Letting the body find its way: skills, expertise, and Bodily Reflection

What forms of consciousness can the subject have of her body in action? This is a recurrent issue in contemporary research on skilled movement and expertise, and according to a widespread view, the body makes itself inconspicuous in performance in favour of the object or goal that the activity is directed to. However, this attitude to consciousness in bodily performance seems unsatisfying for an understanding of skilled action, and the work of several researchers can be seen as responding to this view: Montero, Legrand, Ravn and others in the philosophy of expertise and of dance have developed various notions of consciousness and cognition to account for the mindful processes at play in performance. Two related questions can be distinguished here: (1) Is there an inherent conflict between skilled action and at least more than marginal awareness of that action, or is it possible – and even desirable – to reflect on our own performance without considerably impeding on it? (2) What forms of consciousness pertaining to the body in action must we distinguish in order to answer the first question? This paper gives an overview of this discussion, focusing on the second issue, although the first will come into play in so far as it is linked with the latter question. Drawing on Merleau-Ponty’s analyses of bodily reflection and on dancers’ descriptions, I show that there is, in phenomenological terms, a bodily level of reflection: a fully conscious and exploratory activity that is led by the skilled body, and that is explicitly aimed at by many performers.

I briefly present some other theories of cognition in skilled performance, such as the Mesh account, and then examine Legrand and Ravn's analyses of the various levels of pre-reflective experience of the lived body. Although they acknowledge the possibility of an embodied reflection, I argue that a further level of bodily reflection is needed to account for the fundamental and often strived after experience of letting the body lead that many body experts -especially in the arts -give voice to. Drawing on Merleau-Ponty's analyses of the "kind of reflection" performed by the body (1945,1964) and on dancers' and other performers' accounts, I outline a notion of bodily reflection, that I believe can give an important contribution to the debate about the role of cognitive processes in skilled action.

Absorbed coping versus mindedness: Dreyfus
One leading representative of the idea that there is an opposition between expertise and mindedness is Hubert Dreyfus. On his theory, there are five qualitatively different stages of skill aquisition: Novice, Advanced Beginner, Competence, Proficiency and Expertise. 2 While the Novice draws on conscious rules abstracted from the context, the Advanced Beginner relies both on context-free rules and situational elements. The more experience the learner gets of "coping" with real situations, the more she becomes involved in the tasks, and at the level of Competence, the number of rules that she must consider in a "detached manner" -since they are not yet integrated in her behaviour -can be overwhelming (Dreyfus & Dreyfus 1986, 23). When the performer attains the level of Proficiency, she has becomed profoundly involved in the situation of the task and relies on "intuition" rather than context-free rules; 3 occasionally, however, she still has to think "analytically about what to do" (Dreyfus & Dreyfus 1986, 29), until she finally reaches the level of Expertise where the response is wholly intuitive and no deliberation occurs. In other words, there are various forms of cognitive relations -calculation, deliberation, reflection, monitoring, etc. -to our actions before the highest level is attained. At the Expertise level, however, we are nonminded or "mindless" on Dreyfus's story: absorbed in what we are doing, intuitively responding to the sollicitations the situation affords us. 4 On the lower levels, then, there are still details that the non-expert needs to consciously focus on -the performance is perhaps fluid in certain respects and in determined situations where the challenge is less important, as when the advanced beginner is driving on a road with little traffic on a sunny summer day. At higher levels of skill acquisition, the performer relies less on conscious calculation of alter-2 At first formulated together with his brother, Stuart Dreyfus, in the 1980s, and in some later papers. For reasons of simplicity, I will refer to only one Dreyfus, Hubert, who has continued to elaborate this account in many other works. For details on the five stages, see Dreyfus & Dreyfus 1986, 19 f. and e.g. Dreyfus 2002, 368 f. 3 On the Dreyfusian account, "intuition" is synonymous with know-how, and involves a "holistic similarity recognition" (Dreyfus & Dreyfus 1986, 28). 4 Dreyfus 2007a, 353, 356;, 38. In 2007b, Dreyfus makes a further distinction within the expertise level, between absorbed and involved coping: "absorbed coping … is involved coping at its best". natives and more on intuition, and at the level of expertise, the response is an "immediate intuitive situational" one, according to Dreyfus (2002, 372). It is only when this "absorbed, skillful coping" (Dreyfus 2002, 378) is for some reason interrupted that an expert starts to pay attention, or, conversely, paying attention makes things go wrong even for an expert (Dreyfus 2007b, 377;2013, 22 Box 1.1, 29). It is important to note that there is a qualitative difference between the levels of skill acquisition, that also involves affective transformations -from experiencing frustration, for exemple, to enjoyment of the smooth performance (Selinger & Crease 2002, 254). Thus, it cannot be assumed that the rules applied at lower levels -or the rules experts appeal to in breakdown cases -are simply the same as those used at the higher levels, only "implicitly" (Dreyfus 1992, xxiii). On the qualitatively higher stages, where expertise is more or less attained, the performer relies on a deeply situational understanding that is broken if she resorts to the conscious forms of reflection and calculation that are needed at the lower levels.
