Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action

163 T H E O R I A eISSN 0495-4548 – eISSN 2171-679X Theoria, 2022, 37(2), 163-179 https://doi.org/10.1387/theoria.22572 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action (La sonrisa de Duchenne es una acción no un mero suceso: lecciones del debate sobre las acciones expresivas) Marta Cabrera* Universitat de València ABSTRACT: In this paper, I will argue that, contrary to what is generally assumed in the debate on expressive action, we do not have good reasons to exclude facial and bodily expressions of emotion such as smiling or frowning from the category of actions. For this purpose, I will compare facial and bodily expressions of emotion with simple expressive actions, such as jumping for joy or covering one’s face in shame. I will try to show that simple expressive actions cannot be presented as actions while excluding facial and bodily expressions of emotion from this condition. My contention will then be that either both sorts of behaviour are to be identified as actions or neither is. The latter sounds rather implausible, though, as we would have to assimilate jumping for joy or covering one’s face in shame to spasms, which conflicts with the way we relate to such behaviours. My conclusion will then be that both simple expressive actions and facial and bodily expressions of emotion should be included within the category of actions, at least on the basis of the main assumptions in the current debate on expressive action. KEYWORDS: rationality; instrumentality; voluntariness; responsiveness; evaluative perspective. RESUMEN: En este artículo argumentaré que, a diferencia de lo que suele asumirse en el debate sobre las acciones expresivas, no tenemos buenas razones para excluir a las expresiones faciales y corporales de las emociones, como sonreír o fruncir el ceño, de la categoría de acciones. Para ello, compararé las expresiones faciales y corporales de las emociones con acciones expresivas simples como, por ejemplo, saltar de alegría o cubrirse avergonzado la cara. Intentaré mostrar que no puede presentarse a las acciones expresivas simples como acciones y excluir, al mismo tiempo, a las expresiones faciales y corporales de las emociones de esta categoría. Sostendré, por tanto, que, o bien ambos tipos de comportamiento son acciones, o bien ninguno lo es. Esto último, no obstante, resulta bastante problemático, ya que tendríamos que asimilar el saltar de alegría o el cubrirse avergonzado la cara a meros espasmos, lo cual entra en conflicto con la manera en la que nos relacionamos con este tipo de comportamientos. Mi conclusión será entonces que tanto las acciones expresivas simples como las expresiones faciales y corporales de las emociones han de ser incluidas en la categoría de acciones, al menos, dados los presupuestos principales del debate actual sobre acciones expresivas. PALABRAS CLAVE: racionalidad; instrumentalidad; voluntariedad; capacidad de respuesta; perspectiva evaluativa. * Correspondence to: Marta Cabrera. Department of Philosophy, Universitat de València. Avd. Blasco Ibáñez, 30 (46010València-Spain)– [email protected] – https://orcid.org/0000-0001-6702-3707 How to cite: Cabrera, Marta (2022). «Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action»; Theoria. An International Journal for Theory, History and Foundations of Science,37(2), 163-179. (https://doi.org/10.1387/theoria.22572). Received: 2021-03-05; Final version: 2022-04-25. ISSN0495-4548 - eISSN2171-679X / © 2022 UPV/EHU This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License Marta Cabrera 164 Theoria, 2022, 37/2, 163-179 In this paper, I will argue that, contrary to what is generally assumed in the debate on expressive action, we do not have good reasons to exclude facial and bodily expressions of emotion such as smiling, frowning, crying, clenching one’s fists or hanging one’s head from the category of actions.1 For this purpose, I will compare facial and bodily expressions of emotion with simple expressive actions, such as jumping for joy, covering one’s face in shame, scratching one’s head in frustration or punching the air in delight. I will try to show that simple expressive actions cannot be presented as actions while excluding facial and bodily expressions of emotion from this condition. My contention will then be that either both sorts of behaviour are to be identified as actions or neither is. The latter sounds rather implausible, though, as we would have to assimilate jumping for joy or covering one’s face in shame to spasms, which conflicts with the way we relate to such behaviours. My conclusion will then be that both simple expressive actions and facial and bodily expressions of emotion should be included within the category of actions, at least on the basis of the main assumptions in the current debate on expressive action. In section 1, I introduce the category of facial and bodily expressions of emotion, together with the reasons why participants