Thumos Seminar 2025

The Thumos seminar, which is the main research activity of Thumos, the Genevan Research Group on Emotions, Values and Norms, takes place on Thursdays, 16:15-17:45 at Uni Philosophes (room PHIL 116). The schedule can be found here. The archives of the Thumos seminar are available here. 

Robots don’t care

With the advent of AI, questions concerning the implementation of emotion in artificial agents such as social robots and chatbots acquire renewed interest. Such questions are typically raised in either one of two separate contexts: (a) the question of morally aligned behaviour in artificial agents, or artificial morality; and (b) the question of convenience of interaction with artificial agents. In this paper, I focus on (a) and on the idea that it is important to develop (artificial) emotion in artificial agents so as to ensure their proper functioning as moral agents, a claim relying on some form of moral judgement sentimentalism. I argue against this claim in a variety of ways, all of which in the end relying on the claim that emotions hinge on the capacity to care or be attached to things, which artificial agents lack. In the process, I show that emotions are involved in our moral practice and moral language in ways hitherto neglected and inappropriate for “heartless” robots.

Sentimental Value: Definition and Axiological Perspectives

Sentimental value has largely been overlooked by philosophers. Perhaps they’ve never wondered why their homes are filled with so many objects they cannot bring themselves to throw away. These cherished trinkets are typically considered the primary bearers of sentimental value. In this presentation, I will first examine the few existing conceptions of the subject (Fletcher 2009, Rønnow-Rasmussen 2022), exploring questions such as: Is sentimental value always tied to our past? Does it necessarily involve having felt or feeling sentiments towards a person? Secondly, the discussion will open the door to broader axiological considerations. Assuming the existence of extrinsic final value, I will argue that it is not so clear that sentimental value falls into this category. It raises, however, interesting questions about personal value more generally. In particular, I will show that it can help us understand the attitude of valuing, which seems widely involved when it comes to sentimental items.

Fiction and the Flow of Information

The popular view according to which fictions are prescriptions to imagine is challenged by the request to specify which is the specific imaginative response that fiction prescribes. It has been argued that such response consists in imagining fictional worlds. This paper argues that such response rather consists in imagining gathering information flowing from fictional worlds, which are to be understood as imaginary spatiotemporal systems. It is also argued that gathering information flowing from a fictional world requires imagining having a place in a special region of that world, namely, the narrative periphery.

Aesthetic Discovery and Seeking to Understand

A cornerstone of the noetic conception of aesthetic experience is the claim that such experience is first and foremost a kind of explorative thought-process guided by the object of attention’s aesthetic character. 1 On this approach, aesthetic value gives us
reason to engage in a series of contemplative and reflective processes during which we rely not only on our perceptual, emotional and imaginative abilities, but also on our capacities for sense-making, problem-solving, deciphering, hypothesizing and theory-building. Aesthetic experience is conceived in terms of a seeking to understand whereby, through a process characterised here as a form of discovery, we come to grasp how various aspects of the world are connected to one another, see some phenomenon or character in a new light, or find a novel way of encountering our environment.

A distinction key to what I shall refer to as the ‘learning from art’ model will be drawn, namely between understanding the object of aesthetic appreciation itself (or learning about), and understanding that which the object in question addresses, represents or sets out to convey (or learning through). Although important, this distinction will be held to be too narrow to fully explain the role and scope of
understanding in connection with aesthetic experience in general. An alternative way of accounting for aesthetic understanding is offered, building on the idea that the epistemic benefits in question are not so much the consequences of specific aesthetic encounters with art gained in the aftermath of our aesthetic experience of them, as inherently built into what aesthetic experience is. To support this claim, we will lean on some German Enlightenment philosophers’ attempts to establish the underlying principles of the ars inveniendi, and how they defined the role of wit or ‘thought skills’ in aesthetic experience. The idea that aesthetic experience is both epistemically motivating and epistemically creative 2 will be unpacked.