This step-by-step ascent from various degrees of concentration on rules and advice, of mistakes and perhaps jerky behaviour to a finally more fluid performance, is certainly recognisable by anyone who has tried to learn a new skill. It must also be remembered that the Dreyfusian view of expertise as embodied and the emphasis on phenomenological methods were highly controversial at the time it was first formulated (Selinger & Crease 2002, 272;Boden 2006, 838 f.). Yet, there are several problems with Dreyfus's theory that have been discussed in the literature. 5 Not least the Wittgensteinian allegation that at the expert level, "[w]hat must be done, simply is done" (Dreyfus 2002, 372), or that the judgments the expert makes are based on former experience "in a manner that defies explanation" (Dreyfus & Dreyfus 1986, 36), are somewhat frustrating for someone who wants to understand what is at stake in skilled performance.

Montero: the thinking Expert
One characteristic of Dreyfus's theory of skilled behaviour is the presentation of expertise as a sort of platform that can be attained (at least if we have enough talent), as a final stage where the skill is so much part of us that we often are not even aware of it, and where we are disturbed mainly by external or internal accidents: "things that go wrong" in the world or in our mind. This assumption has been challenged by among others philosopher and dancer Barbara Montero, who sees Dreyfus as an extreme representative of the view -the "myth" -that expert action is automatic, effortless and that mental processes interfere with expert performance (Montero 2013, 307;2016, 35). 6 Accordingly, the expert is someone who, through assiduous practice, has attained a level of proficiency where performance is automatic and "nonthinking" (Dreyfus 2013), "mindless" (Dreyfus 2007a, 353; italics in text). 7 5 Most famously, perhaps, in the debate with John McDowell, see Schear (ed.) 2013. 6 Montero terms Dreyfus's view an "ultra-extreme" one (2016, 36). 7 Komarine Romdenh-Romluc has likewise criticised Dreyfus for not acknowledging that "thought plays an ongoing role in guiding action" (2012, 205), but she does not take issue with the idea that thinking In her monograph, Montero calls this assumption the "just-do-it principle" (Montero 2016): 8 a widespread myth about expert action implying that in great performances of bodily skills the agent does not and should not focus on, attend to or monitor what she is doing. This principle is both descriptive and normative since it supposedly makes claims as to what experts actually do when they perform optimally and what they should do if they want to perform optimally and avoid making mistakes. In Montero's overview of the literature on expertise, a number of mental processes are mentioned depending on the theory concerned: "self-reflective thinking, planning, predicting … conscious control, trying, effort", and more (Montero 2016, 35). She is not merely arguing against the extreme forms of the principle, that exclude all mental processes, but also, for example, those that proscribe only monitoring or trying.
Montero outlines, instead, her own principle: "cognition-in-action". By this, she posits that conscious effort, focus and control are integral to expert action rather than interfering with it. She admits, however, that the principle of "just-do-it" might be valid when it comes to everyday or rote skills, such as tying knots, typing or shifting gears while driving (Montero, 2010, 117-118;2013, 315); however, such everyday skills are not, for Montero at least, interesting when it comes to understanding expertise. Thus, Montero's rejection of the just-do-it principle is part and parcel of her view that expertise is in several respects different from everyday skillful behaviour.
