I am grateful to the insightful comments and helpful suggestions provided by an anonymous reviewer on an early version of this paper and to Simon Mussell for his help improving my English. Work on this project was funded by the German Research Foundation (DFG) (Project; Mental Images and Imagination).
1How to distinguish between emotions such as envy, disgust, and shame from other affective phenomena which resemble attitudes and are occasionally called “sentiments” such as love, hate, and adoration? Contemporary analytic philosophers such as Ben-ze’ev (2000) and Deonna and Teroni (2012) distinguish emotions from sentiments mostly in terms of their temporal structures. While emotions have been regarded as episodic responses with a specific phenomenology, it is widely agreed that sentiments as enduring states cannot be acutely felt. In the “temporal approach” (hereafter TA), as I call it, differences between emotions and sentiments are established according to the way in which each of these states extend over time. In particular, differences in the temporal structure can be explained as differences in the duration, the etiology, and/or phenomenology. These different criteria are sometimes defended in isolation, sometimes in combination. However, as I shall demonstrate, none of the variants can offer a clear-cut distinction between emotions and sentiments. Indeed, not only sentiments but also emotions might be long-term states. Moreover, both emotions and sentiments can be acutely felt.
2Against this backdrop, I turn to the phenomenological tradition and suggest the “intentional approach” (hereafter IA) as a plausible alternative. According to IA, emotions and sentiments exhibit distinct intentional structures. In current philosophy, a version of IA has been put forward by De Monticelli (2006, 2020) who explains the difference between emotions and sentiments mostly in terms of their distinctive axiological structures. While emotions are responses to values, sentiments are dispositions to present their objects as “domains of axiological discovery” (De Monticelli, 2020, p. 284). Yet, though her account provides a clear-cut distinction between emotions and sentiments, I will argue that it requires some refinements in order to truly reflect the intentional nature of each of these affective states. In particular, besides the axiology, other aspects of the intentional structure of both kinds of states need to be analyzed, and the fact that sentiments exist not only in dispositional but also in occurrent form should be considered. In view of this, I will develop a different version of IA according to which emotions and sentiments differ not only in terms of their respective axiological structures but also in how they relate to their targets. Put otherwise, I will argue that emotions and sentiments differ in their respective forms of “affective intentionality”.
3The paper is structured as follows. I introduce different versions of the temporal approach and demonstrate that none of them provides a clear-cut distinction between emotions and sentiments (section 1). Next, I present the intentional approach (IA) as an alternative. To this end, I discuss De Monticelli’s account, before introducing my version of IA according to which emotions and sentiments exhibit different forms of affective intentionality (section 2). The main findings are summarized in the conclusion (section 3).
4Recall the temporal approach:
TA (Temporal Approach): emotions and sentiments can be distinguished from each other in terms of their respective temporal structures.
5In this section, I discuss three criteria behind TA: duration, etiology, and phenomenology. Each of these criteria focuses on a different aspect of the temporal structure of emotions and sentiments.
TA-Duration: emotions and sentiments differ in terms of duration: while emotions are episodes of shorter duration, sentiments stretch over time.
6According to TA-Duration, while emotions are occurrent short-lived mental episodes, sentiments are long-lasting states. TA-Duration has been widely defended among philosophers and in current research it is the main criterion employed to distinguish between both states. Proponents of this approach include Shand (1914), Broad (1954), Ben-Ze’ev (2000), and Naar (2018) (the view is also popular among psychologists, see, e.g., Frijda et al., 1991).
- 1 See, for this claim: Naar, 2018.
7At first sight, TA-Duration appears to be a quite plausible criterion. First, emotions such as envy, disgust, and shame tend to have an episodic character with a definite temporal duration. By contrast, sentiments such as love, hate, adoration, etc. exhibit a certain resemblance to long-term attitudes insofar as they are always enduring states. In addition, emotions have usually been compared or even assimilated to short-lived states such as perceptions.1 Cases of emotions that last ten years would be as puzzling as perceptions that last so long.
8However, on closer inspection, TA-Duration does not offer a clear-cut criterion to distinguish between emotions and sentiments. To begin with, some emotions can also be long-lasting states. Consider for instance the case of “existential envy”. This form of envy, which is depicted in the biblical history of Cain and Abel, is a recurrent literary theme. Existential envy is not an episodic, short-lived occurrence, but an enduring state which marks life-long relations between individuals. Even if existential enviers do not live in a continuous and permanent, acute state of envy, their envy is an enduring state of mind able to motivate action, emotion, and thought. For instance, if A experiences existential envy toward her sister B, A will be prone to devaluate B, to feel anger when B experiences a success, to think that B does not deserve to be appreciated by others, etc. If cases of existential envy are possible, then not all emotions are short-lived occurrences: some emotions might have an enduring nature. Hence, duration cannot be employed as a clear-cut criterion to distinguish between emotions and sentiments.
