Focusing Forgiveness

1 23

��

��

��

1 23

Your article is protected by copyright and allrights are held exclusively by Springer Science+Business Media Dordrecht. This e-offprintis for personal use only and shall not be self-archived in electronic repositories. If you wishto self-archive your article, please use theaccepted manuscript version for posting onyour own website. You may further depositthe accepted manuscript version in anyrepository, provided it is only made publiclyavailable 12 months after official publicationor later and provided acknowledgement isgiven to the original source of publicationand a link is inserted to the published articleon Springer's website. The link must beaccompanied by the following text: "The finalpublication is available at link.springer.com”.

Focusing Forgiveness

Andras Szigeti

! Springer Science+Business Media Dordrecht 2014

1 Introduction

It is clear that forgiveness is closely related to emotions. Bishop Butler’s‘‘forswearing of resentment’’ is still the definition most philosophical works onthe subject take as their point of departure. Some others disagree but usually onlyinsofar as they focus on another reactive emotion – e.g., moral hatred, disappoint-ment, anger – which we overcome when we forgive.1

There is something striking about the negativity of this approach. Forgiveness ischaracterized in the first instance as the renunciation of an emotion, most commonlythat of resentment. It is sometimes expressly stated that forgiveness is itself areactive attitude,2 but even then what we are first told about the emotional aspects ofthis reactive attitude is something negative, namely that it involves the foregoing ofanother reactive attitude, again, typically that of resentment. Roberts too, thoughkeenly interested in the emotional aspects of forgiveness, nevertheless characterizesthe affectivity of forgiveness negatively as the withdrawal of the ‘‘alienatingretributive construal’’ associated with anger.3

A. Szigeti (&)Department of Philosophy, UiT The Arctic University of Norway, Tromsø, Norwaye-mail: [email protected]

1 More specifically, according to Roberts the emotion we overcome in forgiveness is anger, see Robert C.Roberts, ‘‘Forgivingness,’’ American Philosophical Quarterly 32 (1995): 289–306. According toRichards, forgiveness involves overcoming a whole range of negative emotions including sadness,disappointment, frustration as well as resentment; see Norvin Richards, ‘‘Forgiveness,’’ Ethics 99 (1988):77–97. According to Sher, forgiveness involves overcoming blame; see George Sher, In Praise of Blame(Oxford: Oxford University Press, 2006).2 Lucy Allais, ‘‘Dissolving Reactive Attitudes: Forgiving and Understanding,’’ South African Journal ofPhilosophy 27 (2008): 1–23. Espen Gamlund, ‘‘The Duty to Forgive Repentant Wrongdoers,’’International Journal of Philosophical Studies 18 (2010a): 651–671.3 P.F. Strawon, ‘‘Freedom and Resentment,’’ in Freedom and Resentment (London: Methuen, 1974),1–24. See also Gamlund, op. cit., Roberts, op. cit.

123

J Value InquiryDOI 10.1007/s10790-014-9422-4

Author's personal copy

We are of course told quite a lot about what we do because we have forswornresentment (or the other emotions mentioned). It is often said that forgiveness enablesus to reaffirm our personal relationship with the offender,4 and that forgiving is thusnormally accompanied by an offer of reconciliation.5 In addition, we know thatforgiveness is communicated through typical forms of expression: canonicalutterances, bodily gestures and so on. Finally, many emphasize the typical effectsof forgiveness such as the liberating psychological impact of forgiveness on both theperson who forgives and the one who is forgiven,6 the experience of cleansing orcuring, the actual renewal of personal relationships,7 and the undoing of the past.8

But the latter descriptions focus mainly on behavioral manifestations offorgiveness and its positive (and sometimes negative) psychological and behavioralimpact. They say less about forgiveness itself as a reactive attitude. It may well betrue, for example, that resentment and forgiveness are not compatible. But can wesay more about the positive features of forgiveness as a reactive attitude? In thispaper, my main aim is to make a contribution to such a non-privativecharacterization of forgiveness as a reactive attitude.

More specifically, I am interested in two questions. First, I want to ask whetherwe can say more about the connection of forgiveness to emotions than that it ismarked by the absence of one kind of emotion, e.g., resentment or anger? I willargue that we should take seriously the thought that forgiveness is a sui generisemotion type. Forgiveness may be classified as an emotion because it displays asufficient number of the generic features by which we distinguish emotions fromother mental attitudes and episodes. But forgiveness is also clearly distinguishablefrom other emotions in terms of its core evaluative concern, phenomenology andempirical manifestations.

Second, and relatedly, I want to argue that the benefits of focusing forgivenessas an emotion are not just taxonomical but also conceptual. Thus I will suggestthat thinking about forgiveness as an emotion may help us gain traction onwhat we may call the ‘‘hard problem’’ of forgiveness, namely how forgive-ness can be a justified response despite the fact that forgiveness entails andpositively asserts the agent’s blameworthiness9 for having performed a wrongful

4 Jean Hampton, Forgiveness and Mercy (with Jeffrie G. Murphy) (Cambridge: Cambridge UniversityPress, 1988), p. 80.5 Ibid., 42.6 See Roberts, op. cit.7 See Roberts, op. cit. See also Thomas Scanlon, Moral Dimensions: Permissibility, Meaning, Blame(Cambridge, MA: Harvard University Press, 2008), p. 160.8 See Hampton, op cit., p. 86.9 I will use the term ‘‘blameworthy’’ in the following sense: an agent is blameworthy for performing amorally wrong action, i.e., an action for which the agent has no (objective) justification and noresponsibility-undermining excuse. In short, an agent is blameworthy, if she performed an all-things-considered unjustifiable action and is morally responsible for doing so. The term ‘‘blameworthy’’ shouldbe understood here non-affectively and is probably replaceable by the closely related term ‘‘culpable.’’ Onthis understanding, ‘‘being blameworthy’’ is just having an entry in one’s ledger of moral responsibility.See esp. Joel Feinberg, Doing and Deserving (Princeton: Princeton University Press, 1970) on the ledgermetaphor. I believe this usage is intended by Michael Zimmerman when he says ‘‘A person can bepraiseworthy or blameworthy without anyone’s being aware of this, without anyone’s taking note of it,

A. Szigeti

123


action.10 It is not the aim of this paper to work out a solution to this problem.My hope is, however, that by focusing forgiveness the paper can contribute toevaluating existing solutions and clear the ground for developing alternativeproposals as well.