in the debate have tended to exclude such behaviours from the category of actions. In section 2, I present the category of simple expressive actions and briefly examine the way such behaviours have been accounted for in the literature. In section 3, I compare simple expressive actions with facial and bodily expressions of emotion and argue that, on the basis of the main assumptions in this debate, either both kinds of expressive behaviour are to be identified as actions or neither is. Since the latter sounds rather implausible —given the way we approach behaviours such as jumping for joy or covering our faces in shame— I conclude that both simple expressive actions and facial and bodily expressions of emotion should be included within the category of actions. 1. Facial and bodily expressions of emotion Within the debate on expressive action, there is a wide consensus with respect to the idea that facial and bodily expressions of emotion such as smiling or frowning are not actions of an agent (Hursthouse, 1991; Goldie, 2000; Betzler, 2007; Helm, 2016; Bennett, 2016; Müller & Wong, forthcoming).2 The introduction of such category stems from the need to clarify which kinds of expressive behaviour are actions and which are mere happenings an agent 1 By “debate on expressive action” I refer to the debate initiated by Hursthouse with her paper Arational actions (1991). This debate has been variously referred to in the literature as the debate on arational actions, emotional behaviour, actions by emotion, emotional expression or expressive action (Goldie, 2000; Döring, 2003; Betzler, 2007, 2009; Helm, 2016; Bennett, 2016, 2021; Müller & Wong, forthcoming). Following Betzler and Bennett, I will talk about the debate on expressive action when I speak of such debate. 2 In Hursthouse’s terms, facial and bodily expressions of emotion are not intentional actions. Throughout this debate, participants have indistinctively talked about “actions” and “intentional actions”, and have distinguished them from what merely happens to an agent (Goldie, 2000; Betzler, 2007; Müller & Wong, forthcoming), mere evincings (Helm, 2001, 2016), reflexes (Bennett, 2016) or reactions (Bennett, 2021). In this paper, rather than the contrast between intentional and non-intentional action, I will follow the contrast between actions and happenings that appears to underly all such proposals. https://doi.org/10.1387/theoria.22572 165 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action undergoes. As it is evident, whether or not some piece of behaviour is an expression of our agency is not a trivial issue: it is one of the most important distinctions in philosophy and everyday life, for it is clear that we relate very differently to actions than to mere happenings. The clearest statement of the idea that behaviours such as smiling or frowning cannot be seen as actions may be found in Goldie’s description of the category of facial and bodily expressions of emotion:3 I turn now to those expressions of emotion which are not in any sense actions. I have in mind not only facial expressions of emotion (the smile and the frown, the contortion of the face in fear, the opening wide of the eyes in surprise), but also laughter, the flow of tears and the tremor of fear in the voice. They too, like bodily changes, can be causally explained, but they are not actions, in spite of the fact that we think of them as expressions of emotion, rather than as part of the emotion itself. In this category I am mainly interested in those involuntary bodily movements which are involved in expressing some emotions. I emphasize involuntary in order to contrast them with what we can, and sometimes do, directly try to do. We can smile in order to give the impression we are glad when we are not; we can smile when we are glad in order to show that we are; and we can smile because smiling gives us pleasure. But all these ways of smiling are to be contrasted with the genuine or “Duchenne” smile, which involves distinct muscles which we cannot directly try to move (…) we need not feel pushed towards postulation of a motive for his smiling, or of a means-end belief-desire explanation of it. A smile would be an action, explicable like this, only if it were of the sort discussed at the end of the preceding paragraph [when we smile in order to give the impression we are glad when we are not]. (Goldie, 2000, p. 34-5, emphasis added) The idea here is that facial and bodily expressions of emotion are not always performed in view of some particular end – they often simply express or reveal the emotion that someone is feeling, in which case, and for the reasons emphasised above, they cannot be regarded as actions of an agent. When speaking of this kind of behaviour then, we have in mind genuine facial and bodily expressions, and not crocodile tears or the smiles that we perform “in order to give the impression we are glad when we are not” (Goldie, 2000, p. 34) Goldie adds that we should distinguish such