9Against this view, however, the proponent of TA-Duration could try to develop two lines of defense. First, she could argue that when we use an emotional term to refer to a short-lived episode as well as to cases of long-lasting emotional states, the use of the emotional term is ambiguous (for this line of reasoning, see Goldie, 2000 and Naar, 2018). For example, though we employ the term “envy” to refer to “episodic envy” as well as to cases of “long-lasting existential envy”, the term envy is employed here to refer to two distinct kinds of mental state. The same could be applied to cases in which we use the expression “jealousy” to refer to an episodic occurrence of jealousy and to cases in which we are jealous of someone for a long period of time. Similarly, we might speak about being angry occasionally after being insulted and of being angry for ten years. Though in both sentences we employ the expression “anger”, this term is used to refer to two different states.
10Against this line of defense, it can be argued that the similarities between both envies, both jealousies, and both angers are too strong to refer to two different mental states. Indeed, both the occurrent episodic state and the long-lasting state target the same kind of objects, apprehend these objects in the same way, and exhibit a similar phenomenology. Episodic and existential envy target a rival as possessor of an object which the subject considers desirable but which she cannot obtain, the object is presented as valuable and regarded as a good, and in both cases, when it is experienced, envy has a similar negative hedonic valence, is accompanied by particular sensations, etc. Similar cases could be construed for jealousy and anger.
11The second line of defense elaborated by the proponent of TA-Duration consists in arguing that some affective states might come in both configurations. Let’s call this strategy “the two configurations view”. Note that this view concerns only some affective states but not all. In this view, an affective state A, when it is of short duration, has to be considered an instance of emotion, but when A is long-lasting then it has to be regarded as a sentiment. Many proponents of this view tend to assimilate enduring emotions to sentiments. The view that some affective states come in both configurations has been defended, for example, for the case of hate. Psychologists such as Halperin (2008) and Halperin, Canetti & Kimhi (2012) have argued that hate can be an immediate “burning” emotion as well as a “chronic” state. This view is also implicit in Salice’s account of hate (2020). While the first line of defense mentioned above argues that we use the same term to refer to two distinct kinds of mental states (e.g., episodic envy and existential envy are both called envy but they are in fact different kinds of affective state), this line of defense argues that the same affective state might be an emotion as well as a sentiment (e.g., hate is an emotion as well as a sentiment).
12The problem with “the two configurations view” is that we do not have enough evidence for it. Indeed, from the fact that some affective states – as seems to be the case for hate – can be acutely felt and exist for longer periods of time, proponents of “the two configurations view” extract the strong assumption that some affective states can be emotions as well as sentiments. However, we lack support for this strong conclusion. In my view, rather than the aforementioned strong conclusion, it is more reasonable to extract the following two less problematic conclusions. First, emotions tend to be short-lived states but there are also enduring emotions which on certain occasions one might come to feel acutely. Second, sentiments, despite being enduring states, can be on certain occasions acutely felt. That is, though there are no purely episodic sentiments (all sentiments are long-term states), sentiments can be on certain occasions felt (when a sentiment is acutely felt, it is always embedded in a long-term state). I will return to this alternative in section 3. However, my thought is here that TA-Duration does not offer a clear-cut criterion to distinguish between emotions and sentiments.
TA-Etiology: Emotions and sentiments differ in their etiology: while emotions might arise immediately, sentiments are the result of a process of sedimentation.
13According to TA-Etiology, while emotions might arise suddenly, sentiments emerge as a result of a process of sedimentation in which other affective states have taken place. This view has been defended, for instance, by Broad (1954) and more recently by Deonna & Teroni (2012). For these authors, sentiments emerge as a result of a process in which several affective states have been involved. Though this view has been defended often in combination with TA-Duration, it underscores a different aspect of the temporal structure of emotions and sentiments. Rather than a question of length, TA-Etiology refers to the origins of each state.
14The view that sentiments such as love, hate, and adoration presuppose a history between subject and target has been widely stated in the literature. For instance, it has been argued that hate is never the first reaction toward the target and that it has biographical components (see Kolnai, 2007; Sternberg & Sternberg, 2008; Szanto, 2020). Similar views can be defended for love and adoration. There is a wide range of experiences that can lead to a sentiment. Cognitive, affective, and conative elements might prepare the ground for a sentiment to emerge. For instance, repeated thoughts, perceptions, emotions, and desires about a specific object might lead to a sentiment of hate, love or adoration to emerge. By contrast, emotions like envy, disgust or shame can arise quite soon after some changes in our vital situation have taken place.