2 Is Forgiveness an Emotion?

2.1 Forgiveness as a Reactive Attitude

A number of authors refer to forgiveness as a reactive attitude. Ever since PeterStrawson’s seminal essay Freedom and Resentment (1974) the concept of reactiveattitude has played a central role in discussions of certain moral practices, especiallythose involving attributions of moral responsibility. According to Strawson: ‘‘Onlyby attending to this range of attitudes can we recover from the facts as we knowthem a sense of what we mean, i.e. of all we mean, when, speaking the language ofmorals, we speak of desert, responsibility, guilt, condemnation, and justice.’’11

One important issue is to what extent reactive attitudes are affective phenomena.Are reactive attitudes simply reactive emotions? Strawson of course was keen toemphasize that the emotional component of reactive attitudes is essential. Based onthis, some attribute to Strawson the view that reactive attitudes are ultimately notrationally justifiable because emotions are not ultimately rationally justifiable.12 Asa matter of exegesis, I do not think this last point is correct.13 Thus Strawson speaksof the practice of responsibility as ‘‘that whole web or structure of human personaland moral attitudes, feelings, and judgments.’’14

Footnote 9 continuedwithout anyone’s actually praising or blaming him’’ Michael J. Zimmerman, An Essay on MoralResponsibility (Totowa: Rowman and Littlefield, 1988), p. 39.10 Alluding to Chalmers (David Chalmers, ‘‘Facing Up the Problem of Consciousness,’’ Journal ofConsciousness Studies 2 (1995): 200–219), I call this the ‘‘hard problem of forgiveness’’ to distinguish itfrom other worries about forgiveness such as, among others, the question whether forgiveness is ever astrict duty (and not just permissible), or the question whether forgiveness is only justifiable if the offenderrepents and apologizes. The distinction is not meant to imply that the latter issues are easily resolvable.But they do not seem to concern the very essence of forgiveness in the same way as the hard problemdoes. Kolnai and Hieronymi are especially clear on this matter. See Aurel Kolnai, ‘‘Forgiveness,’’Proceedings of the Aristotelian Society 74 (1974): 91–106 and Pamela Hieronymi, ‘‘Articulating anUncompromising Forgiveness,’’ Philosophy and Phenomenological Research 62 (2002): 529–555.11 See Strawson, op. cit., p. 23.12 See Jonathan Bennett, ‘‘Accountability,’’ in Zak van Straaten (ed.), Philosophical Subjects: EssaysPresented to P.F. Strawson (Oxford: Oxford University Press, 1980), 14–47, p. 24 and p. 29. GalenStrawson describes his father’s view as a ‘‘non-rational commitment theory of freedom,’’ see GalenStrawson, Freedom and Belief (Oxford: Clarendon Press, 1986), p. 84.13 Andras Szigeti, ‘‘Revisiting Strawsonian Arguments from Inescapability,’’ Philosophica 85 (2012):91–121.14 P.F. Strawson, Skepticism and Naturalism. Some Varieties (London: Methuen, 1985), p. 39. My italics.


123


In any case, I will side with those15 in this paper who think that reactive attitudes,despite typically being affective phenomena,16 are reasons-responsive and judg-ment-sensitive attitudes susceptible to and requiring rational endorsement. It isprobably worth noting that this does not commit me to cognitivism about reactiveattitudes. Arguably, reactive attitudes can be evaluative without necessarilyinvolving evaluative judgments.17

Forgiveness is closely correlated with moral responsibility. In fact, there is nearconsensus in the literature that forgiveness presupposes moral responsibility becauseforgiveness presupposes blameworthiness and blameworthiness presupposes moralresponsibility (more on this Section 3.2 below). Therefore it is not surprising thatforgiveness is frequently referred to as a reactive attitude. I will accept thischaracterization here.

What I want to focus on in the rest of this section is the affect component of thereactive attitude of forgiveness. I think there is a plausible case to be made thatforgiveness has distinctive affective features. Furthermore, these features can becharacterized in non-privative terms. In fact, I think we should take the thoughtseriously that forgiveness is a sui generis emotion.

2.2 Forgiveness as a Reactive Emotion

How can we tell whether x is an emotion? Following Aaron Ben-Ze’ev, I willanswer this question by looking at the typical characteristics (Section 2.2.1),components (Section 2.2.2) as well as causes and concerns (Section 2.2.3) of x.When considering these items we might aim for a yes/no answer to the questionwhether x is an emotion. Alternatively, our aim may be to consider how x compareswith paradigm emotions.18 I will not distinguish sharply between these twoapproaches here leaving it open to some extent whether forgiveness is aparadigmatic emotion or just relevantly similar to what we take to be theparadigmatic emotions such as guilt or anger.

15 Jay R. Wallace, Responsibility and the Moral Sentiments (Cambridge, MA: Harvard University Press,1994). See also Hieronymi, op. cit., Scanlon, op. cit., etc.16 I think there can be non-emotional reactive attitudes too, i.e., reactive attitudes which do not involvereactive emotions. But nothing crucial hinges in the following on this.17 This means that emotions and a fortiori affective reactive attitudes need not involve a propositionalcognitive component, and perhaps not even cognitive construals a la Roberts. Here I go along with theview that even without such a component emotions can be taken to be representations of evaluativeconcerns. For the idea that emotions are representations of non-propositional evaluative concerns, seeChristine Tappolet, Emotions et Valuers (Paris: Presses Universitaires de France, 2000); Justin D’Armsand Daniel Jacobson, ‘‘The Significance of Recalcitrant Emotion (or, Anti-Quasijudgmentalism),’’ in A.Hatzimoysis (ed.), Philosophy and the Emotions (Cambridge: Cambridge University Press, 2003),127–146; Sabine A. Doring, ‘‘Why Be Emotional?,’’ in Peter Goldie (ed.), The Oxford Handbook ofPhilosophy of Emotion (Oxford: Oxford University Press, 2010), 283–301. If this idea is on the righttrack, then the role of emotions is often similar to non-inferential, non-conceptual perceptual inputs, andpossibly also similar to non-inferential intuitions, see also Tim Crane, ‘‘The Nonconceptual Content ofExperience,’’ in Tim Crane (ed.), The Contents of Experience: Essays on Perception (Cambridge:Cambridge University Press, 1992), 136–157.18 See Aaron Ben-Ze’ev, ‘‘The Thing Called Emotion,’’ in Peter Goldie (ed.), The Oxford Handbook ofPhilosophy of Emotion (Oxford: Oxford University Press, 2009), 41–62.

