” (Eisenberg & Strayer, 1987, p. 151) between the parental figure and child. And indeed, many studies have reported that warm, nurturing parenting predicts children’s current and future empathic responses (for a review, see e.g., Eisenberg, Spinrad, & Sadovsky, 2006). Much socialization research measures empathy indirectly through prosocial behavior, finding a positive correlation between parental warmth and children’s prosociality (for reviews, see e.g., Grusec, Davidov, & Lundell, 2002; Hastings, Utendale, & Sullivan, 2007). In her research on the four parenting styles, Baumrind’s (1991) longitudinal study labeled parents who were high on nurturance and low on restrictiveness as permissive, or Self-Regulation and Moral Emotions 15 democratic; these parents were apparently supportive but not firm, making few demands and generally accepting the child’s desires and actions. Interestingly, their children were competent and prosocial, as would be expected based on parenting that focused on prescriptive morality. However, they were also likely to engage in drug use, again perhaps reflecting parental inattention to proscriptive morality. Children of authoritative parents, whose style included both firm control and nurturance were found to be prosocial and competent as well as low in “problem” behaviors such as drug use. These were parents who attended to both prescriptions and proscriptions and clearly used nurturance while still setting restrictions. The children of neglectful parents, who were low on both nurturance and restrictiveness, displayed the most problem behavior and were the least competent and prosocial; neither proscriptive nor prescriptive socialization was present. Overall, moral socialization involves two parenting dimensions that promote distinct moral orientations, each engaging a different self-evaluative and emotional response from the child. Restrictive parenting fosters a proscriptive focus and the experience of anxiety; in contrast, nurturant parenting fosters a prescriptive focus and the experience of empathy by the child. Moreover, the prevalence of one parenting style over the other instills a predominant moral orientation and emotional response from the child. The two paths can provide an understanding of two emotions central to moral regulation: shame and guilt. The Emotional Consequences of Moral Failure: Shame and Guilt Experiences of shame and guilt, as well as anticipated experiences, are important regulators and evaluators of one’s moral conduct. Currently, the most prevalent perspective to distinguish between the two emotions has focused on their differing Self-Regulation and Moral Emotions 16 appraisals. Most notably, building on the earlier work of Lewis (1971), Tangney and colleagues (e.g., Tangney, 1991; Tangney and Dearing, 2002; Tangney et al., 2007; also see Tracy and Robins, 2006) have argued that the differences between guilt and shame signify those between “specific” and “global” attributions. According to this account, guilt results from a negative appraisal of one’s specific behavior (i.e., I did a bad act), while shame results from a negative evaluation of the global self (i.e., I am a bad person). Guilt and shame have also consistently been found to engage different action tendencies: Guilt motivates reparative actions, including apologizing, confessing, and treating others well, especially relationship partners (Baumeister, Stillwell, & Heatherton, 1994), while shame motivates withdrawal, denial and escape from the shame-inducing event (de Hooge, Zeelenberg, & Breugelmans, 2007; Hoffman, 1982; Lewis, 1971; Lindsay-Hartz, 1984; McGraw, 1987; Tangney, 1993; Tangney et al., 1996; Wicker et al., 1983; Tangney, 1991; Tangney, Miller, Flicker & Barlow, 1996; Tangney, Wagner, Fletcher, and Gramzow, 1992; Niedenthal, Tangney, and Gavanski, 1994; for reviews, see Tangney and Dearing, 2002; Tangney, Stuewig, & Mashek, 2007). For instance, Ferguson, Stegge, and Damhuis’s (1991) found that even among grade-school children, feeling guilt motivated approaching others to make amends while feeling shame motivated hiding from them. There is strong evidence from years of empirical studies linking guilt to reparative actions and shame to withdrawal and escape. An exception is the recent research by de Hooge and colleagues (e.g., de Hooge, Breugelmans, & Zeelenberg, 2008; de Hooge, Zeelenberg, & Breugelmans, 2010). These researchers found that following shame inductions, their “proself” participants were more likely to be prosocial than control participants in a coin-division task (de Hooge et al., 2008); in these Self-Regulation and Moral Emotions 17 studies participants were not given the option of withdrawing. Further, in other research these same authors found that guilt, but not shame, promoted cooperation in a social dilemma game (de Hooge, Zeelenberg, & Breugelmans, 2007). From an attributional perspective, guilt’s focus on a malleable, bad behavior permits change and correction for the transgression, but shame’s focus on the immutable, bad self allows for no such possibility (Tangney & Dearing, 2002; Tracy and Robins, 2006). Instead, a person experiencing shame is motivated to minimize the subjective distress through withdrawal and inhibition. From a self-regulatory perspective, shame and guilt are related to proscriptive and prescriptive regulation, respectively, and arise from different interpretations of one’s own moral failures (Sheikh and Janoff-Bulman, 2010). We experience guilt when we interpret our transgression using prescriptive regulation and focus on positive end-states and moral “shoulds”—we failed to be the moral person we ought to be. Shame, in contrast, is felt when we interpret our transgression using proscriptive regulation and focus on negative end-states and moral “should nots”—here we became the immoral person we ought not to be (see Figure 1). Moreover, a moral asymmetry between proscriptive and prescriptive regulation is reflected in the different experiences of shame and guilt. Just as proscriptive violations elicit greater distress and disapproval from others, so too shame is felt as more distressing than guilt. In our research we have found dispositional and situational evidence for a selfregulatory distinction between shame and guilt (Sheikh & Janoff-Bulman, 2010). Carver and White’s (1994) BIS and BAS Scales, which measure dispositional inhibition and activation, differentially predicted proneness to shame and guilt. Further, priming a Self-Regulation and Moral Emotions 18 proscriptive orientation increased feelings of shame relative to a prescriptive orientation while priming a prescriptive orientation increased feelings of guilt relative to a proscriptive orientation; in other words, focusing on end-states representing the bad behaviors one “should not” engage in, the bad person one “should not” embody, entails shame, not guilt. Focusing on those representing the good behaviors one “should” engage in, the good person one “should” embody, entails guilt, not shame. There are therefore two different types of moral failure, each representing two different regulatory systems and in turn producing two different emotional responses. Both shame and guilt have a negative valence, but are nevertheless associated with different motivational systems. Following a perceived moral transgression, a proscriptive or prescriptive regulatory system is activated, depending on whether one focuses on immorality, specifically the “should nots” directly reflected in our transgressions themselves, or on morality, and in particular the “shoulds” that represent moral alternatives to our transgressions. Consistent with their different foci, shame and guilt appear to be related to parental restrictiveness and nurturance, respectively, and intimately tied to anxiety and empathic concern. Shame The experience of shame has been described as the affect of “indignity, of defeat, of transgression and of alienation” (Tomkins, 1963, p. 118); one may feel physically visible, or imagine others, especially significant others, looking back at one’s self in the state of shame (Lindsay-Hartz, 1984). The response is an act of reducing facial and body movements, lowering one’s gaze and dropping one’s head (e.g., Fridja, 1986; Keltner & Self-Regulation and Moral Emotions 19 Buswell, 1996; Tomkins, 1963). The painful experience motivates attenuation through means of both physical and psychological withdrawal. From a self-regulation perspective, shame is regulated by the proscriptive system, motivating inhibitory behaviors. Thus, in shame we feel that we have failed to inhibit immoral behavior and are confronted by our own immorality; here we are the thief, the liar, the cheater that we ought not to be. Indeed, Tangney and colleagues (Tangney, 1993, 2007; Tangney & Dearing, 2002) argue that shame’s focus is turned inwards, on the flaws or defects of the self. One is focused on herself as a bad, immoral person and on the global, or “bad self” (Lewis, 1971; Tangney & Dearing, 2002). Because shame’s entire focus is on the failure to successfully avoid immorality, we are subsequently motivated to withdraw, to reduce social presence, to keep oneself from engaging in further immorality. Shame has no positive end-state to help redeem one’s moral status, and thus in the throes of shame it is difficult to imagine a better self or state (Lewis, 1971). Additionally, moral evaluation of proscriptive violations would be in terms of punishing, demanding referents—thus incurring harsher, more negative emotional consequences and rendering shame a much more painful experience than that of guilt (Lewis, 1971; Lindsay-Hartz, 1984; Morrison, 1996; Tangney et al. 1992). Reflecting a failure to curb immorality, shame is generally a result of proscriptive transgressions such as failures of inhibition and lack of self-control in the face of immoral “temptations.” Past research has found that some transgressions are more likely to engender shame than others. For instance, Fredrickson and Roberts (1997) have related “body shame” to eating disorders such as anorexia nervosa and bulimia (see also, e.g., Calogero, Davis, & Thompson, 