University of Birmingham Jealousy revisited: recent philosophical work on a maligned emotion

Taking as its starting point a previous work by the author which reviewed early philosophical sources on jealousy and proposed both a conceptual and moral account of this much-maligned emotion, the present article reviews the relevant philosophical literature from the last decade or so. Most noticeable is how scarce those sources still are. Special attention is given, however, to a new conceptual model proposed by Purshouse and Fredericks which rejects the standard architectonic of jealousy as a three-party compound emotion. While the essential contours of the new model are rejected, Fredericks is shown to offer some powerful misgivings about putative instrumentalist defences of jealousy. In addition to this new model, a number of other recent writings about jealousy – historical, conceptual and moral – are subjected to critical scrutiny in this overview article.

) have contributed to the sporadic philosophical discourse. I explore those contributions at various junctures below.
More attention has, historically, been paid to jealousy within the social sciences, although even there, complaints are still made about jealousy being 'one of the least studied emotions in the field of affective science' (Panksepp 2013: 101). That complaint notwithstanding, discussions of jealousy within psychology seem to have become more nuanced of late than they were for most of the 20th century when the prevailing discourse was preoccupied with sexual jealousy, arguably the least philosophically complex and morally interesting form of jealousy (as explained in the following section). Although the equation of jealousy with sexual jealousy persists in some psychological circles (see e.g. the 2008 volume edited by Wurmser and Jarass), the recent Handbook of jealousy, edited by Hart and Legerstee (2013), focuses on developmental aspects of the emotion as they appear, for instance, in the context of sibling jealousy: a considerably richer line of inquiry. The psychological literature on jealousy deserves a review of its ownnot least by way of philosophical critiquebut I focus in this article on philosophical sources and propose to devote a separate article to the psychological literature (Kristjánsson K 2016 A philosophical critique of a psychological study of emotion: The case of jealousy (paper under construction)). Perhaps yet another article would need to be written about jealousy from a neurological perspective, as it is known to appear in diverse forms of dementia and other cerebral disease. Disciplinary pigeon-holing of this sort is not ideal, as we have learnt in recent years that research on particular emotions is most likely to progress when philosophers and scientists engage one another in dialogue and provide grist for each other's mills (see e.g. the forthcoming edited volume by Carr 2016, on gratitude). However, I am forced into the philosophical corner here for reasons of space.
The specific aim of this article is thus to review critically the recent literature on jealousy in philosophy. Having explored the 20th century discourse in an earlier work (Kristjánsson 2002), I limit the purview here mostly to outputs published 2002 or later. For expository purposes, I use my earlier work as a springboard of the discussion, as I hope it will be helpful for readers to see how more recent writing compare and contrast with tentative conclusions reached about jealousy over a decade ago. Before turning to the most recent writings, it is instructive, however, to provide some philosophical and historical backdrops.

The Philosophical Background
In my previous work on jealousy (Kristjánsson 2002: chap. 5), I set out to elucidate responses to three distinct questions: of (1) what jealousy is (or can most serviceably be understood to be); (2) when, if ever, and then to what extent it can be deemed a rational reaction; and (3) when, if ever, and then to what extent it can be deemed morally justifiable.
I proposed a characterisation of jealousy in response to question (1) thatwhile requiring considerable regimentation and tightening of ordinary languagewas meant to respect, as far as possible, the intuitions of discriminating, critically minded English speakers; hence not serving as a mere stipulation of meaning. This conceptualisation took as its point of departure two common conceptions from the philosophical literature, harking back to Dan Farrell's (1980) agenda-setting article: that jealousy is necessarily a painful three-party emotion, and that it is a compound emotion, made up of other, more 'basic' (in a logical, if not necessarily a psychological, sense) emotions. In compound emotions, the more basic emotions are not only experienced simultaneously but rather feed into one another and make up a unique whole.
Developing those conceptions further led to a characterisation of jealousy as a unique composite of envy, anger and righteous indignation.
In jealousy, according to this proposed characterisation, A is jealous of B because of a favour that A conceives B to have received or be about to receive from a third party, C. More specifically, A envies B and wants to take the relevant favour away from B. However, A envies B for a special reason, namely that A thinks A deserves the favour as much or more than B; hence A is righteously indignant (in Aristotle's standard sense of indignation as pain at undeserved good fortune) that B, rather than A, is getting this favour (exclusively or supplementarily) from C. Moreover, A is angry at C for C's unjustified differential treatment or favouritism. In logical terms, this means that jealousy has a quadratic structure; it necessarily incorporates four variables. A is jealous of B because of x with respect to Cwhere A (the jealous person) is the subject of the emotion; x (the perceived undeserved favouring of B over A by C) is its general object; B (the 'rival' or 'interloper') is its specific target; and C (A's desired benefactor of the favour) is its source.