A. Szigeti

123


2.2.1 Typical Characteristics of Emotions

Let us start by considering the main characteristics of emotions. Ben-Ze’ev suggeststhe following list: (i) instability, (ii) great intensity, (iii) partiality, and (iv) brevity. Iwill argue that we can fairly confidently attribute these characteristics to forgivenessas a reactive emotion. Or a bit more cautiously, in some cases it may be disputablewhether forgiveness can be said to have a certain feature. However, even in thesecases the attribution is hardly more controversial than for other emotion kinds.19

Consider (iii) partiality first as it is, I believe, the easiest task. Ben-Ze’ev saysthat ‘‘emotions are partial in two basic senses: they are focused on a narrow target,such as one person or very few people, and they express a personal and interestedperspective… Emotions address practical concerns from a personal perspective.’’(45) It is hard to deny that forgiveness is partial in both of these senses – arguably,more so than some other paradigmatic emotions such as admiration or indignation,for example. The addressee of forgiveness is almost always a single individual, andone who is personally acquainted with the forgiver.20 Furthermore, many even arguethat vicarious forgiveness is not possible, i.e., I cannot forgive an offender on behalfof the victim.21 But even if this is wrong, it is clear that forgiveness is essentiallyinterpersonal. This is also reflected in the grammatical form of the verb ‘‘forgive’’which takes a three-place predicate: ‘‘someone forgives someone for (having done)something.’’22

Take (i) instability next. According to Ben-Ze’ev this means that ‘‘emotionsindicate a transition in which the preceding context has changed but no new contexthas yet stabilized’’ (45). Forgiveness fits the bill well. Most authors writing on thesubject of forgiveness agree that forgiveness involves some sort of re-contextual-ization of the to-be-forgiven wrongdoing and its perpetrator. What this re-contextualization precisely involves and when it is justified is of course disputed, aswe will see shortly. But there seems to be overwhelming agreement that when aperson forgives it is because her view of the action or her view of the agentundergoes a significant change (or both).

Now, it could be objected here that the change of view involved has to berelatively stable for the reactive attitude to qualify as forgiveness. How then doesthis plausible claim sit with Ben-Ze’ev’s view that because of their instabilityemotions are ‘‘intense, occasional, and limited in duration’’ (45)?

19 I will set aside the issue whether those above are indeed all essential characteristics, and whetherperhaps other characteristics should be included in the list or not.20 Richards (op. cit.) and Roberts (op. cit.) discuss cases of forgiving strangers such as that of a carelessdriver who splashes dirt on your clothes. Granted, in such cases there is no acquaintance prior to theoffense. But it is plausible to argue that the offense itself creates the requisite minimal degree of personalcontact between offender and victim. See also the discussion on the personal concern of emotions inSection 2.2.3.21 Murphy op. cit., p. 21: ‘‘[…] I do not have standing to resent or forgive you unless I have myself beenthe victim of your wrongdoing.’’22 Oliver Hallich, ‘‘Can the Paradox of Forgiveness Be Dissolved?’’ Ethical Theory and Moral Practice16 (2013): 999–1017, p. 1004.


123


In reply, we should note that it is somewhat unclear what is meant by Ben-Ze’ev’s qualification. Imagine a perpetrator of horrible war crimes who comes tosee the wrongness of his past deeds and feels guilty about them. As a result of hischange of heart, he publicly apologizes for his crimes and sincerely seeks to makereparation. What does it mean then to say that his guilt is or was ‘‘intense,occasional, and limited in duration’’? Perhaps what is meant is that the first pangs ofconscience were particularly tormenting. But was the initial feeling of guilt sincereif it disappears altogether? Alternatively, the claim could be taken to mean that thewar criminal’s guilt, while causing lasting changes in his behavior, ‘‘comes (back)in waves’’ as it were: ‘‘it is whenever he remembers the scenes from the war, or theprisoner camps, etc.’’ that he is struck again and again by episodes of intense guiltfeelings (while the intensity may or may not decrease somewhat as time passes).

But if that is what is meant, then exactly the same could be true of forgiveness asa reactive emotion. Several authors writing on forgiveness describe how deeplymoving it can be to undergo at first the experience of ‘‘refocusing the wrongdoer’’23

as we do when we forgive. Further, the emotional experience of such a ‘‘gestaltswitch’’ (Roberts uses this expression) can come (back) in waves as well. Imagine avictim of the war criminal just mentioned who finds it in her to forgive him. As theinclination to forgive the offender solidifies, it not only causes lasting changes in thevictim’s behavior towards the offender, it also becomes easier and easier to refocusthe crime and its perpetrator from the perspective of forgiveness. Still, as nearly allthose who write on forgiveness emphasize, this process does not amount toforgetting. So when the victim remembers, say, the scenes from the war or theprisoner camps, etc., and sets these memories against the present image of therepentant offender she can re-experience the moving feeling of forgiveness againand again (while the intensity may or may not decrease somewhat as time passes).

The same considerations apply to the second and fourth characteristics: (ii) greatintensity, and (iv) brevity. Once we disambiguate the meaning of these terms forparadigmatic emotions such as guilt, there seems to be no reason why we could notattribute these features to forgiveness as well. As for (i) instability previously, themost promising way to make sense of these features is to say that (ii) and (iv) applymainly to the felt, phenomenological experience at the onset of the emotion, andpossibly to its re-occurrence if the emotion comes (back) in waves (in somewhatdecreasing measure over the course of time).

In any case, the crucial point is that what decides over the long haul is not theintensity of the first emotional experience. Whether we can ascribe paradigmaticemotions to the subject such as guilt or resentment will be determined by whetherthese emotions yield lasting behavioral changes and alter the subject’s normativeviews. The same points, I submit, apply to forgiveness.

2.2.2 Typical Components of Emotions

Ben-Ze’ev identifies four essential components of emotions: (i) cognition, (ii)evaluation, (iii) motivation, and (iv) feeling. As regards (i) cognition, there can be

23 See Roberts, op. cit., p. 297.

A. Szigeti

123


little doubt that the attitude of forgiveness is representational of and responsive tofacts (primarily facts about the agent and the action to be forgiven). The forgiver ifasked to explain her attitude of forgiveness will cite such facts. It seems equallyclear that this attitude is truth-apt,24 answerable to norms of correctness and seen assuch both by the forgiver and the addressee of the attitude as well as third parties.