15Though TA-Etiology seems plausible at first sight, it too does not offer a clear-cut criterion to distinguish between both states. Indeed, emotions might sometimes also arise after a series of experiences with their target have taken place. Consider again the case of existential envy. Usually, such cases arise among siblings, peers or friends and they presuppose a series of experiences with the target that the subject regards as highly negative. Or consider certain instances of fear. It might happen that an individual who did not experience a particular target as dangerous, after a series of negative encounters with it, develops fear toward it. This emotion of fear has in common with sentiments that it arises after a process of sedimentations of negative experiences with an object has taken place.
16Can a sentiment arise as an immediate response? Examples of sentiments such as love, hate, adoration, benevolence, malevolence, friendship, and enmity presuppose in general that we have made repeated interactions with their respective targets. Yet, on certain occasions we might have the impression to feel love, hatred, adoration, benevolence or malevolence, friendship or enmity toward a target even if we have not interacted with it. For instance, we might love our children from the very moment we discover that we are pregnant, we might hate someone from another ethnic group for the mere fact of belonging to this other ethnic group, and so on. Though in these cases we have the impression that the sentiment arises as immediate response, these are cases in which we have “inherited” from our environment a sentiment toward particular targets. For instance, because of a specific religious uprising we have learned to love an unborn child, or because of being immersed in a context dominated by xenophobia, we have learned to hate persons belonging to another nationality. In these cases, we have the impression that these sentiments arise immediately because through a process of socialization, we have learned certain attitudes toward specific targets. Alternatively, it can be the case that we might have the impression that these sentiments arise immediately because we are not aware of the process of sedimentation that elicited them. For instance, it can be the case that we are unaware that we love the unborn child after a pattern of emotional responses (e.g., surprise, joy, excitement, expectation, etc.) has unfolded over time and that we hate foreigners after a series of other emotional responses (e.g., fear, anxiety, etc.) towards people from other countries has taken place.
17Given that not only sentiments but also emotions might arise after a process of sedimentation, and given that some “inherited” sentiments might arise as immediate responses toward particular objects, the differences in the etiology of sentiments and emotions do not provide a clear-cut criterion to distinguish between both states. As was the case for TA-Duration, TA-Etiology indicates only differences of frequency: while emotions tend to be immediate responses, sentiments tend to presuppose a history between subject and object.
TA-Phenomenology: Emotions and sentiments differ in their phenomenology: while emotions are acutely felt, sentiments are dispositions which can only be experienced in terms of other affective states.
18TA-Phenomenology argues that while emotions as occurrent states can be acutely felt and have an associated phenomenology, sentiments exist only as dispositions to experience other affective states. The reasoning behind this view goes as follows. Proponents of TA-Phenomenology start with the premise that affective states which extend over long periods of time cannot be continuously acutely felt. Indeed, it seems implausible to have an acute episode of hate for five years. Accordingly, they argue that sentiments as long-term states cannot be acutely felt all the time. From these premises they extract the conclusion that sentiments exist only as dispositions to experience other states. For instance, after an incidence that took place five years ago, person A hates person B. For the last five years, person’s A hate exists as a disposition to feel rancor, revenge, anger, indignation, etc. toward person B. This view appears often in combination with TA-Duration and it has been widely stated in the literature. For instance, it is defended by Deonna & Teroni (2012). The meaning of this view has been laid out clearly by psychologists Royzman, McCauley & Rozin for the case of hate. For them, hate “is neither a special emotion nor a blend of emotions but rather a tendency to emote in a number of ways to a number of situations involving the object of hatred” (Royzman, McCauley & Rozin, 2005, p. 6). In this view, the phenomenology or the “what it feels like” of hate has to be explained in term of the emotions that hate disposes us to experience. Similar views could be stated for other sentiments such as love, adoration, benevolence, etc.
- 2 This focus on the phenomenology does not imply that proponents of this criterion regard the emotio (...)
19Without doubt, emotions are occurrent mental states which can be acutely felt.2 However, in my view, emotions can be acute as well as dispositional states. When emotions extend over time as I suggested above with the example of existential envy, then the subject is probably not in a permanent state in which she experiences non-stop acute pangs of envy. When existential envy is not acutely felt, it might exist in dispositional form and motivate other mental states. In sum, though emotions tend to be acutely experienced, they are not always acutely felt. According to this view, existential envy is only experienced acutely under certain circumstances, for instance, when the envier sees or thinks about the envied target and good. Yet, the proponent of TA-Phenomenology can employ my argument to deny the existence of long-term emotions: she might argue that precisely because we do not feel pangs of envy all the time, it is absurd to claim that envy can be a long-term state. Thus, we need a more convincing argument against TA-Phenomenology.