kind of expressive behaviour from what he calls “bodily changes”, that is, those alterations that take place when we experience emotions and that can never be regarded as actions: “autonomic nervous system responses and hormonal changes such as sweating, change of heart-rate, secretion of adrenalin and so forth, and muscular reactions such as trembling, flinching, (...) These changes just happen to us; they are not things which we do or can directly try to do” (2000, p. 26). According to him, then, both bodily changes and facial and bodily expressions of emotion are involuntary reactions (and not responses) that have to be causally explained. The only difference between these two categories seems to be that facial and bodily expressions of emotion are expressions of the emotion while bodily changes are part of the emotion itself. Goldie does not explain, however, how we are to understand such a contrast.4 3 Although Goldie does not give a name to this category, “facial and bodily expressions of emotion” seems to be a convenient label given the kind of behaviour included in such category. 4 The distinction between facial and bodily of expressions of emotion, on the one hand, and bodily changes, on the other, together with the idea that none of these kinds of behaviour are actions is fairly widespread within the debate on expressive action. Müller and Wong, for instance, distinguish between “basic bodily expressions” —smiling, frowning or dropping one’s voice— and “reflex-like Marta Cabrera 166 Theoria, 2022, 37/2, 163-179 Since our focus in this paper is on facial and bodily expressions of emotion, we will leave bodily changes out of our discussion. We may summarise the reasons for the exclusion of facial and bodily expressions of emotion from the category of actions by reference to Goldie’s first quote (2000, p. 34-5): C1. They cannot be explained by a means-end belief-desire explanation or, in other words, they are not instrumental. C2. They are causally explained, that is, they are alien to rationality. C3. They are involuntary. In light of these criteria, we may conclude that, from this view, smiling, frowning, crying, laughing, opening one’s eyes, hanging one’s head, clenching or pounding one’s fists, etc. are analogous, in a relevant sense, to spasms and reflexes: they are mere happenings or symptoms disconnected from our agency.5 That is, they are not things we do, but rather things that happen to us. This view, however, seems to raise an important dilemma within the debate on expressive action. As I will try to show, the claim that facial and bodily expressions of emotion are not actions —following C1-C3— conflicts with the claim that simple expressive actions, such as jumping for joy or covering one’s face in shame, are actions.6 As I will argue, C1emotional reactions” —recoiling in shock or disgustedly spitting out an insect— and claim that none of these kinds of behaviour are actions because they are not instrumental nor rational nor voluntary (forthcoming). Similarly, and following Hursthouse, Bennett distinguishes between “phenomena like blushing, sweating” and “forms of behaviour like smiling”. He claims that none of these kinds of behaviour are actions and that the difference between them is that the latter “often occur involuntarily but (…) it is possible to stop or to suppress [them]” (2016, p. 75), unlike in the case of blushing or sweating. This is precisely Hursthouse’s idea: she says that we often clench our fists, smile or frown without being aware of such movements —in which case they are not actions of ours— while, on some other occasions, we become aware of them while we are already performing them (1991, p. 65). In such case, she says, the thing we do is simply refraining from stopping. From this view, then, genuine smiles can never be regarded as actions – they are mere “manifestations of emotion [that] remain immune to rational explanation” (Bennett, 2016, p. 75). This is why Bennett claims that we should “distinguish the symptoms of some mental state, such as crying or smiling, from those objects (and, I claim, actions) that possess expressive power because they seem to capture or reflect some mental state (or the content thereof)” (2016, p. 84). By contrast, Helm seems to conflate facial and bodily of expressions of emotion and bodily changes in a single category, which he labels “evincings”. Within such category, we find “bits of behaviour that are simply caused by the emotion, behaviours such as changes in respiration or pulse rate or certain facial or other gestures, such as a wince or fist pounding” (2016, p. 97) as well as behaviours such as gritting one’s teeth or hanging one’s head in shame (Helm, 2001, p. 75). This kind of behaviour should be contrasted with what Helm calls “expressions of emotion”, that is, “actions that are rationally motivated by the evaluative content of the emotion”, such as jumping for joy or rumpling someone’s hair out of love (2016, p. 97). 