Similar considerations apply to (ii) the evaluative component. Needless to say, itis subject to considerable dispute what evaluations the attitude of forgivenesstypically implies about the offender and about his actions. It is an even moredifficult question when the evaluations implied by the attitude of forgiveness arewarranted—this is essentially the ‘‘hard problem’’ forgiveness (see Section 3.2.below). But there is little controversy about the fact that the attitude of forgivenessreflects the forgiver’s evaluative concerns. The normativity of forgiveness thusunderstood distinguishes it from the psychological process of forgetting.

The literature is particularly rich in descriptions of what forgiveness makes us do.I have already mentioned this (iii) motivational aspect of the attitude of forgiveness.While there is some debate about what forms of actions and interpersonal gesturespresuppose forgiveness,25 and even some disagreement about what forms of actionsare typically associated with forgiveness,26 it seems obvious that forgiveness willresult in some changes in the desires and behavioral dispositions of the forgiver. It isatypical but conceivable that A forgives B but A does not change her behaviortowards B in any way. This could be the case if an expression of forgiveness isproscribed for more pressing reasons (e.g., it would send the wrong message to otherculprits). More trivially, this could also be the case if B is simply not around. But itis not conceivable that A forgives B but A’s motivational desires do not changewhatsoever with regard to actions potentially affecting B.

At this point, we reach (iv) the feeling component. This component is notoriouslydifficult to describe, although metaphors used to capture how it feels like to forgiveare revealing. As already noted in the introduction, one often invoked image tocapture the qualitative phenomenology of forgiveness is that of cleansing or curingor healing. Interestingly, Hieronymi27 adds that the reactive attitude of forgivenessalso involves a feeling of intimacy. There is also frequent talk of benevolence and ofan atmosphere of harmony.28

The hedonic feeling component of forgiveness may be particularly important,although it is seldom discussed in detail in the philosophical literature. Is forgiveness

24 For an argument that emotions can be characterized as truth-apt cognitive states even in the absence ofpropositional content, see Mikko Salmela, ‘‘True Emotions,’’ Philosophical Quarterly 56 (2006):382–405.25 Perhaps genuine reconciliation between victim and offender is impossible without forgiveness by theformer. But that is not the received view, not at least in psychological forgiveness research, see, forexample, Michael E. McCullough and Charlotte van Oyen Witvliet, ‘‘The Psychology of Forgiveness,’’ inC.R. Snyder and S.J. Lopez (eds.), Handbook of Positive Psychology (New York: Oxford UniversityPress, 2002), 446–458, p. 447.26 Thus some (e.g., Richards, op. cit., p. 79) doubt that Kolnai is right to say that gestures of ‘‘(re-)acceptance’’ typically follow forgiveness.27 See Hieronymi, op. cit., p. 551.28 See Roberts, op. cit.


123


pleasurable or pleasant? Is the adoption of the attitude of forgiveness subjectivelyexperienced by the forgiver as feeling good? I think there is a strong case to be madethat unless the forgiver feels good about forgiving in this hedonic sense the attitude offorgiveness is not sincere. Or at least, without this positive feel the attitude cannot besaid to be final and settled. It is forgiveness malgre soi. As such it may well point tolingering doubt, resistance, residual anger or insuppressible resentment.29

2.2.3 Typical Causes and Concerns of Emotions

‘‘Emotions typically occur when we perceive positive or negative significant changesin our personal situation, or in the situation of those related to us. A major positive ornegative change significantly improves or interrupts a stable situation relevant to ourconcerns.’’ It is important that the relevant change eliciting the emotional responsecan also be psychological – a ‘‘change in our psychological environment’’30 –occurring when one remembers, imagines or re-evaluates past events.

As already noted, there is an overwhelming consensus that forgiveness is broughton by a perceived change. The most typical and significant change is the repentanceand apology of the offender. However, many authors admit other kinds of changesas relevant as well. Roberts, for example, argues31 that the perception of theoffender’s suffering, or the recognition of the offender’s moral commonality withthe forgiver, or the acknowledging of a special relationship to the offender can alljustify forgiveness. In any case, most standard accounts of forgiveness agree thatwhat prompts and warrants forgiveness is a change in how the offender is related tothe offense and a concomitant change in the significance of the offense for thevictim. For example, repentance and apology may be portrayed as severing somecrucial link previously obtaining between the offender and his wrongdoing.32 Oncethat link is cut the significance of the past offense changes for the victim whoacknowledges this by forgiving the offender for his wrongdoing.

It is also important for our present purposes that the change prompting emotionalresponses is typically of personal significance. Thus according to Ben-Ze’ev (44):‘‘Our attention may be directed to any type of change, but in order for the change togenerate emotions it must be perceived as having significant implications for us orfor those related to us.’’ In short, the typical concern of emotions is personal.33

29 I will come back to this point in Section 3.3.30 See Ben-Ze’ev, op. cit., p. 43.31 See Roberts, op. cit., p. 293.32 An alternative approach is to emphasize the role of repentance and apology in changing thesignificance or standing of the offender (rather than that of the offense) in the victim’s eyes (see, forexample, Roberts, op. cit. and Richards op. cit., pp. 87–92 on the second approach). On this secondapproach, repentance and apology change primarily how the victim relates to the offender rather than howthe victim relates to the offense. Hieronymi, who favors the first approach, is clear about this distinction(op. cit., p. 547).33 It may also be true that emotions, or at least reactive emotions, can also create interpersonal relations.Bennett makes this point about reactive attitudes in general. The reason, he says, why reactive attitudeshave such a relationship-generating potential is that reactive attitudes are ‘‘forms of address’’ directed atother members of the moral community, see Bennett, op. cit., pp. 29–30.

A. Szigeti

123


I have already elaborated on the personal nature of forgiveness when discussingtypical characteristics of emotions. A few clarifications are in order, however.Although there is some controversy about this, it is plausible that the personalconcern of forgiveness does not entail that there must be a live two-way relationshipbetween offender and forgiver. For example, the offender may quite simply be dead.Thus we can agree with Kolnai34 that there is ‘‘meaning, in say, forgiving themisdeeds of Napoleon I’s but not those of Napoleon III, Frederick the Great’s butnot Hitler’s, and so forth.’’35

So the point of emphasizing that forgiveness tracks a personal concern is tohighlight that the intentional object of forgiveness is what the action to be forgivenmeans for the forgiver. It is in this sense that vicarious forgiveness is probablyimpossible. I cannot forgive an offender on behalf of someone else unless I regardmyself too as a victim because I understand the meaning of the wrongdoing toinclude me in some sense (e.g., because the victim is a family relation or because Icould easily have been in the victim’s place, and so on). But if I am not implicatedin any way by the wrongdoing I do not have standing to forgive (while of course it isoften debatable whether I rightly see myself as sufficiently implicated in thewrongdoing to be entitled to forgive it).