20A compelling argument against TA-Phenomenology can be developed by focusing on how we experience sentiments. I do not see why long-term states such as sentiments cannot on certain occasions be acutely experienced. The best proof of this view is provided by our own experience. There is something that it feels like to hate, to love, to admire, to adore, to be benevolent and so on, just as there is something that it is like to feel shame, envy, disgust or fear. In brief, sentiments too have a phenomenology and their “what it is like” cannot be reduced to the “what it is like” of other states they might motivate.
21In my view, regarding their phenomenology, the only difference between emotions and sentiments is that while we can have an acute episode of an emotion even if the emotion does not exist as a long-term state, to have an acute episode of a sentiment, the sentiment must exist as long-term state. Thus, to experience an occurrence of hatred, love, adoration and so on presupposes that these sentiments exist already as enduring dispositions (either because of a personal history or because we have “inherited” them from our environment). Every acute episode of a sentiment is embedded in an enduring disposition: there are no purely episodic sentiments. By contrast, acute episodes of emotions may or may not be embedded in an enduring disposition.
22According to the model I am defending here, when emotions occur as mental episodes, they are acutely felt (e.g., in seeing my neighbor’s car, I experience a pang of envy). Yet, when they exist as dispositions, they make us prone to experience certain emotions (e.g., my existential envy might prompt me to experience episodes of anger, indignation, etc.). Sentiments exist as dispositions (e.g., my love for my son might prompt me to experience episodes of joy, tenderness, etc.) but they can also be occurrent mental states with an associated phenomenology (e.g., there is something that it feels like to love: I can have an acute experience of love in the presence of a loved one).
23If the proposed model is plausible, then TA-Phenomenology does not offer us a clear-cut criterion to distinguish between emotions and sentiments. In fact, TA-Phenomenology is based on a flawed reasoning. The premise according to which affective states of long duration cannot be uninterruptedly felt and the premise that sentiments as long-term states cannot be continuously felt are correct. However, from these premises it does not necessarily follow that sentiments exist only as dispositions to emote. Like emotions, sentiments can be acutely felt as well as exist in dispositional form.
24So far, I have argued that none of the versions of TA can offer us clear-cut criteria to distinguish between emotions and sentiments. In what follows, I will discuss the intentional approach:
IA (Intentional Approach): Emotions and sentiments differ in their respective intentional structures.
25I will start by discussing De Monticelli’s version of IA and present, in a next step, what I take to be a more plausible version of IA.
IA-Axiology: emotions and sentiments differ in their axiology: while emotions respond to values, sentiments are dispositions to present their objects as domains of axiological discovery.
26Drawing on Scheler’s (1973) model of the stratification of the emotional life according to which affective states can be placed at different levels of depth according to the rank of values they respond to, De Monticelli (2006; 2020) argues that there are different layers of sensibility which reflect the importance of the values concerned and the degree of personal involvement (2020, p. 284). In this context, De Monticelli distinguishes between feelings and sentiments along the following lines.
27Like Scheler, she argues that feelings can be connected with the sphere of the sensory, vital, personal, and divine values, which are revelatory of “how” we are. This happens in sensory pleasures and pains, in vital feelings such as fatigue or drowsiness, as well as in emotions and moods. Therefore, in De Monticelli’s model, the emotions (e.g., shame, envy) can be regarded as belonging to the broad category “feelings”. As such, emotions are responses to values.
28By contrast, according to De Monticelli, sentiments (e.g., admiration, contempt, love, hate) tell us “who” we are. As she puts it, sentiments are “relatively abiding dispositions to assent to or dissent from the very being and value of someone or something. The peculiarity of sentiments is to present their objects to us as domains of axiological discovery” (De Monticelli 2020, p. 284). In her model, sentiments motivate and are the “ground” for long-term desires, passions, intentions, choices, as well as high-level emotions. “Sentimental dispositions” – as she calls them – belong to the personal layer of our sensibility and lead us to discover or change value preferences.
29In a more pointed way, in De Monticelli, the difference between emotions and sentiments can be explained in terms of the axiological element of their respective intentional structures: while emotions respond to values, sentiments are dispositions to present their objects as domains of axiological discovery. IA-Axiology, as I call it, has the virtue of offering a clear-cut criterion to distinguish between both states. While emotions such as fear, disgust, envy, shame, and so on respond respectively to the dangerous, to the disgusting, to a value possessed by the rival, to a potential disvalue in our own self, etc., sentiments predispose us to discover values in the target and to respond to them with emotions, choices, actions, desires, etc., of a certain kind.