From Recalcitrant Emotions To Response-Dependent Values

Despite extensive discussion, the puzzle of recalcitrant emotions—emotions felt despite conflicting judgments—has not yet been fully resolved. The literature on emotions has been preoccupied with a different puzzle: how recalcitrant emotions can be irrational without being contradictory. However, this normative challenge takes the very existence of recalcitrant emotions for granted, shifting the focus from how recalcitrant emotions are possible to determining whether existing theories of emotions can adequately account for their precise form or degree of irrationality. This neglect is particularly striking when compared to the long-standing debate on weakness of the will, where the very existence of the phenomenon has always been a central point of contention.

This paper has three aims. First, it presents a novel challenge regarding the possibility of recalcitrant emotions that is even more pressing than that of weakness of the will (Section 1). Second, it demonstrates that theories of emotions that take them to be either experiences of or responses to values fail to adequately address this challenge (Section 2). Third, it argues that properly addressing this challenge implies that values are response-dependent properties, thus, emotions are both experiences of and responses to values, and that evaluative knowledge, at its best, is emotional knowledge: I know perfectly well that the decision is just, but I fail to internalize it (Section 3).

Fear of Missing Out

What is the fear of missing out (FOMO)? The simple answer, in keeping with its name, is that it is a kind of fear. In particular, it is a kind of fear which represents its object as absent; we feel FOMO about not having such-and-such experience. But this raises a problem. Philosophers understand fear as representing objects as dangerous, but it is not at all obvious how the absence of experience poses a threat to the subject. I will consider two solutions to this problem which align FOMO with other instances of fear that flout this paradigm. Namely, I consider whether FOMO is a case of objectless fear or recalcitrant fear but find that neither solution adequately captures the phenomenon. Instead, I argue that we can reconcile FOMO and dangerousness – FOMO involves a representation of some absent experience as posing a threat to one’s social connections. 

Attention, Memory, and ‘Reliving Emotions’

Say that one is remembering having listened to the opening of the Adagietto in Mahler’s 5th symphony yesterday evening. What is one attending to when one remembers in this way? One natural answer is that one is attending something in the past – something which has the temporal qualifier ‘past’ predicated of it. However, this needs disambiguating. Is one attending to something <the Adagietto in Mahler’s 5th symphony> as a past event not currently given in auditory perception or is one attending to one’s past auditory experience of <the Adagietto in Mahler’s 5th symphony>. If there is pressure to answer in the second way, then it seems attention to things past – prototypical first-person memory experience – has the intentional structure of attention to our experience of things in the past (see Peacocke 1985; Martin 2001). As such, we don’t, at least in ‘quasi-perceptual’ or imagistically driven episodic ‘autobiographical’ memory, directly remember the past, but rather in attending to things past, past events say, we have to in some respect or other relive the past by attending to our experience of things past.

In this paper I argue that any plausible account of attentive-remembrance – of autobiographical memory conducted in the mode of attention or advertence to its intentional object – needs to respect this condition, which I call the reliving condition. However, there are a number of challenges that are faced in accepting it, despite its initial plausibility. First it seems to make the intentionality of attentive-memory fundamentally experience-directed rather than world-directed, such that our past experiences serve as mediators for our contact with past events. Accepting the condition also faces the challenge of explaining common-place ascriptions of misremembering: when I accuse someone of misremembering an argument we had last week am I accusing them of getting facts about the non-experiential world wrong (say the relevant steps that were taken in the argument), or I am accusing them of failing to accurately re-present to themselves a past experience of theirs. The condition also gives rise to certain puzzles concerning the way attention in deployed in memory: can I fail to pay attention to details of my experience of some past event in the same way as I can fail to pay attention to details of the past event itself, and can misremembering be ‘corrected’ by attending more carefully not just to some past event but my ‘witnessing’ of it. Here I suggest that a properly fleshed-out phenomenological account of attentive-memory can meet these worries while respecting the reliving condition. 