The focus on the personal and partial concern of forgiveness also leads to deeperand philosophically more troublesome questions.36 It is plausible and almostuniversally acknowledged that forgiveness is a reasons-responsive attitude (seeSection 2.1. above). That is, the personal and partial focus of forgiveness should notbe taken to entail that forgiveness would not be subject to rational norms. As we willsee, the reason-sensitivity of the attitude of forgiveness leads to what I call the hardproblem of forgiveness.37 But if forgiveness is personal and partial in the way justdiscussed what sort of reasons are the reasons to forgive?

First, how strong reasons are they? Does the personal and partial focus offorgiveness mean that forgiveness can never be a strict duty, as many seem tobelieve?38 Or can one be blameworthy for failing to forgive? Second, and relatedly,what is the scope of these reasons? The personal and partial focus of forgivenessdoes appear to entail that the reasons to forgive are in some sense necessarily agent-relative. That is to say, the reasons to forgive cannot be fully stated without indexingthose reasons to the forgiver. The reasons to forgive are the forgiver’s reasons. Butwhat exactly are the implications of the agent-relativity of the reasons to forgive?Does the agent-relativity of reasons to forgive mean that some consideration C gives

34 See Kolnai, op. cit., p. 104.35 It is a more complicated question which I cannot discuss here whether it follows from the personalnature of forgiveness that the addressee of the reactive attitude can only be a single person (or a fewclearly identified persons such as the three thugs who robbed me). Can we forgive a corporation or apolitical party for causing some harm? Can we forgive a large random group of people on the beach forfailing to prevent the drowning?36 The same questions arise for other emotions that resemble forgiveness in having a partial and personalconcern in the sense discussed above.37 See esp. Kolnai, op. cit.; Hieronymi, op. cit.; 2002; Hallich, op. cit.38 See Gamlund, op. cit.; and also Espen Gamlund, ‘‘Supererogatory Forgiveness,’’ Inquiry 53 (2010b):540–564.


123


agent X a reason to forgive (offender O for doing A) but the same considerationC need not constitute a reason for agent Y to forgive (offender O for doing A)?

Fortunately, these questions do not have to be answered here to draw my interimconclusion. The upshot of this section is that there is a good case to be made thatforgiveness is an emotion as forgiveness appears to have the typical characteristicsand components of paradigmatic emotions. Furthermore, forgiveness resemblesparadigmatic emotions also in terms of its typical cause (which is perceived change)and concern (which is personal and partial). So why not say that forgiveness is a suigeneris emotion? ‘‘If it looks like a duck, swims like a duck, and quacks like a duck,then most probably it is a duck.’’

3 What Sort of Emotion is Forgiveness?

It is one thing to argue that forgiveness is plausibly characterized as an emotion, it isanother thing to ask what sort of emotion it may be. How can we distinguishforgiveness as an emotion from other cognate emotions such as pity, compassion,love, sympathy, and charity?

Emotions are usually distinguished from one another in three ways. It is assumedthat the three criteria will typically yield a convergent classification. The first way tocategorize emotions focuses on empirical features. Emotions are differentiatedthrough specific descriptions of their eliciting conditions, processes of socialization,corresponding action tendencies, associated social norms, evolutionary histories,neural architectures, concomitant bodily changes, and so on.

The second way to categorize emotions is by establishing a ‘‘core relationaltheme’’ for each emotion type.39 These characterizations of a given emotion typeseek to capture the distinct evaluation in terms of which that emotion presents theworld to its subject. For example, the core relational theme of sadness is‘‘irrevocable loss’’.40 Here are a few more examples:

• ‘‘Jealousy monitors the social environment for potential losses of affection orallegiance.’’41

39 There is a difficulty here about how to arrive at the core relational theme of an emotion type. On someapproaches a sufficiently fine-grained application of the first method will produce the core relationaltheme so there is no real difference between the first and second ways of distinguishing emotions. This ishow I read the kind of metaethical sentimentalism advocated by Jesse Prinz, see Jesse Prinz, ‘‘TheEmotional Basis of Moral Judgments,’’ Philosophical Explorations 9 (2006): 29–43; Jesse Prinz, TheEmotional Construction of Morals (Oxford: Oxford University Press, 2007); Jesse Prinz and ShaunNichols, ‘‘Moral Emotions,’’ in J.M. Doris & The Moral (eds.), Psychology Research Group The MoralPsychology Handbook (Oxford: Oxford University Press, 2010), 111–146. By contrast, rationalsentimentalists, who accord a decisive role to the rational endorsability of emotions, need to rely on bothmethods whereby the second will be a necessary corrective of the first (see works by D’Arms & Jacobsoncited below).40 Prinz and Nichols, op. cit., p. 119.41 D’Arms and Jacobson, op. cit., p. 139.

A. Szigeti

123


• ‘‘One’s envy portrays a rival as having a desirable possession that one lacks, andit casts this circumstance in a specific negative light.’’42

• ‘‘[F]ear […] speak[s] to whether or not something is a threat.’’43

• ‘‘Shame presents something as a social disability.’’44

The third way is to distinguish emotions by a range of phenomenological featuressuch as qualitative feel, intensity and valence. For example, it may be argued thatenvy/admiration, thrill/fear, lust/disgust are most saliently distinguished by theirvalence. Again, these three ways of distinguishing emotions are thought to yieldconsistent results at least in the long run.45

The view I defend here is that forgiveness can be clearly distinguished from otherparadigmatic emotions in terms of these three classificatory criteria (or moreprecisely, these three sets of classificatory criteria). I also hope to show in thefollowing that there is more to be gained from classifying forgiveness as a suigeneris emotion than just taxonomical clarity.