According to Montero, it is precisely because Dreyfus applies an analysis based on everyday skills -such as tying one's shoestrings or driving to work -to expert action that his view on expertise is mistaken (Montero 2016, 53). Whereas everyday skills may be effortless and automatic, as Dreyfus believes, this is far from always the case with expert action as Montero sees it. For Montero, an expert is someone who has "engaged in around ten or more years of deliberate practice, which means close to daily, extended practice with the specific aim of improving, and are still intent on improving" (Montero 2016, 64;italics removed). With expert action thus defined, conscious thought or monitoring does not interfere with performance, but is, on the contrary, often what promotes outstanding performance: "at least the right kind of self-directed thinking is compatible with performing at one's best" (Montero 2016, 94).
Montero draws on a number of studies and reports from experts that exemplify various forms of cognition-in-action. Musicians and tennis players use mental techniques to cope with anxiety, such as having a "mantra-like phrase" they repeat in their heads (2016,101). Research on nurses shows that they engage in step-by-step reasoning and musicians as well as dancers -including Montero herself -testify to the importance of developing "strategies for engaging the conscious mind" as part of training, so that attention can be upheld during performance (141).
"disrupts" absorbed coping (202). Further, Romdenh-Romluc's notion of "thought in action" amounts simply to having "conceptual representations" throughout, which means that she does not acknowledge the important qualitative differences between levels of skill/expertise. 8 It is called "the principle of automaticity" in Montero 2013, 304 f. Of course, the principle is an analytic reconstruction of various interrelated views of expertise.
The goal of Montero's analysis is not to present a fullfledged definition of the general term "expertise", but to give an account of one important form of expertise -professional-level -to which the "just-do-it principle" does not apply. Moreover, she does not dispute that expert skills are largely automatic; instead, Montero stresses that what characterises expertise is not primarily that performance at this level happens on its own, but that it involves the aim to constantly push beyond one's limits. 9 Many times, this means focusing on and struggling with one's automaticised behaviours or habits. Montero contends that automaticity is needed when things go wrong, not when they go right (Montero 2013, 316).
Montero here relies upon among others the psychologist Anders Ericsson's extensive work on expertise. Ericsson has shown that outstanding performance is the result of precisely the continuous and systematic effort to go beyond one's comfort zone -the automated level of skill where performance is smooth and undemanding -whereas simply remaining at that level and repeating the same behaviour often rather leads to a deterioration in performance (Ericsson, 2008, 989;2018, 4;Ericsson & Pool 2016, 13). Thus far, his view is in line with that of Montero; nevertheless, it does not apply only to professional-level expertise, but to any skill that an individual wants to enhance Pool 2016, Ericsson 2018). The important point is precisely the aim to improve, and so everyday skills will be characterised by stability, automaticity and arrested development, whereas expert skills involves the aspiration to improve and thus to resist automaticity (Ericsson, 2008, 991).