30Yet, though agreeing with the idea of differentiating between emotions and sentiments in terms of their respective intentional structures, some aspects of De Monticelli’s account are problematic. First, the difference between emotions and sentiments should be approached without linking them to levels of depth. The notion of depth involves a rigid understanding of our affectivity according to which some affective states are closer to one’s personal core, i.e., to “who we are”, while others express only “how we feel”. However, not all sentiments are revelatory of our personal identity. Consider, for instance, the case in which person A who has a “good heart” experiences hate toward person B for having killed her family. Is this hate constitutive of A’s identity? I do not think so. This hate does not chime with A’s character nor does it reveal who she is (though it might reveal what she cares about). In fact, person A might suffer precisely because the hate does not fit in with her psychology, which is mostly dominated by love and benevolence. Thus, sentiments are not necessarily constitutive of the person we are, though, like emotions, they reveal how we feel.
31A second problematic aspect concerns De Monticelli’s understandings of sentiments in terms of “dispositions” to make choices, act, behave, and emote. As I argued above, sentiments are dispositions to experience other mental states but in the view presented above dispositions can also be acutely felt. There is no doubt that sentiments predispose us to have other mental states but it is also certain that they can be experienced as occurrent states. Love, hate, adoration, veneration, etc. each feel a particular way: each of these sentiments has an associated characteristic phenomenology.
32Finally, De Monticelli’s model requires a refinement in two respects. First, while the idea that emotions are responses to values was widely held in classical phenomenology (Scheler 1973) and has been object of recent defenses (e.g., Mulligan 2009; Vendrell Ferran 2008), the specific axiological nature of sentiments must be the object of further research. De Monticelli’s claims that sentiments present their objects as “domains of axiological discovery” and that they “instigate a kind of search as exciting and fallible as any search for truth” (2020, p. 285) have strong intuitive appeal, but they require further elucidation (how to interpret the “axiological discovery”, the “search”, etc.?). Second, though axiology is a central element of affective states, this element alone does not exhaust their intentional structure. Indeed, the fact that emotions and sentiments are directed to objects (understood in the broad sense as encompassing things, animals, persons, states of affairs, etc.) should also be taken into account.
33In what follows, I will develop a version of IA by analyzing two different elements of their respective intentional structures. The analysis will lead to an alternative version of IA according to which emotions and sentiments exhibit different forms of affective intentionality.
- 3 According to “impure” intentionalists, such as Crane, intentional states cannot be defined exclusi (...)
34As argued by Crane (1998), drawing on Brentano (2015) and the phenomenological tradition, intentional states can be defined by two key elements: they target an object and they apprehend this object in a certain way.3 First, they exhibit a relational nature, i.e., they are directed toward an object. In perception something is perceived, in imagination something is imagined, in remembering something is remembered, in judging something is judged, etc. Second, intentional states exhibit a fine-grained nature by virtue of which when an object (in the broad sense stated above) is apprehended by an intentional state, it is apprehended in a particular way. For instance, in perception the targeted object is presented as being there for the senses, in imagining as being represented in image, in judgment as being true and false, etc.
- 4 Here, I focus on two elements – the material and the formal object – which have been widely analyz (...)
35Both elements can be employed to examine the nature of affective states such as emotions and sentiments. First, affective states are directed toward objects of different kinds, which the literature usually refers to as “material objects” (Kenny, 1963). Envy targets the rival as the possessor of a coveted good, love targets a loved person, cheerfulness is directed toward everything in my surroundings, etc. I will refer to this constitutive moment of affective states in terms of “relationality”. Second, affective states present their objects under a particular aspectual shape. More specifically, the objects are apprehended as being good or bad, that is, as imbued with an evaluative property or value and as inviting us to adopt a pro- or a contra-attitude toward them (this view can be found in Brentano, 2015; Scheler, 1973; and other phenomenologists; for an overview see Vendrell Ferran, 2008). These evaluative properties or values are the “formal objects” of affective states. While the material objects are not intimately linked with the affective states, the formal objects of our affective states are limited. While my fear can target animals, persons, objects, and situations of very different kinds, these targets always present as having the evaluative property or value of the dangerous (formal object).4 This constitutive moment of affective states is what can be called (as I did it above) their “axiology”. This phenomenological view of affectivity is particularly important in the context of current research where philosophers are still discussing when a state should be considered as belonging to the family of the “affective”. In fact, the phenomenological view suggests as a hallmark of all affective states the fact that they present their objects not as being neutral but rather as being related to values.