This characterisation indicates that jealousy belongs to a category of conceptually and psychologically complex self-conscious emotions which include oneself (or, more theoretically speaking, one's own 'self') as an intentional and attitudinal object (cf. Kristjánsson 2010a: chap. 4). I leave out of consideration here the interesting possibility that jealousy can be felt vicariously, on behalf of someone elseeven a large social groupconsidered to be on the receiving end of C's differential treatment. I assume that in such cases, A identifies fully enough with the perceived victims for them to form part of A's own self-concept. Furthermore, jealousy belongs on this account to another large category, of desert-based emotions: emotions that run deeper, developmentally and logically, than those focused on justice qua institutional entitlement (cf. Kristjánsson 2006Kristjánsson , 2015. The appeal of this characterisation lies, arguably, in the fact that if one removes an emotion from the compound or adds one to it, the resulting emotion compound can, more usefully, be described as something other than jealousy. For example, if A believes that B truly deserves the favour more than A, or does not think of it in terms of just deserts at all, A may be better described as sad, disappointed or despairing rather than jealous. The reason for this is that jealousy appears not to be a passive emotion of mere resignation but a call for repairs, or at least a statement of the deservingness of such repairs, even if they are beyond hope (cf. Toohey 2014). Of course, this does not mean that A cannot be sad, disappointed and despairing about the relative lack of favouring by C in addition to being jealous. Alternatively, if A is not angry at C, but is exclusively focused on the undeservingness of B's relative fortune and how to deprive B of it, the emotion is simply one of two-party indignant envy (cf. Roberts 2003: 264). Or if we add a primary focus on B's perceived unsavoury efforts in 'luring' C away from A, the emotion becomes one of anger at B, rather than (or in addition to) jealousy.
Despite this 'appeal' of the characterisation in question, it remains inherently controversial in philosophical circles, as becomes clear in the section 'The recent work on jealousy' below. Not only is it controversial which specific emotions make up the emotion compound, some philosophers even reject the three-party, compound architectonic across the board as we see later. Regarding the first controversy (about the ingredients in the compound), a very odd conceptualisation about the difference between jealousy and envy has crept into some philosophical writings. It assumes that 'in envy we wish to obtain something that the other has and in jealousy we fear losing something that we already have to someone else' (Ben-Ze'ev 2013: 41). I find it difficult to understand how jealousy can fail to be envious; if A does not resent C's relative favouring of B and want to take it away from B, then a core element in the jealousy compound is missing. Let us focus rather on the fear factor. On the conceptualisation proposed at the beginning of this section, for jealousy to be rational, A must have a reason to believe that B has taken, or is going to take, away C's favouring. It is not enough that A fears that this may possibly happen. Briefly put, fearing that you may, at a future point in time, find a reason to be jealous is not to experience the emotion of jealousy, but simply to be fearful or suspicious.
Someone might complain that I am being too constrictive here. After all, a distinction between prospective deterrent shame and retrospective post-mortem shame is well entrenched in the shame literature (Kristjánsson 2014); should we not also, as Rydell and Bringle (2007) suggest, distinguish between two kinds of jealousy: suspicious and reactive? There is a stark disanalogy here, however. Prospective shame is still shame. It is not simply fear of doing something shameful in the future but, rather, shame over the very fact that one considers a possible shameful action as an option here and now. In contrast, so-called suspicious or anticipatory jealousy is not jealousy here and now but fear that some deprival of favouring will happen in the future that will give one a reason to experience the relevant composite of anger, indignation and envy (Kristjánsson 2002: 149-150). That said, A could be suspicious without any good reason that C has already started to favour B; in that case, A is experiencing genuine jealousy, albeit irrational, not only fear of future jealousy. Notably, careful philosophical analyses of jealousy by people like Roberts (2003) avoid the conflation of fear and jealousy. Roberts talks about the favour being construed as 'in the process of being lost to the rival, or as already so lost, or as about to be lost ' (2003: 257). Moreover, the most painful experiences of jealousy surely involve cases where there is no hope of a reversal of fortunesnot when one has got something that one still hopes to retainnamely cases where C's favouring has been irrevocably lost over to B.
In response to question (2), about the putative rationality of jealousy, we need to begin by avoiding the common 'moralistic fallacy' of equating moral with rational appropriateness (D'Arms and Jacobson 2000). The rationality of an emotion is a necessary, but not a sufficient, condition of its moral justifiability. Indeed, almost all philosophical emotion theorists will agree with that assumption, apart from small pockets of hard rationalists, who equate the rational with the moral, and hard sentimentalists, who deem emotions incorrigible and self-justifying (see Kristjánsson 2010b). Irrational emotions are either illogical or they involve epistemological missteps such as disregard for facts, negligent and hasty judgements or purposeful self-deceptions. If a case of jealousy involves none of those missteps and the emotions of envy, anger and indignation are intelligible in the given case, individually and collectively, the jealousy can be deemed rational. Nevertheless, it remains an open question if it should be felt morally, all things considered, in the given context. Notice that it is not necessary for the rationality of jealousy that C does in fact favour B over A; it suffices that A has good reasons for judging/construing that to be the case. Speaking more generally, then, the rationality of an emotion has to do with 'reasonable warrant' rather than 'truth'.