"Just-Do-It", experts and amateurs
Montero's argument that expert action often requires rather than excludes conscious effort is convincing, and Ericsson's work among others' support it. Nevertheless, there are a few points that I have concerns about. Firstly, while it is judicious to distinguish between everyday skills and professional expertise, rather than simply assuming that an account of the skills required of an everyday commuter can simply be transfered to the race-driver, there must be some further argument behind the contention that the former is not an expert. 10 For Ericsson, the everyday driver "who's been at it for twenty years" (Ericsson & Pool 2016, 13) is certainly not an expert in Montero's sense, but not merely for the reason that she is not professional (taxi drivers are professional but they aren't all very good drivers); rather the reason is that she has no aim to become a better driver. To simply claim that everyday skills such as driving are "not something that we train to improve" (Montero 2016, 242) is disputable: contemporary requirements on new drivers (at least in northern Europe where energy-efficient driving is enforced as long as safety concerns do not 9 This aspect of expertise is explored further by Montero with Toner et al. in (2022). 10 Also, it is not true that Dreyfus makes no distinction between "ordinary" expert performers and professional ones: he acknowledges that his account is idealised, and that professional experts sometimes rely on a form of thinking that he calls "deliberative rationality", and that is distinct from the calculative rationality used at the lower stages (Dreyfus & Dreyfus 1986, 31-32, 39-40). In his later papers, however, Dreyfus gives a cruder account, and deliberation is assimilated with conceptual thought that undermines the flow of absorbed coping (Dreyfus 2013, 22). Cf. Zahavi 2013, 326. take precedence) are demanding, and must in many respects be improved over years of practice, something that requires conscious monitoring and effort. To learn how to drive economically and safely in changing weather conditions, on icy and snowy roads, through forests where the driver is at risk from animals crossing the road, is not learned in a few driving lessons. Here, as well, conscious control is needed at all levels of competence.
Secondly, I am not convinced by the claim that the "just-do-it principle" derives from everyday skills and applicable in this context, but then fallaciously transfered to the level of expertise. Montero writes, for example: "Ask yourself […] why you lift your feet approximately 1.3 inches above each stair as you climb" (Montero 2016, 213), to show that a thought of that kind can interfere with everyday skills. But it is not clear how this slightly convoluted thought would hinder my climbing the stairs, and even if it did, this statement about the height of our feet with respect to the stairs has no obvious connection to the conscious rule or advice we might try to follow if we wanted to climb the stairs more elegantly or more ergonomically, thus aiming precisely to improve our climbing the stairs. 11 In other words, it does not exemplify the kind of cognition-in-action that she argues for where the cognitive act is supposed to somehow foster performance of the skill in question, such as "lift your feet a bit higher" or "engage your whole body".
Thirdly, once we have acknowledged that expertise is partly characterised by cognition-in-action, both during training and performance, the interesting question becomes that of distinguishing the types of conscious processes that experts engage in, in which contexts, and also what forms of conscious engagement might actually to hinder optimal performance. Montero's work abounds in stimulating examples; nonetheless, also her book Thought in Action (2016) lacks a more elaborate story of what cognition-in-action is and why thinking is often believed to impede skilled performance. In addition to her argument that the detrimental character of cognitive control to everyday skills is wrongly supposed to be valid also for expert skills, she sometimes suggests that misdirected focus, "extreme confidence-undermining negative thoughts" or "any sort of focus that takes away from what the expert ought to be focusing on" might compromise performance quality (Montero 2016, 94). But in general, Montero claims that cognition counters bodily automaticity and promotes performance at an expert level.
It seems, however, that the aim to reject the so-called "just-do-it principle", to argue for the importance of thinking in performance and uphold a firm distinction between expertise and everyday skills, leads Montero to over-emphasise top-down cognitive processes such as thinking, reflecting, monitoring, deliberating, analysing. At the same time, experiences that are widely shared by expert performers, such as letting the body take over or being fully absorbed in the performance, is not really allowed for by Montero. Instead, she explains them away. For example, she mentions the ballerina Violette Verdy who once had a performance where she "wasn't even there", and that was a great achievement (Montero 2016, 163). Montero suggests that it might be due to how "she interpreted her own mindful performance" (Montero 2016, 164), and the notion that "the body takes over" is seen as equivalent to automa-ticity: it means that the mind is left "free to contemplate something else" (Montero 2016, 235). Montero's general picture of the expert's mind here remains one of control, analysis and intentionally keeping a phrase or keyword in focus.

Bodily intellection, "Mesh", "Arch" and aesthetic enjoyment
While Montero is certainly right to emphasise that conscious control is in many cases not only compatible with but also necessary to expert performance, she does not leave much room for cognitive processes that are at one with the body: not cognition as directed to action, but cognition within and through embodied performance itself. In this respect, her account is similar to Dreyfus's, and although the latter claims to find support for his view in Merleau-Ponty's thought, both philosophers misunderstand the point of the French phenomenologist's account. For example, both Montero and Dreyfus refer to a section in The Structure of Behaviour 12 that I will quote in full.