36De Monticelli’s account is based precisely on this axiological element which determines the fine-grained nature of emotions and sentiments (though the axiology of sentiments requires a more detailed analysis). Yet, in my view, the first element regarding the relational structure of affective states should also be explored. Only when we take into account both elements will we be able to develop clear-cut criteria to distinguish between emotions and sentiments on the basis of their respective intentional structures. Therefore, IA-Affective Intentionality will employ a twofold criterion which takes into account the relationality as well as the axiology of both states. In what follows, I will explore the thought according to which emotions and sentiments substantially differ in the way in which they target their respective objects and present them under a certain evaluative light. As I will suggest, each of these states exhibits a distinctive relational and axiological structure and these differences provide us with clear-cut criteria to distinguish between the two.
37According to the first criterion, emotions and sentiments differ in terms of the relational structure of each of these states. How to explain the distinctive relational structure of emotions and sentiments?
- 5 Here and in 2.2.2. I work with the phenomenological distinction between “content” and “mode”.
- 6 This view is not shared by all early phenomenologists. See Pfänder (1913/1916) and Kolnai (2007).
38Let’s focus first on the possibility that each of these states differ in terms of their respective “contents”.5 To begin with, one possibility consists in arguing that emotions and sentiments target objects of different kinds. Interestingly, within early phenomenology, Stein argued that while emotions might target items, animals, persons, events, etc., sentiments target only persons (Stein 1989, 101; she also defends a further claim to which I will return below according to which sentiments are directed exclusively to personal values).6 I call this “the only persons view”.
39“The only persons view” is, however, quite controversial. Indeed, sentiments might have targets other than persons. Whether a dog, a cat or a bird, we love our pets regardless. We really hate racism, xenophobia, and misogyny. A child might adore fictional characters such as Superman. If we can love animals, hate thoughts, adore fictional entities, abstract artifacts, etc., then sentiments do not only have persons as targets. Therefore, “the only persons view” is false.
40Moreover, even if Stein were right and sentiments only have persons as targets, this would not provide a clear-cut criterion to distinguish them from the emotions because emotions might also target persons. We are envious of a friend, angry at a neighbor, etc. Indeed, if a subclass of targets of the emotions is the same as the class of objects targeted by sentiments, the difference between both states is not clearly defined.
- 7 For Pfänder (1913/1916), sentiments, like emotions, can exist in different temporal forms. As he p (...)
41An alternative way to spell out differences in the relational structure of emotions and sentiments consists in focusing on “how” each of these states targets its respective object. This suggests a focus on the “mode” in which emotions and sentiments respectively apprehend their contents. In classical phenomenology, Pfänder argued that both emotions and sentiments are directed to intentional objects which are presented via a cognitive state (mainly a perception, imagining, memory or belief). In his view, emotions and sentiments are oriented differently toward their targets. While emotions are states with a hedonic valence, sentiments are described as “bridging the gap” between subject and object, showing a “centrifugal” direction toward their targets, and “streaming” from subjects toward objects (1913, p. 332–335).7 This metaphorical use of the language suggest that sentiments for Pfänder exhibit a particular link between subject and object which cannot be found in the emotions. Rather than being merely affected by the object, the subject of a sentiment does something with the object: it affirms or denies it, it aims at uniting with it or creating a distance with it.
42Though Pfänder provides something of a metaphorical description of sentiments, I suggest that we read him in the following sense. Sentiments, but not emotions, presuppose an active involvement (what he calls a stream, a bridge) of the subject toward the object. The subject is actively directed toward the object. This interpretation is also coherent with De Monticelli’s observation that in sentiments there is a “search” in the target. Yet, I suggest here to understand this aspect in the sense that sentiments are a “form of regard”, an expression coined by Mason to describe affective states such as contempt (2003, p. 247). By contrast, emotions involve the opposite movement from the object to the subject, i.e., they are a way of being affected by the object and respond to it. This difference is a difference in how the subject relates to the object of the sentiment in terms of being actively directed to it (as is the case for the sentiments) or being passively affected by it (as is the case for the emotions).
43Now we can specify in what sense the relational structure of emotions differs from that of sentiments in the following terms:
IA-Relationality: Emotions and sentiments differ in their relational structure: while emotions are responses to the target, sentiments are a form of regarding it.
44As I shall show, the relationality of both states is intimately linked to their axiology which I will interpret in terms of the mode in which each of these states evaluatively apprehends its target.
45Let’s explore more closely the axiological element of the intentional structure of emotions and sentiments. This aspect was central to De Monticelli’s account. Here, I agree with the view that emotions are responses to values. However, as I mentioned above, her claim that sentiments present their targets as objects of “axiological discovery” and involve a “search” needs to be elucidated further. How to understand, then, the axiological dimension of sentiments in comparison with emotions?