2005; Sanftner, Barlow, Marschall, & Tangney, 1995; see Self-Regulation and Moral Emotions 20 also, Nussbaum, 2005). Alcohol, drug abuse, and excessive gambling have been found to be positively correlated with shame, while negatively or negligibly correlated with guilt (Dearing, Stuewig, & Tangney, 2005; Tavares, Martins, Zilberman, & el-Guabaly, 2002). These behaviors are “excesses” or “indulgences” that reflect proscriptive immorality— and thus are associated with feelings of shame. In fact, our research has also found that scenarios of proscriptive transgressions, which included scenarios of gambling and overeating, elicited significantly more shame than guilt (Sheikh & Janoff-Bulman, 2010). The evaluation related to shame is similar to messages transmitted by parental figures through punitive, restrictive measures: “You are an immoral person, capable of and inclined to engage in immoral behavior.” And, indeed, developmental researchers have found that adolescents of authoritarian and punitive parents, who use putdowns and are emotionally abusive, exhibit proneness to high levels of shame (Lutwak and Ferrari, 1997; Hoglund & Nicholas, 1995; for summary, see Tangney and Dearing, 2002), and parents’ own proneness to shame predicts their use of psychological restrictiveness (Freeman, Clara, Frank, Walling & Mak, 2007). Recognizing one’s own proscriptive failure is likely to engender shame, and because of its links to proscriptive morality, anxiety is likely to occur as well. Shame has been noted to be self-oriented (Leith and Baumester, 1998; Tangney, 1991), and it seems likely that shame’s co-occurrence with anxiety is what makes it focused on the self instead of others. Indeed, past psychologists have related shame to anxiety (Tangney, Wagner, Fletcher & Gramzow, 1992) as well as neuroticism (Darvill, Johnson, and Danko, 1992). Although anxiety often appears with other “maladaptive” phenomena, such as depression and low self-esteem, we believe that anxiety in particular is the Self-Regulation and Moral Emotions 21 developmental precursor to feelings of shame; anxiety is central to proscriptive (avoidance) regulation, and shame occurs as a result of proscriptive failure. One manifestation of anxiety that relates to shame is what researchers have labeled “personal distress,” a negative state resulting from a recognition of another’s suffering (Eisenbeg & Miller, 1987; Eisenberg, Fabes, Murphy, Karbon, et al., 1994; Eisenberg, Fabes, Schaller, Miller et al., 1991). Unlike empathy, personal distress is selforiented and motivates withdrawal from the distress-eliciting situation, attenuating helping behaviors towards the sufferer (Eisenberg & Fabes, 1990; Eisenberg et al., 1994; Batson & Shaw, 1991; Schroeder, Dovidio, Sibicky, Matthews, & Allen, 1988). Past researchers (Tangney and Dearing, 2002; Leith and Baumeister, 1998) have noted that shame and personal distress often occur together, and that during shame “the tremendous occupation with the self draws one’s focus away from a distressed other.” We suggest that shame and personal distress occur together because of their similar underlying regulatory orientation: both are focused on an undesired outcome and involve the motivation to mitigate one’s current, painful state. Guilt Guilt involves a focus on “shoulds” rather than “should nots,” on the moral alternatives to our transgressions rather than the immorality reflected in the transgressions themselves. In guilt, we feel that we have failed to be the reliable friend, the charitable person, the loyal partner that we ought to be. Guilt pushes us towards the possibility of redemption; it motivates us to regain our morality through reparative actions such as confession, apology, and correction. Self-Regulation and Moral Emotions 22 Because guilt is felt when we have failed to enact morality, prescriptive violations are most apt to engender guilt. Actions that damage close, interpersonal relationships (e.g., Baumeister, Stillwell & Heatherton, 1995) and failures to advance ourselves through hard work and industriousness (Sheikh & Janoff-Bulman, 2010) are most likely to produce feelings of guilt. In both cases, a desired goal is likely to be present to motivate “activation” and reinstate one’s morality. Transgressive omissions, failures to do what one should have done, are also apt to engender guilt because of the likelihood of focusing on the (omitted) moral behavior (Sheikh & Janoff-Bulman, 2010). Guilt is linked especially to communal relationships between people (Baumeister, Stillwell, & Heatherton, 1994; 1995; Tangney, 1992). If part of prescriptive regulation, guilt’s relational antecedents should include nurturing, warm parental figures, and in fact Zahn-Waxler and Kochanska (1990; Zahn-Waxler, Kochanska, Krupnick, & McKnew, 1990) conclude that parental warmth and affection are the precursors to the development of guilt. Such parental figures promote the recognition that one is a moral person, capable and inclined to care for others and willing to work hard to maintain loving, warm relationships with others. Further, empathy, associated