Merleau-Ponty writes:
For the player in action the football field is not an "object", that is, the ideal term which can give rise to an indefinite multiplicity of perspectival views and remain equivalent under its apparent transformations. It is pervaded with lines of force (the "yard lines"; those which demarcate the "penalty area") and articulated in sectors (for example, the "openings" between the adversaries) which call for a certain mode of action and which initiate and guide the action as if the player were unaware of it [comme à l'insu du joueur]. The field itself is not given to him, but present as the immanent term of his practical intentions; the player becomes one with it and feels the direction of the "goal", for example, just as immediately as the vertical and the horizontal planes of his own body. It would not be sufficient to say that consciousness inhabits this milieu. At this moment consciousness is nothing other than the dialectic of milieu and action. Each maneuver undertaken by the player modifies the character of the field and establishes in it new lines of force in which the action in turns unfolds and is accomplished, again altering the phenomenal field. (Merleau-Ponty, 1942 182-183;1983, 168-169) Dreyfus sees in this passage a description of "the world of total absorption" (2013, 17) while Montero interprets it as exemplifying a just-do-it position: "as Merleau-Ponty sees it, if we were to focus on [our unreflective bodily actions], they would degenerate into the absurd" (2016, 24). 13 Now, Merleau-Ponty's primary interest in this passage is not in describing expertise in Montero's sense -there is no indication that the football player is a professional -but in our fundamental relation as humans to the world that appears more clearly in skilled -in the everyday sense -performance. Here, the field we relate to does not consist of independent things in the sense of objects whose characteristics would be exhausted by a formula that makes no allowance for embodied subjectivity, and consciousness is immersed in "the dialectic between environment and action" (Merleau-Ponty, 1942, 183;1983, 169).
On Merleau-Ponty's account, our consciousness is embodied and intertwined with the environment -here the football field. It is important to note that the "lines of force" mentioned are not only those directly relating the player's operations and the field, as invisible outlines of the trajectories of the ball and the intended movements. Also the concrete patterns of the "yard lines" and the lines of the penalty area are included, and even if they call for certain actions on the part of the player "as if it was without the player's knowing [à l'insu du joueur]", they are named, and thus conceptualised and part of the player's background knowledge that he can evoke if needed. The "as if" is important: Merleau-Ponty does not claim that this happens unconsciously, but here as in other texts he points to a state of mind in between the theoretical observing relation and the mechanical (automatic) non-minded one, where there are practical intentions, motor significations (Merleau-Ponty 1945, 128), that impells us to "recast" our notion of consciousness (Merleau-Ponty, 1942, 183;1983, 169). It is a bodily form of intellection, rather than conscious thought directed to bodily action.
So while Montero would probably accept Merleau-Ponty's description above as applied to everyday skills, it is not, for her at least, an adequate account of expert soccer playing to claim that "the conscious mind … dissolves into a relation with the environment" (Montero 2016, 25). But it is easy to imagine that the football player, as he is described by Merleau-Ponty, then comes to "focus on" some aspect of his performance: His coach might give him the advice to better coordinate his movements with the other players, to focus on the distribution of the defence on the field, etc. Perhaps something similar might happen as with classical guitarist Tobias Schaeffer, discussed by Montero, who was advised to focus on the movements of his hands while playing. At first this slowed down his playing -an example of thinking interfering with performance. With practice, however, he managed to think about what he was doing while playing, and more importantly, this improved it (Montero 2016, 99).
These are examples of thoughts becoming habitualised -as the yard lines and penalty area in Merleau-Ponty's description -for both amateurs and experts. How should this level of consciousness, highly integrated in bodily performance, be characterised?