5 My claim is not that participants in this debate hold that facial and bodily expressions of emotion are identical to spasms and reflexes, but, rather, that they grant them the same status as mere happenings. 6 Someone might wonder whether C1-C3 are necessary or sufficient conditions for a behaviour to count as an action, or whether they are separate conditions at all. I myself have no clear answer to this question, however, whether C1-C2 are necessary or sufficient conditions for agency does not affect my argument, since my claim is that bodily and facial expressions of emotion and simple expressive actions satisfy the same criteria and, therefore, should be included within the same category, regardless of whether such criteria impose necessary or sufficient conditions. https://doi.org/10.1387/theoria.22572 167 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action C3 cannot justify our classifying these two kinds of expressive behaviour differently, for it seems that: — Neither can be instrumentally explained. — Both can be rationally explained. — Both can involve muscles we cannot directly control. If the arguments that I will present in favour of such view are sound, it seems that participants in the debate on expressive action are faced with an important dilemma: either both sorts of behaviour are to be identified as actions or neither is. As I will argue in coming sections, the latter sounds rather problematic, as we would have to assimilate simple expressive actions, such as jumping for joy or covering one’s face in shame, to spasms, which conflicts with the way we relate to such behaviours. I will conclude, then, that —on the basis of the main assumptions in the current debate— both simple expressive actions and facial and bodily expressions of emotion should be included within the category of actions. Insistence on the idea that facial and bodily expressions of emotion cannot be regarded as actions, would seem to require that new reasons should be provided to help differentiate such pieces of behaviour from simple expressive actions. However, before we can compare facial and bodily expressions of emotion with simple expressive actions and argue for the idea that they should both be included within the category of actions, we must examine in some detail what simple expressive actions are and the way they have been accounted for in the debate on expressive action. 2. Simple expressive actions At first sight, jumping for joy seems like one of the simplest things we can do. Compared to actions such as cooking a meal, writing a novel or building a bridge, happily jumping up and down when receiving good news seems to be an action that does not require from us a too complex analysis. The same goes for something apparently as simple as covering one’s face in shame —when remembering, for instance, a particularly shameful moment— or scratching one’s head in frustration.7 Such behaviours, however, as Hursthouse (1991) famously noted, cannot be regarded as actions by the lights of the standard account of agency. According to such account, (1) a behaviour counts as an action if it can be rationalised, and (2) to rationalise an action is to present it as something instrumentally valuable from the perspective of attaining some end. That is, if A is an action performed by subject S, we should be able to explain A by offering a desire-belief pair (D-B) such that satisfying desire D is the aim of S and S has the belief B according to which A allows her to satisfy D.8 Running away from a predator in fear, for instance, clearly admits of a rationalisation of this kind and, consequently, can be considered an action: the subject has the desire to protect herself and the belief that running away 7 We are considering cases in which an agent genuinely jumps for joy, covers her face in shame or scratches her head in frustration. In such cases, the agent does not perform such behaviours in view of some end – she does not jump, for instance, in order to communicate her happiness; she just joyfully jumps at some good news she has received, for instance. 8 See Davidson (1963). Marta Cabrera 168 Theoria, 2022, 37/2, 163-179 is the best means of satisfying such desire. If we take this view, therefore, we cannot regard jumping for joy or covering our faces in shame as actions: when we jump for joy, we are not trying to instrumentally satisfy an end through our jumping. Likewise, when we cover our faces in shame, we are not trying to instrumentally satisfy an end through our covering our faces. According to this, and against our initial intuition, then, it does not look like there is something we do in such cases – such behaviours are mere happenings we undergo. The standard account of agency, though, was put into question by Hursthouse, who defended the idea that there is a class of actions —arational actions— which are clearly actions despite not being instrumentally rationalisable.9 This would be the case of jumping for joy, covering one’s face in shame, rumpling