3.1 Empirical Classifications

One of the advantages of focusing forgiveness as an emotion is that this conceptualapproach yields a fruitful empirical project. That is to say, once we no longer regardforgiveness as merely the absence of an emotion, we can begin to study theempirical characteristics of the affective response and its relationship to relatedphenomena. In fact, although often ignored by philosophers on the subject, thisempirical project is well under way in experimental social psychology, evolutionaryanthropology, sociology and elsewhere.46

I have already talked in abstract terms about the eliciting conditions offorgiveness (perception of subjectively significant changes) and correspondingaction tendencies (various gestures of reconciliation and acceptance). Appliedethics, cultural anthropology and sociology have put more flesh on the bones ofthese abstract statements.47

I will rest content here with just listing other relevant empirical issues for the studyof forgiveness distributed across the other disciplines mentioned above. Importantquestions include the following: Are there characteristic somatic changes associatedwith the forgiveness response? Can we provide a plausible evolutionary account forthe emergence of forgiveness as a response distinct from seemingly innate human

42 Justin D’Arms and Daniel Jacobson, ‘‘Sentiment and Value,’’ Ethics 110 (2000): 722–748, p. 66.43 D’Arms and Jacobson, op. cit., p. 108.44 Justin D’Arms and Daniel Jacobson, ‘‘Anthropocentric Constrains on Human Value,’’ in Russ Shafer-Landau (ed.), Oxford Studies in Metaethics, vol. I (Oxford: Clarendon Press, 2006), 99–126, p. 109.45 Although, as usual, the devil is in the details. Can two emotion types differ only in their valence? Thatdepends on what exactly we mean by emotional valence (I will come back to this in Section 3.3).46 It is even said that we are ‘‘entering a golden era of [empirical] forgiveness research,’’ McCulloughand Witvliet, op. cit.47 The list of potential references is too numerous here. In cultural anthropology, see works by JoelRobbins (e.g., Becoming Sinners: Christianity and Moral Torment in a Papua New Guinea Society(Berkeley: University of California Press, 2004)). For issues in applied ethics, McCullough and Witvliet,op. cit., is a good place to start.


123


responses such as revenge and avoidance? What cognitive systems implementforgiveness responses? What is the underlying neurophysiological architecture offorgiveness? How is forgiveness socialized and learnt? What sort of social structuresand institutions promote forgiveness?

Since I am mainly interested in the philosophical uses of my thesis, let me leavethese questions behind for the time being. In what remains, I will return toinvestigating the potential philosophical implications of the proposal made in thispaper.

3.2 The Core Relational Theme of Forgiveness

We have noted above that the core relational theme of a reactive attitude is supposedto capture the kind of evaluation specific to the reaction.48 Roughly, fear is aresponse homing in on the dangerous, sadness on the loss of something important,etc. In other words, the core relational theme serves to establish what all tokenemotional reactions of an emotion type have in common. We fear many things:dogs, accidents, illnesses, inflation, and social death. But what is common to allthings feared is that they are all perceived to be a threat.

3.2.1 What Forgiveness Is Not

The philosophical challenge now is to find the core relational theme of forgiveness.It is at this point, I submit, where a positive description of forgiveness as anaffective reactive attitude is particularly helpful. Can we single out an evaluativeproperty shared by all forgiveness reactions? In order to do this, we need todistinguish the reactive attitude of forgiveness from some other cognate yet distinctresponses to wrongful actions.49 How to do this was wonderfully demonstrated bythe pioneering theorist of the subject, Aurel Kolnai (1974). The exercise has beenably repeated many times since.50 The point of repeating it here is that bydistinguishing forgiveness from other responses we hope to get a clearer grasp of thecore relational theme of forgiveness as well.

The first thing to remember is that forgiveness is a reactive attitude. Similarly toother reactive attitudes, this attitude too has characteristic forms of expression inlanguage and action. However, forgiveness need not issue in action. Consequently,and this is generally agreed, forgiveness is different from pardoning, mercy andreconciliation. These are all forms of treating the wrongdoer in certain ways.Moreover, neither pardoning nor mercy presuppose forgiveness: they can bejustified even when forgiveness is not. Whether true reconciliation is possiblewithout forgiveness is a bit more contentious issue as we have already noted. But in

48 Many use the term ‘‘formal object’’ to refer to the emotion’s core relational theme. See, for example,Fabrice Teroni, ‘‘Emotions and Formal Objects,’’ Dialectica 61 (2007): 395–415.49 For simplicity’s sake, I talk about forgiving actions in this paper. We need not answer the questionhere whether forgiveness can also be directed at other people’s attitudes or character.50 See, for example, Hampton, op. cit., and Hallich, op. cit.

A. Szigeti

123


any case acts of reconciliation must be distinguished from the attitude offorgiveness.

The second, also relatively uncontroversial, point to remember is that forgivenesspresupposes blameworthiness. This crucial qualification distinguishes forgivenessfrom all reactive attitudes which deny that agent is morally responsible or that theagent has done wrong. When we respond to some harmful action by providing anexcuse for the agent – for example, the excuse that the agent could not have doneotherwise – we thereby deny the agent’s moral responsibility for that action. Whenwe respond to some harmful action by providing a justification for the agent – forexample, that performing the action prevented even greater harm – we thereby denythat the agent has done wrong.51

Interestingly, this second condition is also essential to distinguishing forgivingfrom forgetting. Admittedly, there may be differences between the two in terms ofvoluntariness as well. But the crucial difference is that forgiving positively assertsthe agent’s blameworthiness. As Kolnai puts it: ‘‘Forgiving is not only not‘forgetting’ – in spite of the popular use of that metaphor – but incompatible withforgetting… for the object of forgiveness is a subsistent guilt, not a guilt that (in theforgiver’s eyes) has ceased to exist.’’52

We are now getting closer to identifying the core relational theme of forgivenessas a reactive attitude. Forgiveness, homes in on the wrongful action of ablameworthy agent. However, it turns out that this gloss of the supposed corerelational theme of forgiveness is not specific enough yet. For other reactiveattitudes also appear to have the same core relational theme at this level ofspecificity. And so a more fine-grained analysis is required.

Although this is somewhat controversial, perhaps we can rule out anger (paceRoberts). This is because, while anger is also a response to harm we suffer, andtypically suffer at the hands of others, anger does not appear to presuppose that thetarget of the anger is morally blameworthy. And although this is also controversial,perhaps we can also rule out sadness, frustration, contempt and disappointment forsimilar reasons (pace Richards). Again, the justifiability of these reactive attitudesdoes not seem to be predicated on the blameworthiness of the target person for awrongful action. That is to say, the characteristic evaluative concern of thesereactive attitudes is broader. They may be but are not necessarily responsesaddressed at blameworthy agents for their wrongful actions. Their evaluative focusseems to lie elsewhere.53

51 I take condoning to be just a blatantly objectionable attempt to justify wrongdoing, and so not toconstitute a separate category of response.52 See Kolnai, op. cit., p. 100.53 Some people may argue that the core relational theme of these reactive emotions is not just broaderthan that of resentment, but also that there is no overlap between their core relational theme and that ofresentment. That is, if anger is a fitting response, resentment cannot be. Strawson (op. cit., p. 4), I think,comes close to saying this here: ‘‘If someone treads on my hand accidentally, while trying to help me, thepain may be no less acute than if he treads on it in contemptuous disregard of my existence or with amalevolent wish to injure me. But I shall generally feel in the second case a kind and degree ofresentment that I shall not feel in the first.’’ We do not need to settle this issue here.