with parental nurturance, is also a developmental precursor of guilt. Indeed, research from various areas of psychology links empathy to guilt (Leith and Baumeister, 1998; Tangney, 1991; Tangney and Dearing, 2002; Hoffman, 1982; Zahn-Waxler & Robinson, 1995). Guilt’s other-orientation is largely attributable to its relation to empathy, which, by definition, is other-oriented. Hoffman (1982) has long suggested that the developmental roots of guilt lie in empathy and, in particular, has pointed to a number of characteristics similar to both guilt and empathic concern: both entail concern for someone else’s suffering and a feeling of Self-Regulation and Moral Emotions 23 responsibility towards easing that suffering. A fundamental basis for these similarities between guilt and empathy, we believe, is their common regulatory nature: both motivate activation and approach of desired outcomes, regulating prescriptions in particular. More specifically, empathy underlies the experience of guilt, just as anxiety underlies the experience of shame. Some Final Thoughts on Shame and Guilt Although certain types of violations are more apt to elicit either shame or guilt, it is nevertheless the case that many transgressions could elicit either emotion. The experience of one emotion rather than the other will depend on whether the transgression engages proscriptive or prescriptive regulation. For example, the act of cheating can elicit a focus on the immorality of being a cheater and having an affair, thus engendering shame. Or it can elicit a focus on the failure to be a loyal partner, involving greater attention to one’s partner, thus engendering guilt. Both foci may also be present after a transgression, which may account for the positive correlations often found between the two emotions in research studies (see, e.g., Tangney et al., 1992). The two emotions often co-occur, likely reflecting the socialization of both moral systems and the engagement of both in everyday life (see endnote 1). It is the predominance of one moral system over the other that results in a proneness to one particular interpretation and subsequent emotional response. A self-regulatory perspective on shame and guilt fits with other conceptions of the differences between the two emotions, including the prevalent attributional perspective that posits a self versus behavior distinction between shame and guilt respectively (e.g., Lewis, 1971; Tangney et al., 2007). Guilt’s focus on rewarding, moral outcomes is apt to Self-Regulation and Moral Emotions 24 limit the negative evaluation associated with transgressing to the specific behavior, thus allowing for the possibility of alternative, positive behaviors. Shame’s focus on punishing, immoral outcomes, on the other hand, leads to harsher, more negative evaluations that then generalize to the global self. The more condemnatory proscriptive system would be more likely to focus on the entire self, whereas the less strict prescriptive system would be more apt to focus only on the behavior. Overall, our theory does not contradict the attributional perspective; rather, it provides an interpretive framework for why shame and guilt have different attributional responses. Conclusion Considerable work has demonstrated the relevance of self-regulation, in particular the approach-avoidance distinction, to emotion. Most notably, Carver (2001; Carver & Harmon-Jones, 2009; Carver and Scheier, 1998; 2008 Carver, Sutton, and Scheier, 2000) has argued for a bipolar dimensional structure of emotion (see also, e.g., Higgins, 1987, 1997) where each regulatory system is associated with its own positive-negative affective dimension. A bipolar dimensional view suggests that there are two different types of successes and failures, each of which incurs a different emotional response: succeeding at approaching a desired goal, succeeding at avoiding an undesired goal, failing to approach a desired goal, failing to avoid an undesired goal. An exploration of shame and guilt allows us to disentangle motivation and emotional valence: both shame and guilt have a negative valence, but guilt is an approach-based moral emotion, whereas shame is an avoidance-based moral emotion. Because shame and guilt both follow from moral failures, they have been traditionally thought of as similar and grouped under “negative self-focused” emotions Self-Regulation and Moral Emotions 25 (e.g., Haidt, 2003). Self-regulation and a bipolar dimensional perspective, however, offer insights to key differences between the two emotions in terms of development, in relation to other emotions, and in the type of moral focus. The socialization (restrictiveness versus nurturance) and affective (anxiety versus empathy) origins of shame and guilt differ as a function of the regulatory system involved. Ways of thinking about morality also relate to different emotions as a function of regulatory system: Is morality about inhibiting indulgent, “tempting” behaviors or about activating good, praiseworthy behaviors? Different conceptions of morality and immorality--of