the hair of a loved one out of affection, rolling in grief in the clothes of a loved one who has died, destroying things in a fit of rage, gouging out the eyes of a rival in a picture out of hatred, etc. In all these cases, we are unable to explain what the agent does by reference to a belief-desire pair that shows her action as something instrumentally valuable. According to Hursthouse, these actions can only be explained by saying that “in the grip of the emotion, the agent just felt like doing them” (1991, p. 61). From this view then, we can admit that some of the things we do are disconnected from instrumental rationality —and are, hence, arational— but are still actions of ours. If this is correct, it seems that the standard account of agency is in trouble and that our intuition that jumping for joy and covering one’s face in shame are actions can be preserved. Many have tried, nonetheless, to show that the standard account is immune to Hursthouse’s attack because the class of actions that she presents —or, at least, most of the actions she presents— can be rationalised after all (Smith, 1998; Goldie, 2000; Helm, 2001, 2016; Betzler, 2007, 2009; Döring, 2003; Scarantino& Nielsen, 2015; Bennett, 2016, 2021; Müller & Wong, forthcoming). The shared thought here seems to be that, if some piece of behaviour can be regarded as an action, then such behaviour must be connected to rationality in some sense. From this point of view, then, the idea that there is something like arational actions cannot be correct. Before examining the way in which some such views have accounted for the behaviours we are interested in —jumping for joy, covering our faces in shame, etc.— we should briefly sketch two crucial features of how the debate has evolved since Hursthouse’s seminal paper. Firstly, the current debate has produced a number of distinctions within the class of so-called arational actions. Although not everyone explicitly agrees in this respect, the common assumption is that not all cases in Hursthouse’s list belong to the same kind (Goldie, 2000; Betzler, 2007, 2009; Scarantino& Nielsen, 2015; Pineda, 2019; Müller& Wong, forthcoming) and should be divided into three types: symbolically displaced actions, radically displaced actions and simple expressive actions.10 The cases we are inter9 In Hursthouse terms, such actions are clearly intentional actions. 10 Symbolically and radically displaced actions are actions in which the object of an emotion is displaced to a different object that emerges as the explicit object of the action. In the case of symbolically displaced actions, “there is some relation between object of emotion and object of expressive action onto which the agent’s mind might latch” (Goldie, 2000, p. 31). This is what happens when one rolls in grief in the clothes of a loved one who has died or when gouging out the eyes of a rival in a picture out of hatred. Since the object of the agent’s emotion is not available for her to act on it —in the first case, because the loved one has passed away and, in the second, because there are social conventions discour- https://doi.org/10.1387/theoria.22572 169 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action ested in —jumping for joy, covering one’s face in shame, punching the air in delight or scratching one’s head in frustration— belong to the latter type. Secondly, attempts to rebut Hursthouse’s case against the standard account have mainly focused on symbolically displaced actions.11 As a result of this, simple expressive actions and radically displaced actions have gone under-explored. In fact, simple expressive actions tend to be presented as residual, problematic or even borderline cases (Goldie, 2000; Betzler, 2007, 2009; Bennett, 2016). The usual strategy has been to show that symbolically displaced actions do have a purpose and, therefore, that we can easily rationalise them in instrumental terms.12 Hence, such kind of behaviours are not arational actions – they are actions perfectly explained by the standard account. Furthermore, some of the participants in the debate seem to be convinced that this line of reasoning not only applies to symbolically displaced actions, but to all actions included in Hursthouse’s list and, therefore, that her case against the standard account of agency fails. This last strategy, however, seems to have devastating consequences for simple expressive actions, since instrumental explanations of behaviours such as jumping for joy or covering one’s face in shame clearly over-intellectualise them. Bennett argues, for instance, that so-called arational actions do have “a purpose —that of doing justice, or giving adequate external form to one’s sense of the situation” (2016, p. 74, emphasis added)— which means that they can be instrumentally rationalised – and that even the “spontaneous case [jumping for joy] is susceptible of a ‘reading’ that might show why it would be appropriate to select [such expressive behaviour] as a fitting vehicle