A number of researchers have aimed to elucidate skilled performers' conscious control and awareness of their action, among them John Sutton, Wayne Christensen and colleagues (2011Christensen and colleagues ( , 2016Christensen and colleagues ( , 2019 who formulated what they call a Mesh account, where cognitive control is closely integrated with automatic processes in skilled action, and the degree of automaticity versus control is related to the complexity of the task and to the situation. Based on empirical and phenomenological evidence, the theory explains some of the features of Montero's account, relating various kinds of conscious control to the degree of challenge that the skill involves. Christensen, Sutton et al. (2019) distinguish between smooth control (the only one described by Dreyfus, although misinterpreted as noncognitive), adaptive control and effortful problem solving in difficult situations. As they put it, attention can be disruptive if misdirected at "automated aspects of skill control" (176), such as when an experienced driver heeds to "the details of the movements involved in shifting gear" (174). Christensen et al. are not explicit as to why this should be the case, but a charitable interpretation might be that attention can be disruptive in this case if it should be directed elsewhere: on the traffic, the weather conditions, etc.
Simon Høffding and Glenda Satne (2021) offer an extension of this theory that they call "Arch", arguing that the hybrid character of Mesh fails to explain a more thoroughly integrated skilled performance, as in shared musical activity. They bring attention to cases where expert performers -prominent musicians in this case 14state that overthinking can deteriorate performance from an aesthetic point of view (Høffding & Satne 2021, S435). The solution is not relying on automaticity, but rather achieving an intense togetherness and responsivity to the other members of the group, a "hive-mind" that is scaffolded by the music (Høffding & Satne 2021, S434).
Montero has recently, along with John Toner and Alan Moran (2022), developed a more nuanced account of the bodily awareness of skilled performers, marshalling both empirical research and phenomenological studies, elucidating how the body and the mind are intertwined in the experts' effort to continuously improve. In (2021), Montero has also developed her earlier account of the role that proprioception can have as an aesthetic sense in dance, for the dancer -and for the observer through a kind of proprioceptive sympathy. Focusing on her own movements, the dancer can, as Montero suggests with Toner et al., experience her body as a "site of aesthetic pleasure" (2022,136). In the latter work, a distinction is also made between self-consciousness and self-awareness, to explain the possible negative influence of conscious attention to our performance (Toner et al. 2022, 132). Self-consciousness denotes a worry about one's performance or appearance, that can generate anxiety and misdirected focus, while self-awarenss gives "important information about the fine nuances of our involvement in the activity" (Toner et al. 2022, 133). Here proprioception is crucial, either working at a prereflective level or -as when the dancer enjoys her movements -being in focus. This distinction between reflective and prereflective forms of consciousness is derived from the phenomenological literature, and in particular from the work of Dorothée Legrand, Susanne Ravn and Giovanna Colombetti, whose contributions I will now detail. 15 5 Legrand: the performative body and the opaque body Legrand (2006Legrand ( , 2007aLegrand ( , 2007bLegrand ( , 2010 and colleagues (Legrand et al., 2007, Legrand & Ravn 2009) have analysed embodied consciousness and reflection in dancers' skilled performance, in part drawing from Ravn's extensive study of dancers' bodily experience based on ethnographic interviews with dancers from various genres and analysed with phenomenological methods. While Christensen et al. propose a systematic theory of skill acquisition and control covering everyday skills as well as expertise, that Høffding and Satne aim to extend, Legrand and Ravn focus, as Montero, on dancers, but emphasise to a higher extent the continuity between the experiences of 14 Drawing from Høffding's study of musical absorption (2019), based on intense qualitative work with The Danish String Quartet. 15 Toner et al. draw to a large extent on Colombetti's work, but since Colombetti bases her analysis of this precise issue on Legrand's and Ravn's research, I will focus on the latter here. non-experts and experts. Meanwhile, the aim of Legrand and Ravn is not to debunk a myth about experts, as with Montero, nor to advance full-fledged theories of expertise, but rather to bring forward the phenomenology of skilled performance, in particular of dancers. They argue that there are, in skilled action, several levels of bodily consciousness that cannot all be reduced to deliberative thinking or reflection -what Legrand and Ravn refers to as "reifying scrutiny" (Legrand & Ravn 2009, 393).