The reason why romantic or sexual jealousy makes for such a morally (as distinct from psychologically) unexciting caseeven if it may, historically, constitute the linguistic archetype of jealousyis that it seems to fall flat on the first hurdle of appropriateness, that of rationality. The sexually jealous person overlooks the fact that love is not a matter of will and no one deserves to be sexually attractive to another. We cannot decide to love someone (romantically/sexually) because we think the person deserves or owes our love. That said, A may well rationally object to C's breach of commitment in favouring B sexually/romantically over A if A and C are already in a relationship with explicit (as in marriage) or implicit (as in stable relationship) commitments to one anotheror A may rationally bear a grudge against B for tempting C, if A has good reasons for holding that to be the case. But in the first of those scenarios the emotion in question is not jealousy but rather anger towards C, and in the second it is anger towards Btypically mixed with envy (cf. Roberts 2003: 261). This is not to say that genuine jealousy cannot be felt in sexual contextsas irrational emotions need not be less genuine than rational onesbut rather that from a philosophical point of view, the question of its moral justifiability fails to emerge as the emotion does not satisfy the prior rationality condition. Clearly the same cannot be said for, say, sibling jealousy or classroom jealousy, for we rightly consider differential treatment by parents and teachers morally unjust, other things being equal. The child and the pupil deserve not to be victims of favouritism.
In response to question (3), I previously went against the grain of popular and academic opinion by arguing, from a quasi-Aristotelian virtue ethical perspective, that jealousy does admit of a medial, morally justifiable, trait-like condition and can, in this medial form, be understood as a moral virtue (Kristjánsson 2002: chap. 5). This proposed justification bears the standard hallmarks of an Aristotelian virtue-based rationale. Emotions can, in Aristotle's much-rehearsed view, just as actions, have an 'intermediate and best condition [...] proper to virtue'when they are felt 'at the right times, about the right things, towards the right people, for the right end and in the right way ' (1985: 44 [1106b17-35]). Persons can be fully virtuous only if they are disposed to experience emotions in this 'intermediate' way on a regular basis. Virtuous emotions are in this Aristotelian model not only instrumentally beneficial to a flourishing life (eudaimonia); they are an intrinsic, indispensable part of it. Because of the relevance of this Aristotelian turn in the potential justification of jealousy, a quick detour into Aristotelian territory is neededalso alluding to some of the recent historical writings about jealousy by philosophers.

Some Aristotelian Backdrops
There are three main reasons why it may be considered wise to empty a putative justification of jealousy into the time-revered Aristotelian bottle. The most general reason is that most successful latter-day moral explorations of emotions have been couched in those very terms (e.g. Nussbaum 2001, on grief and compassion). Another and more specific reason is that Aristotle already offers well-known and plausible moral justifications of medial forms of two of the three emotions in the proposed jealousy-triad compound, proper anger (aka 'mildness of temper') and indignation (nemesis), and if one complements those with his justification of healthy pride and self-respect as ingredients in the master moral virtue of great-mindedness (megalopsychia), the justification of jealousycharacterised along the above lines as a moral virtuefollows somewhat naturally (Kristjánsson 2002: esp. 162-166). The third reason has to do with Aristotle's useful distinction between episodic and trait-like forms of emotions. He famously argued that people are praised or blamed for their emotions as dispositional virtues and vices, but we 'do not blame the person who is simply angry ' (1985: 41 [1105b20-1106a7]). The underlying idea is that we cannot control (and hence not be responsible for) the experience of episodic emotions once the relevant emotional disposition to experience them is in place. When I talked about Aristotle's moral justification of emotions such as anger, indignation and pride above, those were all justifications of emotions as virtuous traits.
That said, contemporary emotion-regulation theory indicates that the Aristotelian view on responsibility for emotions as confined to traits is overly simplistic (see e.g. various chapters in Gross's 2009 edited volume). Even if A has cultivated a certain trait of jealousy, or allowed it to take root in her psyche, this does not mean that A is prey to an ungovernable passion from then on. A can be expected, psychologically and morally, to control her episodic jealousy in certain contexts. Conversely, even if A has a very weak jealousy trait, this does not mean A may not experience intense pangs of jealousy in certain extreme circumstancesfor even a worm will turn. So although ascriptions of responsibility and moral justifiability are mainly directed towards trait-forms of emotions like jealousy, such ascriptions are not necessarily out of place in the case of episodic experiences. I shall not pursue those complexities further, for present purposes. Hence, whenever I refer to the moral justification of jealousy below, I am referring to 'jealousy' in a trait sense rather than as an episodic emotion.