for that emotion if (…) one is in the business of deliberatively selecting a vehicle for one’s emotion” (2016, p. 91-2). The problem is that it does not look like we deliberatively select a vehicle for our joy when we jump for joy. Smith (1998), on the other hand, does not mention examples of simple expressive actions but assumes that his proposal can explain away all cases presented by Hursthouse. aging us from gouging people’s eyes out— her action takes a different object that bears some symbolic relation with the object of her emotion. There is also a displacement of the object of one’s emotion in radically displaced actions but such displacement involves no symbolic identification. The connection between the object of one’s emotion and the object of the expressive action is arbitrary, as when one destroys an object in a fit of rage or kicks a chair out of anger. In such cases, it seems that any other object at hand —a table, a bag, etc.— could have just as well been the object of one’s action. Finally, simple expressive actions are those actions where neither the object of one’s emotion is displaced nor a symbolic identification is at issue. These are cases such as jumping for joy or hiding one’s face out of fear or in shame. I have borrowed the terminology “symbolically displaced action” and “radically displaced action” from Scarantino and Nielsen (2015) and “simple expressive action” from Pineda (2019). 11 Gouging out the eyes of a rival in a picture out of hatred (Hursthouse, 1991; Goldie, 2000; Döring, 2003), rolling in grief in the clothes of a loved one who has died (Smith, 1998) and carrying one’s grandfather’s coffin during his funeral out of love (Bennett, 2016), for instance. An exception to the general tendency can be found in Helm (2001, 2016) and Müller and Wong (forthcoming). 12 Not everyone agrees that the mental states that rationalise the agent’s behaviour are beliefs and desires. Döring (2003) and Scarantino and Nielsen (2015), for instance, argue that the end of the agent’s action is provided by the emotion she experiences and not by a desire that she may have. It will be enough for our discussion, though, that all of them defend that symbolically displaced actions have a means-end structure in virtue of which such actions can be rationalised and adequately accounted for as actions by the standard account. See Bennett (2021) for an exception to this tendency. Marta Cabrera 170 Theoria, 2022, 37/2, 163-179 According to him, when a man rolls in grief in the clothes of his dead wife “the man is doing what he is doing because he desires to roll around in his dead wife’s clothes and believes that he can do so by doing just what he is doing: that is, by rolling around in those particular clothes that he is rolling around in” (1998, p. 22). Smith adds that in order to understand the desire of the agent, we must make reference to his emotion, but that the belief-desire pair suffices to rationalise his action. If we turn to the case of jumping for joy, however, it does not seem that one jumps for joy when receiving good news because one desires to jump and believes that one can do so by just doing what one is doing, that is, jumping. Such an explanation does not look like a good description of what goes on in this case.13 Given these difficulties, Goldie (2000) and Betzler (2007, 2009) have speculated that, contrary to our intuitions, simple expressive actions might not be actions at all. Goldie says that “jumping for joy, scratching your head in frustration, and punching the air in delight (…) [are] things which one does [that] are unlike the genuine smile in that the latter involves a movement of certain muscles which one cannot directly try to move. But still, surely a genuine spontaneous jump for joy […] no more involves a belief than does the genuine spontaneous smile” (2000, p. 36). In a similar line, and after concluding her alternative account of so-called arational actions, Betzler wonders whether her ideas could be extended to simpler kinds of behaviour: In contrast, hiding one’s face from fear, blushing in shame, jumping for joy, or posturing in front of adversaries are cases that might be thought to be intelligible without assuming any particular point of view of the respective agent. Some of these expressive actions may be in the vicinity of mere bodily movements. Others can be explained simply by referring to the instinct of survival that all agents share as members of a species. A concession in such cases, however, is compatible with the rationalization applying in many other cases. (Betzler, 2009, p. 284-5, emphasis added) These ideas, as I noted, are merely speculative and stem from the view that what we have been calling simple expressive actions are residual and problematic cases. Assimilating such behaviours to mere happenings, though, seems to have some rather implausible consequences that should be properly addressed by anyone who wants to defend this view. In particular, if jumping for joy