The phenomenological distinction between reflective and pre-reflective forms of consciousness is crucial to Legrand and Ravn. 16 I can experience my actions in an intentional act directed to the body as an object, as when I look at myself in a mirror or touch a part of my body. In that case, my body is an intentional object, and I am conscious of "the self-as-object" in Legrand's terminology (2007a, 589). At the same time, however, I have a pre-reflective consciousness of myself as subject and this bodily subject is a "constant structural feature of conscious experience" (Legrand 2007a, 583). The qualification of this experience as "pre-reflective" means that selfconsciousness is here given within intentional acts, so that the intentional object varies while self-consciousness remains as an invariant structure of experience. The experience is not only, for example, of a sunny winter day or of a ruby red wine with resolved tannins, but it is also, at the same time, given as my experience. It is "an awareness of oneself as subject" (2007a, 586). Put this way, the body-as-subject is the agent of perception and action, and the "paradigmatic form of bodily selfconsciousness is non-objectifying, in the sense that one usually does not take one's body as an object of experience but rather experiences oneself as a bodily anchored subject" (Legrand et al., 2007, 695). If instead we reflect on the body, this intentional act transforms it into an object of consciousness, thereby missing the body as specifically experiencing (Legrand, 2006, 99).
In another paper, Legrand (2007b) makes a more detailed analysis of bodily consciousness, distinguishing between several forms: the invisible body, the transparent body, the performative body and the opaque body. "The invisible body" appears as a limit concept of sorts, derived from a pathological case: a person suffering from a form of deafferentiation (loss of sensory input from a part of the body). This patient has lost proprioception and touch from below the neck, and his body is thus imperceivable, "invisible", when he does not observe his body, for example when he closes his eyes (Legrand 2007b, 500). 17 Another form of experience that this deafferent person can have of his bodily subject is as an "opaque body", which is when it is the object of observation: in order to move and act, he must fully concentrate on his own limbs. Legrand sees normal pre-reflective self-consciousness, the transparent body and the performative body, as lying between these two (Legrand 2007b, 500).
The experience of one's own body as an object of attention, "the opaque body", is shared by dancers and non-dancers, for example in situations where the non-dancer begins to learn a skill or the dancer a new choreography (Legrand 2007b, 501). This observational attitude can perhaps be compared to Dreyfus's reflection or mindful 16 They rely here on Zahavi's extensive inquiries into the structure of self-awareness (see e.g. Zahavi 1999Zahavi , 2005. 17 The patient is Ian Waterman, described by Jonathan Cole (1995). His case is discussed by Gallagher (2005, 43-45, 109-126). deliberation, and Legrand agrees that it can "even be counterproductive" at another level of pre-reflective self-consciousness, namely the one where "the expert dancer embodies the dance" (ibid.). The latter level Legrand calls "the performative body", and it is an experience where "the body and its morphocinetic actions come 'at the front'" (Legrand 2007b, 502); it is most common with bodily experts such as dancers.
"The transparent body" is another form of pre-reflective experience of the body. It is a fundamental experience of our body in the world, had by both normal people and experts, where we look "through [the body] to the world" (Legrand 2007b, 504; italics in text), and seems to correspond to the pre-reflective experience of oneself as subject described in the earlier papers. It is a pre-reflective experience of the world in a bodily way (2007b, 506), and can be had also by the deafferent patient described (507). 18 The experience that corresponds to reflection on one's bodily actions in Montero's picture, then, appears to be "the opaque body". In a later publication, however, Legrand makes the distinctions in other terms, aiming to mark out various levels of "intermingling of subjective and intentional aspects" (Legrand 2010, 213). She describes them as they unfold in an intentional act, exemplified by the perception of a rose: (1) The situation where I see the rose within reaching distance: the bodily subjectivity appears pre-reflectively as a voluminous, oriented self, "self-as-subject" (Legrand 2010, 214 f.). This corresponds to the transparent body in the earlier account. (2) If I reach out my hand towards the rose, I will still have a pre-reflective experience of my "self-as-subject", but there is a slight difference in emphasis from the former in that I am aware of my bodily self as bearer of sensations, even though the intentional object is still the rose (Legrand 2010, 219 f.). (3) I move closer to the rose, paying attention to the movements of my hand so that I do not hurt myself on the thorns, but when I squeeze its stem I "feel a sudden pain at the tip of my finger" (Legrand 2010, 213). At this point, the bodily self is perceived as an object, "self-as-object", but not in a reifying way. Rather, it can be characterised as a form of "double touch" in the sense of Husserl, who famously wrote that when I touch my left hand with my right hand, I have not only sensations of the left hand as an object of flesh and bones, with its tactile qualities of smoothness and form that I feel in my right hand, but I also have sensations of touch within my left hand (Husserl, 1952, 145). Finally, (4), there is a reifying experience of the self-as-object, as when I look closer at my bleeding finger and try to detect the thorn. This scrutinising observation of my body as part of physical nature, is of a form that I can easily transfer to another bodily subject, examining her finger. Legrand claims that it is also "the most commonly studied form of selfconsciousness" (Legrand 2010, 214).