46For starters, one possibility put forward by Stein consists in looking for a difference in the kind of values targeted by each of these states. Indeed, for Stein, sentiments are only directed toward personal values (Stein, 1989). Though she is not fully clear about what is meant by “personal values”, it seems that she is referring to the values which are constitutive of the person one is. Let’s call this “the only personal values view”. Note that this view distinguishes between emotions and sentiments in terms of their respective targeted contents.
47However, this view cannot offer us a clear-cut criterion to distinguish between both states. Emotions might be directed to personal values as well. Consider for instance the case of being morally disgusted toward a person who exhibits racist behavior. If here we have an emotion which targets a personal disvalue, then the axiological difference cannot be explained in terms of a difference in the kind of targeted values.
- 8 Note that while phenomenologists have usually been objectivists about values, here I defend the vi (...)
48Having rejected the claim that emotions and sentiments differ in terms of the kind of values instantiated by the target, my suggestion here is to search for the difference between the two states in how each of them evaluatively presents its target. That is, a focus on the “mode” in which the target is evaluatively presented is needed. As mentioned above, differences in the axiological structure of emotions and sentiments are related to differences in their relational structure. This means that while I noted before that emotions are a form of being affected by changes in our environment and sentiments are forms of regard, we can now make this description more specific in evaluative terms. More precisely, my thought is that while emotions are a form of being affected by evaluative properties of the target, sentiments are responsible for presenting the target under a certain evaluative light. In other words, sentiments project onto the target evaluative properties that are congruent with the sentiment we are experiencing. It is in this sense that they present their objects as “domains of axiological discovery” and involve a “search” in their targets, as De Monticelli puts it. Yet, note that unlike De Monticelli, I do not think that sentiments reveal the person we are, but simply express how we feel about something. Moreover, while De Monticelli describes sentiments in terms of a search for truths in the target and of domains of axiological discovery, I describe sentiments in terms of projection of what we feel in the target. That is, the fact that in hate the other seems to me to be odious is a product of my projection and does not necessarily presuppose that the other is such.8
49To illustrate this point, consider for instance envy and hate as instances of emotion and sentiment respectively. Envy is a response to the fact that another person, the rival, is presented to us as worthy insofar as she has a good that we also covet. By contrast, hate is not a way in which we are affected by a value instantiated by the target but rather a way to see the target as evil. Put otherwise, hate is never a form of being affected by negative features of the target. Why? First of all, we can hate persons with positive qualities and hate them precisely for having these positive qualities. This occurs when someone hates another person for being more intelligent, beautiful, etc. Second, though hate might arise after repeated negative interactions with the target have taken place, the apprehension of negative properties in the target does not necessarily elicit hate as a response. In fact, we could react toward a disvalue of the other with indignation, anger, etc., even with compassion and pity. According to the picture that I am suggesting here, rather than a form of being affected by an evaluative property of the other, like the emotions, sentiments are a form of regarding the other under a certain evaluative light. Finally, while emotions are something we suffer (as the Latin name passio suggests), sentiments involve an active search for evaluative properties in the target (for this view on love and hate, see Scheler, 1973; Kolnai, 2007; Ortega y Gasset, 1988). More precisely, in hate we project disvalues on the target such as being morally depraved, evil, menacing, etc., so that the other is presented as odious; in love we project positive values such as being tender, attractive, cute, etc., so that the other is presented as lovable.
50With this development in hand, we can now specify in what respect emotions and sentiments differ in axiological terms:
IA-Axiology: Emotions and sentiments differ in the way in which they evaluatively apprehend their respective objects: while emotions are responses to values of the target, sentiments project onto the target evaluative properties congruent with the sentiment.
51These differences in the axiology of both states provide clear-cut criteria to distinguish between emotions and sentiments. Responding to the value of something is not the same as regarding this something with a particular evaluative intention.
52Let me sum up the findings of the last two sections (2.2.1. and 2.2.2.). As a result of these differences regarding the relationality and axiology of both states, according to the proposed model, emotions and sentiments exhibit different forms of affective intentionality. More specifically, according to the version of IA developed here:
IA-Affective Intentionality: While emotions are responses to values of the target, sentiments are forms of regard which project on the target values congruent with the sentiment we are experiencing.
53Note that the advantages of IA-Affective Intentionality in comparison to De Monticelli’s IA-Axiology are that it can explain the difference between emotions and sentiments without resorting to stratified models, it allows for sentiments to be not only dispositions but also occurrent episodes, and it offers a specific interpretation of the axiological dimension of sentiments in terms of projections of evaluative properties that are congruent with the sentiment in question. Accordingly, it can explain why in hate a lovable object might appear to us as odious, why in love we regard something odious as lovable, etc.