or covering one’s face in shame were nothing more than mere happenings —or, in Betzler’s terms, mere bodily movements— we would need an explanation of the way in which we usually approach such behaviours. When we see that someone who has just received some news starts jumping up and down, we immediately understand that the news she has received is good – or, at least, that it is good from the agent’s point of view. This means that we approach her behaviour as a response to something that she evaluates as worth celebrating. Her response, moreover, can be assessed as appropriate or inappropriate, or as proportional or disproportional. In fact, if jumping were, for whatever reasons, inappropriate in the given circumstances, the subject could be blamed for not having waited for a more convenient time to express her happiness. As we can see, though, this is not the way we relate to mere happenings —which are disconnected from our agency— 13 Smith’s proposal has, in fact, been widely rejected as a solution to Hursthouse’s problem since it does not seem to provide a good rationalisation of any of the cases presented in her list (Goldie, 2000; Döring, 2003; Betzler, 2007; Bennett, 2016; Pineda, 2019; Müller & Wong, forthcoming). https://doi.org/10.1387/theoria.22572 171 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action and therefore Goldie’s and Betzler’s suggestions do not seem to provide, in principle, a good explanation of these cases. To recapitulate, we started by saying that —compared to actions such as cooking a meal, writing a novel or building a bridge— jumping for joy, covering one’s face in shame, punching the air in delight or scratching one’s head in frustration seem to be among the simplest actions we can perform. This intuition, however, seems to be questioned by the standard account of agency, according to which, behaviour that cannot be instrumentally rationalised cannot be regarded as the actions of an agent. Hursthouse’s critique of the standard account, however, allows us to regard such cases as actions, in particular as actions that are disconnected from rationality and that can only be explained by saying that “in the grip of the emotion, the agent just felt like doing them” (1991, p. 61). This last idea is nonetheless rejected by many participants in the debate who try to show that the cases presented by Hursthouse can be rationalised and thus be considered actions by the lights of the standard account. As we have seen, however, instrumental explanations of simple expressive actions tend to over-intellectualise them, therefore explanations of this kind do not succeed in showing that jumping for joy or covering one’s face in shame are not arational actions. Alternatively, a defendant of instrumental rationality can question whether such behaviours are actions at all —since they do not involve any meansend belief-desire pair— but this leads us to a very difficult position: jumping for joy or covering one’s face in shame are assimilated to spasms and reflexes, which conflicts with the way we deal with such behaviours. It appears, then, that simple expressive actions are still problematic cases for the standard account: we have the intuition that they are actions but they seem to be disconnected from rationality – or, at least, from the kind of rationality presupposed in the discussion so far. As Helm (2001, 2016) and Müller and Wong (forthcoming) have convincingly argued, all previous views assume that rational action is necessarily instrumentally rational action. However, they claim, an alternative conception of rationality is needed if we are to explain the sense in which some expressive behaviours are not a matter of mere accident:14 We should not think that all intentional action is instrumental, done for the sake of achieving some end, for all that’s needed for an action to be rationally explicable is that it has a point the explanation reveals to be worthwhile (…) the point of jumping for joy is that it is a celebration, and in feeling joy one feels its target to be a good worthy of celebration (…) kissing someone or rumpling his hair or, depending on the circumstances, holding him or even just sitting quietly with him all have a point in that they are all ways of being solicitous, and in feeling occurrent love one feels someone as worthy of solicitude. And tearing one’s hair or clothes has a point insofar as these are all ways of mourning, and to feel grief is to feel a loss to be worthy of mourning. In each of these cases, the emotional expressions are not means undertaken to achieve the end of celebration, solicitude or mourning; they are the celebration, solicitude and mourning. (Helm, 2016, p. 98, emphasis added) The rejection of instrumentality as a criterion for agency, then, allows us to see a way in which simple expressive actions might be connected to rationality – and are, therefore, not 14 See also Bennett (2021) for a non-instrumental explanation of expressive action. In said