123


3.2.2 What Forgiveness Is

But what about resentment? As a number of authors have pointed out,54 the problemis that the evaluative concern of resentment is also restricted to blameworthiness.That is, resentment seems to be a reactive attitude that tracks the wrongful actions ofblameworthy agents.

Needless to say, resentment is clearly different from forgiveness. Resentmenttakes a negative, even retributive stance towards its object, while (at leastcomparably speaking) forgiveness is far more positive. In fact, it seems to be anessential feature of forgiveness that it is incompatible with resentment (while paceRoberts and Richards it may well be compatible with sadness, disappointment,anger, etc.). The combination of two significant facts about forgiveness andresentment – (i) that they are incompatible, and yet (ii) appear to have overlappingcore relational themes – well explains why most theorists accept as their point ofdeparture Bishop Butler’s definition of forgiveness as the forswearing ofresentment.

But these two significant facts about the relationship of forgiveness andresentment also lead to the crucial theoretical difficulty about forgiveness – what Ihave been calling the ‘‘hard problem’’ of forgiveness. Suppose resentment isjustified because the agent is indeed blameworthy for some wrongdoing. Whatjustifies the move from resentment to forgiveness? That is, what justifies the moveto a different, positive reactive attitude, which asserts the same things about theaction and the agent as resentment, namely, that the action was wrong and the agentblameworthy for having done it?

As we have seen in the previous section, solutions which deny that the agent wasmorally responsible or the action wrong and so deny that resentment was everjustified as a response, explain away rather than justify forgiveness (and so of coursedo solutions which deny that forgiveness is rational). Therefore, it is clear that anysolution to the crucial difficulty about forgiveness will look for some factor thatstarted to obtain at t2 some time after the committing of the offense at t1. It is after t2that it becomes at least permissible and sometimes perhaps even obligatory torespond by forgiving instead of resenting. I have already canvassed such solutions.Some look for a factor that alters how the offender is related to the offense. Forexample, repentance and apology may be argued to fundamentally alter the natureof the link between the offender and his wrongdoing. Others look for a factor thatalters how the offender is related to the victim.55

On any one of these solutions, the core relational theme of forgiveness must beexpanded. The sought-after addition should differentiate the core relational theme offorgiveness from that of resentment. For example, if we accept (one of) Roberts’

54 See esp. Kolnai, op. cit., and Hieronymi, op. cit.55 The ‘‘paradox of forgiveness’’ eloquently presented by Kolnai (op. cit.) consists in the fact that, on theone hand, it often looks like that once that factor obtains (after t2) there is really nothing left to forgive –after all the offender is now a different person. On the other hand, before that factor obtains (from t1 to t2)forgiveness is unjustified and morally objectionable since by stipulation nothing of moral significance haschanged since t1.

A. Szigeti

123


solution(s) that the recognition of ‘‘moral commonality’’ with the offender can be agood reason for exercising forgiveness, then the core relational theme of forgivenessmust be further specified by paraphrasing it in something like the following way:‘‘forgiveness tracks wrongful actions committed by blameworthy agents with whomthe forgiver has some important moral feature in common.’’

Alternatively, we might accept Hieronymi’s solution to the problem offorgiveness. This solution is restricted to cases of apologetic and repentantwrongdoers. The proposal, roughly, is that once the offender has apologized hisoffense no longer constitutes a threat to the victim’s moral standing. If we embracethis solution, then we must paraphrase the core relational theme of forgiveness asfollows: ‘‘forgiveness tracks wrongful actions committed by blameworthy agentswhereby the action is such that (given the offender’s apology and repentance) it nolonger constitutes a threat to the victim’s moral standing.’’

So a further advantage of focusing forgiveness as a reactive attitude comes intoview here. Focusing the reactive attitude in this way will help us see what it wouldtake for a solution to the problem of forgiveness to be successful. Whether any ofthese solutions goes through will depend on how convincingly the proposedspecification of the core relational theme of forgiveness will distinguish it fromother emotions.56

A detailed discussion of the respective merits of these solutions is beyond thescope of this paper. My aim here was simply to indicate how the core relationaltheme of forgiveness might be expanded to distinguish it from other emotions and tocapture the central evaluative concern of this reactive attitude. I now turn to thethird way of distinguishing forgiveness from other reactive attitudes.

3.3 Valence

I hope to have made a persuasive case thus far that the ‘‘forswearing of resentment’’formulation in itself reveals too little about substantive, non-privative characteristicsof forgiveness. We need to say more about forgiveness as a reactive attitude thanthat it is incompatible with another one. At the same time, this is not to deny that theincompatibility of forgiveness and resentment may well turn out to be central to afocused definition of forgiveness as a reactive attitude. Here is how.

It is generally agreed that most emotions are valenced. The familiar point is thatthere are two basic categories of emotions: positive (e.g., admiration, praise,happiness, relief) and negative (e.g., guilt, shame, sadness, anger).57 On closerscrutiny, it is somewhat less clear what exactly is meant by this claim. A useful

56 For example, the solution taken from Roberts here is unlikely to distinguish forgiveness fromsympathy.57 Surprise and wonder are the most frequently discussed exceptions. In fact, some argue that becausesurprise and wonder are not valenced they are not emotions. See Kevin Mulligan, ‘‘Emotions andValues,’’ in Peter Goldie (ed.), The Oxford Handbook of Philosophy of Emotion (Oxford: OxfordUniversity Press, 2009), 475–500.


123


distinction is between affect valence and emotion valence.58 The former term refersto the valence of a consciously felt emotion token: whether experiencing theemotion feels good or bad. By contrast, emotion valence refers to the valence of agiven emotion type.