Slightly inimical toif not embarrassing foran attempt to couch a moral justification of a medial jealousy trait in Aristotelian terms is the fact that he himself does not mention this emotion, let alone justify it. A word commonly translated as 'jealousy', namely zêlotupia, did exist in ancient Greek. However, its meaning seems to have been vague, somehow straddling that of 'envy' and 'emulation', and it does not explicitly correspond to 'jealousy', either as characterised above or to any alternative contemporary specification of it (see Konstan 2006: chap. 11). The great underlying mystery here is that the ancient Greeks do not seem to have had any single term for the concept of jealousy. To be sure, it does not prove that they did not possess the concept; after all, in his corpus Aristotle keeps mentioning states of character that have no fixed names in Greek. However, those tend to be obscure aberrations (excesses or deficiencies) of well-known traits, invoked to satisfy Aristotle's penchant for a systematic virtue-and-emotion architectonic. We moderns tend to think of jealousy, however, as more or less the same everywhere and as such an invariable facet of the human condition that we are at a loss to understand a culture that did not have a name, perhaps not even a concept, for it.
This 'mystery' is music to the ears of cultural relativists about emotion such as Konstan who claims that 'the emotions of the ancient Greeks were in some significant respects different from our own' and their repertoire of emotion words does, therefore, not map neatly onto ours (2006: ix-x). This assumption will not bother social relativists about moral virtues. For those who read Aristotelian virtue ethics through a universalist lens, however, some elucidation is due. Roberts does explain persuasively how different cultures may 'hypercognise' or 'hypocognise' emotion types and hence have varyingly fine-grained vocabularies to describe them (2003: 12). Nussbaum warns against 'the common error of supposing that if there is no single term in a language for an experience, that experience must be lacking ' (2001: 155). Similarly, Sanders (2014) blames theorists like Konstan for invoking a lexical method of looking for directly analogous terms for emotion concepts; he thinks that emotions (such as jealousy in ancient Greece) can be inferred from moral scripts that become apparent in expressed values and actions. While all this may be true, it still remains a mystery why such a common emotionat least nowadaysas jealousy did not have a clear linguistic designator in Aristotle's time.
At all events, the fact that Aristotle did not have a word for jealousy in the contemporary senseand possibly not a concept eitherdoes not tell against the well-foundedness of offering a reconstructed moral justification of jealousy along the same argumentative lines as he used for the emotion traits of anger, indignation and pride. Indeed, there is a helpful discussion in Aristotle of two emotions that are related to jealousy as parts of the larger envy-family: begrudging spite and emulation. Begrudging spite (epēreasmos) is A's pain at B's possession of a valued thing and the desire to remove it from B, without any moral reason, and (contra envy) without A wanting it for herself (2007: 117 [1378b16-17]). Emulation (zêlos) is a mixed emotion characterised by A's pain 'at the apparent presence among others like him by nature, of things honored and possible for a person to acquire, [with the pain arising] not from the fact that another has them but that the emulator does not ' (2007: 146 [1388a30-35]). Emulation is mixed because it also includes A's pleasure at B's possession of this valued thing, and (contra envy) lack of any desire to take it away from B. Morally speaking, it is a decent emotion and even a (developmentally relative) virtue for young moral learners. It bolsters the case for including envy in the jealousy-triad compound to observe that if we replace envy there with either begrudging spite or emulation, the compound emotion is clearly no longer one of jealousy. So if A does not really care for C's favour but only wants it removed from B (as in spite), or does not want to take anything away from Beven relatively speakingbut simply wants to get C's favour also (as in emulation), it becomes counter-intuitive to speak of A as being jealous any more.

The Recent Work on Jealousy
Two philosophers have recently offered analyses of jealousy which merit scrutiny (Purshouse 2004;Fredericks 2012), and at the end of this section I explore a new book with strong philosophical dimensions to it (Toohey 2014). Fredericks's work is a PhD thesis but of such unusually high quality that it will no doubt issue in publications in respected outlets soon. Strikingly, the two authors agree more or less on the conceptual contours of jealousy; Fredericks amends those slightly from Purshouse's account, but her amendments concern minor nuances that can mostly be left out of consideration here. Hence, for simplicity, I refer to these two analyses collectively as the PF-model. This model is radically different from the one I offered in 2002; indeed, Fredericks takes explicit exception to my account.