paper, however, he exclusively focuses on cases of symbolic action, such as Achilles’ dragging Hector’s body around the walls of Troy or the kneeling of the believer in church. Marta Cabrera 178 Theoria, 2022, 37/2, 163-179 4. Conclusions I started by introducing the category of facial and bodily expressions of emotion, as well as the reasons that several philosophers have provided in favour of their exclusion from the category of actions. As we have seen, in the current debate on expressive action, behaviours such as smiling, frowning or crying are regarded as mere happenings we undergo because they are thought to be neither instrumental nor rational nor voluntary. I then presented the category of simple expressive actions and the various ways in which they have been accounted for in the literature. The main claim I have argued for —through A1-A3— is that simple expressive actions cannot be presented as actions while excluding facial and bodily expressions of emotion from this condition: neither of these two kinds of expressive behaviour can be instrumentally explained, both can be rationally explained and both can involve muscles we cannot directly control. As a result, I have concluded that, on the basis of the main assumptions in the current debate on expressive actions, simple expressive actions and facial and bodily expressions of emotion cannot be classified differently: either both sorts of behaviour are to be identified as actions or neither is. Classifying behaviours such as jumping for joy or covering one’s face in shame as happenings, however, conflicts with the way we relate to such behaviours. Therefore, I have concluded that both simple expressive actions and facial and bodily expressions of emotion should be included within the category of actions. Acknowledgments I am very grateful to Josep Corbí, Jordi Valor, Christopher Bennett, Jonathan Mitchell, Pia Campeggiani, Carlos Moya and two anonymous referees for their valuable comments on earlier versions of this paper. I am also indebted to the audiences in the Valencia Colloquium in Philosophy and the 2nd VLC Philosophy Workshop: Sharing Reasons. This work has been supported by the research projects PID2019-106420GA-I00 and PID2020-119588GB-I00 and by the grant BES-2017-081537 funded by MCIN/ AEI/10.13039/501100011033 and “ESF Investing in your future”. REFERENCES Bennett, C. (2016). Expressive actions. In C. Abell & J. Smith (Eds.). Emotional expression: philosophical, psychological and legal perspectives (pp. 73-94). Cambridge: Cambridge University Press. Bennett, C. (2021). The problem of expressive action. Philosophy,96(2), 277-300. Berlin, I. (1958). The crooked timber of humanity. Princeton: Princeton University Press. Betzler, M. (2007). Making sense of actions expressing emotions. Dialectica, 61(3), 447-466. Betzler, M. (2009). Expressive actions. Inquiry, 52(3), 272-292. Corbí, J. E. (2012). Morality, self-knowledge and human suffering: an essay on the loss of confidence in the world. New York: Routledge. Davidson, D. (1963). Actions, reasons and causes. Journal of Philosophy,60(23), 685-700. Döring, S. (2003). Explaining action by emotion. The Philosophical Quarterly, 53(211), 214-230. Goldie, P. (2000). Explaining expressions of emotion. Mind, 109, 25-38. https://doi.org/10.1387/theoria.22572 179 Duchenne smiles are actions not mere happenings: lessons from the debate on expressive action Helm, B. (2001). Emotional reason: deliberation, motivation and the nature of value. Cambridge: Cambridge University Press. Helm, B. (2016). Emotional expression, commitment and joint value. In C. Abell & J. Smith (Eds.). Emotional expression: philosophical, psychological and legal perspectives (pp. 95-114). Cambridge: Cambridge University Press. Hursthouse, R. (1991). Arational actions. Journal of Philosophy, 88, 57-68. Hustvedt, S. (2011). The summer without men. London: Picador. Müller, M. J. & Wong, H. Y. (forthcoming). Which emotional behaviors are actions?. In A. Scarantino (Ed.). Routledge handbook of emotion theory. London/New York: Routledge. Pineda, D. (2019). Sobre las emociones. Madrid: Cátedra. Scarantino, A. & Nielsen, M. (2015). Voodoo dolls and angry lions: how emotions explain arational actions. Philosophical Studies, 172, 2975-2998. Scruton, R. (1986). Sexual desire: a philosophical investigation. London: Weidenfeld and Nicholson. Smith, M. (1998). The possibility of philosophy of action. In J. Bransen (Ed.). Human action, deliberation and causation (pp. 17-41). Dodrecht: Kluwer. Strawson, P. F. (1962). Freedom and resentment. Proceedings of the British Academy, 48, 187-211. Williams, B. (1981). Moral luck. Cambridge: Cambridge University Press. Marta Cabrera is a PhD candidate at the Universitat de València. Her main areas of interest are philosophy of emotions, philosophy of mind and action theory. address: Department of Philosophy, Universitat de València. Avd. Blasco Ibáñez, 30 (46010 València-Spain). Email: [email protected]. ORCID: 0000-0001-6702-3707