Once this is established, I explore a further complex case, namely one in which in emotion comes into the picture. Say that one is not merely remembering having listened to the opening of the Adagietto in Mahler’s 5th symphony yesterday evening, but is also in some sense remembering one’s affective response to the music, say one’s being overcome with a kind of aesthetic admiration for the grace and beauty of the music, as well as its tragic overtones. A number of further questions arise in this context: in such cases do we direct attention towards the past emotion as well as the past perceptual experience? Can we somehow focus on both our past emotion and the relevant past event perception simultaneously? Further to this, what kind of emotional phenomenology is involved in a case like this; do we in some respect undergo the emotion again but in a modified way? Here I explore various options for theorizing about these kinds of cases, arguing that at least some of the time we relive our emotions.

Humean constructivism and the authoritative ought

What makes it true that we authoritatively ought to perform an action? I examine Humean constructivist answers to that question, according to which what makes it true that we authoritatively ought to perform an action is that we would, were we to reflect properly, have a certain kind of positive response toward performing the action. I distinguish between two kinds of Humean constructivist views: substantivism and formalism. Substantivists believe that proper reflection will lead all of us to the same substantive practical principles—to principles with content, to principles that prescribe particular types of action. Formalists deny that we are warranted in thinking that proper reflection will lead all of us to the same substantive principles; ccording to formalists, we can identify the form of authoritative oughts, but we cannot identify the authoritative ought with any particular content. I argue for the formalist version of Humean constructivism.

The plurality of the values of amusement

Like any emotion, amusement has an important link with a value (Deonna and Teroni, 2012), known as the comic or amusing. However something can be laughable without being perceived as comic, but rather ridiculous. An important problem then seems to emerge: the comic and the ridiculous cannot fall under the same type of thick value, because they do not refer to the same type of thin value (Willliams, 2006). Indeed, comic seems to refer to a positive value, while ridiculous is understood more as a negative value. Amusement thus seems to refer to at least two different values, which threatens its unity as an emotion: what could be the standards of correctness of amusement if it can refer to two opposite values? Similarly, the attitudes implied by these values seem to be different, in the sense that they do not involve the same kind of action. When we find something amusing, we tend to joke about it, whereas when we find something ridiculous, we tend to adopt an attitude of mockery. So we will ask how we can understand the link between amusement, these values and these attitudes. The aim is to show that, despite their disparity, they are part of the same emotion, and can be understood without threatening the unity of amusement, since they always refer to a way of doing humor.

Valuings as Sentiments

We are valuing beings, beings who possess the capacity to value things. But what is it ‘to value’ something? The most common accounts in the literature hold that to value an item is either to have a first-order or a second-order desire towards it; or to believe that item to be valuable; or to care about that item; or to have a combination of all these mental states. In our paper, we raise some objections against all these accounts and defend a new affective account of valuings. Unlike standard affective accounts, according to which the term ‘valuing’ refers to a single type of affective state, such as care, we hold that ‘valuing’ refers to the members of a class of affective states, namely, the class of sentiments. On our view, to value something is to have a particular sentiment towards it. Since sentiments can be of different types, our account implies that there are as many ways of valuing things as there are types of sentiments.

Morality and the Value of Fitting Emotion

How do we determine if an emotion is fitting? The standard way involves determining the correctness conditions of an emotion and then seeing if those conditions are instantiated. The problem with this methodology is that it struggles to distinguish between formally normative and authoritatively normative standards of fitting emotion. In this paper, I argue that our reflection on the fittingness of emotion is incomplete if we do not consider the value of fitting emotion. Critically expanding on Rowland’s (2022) recent work, I suggest that we can explain which emotional standards are authoritatively normative by reflecting on the non-instrumental value of our emotions. An important and pervasive source of the non-instrumental value of emotion, I argue, consists in the constitutive role they play in participatory practices. My proposal also has significant upshot: it makes room for morality to determine authoritative standards of fittingness.

TBA