At least most of the time, we are fairly confident in determining the valence of aconsciously felt emotion token.59 Determining the valence of a given emotion typeis less straightforward. This is because different aspects (e.g., behavior, evaluation,facial expression, adaptive value, etc.) of the same emotion may be differentlyvalenced. It is not immediately obvious how the valence of the emotion tout court isto be determined.60

I have noted already that some emotions are distinguished from one anotherprimarily by the fact that they have different valence. I listed envy vs. admiration,thrill vs. fear, and lust vs. disgust. The suggestion is that what distinguishes betweenthe two members of each pair is their valence, while their core relational themes areotherwise largely overlapping. On this account, what distinguishes the thrill ofstanding on the mountaintop from the fear of standing on the mountaintop is therespective positivity and negativity of the felt emotion.

I believe that we should take seriously the thought that the relationship ofresentment and forgiveness is similar to that between other emotions of oppositevalence such as the already mentioned envy/admiration, thrill/fear, and lust/disgustpairs. According to this suggestion, forgiveness is the positively valencedcounterpart of resentment. The overcoming of resentment, which is generallyassumed to be distinctive about forgiveness, consists on this proposal in the reversalof resentment’s emotional valence.

In this way, we obtain a non-privative characterization of the reactive attitude offorgiveness without denying the importance of the connection of this attitude toresentment. Moreover, the suggestion would also provide an elegant explanation ofthe two significant facts we have identified previously about the relationship offorgiveness and resentment: (i) that they are incompatible, and yet (ii) appear tohave overlapping core relational themes.

58 See Louis C. Charland, ‘‘The Heat of Emotion: Valence and the Demarcation Problem,’’ Journal ofConsciousness Studies 12 (2005): 85–101, and Giovanna Colombetti, ‘‘Appraising Valence,’’ Journal ofConsciousness Studies 12 (2005): 103–126.59 Although Charland (op. cit.) has offered important objections to the idea that the valence of the tokenemotional experience is determinate. His suggestion is that the valence of affect is crucially influenced bythe prior evaluative commitments of the person who introspects them. See also Colombetti op. cit.,pp. 117–118.60 See esp. Colombetti, op. cit. In philosophical discussions, the valence of the emotion type overall isusually taken to be determined by the valence of the core relational theme of the emotion (or by thevalence of its formal object if you wish). That is, the valence of the evaluative aspect is thought todetermine the valence of the emotion tout court. Thus guilt would be a negative emotion on this viewbecause it tracks moral wrongdoing.

A. Szigeti

123


4 Conclusion

Based on the above considerations, I submit that the reactive attitude of forgivenessqualifies as a sui generis emotion. It appears to have the characteristics,components, causes and concerns shared by paradigmatic emotions. This approachhas also enabled us to provide a non-privative definition of the core relational themeof forgiveness and explain the incompatibility of resentment and forgiveness.

Let me conclude by summarizing why this finding could matter. First, providinga substantive characterization of forgiveness as a reactive attitude forges what Ibelieve is a helpful link to empirically oriented emotion research. Once we get pastthe negative characterization of forgiveness as merely the absence of some otheremotion, we can study more easily various empirical features of this reactiveattitude.

But the recommended approach may help us to get more traction on traditionalphilosophical quandaries concerning forgiveness as well. So, second, although inthis paper I have not attempted to develop a solution to the hard problem offorgiveness – i.e., what can justify the shift from resentment to forgiveness –focusing forgiveness as an emotion can be useful in evaluating existing attempts tosolve this problem. We have seen that the feasibility of any such attempt willdepend on whether it succeeds in clearly distinguishing forgiveness from otherreactive attitudes.

Third, focusing forgiveness as an emotion can provide a useful point of departurefor developing what might be a more viable alternative than existing solutions. Itmay be that at least some of the existing solutions have moved too quickly fromintuitions that a certain reactive attitude, notably resentment, towards the offender isnot appropriate to concluding that certain evaluative judgments must be true of theoffender as well. One future avenue to explore is to no longer regard theappropriateness of the reactive emotions of resentment and forgiveness as a finalcourt of appeal for the correctness of certain judgments about the offender and hisoffense. Arguably, just as fear is at best a ‘‘rough-and-ready indicator’’61 of whethersome thing or situation instantiates the property ‘‘dangerous,’’62 the emotion offorgiveness may be only a ‘‘rough-and-ready indicator’’ whether an agentinstantiates some property which insulates him from being called to task for apast action of his.

The fourth and final point is related to the previous one. Once we start to think offorgiveness as an emotion we can discover further commonalities betweenforgiveness and other emotions. Of particular interest is the epistemic role emotionscan play as an independent source of knowledge. Emotions can deliver informationconfirming our reflective judgments or conflicting with these. An important recent

61 See Francois Schroeter, ‘‘The Limits of Sentimentalism,’’ Ethics 116 (2006): 337–361; and for asimilar idea, Andras Szigeti, ‘‘Emotions and Heuristics,’’ Journal of Ethical Theory and Moral Practice16 (2013): 845–862.62 Which is not necessarily to say that fear is not a reasons-responsive attitude subject to norms ofcorrectness. But whether fear is appropriate will depend on whether its object is fearsome, not on whetherit is dangerous. This is how I understand Francois Schroeter’s critique of sentimentalism in Schroeter, op.cit.


123


theme in the philosophy of emotions is that our emotions may sometimes be morereliable than our reflective judgments.63 When reflective judgments and ouremotional responses conflict we are sometimes better off trusting the latter.64 In thisspirit, we can direct our attention to the epistemic function of forgiveness. We canask whether owing to its affective character forgiveness can deliver insights whichwe have reason to trust, sometimes even when those insights conflict with ourreflective judgments.

Needless to say, all four points identified here require further scrutiny. Inparticular, the last two points (the feasibility of which need not necessarily affectthe other conclusions of this paper) require a closer engagement with sentimentalistapproaches to metaethics. My hope is that the paper could nevertheless contribute toclearing the ground for future work on these tasks and could indicate more optionsfor resolving them.

63 See Doring, op. cit., and Karen Jones, ‘‘Emotion, Weakness of Will, and the Normative Conception ofAgency,’’ Royal Institute of Philosophy Supplement 52 (2003): 181–200.64 In these respects, emotions are comparable to intuitions, see esp. Sabine Roeser, Moral Emotions andIntuitions (Basingstoke: Palgrave Macmillan, 2011). We should also recall here the two points madeearlier (i) about emotions being representations of non-propositional evaluative concerns, and (ii) aboutemotions being truth-apt even in the absence of propositional content. These two points explain not justwhy conflicts between reflective judgments are likely to occur, but also why in some cases emotions canbe more reliable epistemically speaking.

A. Szigeti

123


Focusing Forgiveness

Documents