moral rules and standards--are associated with different emotional experiences. Other moral emotions, such as anger and disgust, may also reflect different sorts of moral regulation. Another’s prescriptive violation, such as perceiving that oneself or another has been treated unfairly or unjustly, may incur moral anger, motivating a focus on the prescriptive behavior that should have occurred (e.g., “there is a violation of what ought to be...” Carver & Harmon-Jones, 2009, p.184) and promoting an action-based response. Indeed, the self-regulation literature has implicated anger as an approach motivator (e.g., Carver and Harmon-Jones, 2009; Corr, 2002; Frijda, Kuipers & ter Shure, 1989; Harmon-Jones, 2003; Harmon-Jones, 2003; Harmon-Jones & Allen, 1998; Harmon-Jones & Sigelman, 2001). Disgust, on the other hand, may involve proscriptive regulation, focusing the individual on immorality and eliciting withdrawal and inhibitory behaviors to enlarge the space (both physical and psychological) between oneself and the disgust-inducing object. Although our dual system perspective proposes conceptual relations between moral regulation and emotions, developmental differences across individuals may also Self-Regulation and Moral Emotions 26 predict chronic sensitivities to one form of regulation over the other. The tendency to adhere to one moral system over the other, and thus the likelihood of experiencing one set of emotions over the other, may result from a general approach or avoidance sensitivity. The temperament or orientation of the child may also indeed pull for one style of parenting over the other (e.g., Baumrind, 1967), thus leading to socializing a focus on one form of moral regulation over the other. Moreover, individual orientation may change over the life-span through other relational influences. Thus, although this paper focuses on relational influences of parental figures, we expect other relationships, such as friendships and romantic partnerships, to have regulatory effects as well. A loving, nurturing partner, for example, is likely to promote prescriptive regulation through empathic concern, and feelings of guilt as a result of prescriptive failure. A restrictive, punitive partner, on the other hand, may increase proscriptive regulation, feelings of anxiety, and therefore shame as a result of proscriptive failure. Proscriptive and prescriptive forms of moral regulation are in part socialized through restrictive versus nurturing parenting styles, which respectively induce anxiety versus empathy. Moral failures associated with the two systems predict the presence of shame versus guilt. Anxiety and a focus on “should nots” underlie shame, which is likely a result of failures of inhibition and restraint of immoral “temptations.” Empathy and a focus on “shoulds” underlie guilt, which is likely a result of failures to activate positive, moral outcomes. It appears that a seemingly simple distinction between two motivations may nevertheless illuminate important links between patterns of parental socialization, morality, and emotions. Self-Regulation and Moral Emotions 27 Endnotes 1. Self-regulation researchers (e.g., Carver, 2001) have acknowledged that, outside the laboratory, events are complex and the two motivational systems may be activated in confluence. Oftentimes, motivations to avoid a negative outcome and inhibit behavior occur in tandem with the motivation to approach a positive outcome and activate behavior. However, the two motivations nevertheless function as distinct regulatory systems. 2. It is interesting to note the relation between past notions of the “objects” (i.e., mental representations, Russell, 2003; Oatley and Johnson-Laird, 1987) of emotions and their end-states. For instance, shame has been described as a failure to live up to an ideal standard or those of an idealized “other” (e.g., Higgins, 1987; Lewis, 1971; Piers & Singer, 1953). Interestingly, Lindsay-Hartz (1984, p.696) suggests that an ideal other may reflect a negative end-state, such as a bad self: “Viewing ourselves though the eyes of another is like looking at ourselves in a mirror. The other person’s viewpoint serves as a means for revealing to us a very negative view of ourselves...” (p. 696). As such, the phenomenal experience of shame may indeed involve a representation of an idealized “other” or ideal image that serves as a mirror to highlight the negative aspects of the self. Here, representations of the idealized “other,” the “ego ideal,” or an ideal standard more generally do not function as positive goals to approach, but rather as a mirror to emphasize the immoral aspects of oneself that require inhibition. 3. Some researchers have thought of guilt as “self-focused” (e.g., Iyer, Leach, & Crosby, 2003); however, these studies often control for empathy when investigating the effects of Self-Regulation and Moral Emotions 28 guilt. We argue that guilt is “other-focused” because of its relation to empathy, and removing guilt’s connection to empathy would also remove its “otherness.” Self-Regulation and Moral Emotions 29