While acknowledging, implicitly at least, that jealousy is a cognitively complex (as distinct from a structurally 'basic') emotion, the PF-model explicitly rejects two received wisdoms about jealousy: that it is a compound emotion and that it is necessarily a three-party emotion with a quadratic structure. Clearly, such radical departures from the mainstream need to be well motivated. To start with the second, it seems to be based predominantly on a strong ordinarylanguage intuition that Purshouse and Fredricks are unwilling to abandon (cf. Fredricks 2012: 133), although they are happy to deny bedrock status to some other prevailing ways of using words. This is the intuition that it is reasonable to call A, the collector of rare coins, jealous when A experiences pain at the realisation that a rival B owns the most valuable collection which A desperately desires (Fredericks 2012: 48;cf. Purshouse 2004: 185). Now, everyone will agree that there are playful, metaphorical cases where 'jealousy' is used in place of 'envy' or 'admiration', as in 'I really feel jealous of your success!' where this is simply supposed to mean 'I really admire your success!'. The example of the coin collector does not fall into that category. Let us also exclude the possibility herewhich would not pose any threat to the three-party modelthat A is jealous of B with respect to some elliptical agency, C, say God or Providence, for providing B with superior fortune in coin-collecting (although that could well be what A thinks). Nevertheless, I will contend that in this scenario, an ascription of jealousy is out of place and that if this is, indeed, the verdict of ordinary language, then it must be corrected in the service of conceptual rigour. In the model suggested earlier in this article, if A judges B to have gained an advantage in collecting through unfair play, then A is angrily envious of B; if A judges B to be undeserving of the advantage over A, then A is indignantly envious; if A is simply unhappy, full stop, about the unfavourable comparison with B and wants to get some of B's coins, then A is invidiously or maliciously enviousor, if you like, 'just' enviousof B. All these possibilities present standard cases of envy; it simply muddies the conceptual waters to bring jealousy into the equation. From a philosophical perspective, nothing compels acceptance of the linguistic intuition on which the PF-model relies here but much militates against it, especially when viewed against the background of the realisation that indiscriminate language speakers often confuse 'envy' and 'jealousy', with an increasing recent tendency being to replace the former indiscriminately with the latter (Kristjánsson 2002: chap. 5). Critically minded philosophers should resist such a tendency rather than condoning it. The PF-model considers a compound theory of jealousy redundant if we acknowledge three necessary conceptual-evaluative conditions (rather than more cognitively 'basic' emotions) undergirding jealousy: that (a) A desires to possess a good, possibly to a certain extent, or in a certain way, say, exclusively or pre-eminently; (b) A regards the actual or potential possession of this good by another person, the rival, as inconsistent with the fulfilment of his desires; and (c) A has in mind some (possibly imagined) set of circumstances in which the desire would have been satisfied (Purshouse 2004: 195; Fredericks adds the condition that the good in (a) be non-replicable, 2012: 67; for Purshouse it is just 'generally ' so, 2004: 198).
The first observation about this conceptualisation is that it seems, at first sight at least, to be too capacious and fail to distinguish jealousy from envy. Purshouse argues that although it does not exclude envy across the board, each component rules out a certain kind of envy (2004: 195-198). Thus, (a) rules out 'destructive envy' where A just wants to take the good away from B without wanting to possess it herself; (b) rules out emulative envy where A only wants to match B without taking anything away from B; (c) rules out the sort of (childish) envy where possession of the envied good is not even envisaged realistically, like being the president of the USA for a non-American. All this is cold comfort for the PF-model, however, as the sorts of envy that Purshouse claims are ruled out in (a) and (b) are not really proper instantiations of envy at all, but rather what Aristotelians consider the separate emotions of begrudging spite and emulation (Kristjánsson 2006: chap. 3). Moreover, I am not sure that jealousy always has to refer to a set of possibly imagined circumstances (with 'possible' understood in an actual rather than a merely logical sense). The fact that an inferior philosopher A is putatively jealous of Plato (B) for being so much admired by the philosophical community (C) may indicate to us that A's jealousy is irrational; but why should it not constitute genuine jealousy (cf. Fredericks 2012: 79, where she seems to confuse an emotion's genuineness with its rationality)? All in all, I do not think the PF-model does justice to the uniqueness of jealousy as an emotional reaction; it offers a conceptualisation that is too permissive and fails in its attempt at conceptual clarification and discrimination.
Fredericks makes the opposite complaint about my 2002 model, namely that it is too narrow (2012: 208) and fails to account for a number of prevailing intuitions. She gives examples of A who has extremely low self-esteem and does not feel any anger at C's favouring of the rival B, because A thinks she is so much inferior to B, and of another A who thinks that B is such a great person that B really deserves C's attentionyet, both As could be jealous, Fredericks contends. In general, she claims that I do not leave room for jealousy that closely resembles fear or sadness, rather than anger and indignation (2012: 63-64). I have given a separate account above of why I think it is unwise to equate jealousy with fear of jealousy.