54In this paper, I have argued that the Temporal Approach (TA) that distinguishes emotions from sentiments on the basis of differences in their respective temporal structures (in terms of duration, etiology, and phenomenology) cannot offer clear-cut criteria to distinguish between both states. I have argued that the Intentional Approach (IA) offers a more promising alternative to distinguish between both states. I have discussed De Monticelli’s IA-Axiology as a current version of the Intentional Approach (IA) and presented IA-Affective Intentionality as a more plausible alternative. According to the proposed version, while emotions are responses to values of the target, sentiments are forms of regard which project on the target values congruent with the sentiment we are experiencing.
55Note that while the term “affective intentionality” has been widely employed in current research to refer to the genuine intentionality of the emotions, the way in which I am using the term here is much broader because I use it also for other affective states such as sentiments. While today’s researchers have widely argued that emotions are intentional states which present their targets as imbued with values (though the ways in which this relation has been interpreted differs from author to author: for some it is a perception of value, for others it is a response to values apprehended by other means, etc.), they usually remain silent about the affective intentionality of states other than emotions. Here I offered an account of the affective intentionality exhibited by sentiments in comparison to the affective intentionality exhibited by emotions.
56Both forms of affective intentionality might be closely related. On the one hand, being repeatedly negatively affected by someone might lead us to approach her in future encounters with a negative attitude. For instance, after repeated negative emotional experiences of indignation, anger, etc. we might come to develop a sentiment of hatred toward the other. On the other hand, the sentiment we have toward someone might make us prone to certain forms of being affected by the target. For instance, my love for a person presents this person to me under a positive light such that I am more prone to being affected by her positive values and experience joy, happiness, etc. However, both the emotion and the sentiment have to be conceptually distinguished as exhibiting two distinct forms of affective intentionality.
57Before concluding, I want to note two important implications of the proposed model. First, the proposed approach sheds light on two different ways in which affective states are linked to values. In particular, it suggests that affective intentionality is not unique for all affective states and that it comes in different forms. Two of these forms have been explored in this paper for the case of emotions and sentiments. As such, while Brentano and some contemporary authors consider affective intentionality to be something unitary, here I have suggested that we have good arguments to think that it comes in different forms.
58Second, and in much broader terms, the proposed approach offers a specific tool to elaborate a taxonomy of the affective realm. The differences between emotions and sentiments identified above suggest that though it is the hallmark of all affective states that they are linked to values, it is possible that not only do emotions and sentiments differ in their intentional structure, but also that each kind of affective state exhibits a distinct form of affective intentional structure. Why not, then, examine whether other affective states also exhibit distinctive forms of affective intentionality and elaborate on this basis a taxonomy of the affective realm? This would suppose that we analyze the affective intentional structure of moods such as euphoria and melancholy, general feelings such as languor and vigor, intentional feelings such as feeling unfairness and feeling disgust, etc., by focusing on the mode in which they are directed to and apprehend their respective targets. In my view, a version of the intentional approach for the case of moods is developed by Kriegel (2019). According to him, moods as modes of apprehension exhibit a sui generis intentionality. The idea that moods exhibit a sui generis intentionality can also be found in Ratcliffe (2013) who regards moods as “pre-intentional”. Elsewhere, I have provided an analysis of vital feelings focusing on their sui generis affective intentional structure (2021). Yet, to my knowledge, a focus on relational and axiological elements of the intentional structure of different affective states has not been employed yet to elaborate a taxonomy of the affective realm.
- 9 For an analysis of Stein’s account, see Quepons 2015.
59Early phenomenologists such as Scheler (1973) and Stein (1989) as well as more recent accounts such as that provided by De Monticelli (2006, 2020) defend a stratified model of the affective mind, according to which these phenomena belong to different strata of the person and target values of different kinds (for instance, vital feelings target vital values, emotions target aesthetic and moral values, etc.). In some of the early phenomenologists such as Stein (1989), the stratification thesis could be interpreted as suggesting that each of these layers exhibits a different form of intentional reference. For instance, Stein describes the intentionality of moods as a “background intentionality” compared to the intentionality of the emotions, which is object-directed.9 Independently of the stratification thesis, Scheler (1973) and Reinach (2017) elaborate the distinction between emotions and intentional feelings on the basis of differences in their respective intentional structure. Yet, note that the thesis of different forms of affective intentionality is not necessarily linked to stratified models in which differences between affective states are not stated in terms of differences in the intentional structure but in terms of ranks of values and degrees of involvement.
60The line of thought inaugurated in this paper opens the possibility to elaborate a taxonomy of the affective mind based on different forms of affective intentionality, rather than in terms of different temporal structures, as has been common practice in current research, or in terms of levels of sensibility related to the hierarchy of values and the degree of personal involvement, as has been usual in classical and more recent phenomenological accounts.
RSS feed