The real sticking point is what Fredericks might want to call 'sad jealousy'. In the end, I think this debate is about something more than a clash of linguistic intuitions that I or Fredericks or the majority of ordinary-language users hold. It may well be that 'jealousy' is often used, in ordinary language, to describe cases where A feels she is entirely lacking in deservingness compared to B, yet feels sad over C's favouring of B. However, the crucial question will be if such uses are reasonable and serviceable for the individuation of jealousy from other emotions in the emotional terrain. My worry is that, in the cases Fredericks suggests, there is nothing left to distinguish jealousy from mere sadness over A's loss of C's favouring to B. We have reason to see jealousy standing out as a unique emotion precisely because it involves the expression of a moral grievance rather than mere resignation. This essential element is well brought out in a case study that Wurmser and Jarass discuss (Wurmser and Jarass 2008b: 4), of the jealous woman Jane who, when interviewed, expressed the concern that 'she had been wronged, that she had suffered grievous injustice'and made a claim for distribute justice to be restored. The authors use this case as an illustration of irrational, pathological jealousy, which it may well be as it hails from the context of sexual jealousy where considerations of distributive justice seem out of place. But that does not change the fact that Jane's emotion counts as jealousy, however irrational, precisely because she is issuing a moral grievance rather than just wallowing in misery over lost love.
The PF-model does not go far enough towards a normative analysis of jealousy and seems in the end too reliant on putative ordinary-language intuitions, although the authors have not, to the best of my knowledge, conducted any social science studies of thosenor do they refer to such studies done by others. Even for those who consider a psychological study of lay intuitions the natural starting point for conceptual inquiries in emotion research (Morgan et al. 2014), such studies have already shown linguistic intuitions on jealousy to differ radically (Kristjánsson 2002: chap. 5). This holds even among those language users that Roberts refers to as 'insightful' and 'sensitive ' (2003: 57). Some users, like 'Jane' above, seem to understand jealousy along the indignation lines I have suggested, while others take it to incorporate fear and/or sadness. In such cases, there is no substitute for normative philosophical regimentation at least if we aim to say something substantive about the moral standing of the emotion. Otherwise, we will just be talking at cross purposes. I have suggested before (2002) and in this article that a certain characterisation of jealousy is, for reasons of conceptual clarity and economy, the most serviceable one around.
In the end, I consider the crucial 'competition' in the conceptual field to be not between my model and the PF-model, but between my model and an account that retains the generally accepted three-party, compound-structure of jealousy but that replaces the indignation component with fear/suspicionand/or sadness. I have attempted to argue for the superiority of my model by pointing out that whereas the emotion I describe cannot be called anything other than 'jealousy', the alternative emotion can helpfully be referred to by other entrenched emotion terms. I am fully aware, however, that this argumentative move is not conclusive and that a satisfactory conceptual analysis of jealousy remains work in progress. I am also aware of various logical and psychological problems associated with the very idea of 'compound emotions'. Logically, it is difficult to know where to stop in reducing complex emotions to more simple ones; even such a fundamental 'primary' emotion as anger would, in some cases at least, seem to be reducible to even simpler emotions such as frustration or despair. Psychologically, a compound-model of a negatively felt emotion such as jealousy begs the question of whether the pain felt in jealousy is specific to that compound emotion or is somehow made up of the pain felt in the (three) more basic emotions. Or are negatively valenced emotions not set apart at all by different sorts of felt pain but simply by their cognitive consorts? If that is the view on offer, some objectors may complain that debates about which emotions make up the jealousy-compound divert attention from the fact that the most conspicuous phenomenological aspect of jealousy is the felt pain.
To return to the PF-model, Purshouse makes do with a conceptual account of jealousy; Fredericks, however, moves on to explore jealousy's moral standing. In that endeavour, I find her contribution more rewarding; andfor what it is worthshe has at least persuaded me to change my mind on one significant issue. In my 2002 book I argued that jealousy could be justified on two grounds. One is through its intrinsic value for the well-rounded moral life; I return to that point later. The other is through its instrumental value in strengthening commitments and enriching relationships (Kristjánsson 2002: 160) as a sort of a value-signalling protective mechanism (cf. Clanton 1996: 177). Fredericks explores such a functional justification and finds it wanting. She argues plausibly that jealousy offers, in general, an ineffectual, unreliable and often counter-productive means for acknowledging valuable people and strengthening caring or non-caring relationships with them (2012: chaps. 2 and 3). Fredericks marshals powerful and subtle arguments here, showing that even what I would call 'rational' jealousy creates a loss in the moral value of relationships and prevents them from becoming more valuable. Those arguments have persuaded me to drop instrumentalism from the moral justification of jealousy. More generally, those concerns may make us wonder if instrumentalist arguments for the value of emotions ever work. Let us suppose that a new persuasive social scientific study appeared which found jealousy to have, in fact, a positive effect on caring relationships. Would those findings suffice to undergird a moral defence of jealousy? I am not sure, for the same reason that I am not sure that we would find A's over-the-top angry tantrums, because B had failed to clean the toilet, morally justified even if it turned out that angry over-the-top tantrums strengthen caring relationships because of, say, the catharsis that often follows them. Something more is needed to justify an emotion trait.
That something more is, I argue, provided by the sort of intrinsic justifications that Aristotle gives for virtue-based emotion traits of character. The clearest examples he offers are of the desert-based emotions, such as compassion and righteous indignation. It is a sign of a wellrounded moral character to feel pain at undeserved outcomes and pleasure at deserved ones, other things being equal, irrespective of the extrinsic rewards that these emotions may produce. A desert-based sense of distributive justice in human affairs is thus incorporated into Aristotle's very conception of eudaimonia (Kristjánsson 2006;cf. Nussbaum 2001: 32). These considerations underwrite the proposed intrinsic justification of jealousy as the insignia of healthy pride, admirable self-respect and a keen sense of justice (Kristjánsson 2002: 162): a justification that still survives even if the instrumentalist one is sloughed off. Not being inclined to utter a moral grievance when C undeservedly favours B over younot being inclined to resent C's favouring and to remove the favour from Bis the sign of such a lack of assertiveness and self-respect, such a cringing spirit of tolerancenot to mention lack of sensitivity to injusticethat it can only be deemed a moral failure on an Aristotelian account: a vice.
My proposed conceptual and moral account of jealousy places it firmly in the category of desert-based emotions, and it is as such that I justify its proper incarnation, qua trait, as a moral virtue. Notice that nothing in this justification entails that jealousy is most often felt in a virtuous way. Indeed, given the prevalence of irrational sexual jealousy and of excessive forms of sibling or friendship jealousies, of which world literature is full, there is every reason to believe that experiences of virtuous jealousy are, by comparison, rare. Even when an experience of jealousy can count as rational, there are often other complicating factors in the situation which should steer a person of phronesis-guided overall virtue away from being jealous. In any case, I hope to have made it abundantly clear that any justification of jealousy as intrinsically valuable can only be of what Clanton calls 'appropriate jealousy, constructively expressed ' (1996: 183).
Larger questions than can be addressed in this article loom, however, about the extent to which any such intrinsic justifications of emotion traits can have universal moral value. Roberts (2013) argues that they will always remain relative to historic moral frameworks (as distinct from moral theories), such as Aristotelianism, Christianity or Buddhism. For example, a Christian or a Buddhist could choose never to be jealous, in line with those frameworks, without such a jealousy-deficit counting as morally (let alone rationally) inappropriate. While one could question Roberts's general framework-relativism, I shall make do with the more parsimonious claim here that I have proposed an intrinsic justification of jealousy from within an Aristotelian moral framework.
Fredericks objects to any such intrinsic justification of jealousy, for example conducted along Aristotelian lines (2012: 187-198, 210-211), but her arguments are much weaker here than in the case of the instrumentalist justification, primarily because she refuses to acknowledge the relevant moral grievancethe righteous indignationas part of a jealousy compound. I hope, however, to have provided a strong philosophical argument why (a) jealousy is best understood as a compound emotion and (b) that indignation forms part of the compound.
The final work worth considering in this section is the recent book by Peter Toohey (2014). That book defies an easy classification into academic pigeon-holes. Written by a professor of classics, who (as could be expected) makes clever and sustained use of literary examples of jealousy, the book relies otherwise mostly, if slightly surprisingly, on social scientific sources, mixed with the author's own armchair philosophy. Despite being written from a broad humanities perspective, Toohey does not engage at all with the philosophy literature on jealousy; for example Farrell's (1980) classic landmark piece is completely overlooked. Toohey's own common-sense philosophymanifested in his conceptual and moral musings about jealousyis credible enough, however, to give this work considerable philosophical traction.
The approach taken by Toohey to conceptual questions about jealousy is excitingly refreshing. The basic question of what jealousy 'is' gets answered by a thorough exploration of visual and literary works of acts, depicting jealousy. Indeed, a vast number of the former (from medieval artists to Gauguin and Munch) are reproduced throughout the book and analysed in detail. The author's idea here is that to understand jealousy, we need to immerse ourselves in its symbolism (for example of colours and of eyes and ears) and to study fleshand-blood situations, in contrast to the flimsy and truncated cases academics typically elicit of some lifeless As, Bs and Cs. To cut a long story short, Toohey's analysis leads him to embrace the three-party (or what he calls 'triangular') model of jealousy and to conclude, for reasons similar to those given earlier in this article, that some of the most intriguing cases of jealousy are non-romantic. He offers particularly insightful observations on jealousy among artists, academics and within families (between siblings and between parents and children). While relying heavily on social scientific sources, he also rejects their common envy-jealousy dichotomyalthough he does suggest, somewhat quirkily, that envy may be a form of jealousy rather than vice versa (2014: 20-21).
Toohey does not underestimate the destructive side of jealousy. Yet he expends considerable energy in undermining the view that 'the emotion is utterly abhorrenta product of a warped character, unhinged fury or actual mental illness ' (2014: 81). He ends up defending jealousy's role in 'protecting relationships, maintaining fair treatment, encouraging creativity and competitive achievement ' (2014: 221). The problem with Toohey's rationales is that they typically straddle the instrumentalist-intrinsic dichotomy; hence, it is not entirely clear when he is arguing for the role of jealousy as an intrinsic part of the good life versus its positive role in producing extrinsic benefits. Some of the arguments for the latter may be susceptible to the sort of criticism that Fredericks (2012) has successfully mounted. Yet what remains intact is Toohey's argumentwhich harmonises with the main theme of the present articlethat what typically motivates jealousy is 'inequity aversion': the desire to identify and censure inequity in human relations and 'hopefully to re-establish fair treatment ' (2014: 188-189). He explains, for example, the jealousy often felt towards benefits cheats (with the government as the benefactor: C) as grounded in the realisation that they are being rewarded for no effort of their own (2014: 192). In general, jealousy has 'its eyes set firmly on what it reckons is your due' (2014: 194)or as I have put it above, jealousy is a desert-based emotion. Toohey reaches a measured moral verdict about jealousy: It 'stands at the crossroads between selfishness and fairness, it has benefits and costs, it encourages the best and worst in people'but in its proper and best incarnations, it 'can be a very beautiful thing' (2014: 185 and 223).
Toohey's book makes for by far the most enjoyable read of all the works canvassed in this article, and I whole-heartedly recommend it. Yet precisely because of the erudition and insightfulness of its author one cannot help sensing the disappointment of an opportunity lost by his failure to engage with the existing philosophical literature, and to start building the academic bridges that could elevate the discourse on jealousy to a new level of profundity.

Concluding Remarks
It remains a matter for some surprise and disappointment how rarely jealousy shows up on the academic radar in journals such as the present one. I hope that my exploration of the philosophical discourse on jealousy has repaired some of the shameful dearth of philosophical attention given to this emotion. At the same time, I have indicated how a study of jealousy can serve as a helpful pathway to a fuller understanding of the emotional value embodied in desertbased emotions. Philosophers have a knack for disposing cavalierly of objections to their views so as to leave them comfortably the same as before. I hope that the revision I have made to my previous account in light of Frederick's careful argumentation signals willingness to make amends in light of objections. I have, however, provided arguments for not engaging in the sort of conceptual reshuffle that Frederick's and Purshouse's radical departures from the earlier proposed conceptualisationand indeed from any three-party, compound-emotion account of jealousywould entail.
It would be amiss to fail to mention here emotion educationin an article inspired by Aristotelian considerationsas Aristotle considered questions of moral justification inseparable, in practice, from questions of moral cultivation. Let me simply suggest that jealousy may be one of those virtues where we need to teach students to err on the side of one, rather than the other, extreme in order to successfully hit the golden mean: namely, to steer clear of the more contrary, but more common, extreme of the mean by dragging themselves off in the opposite direction, as one does 'in straightening bent wood' (Aristotle 1985). Although, as I have argued, moral character can be crippled by under-reactions of jealousy, it is probably more often endangered by over-reactions. In line with what Nussbaum has suggested more generally (2001: 236), the arts and literature will serve as the most valuable source of moral understanding and guidance in proper-jealousy education, since they provide such rich sources of narratives of both well-formed and malformed jealousy (cf. Toohey 2014). In addition to educational concerns, which deserve more sustained scrutiny, jealousy clearly has wider socio-political implications than those indicated by the relatively narrow understanding of 'moral' in the present article. For reasons of space, however, those implications will have to remain a topic for another day.
To wrap up, many readers will have heard of H. G. Wells's tongue-in-cheek definition of moral indignation as jealousy with a halo. If the account proposed in this article bears scrutiny, jealousy admits of an equally sarcastic definition as moral indignation with a stigma. I have tried, however, to remove the stigma of categorical repudiation and ready condemnation from this emotion and indicate its potential salience for flourishing lives. Engaging with the most recent philosophical sources on jealousy has helped nuance that aim.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.