Legrand and Ravn: embodied reflection
It is not entirely clear how these accounts of the experiential levels of the body relate to one another, but as I understand it, the second level above (I reach my hand towards the rose …) corresponds in certain ways to what was called the "performative body" in the earlier paper, but in this case as it may appear for non-bodily experts. The two forms of experience of the "self-as-object", on the other hand, seem as a distinction within the experience Legrand earlier called the "opaque body". Below, I will first focus on the third level (I move closer to the rose, paying attention to the movements …), and the idea of embodied reflection that is also explored by Legrand & Ravn (2009).
In Legrand's earlier papers, not much was said about the question whether different forms of reflection correspond to the levels of pre-reflective consciousness she distinguishes. There is mainly a distinction being made between the pre-reflective consciousness that structures all intentional experience and that presents the bodysubject in a transparent way, and observation, an intentional experience that can also be directed towards my own body. In the 2010 paper, Legrand addressed this issue, in terms of the "subjective access to the self-as-object" (level 3), an experience of the bodily subject that is non-reifying, and referred to it as an embodied form of reflection. Leaving aside the question whether "reification" is an appropriate term here, I will examine this third level of "non-reifying", reflective experience of the body, as she explored with Ravn in (Legrand & Ravn 2009).
In the latter paper, this experience is said to give access at the same time to our body in its physicality, although, again, non-reified, and our subjectivity. Legrand and Ravn give examples of dancers' concrete experiences that flesh out Husserl's thought experiment mentioned above, showing that part of their practice is constituted by an attention to their own body in its physicality that is at the same time non-reifying. As I understand it, this means that the experience is of the physical aspect of one's body as belonging to the world, but given in a way that preserves the dancers' subjectivity. It is also described as a form of attention that is directed to "the experience of the body in movement" (Legrand & Ravn 2009, 398), or even deeper, to "the depths of the body" (2009,399).
This non-reifying experience of our bodily subjectivity is exemplified by various dancers, and towards the end of the paper called "embodied reflection" (Legrand & Ravn 2009, 404 f.), with reference to the view of Varela, Thompson and Rosch (1991, 27), where "embodied" implies that "mind and body are brought together". In other words, embodied reflection here contrasts with theoretical, or perhaps "intellectual" reflection. Legrand and Ravn take a step further when they analyse theoretical reflection as reifying.
A dancer can use such reifying reflection, as when she corrects her position in a mirror, or when she scrutinizes her appearance. As I understand the notion, an everyday example might be when I take a look at my hands and observe that they look really white in the light, or when I am struck by the thought "it's time to cut my nails". Here, I use a reifying form of reflection, looking at my body as from the outside.
Husserl's description of the double sensation, by contrast, concerns the manner that our body appears to us as both a physical object of sorts that I can perceive, and from the inside as perceive-ing. In thematising what is going on in the experience of touching a part of my own body, and how we can have at the same time a sensation of the hand as a tactile object and sensations of touching, it appears that the sensation is "doubled" (Husserl, 1952, 145). Embodied reflection on this account is a reflection on the double touch: a reflection on the fact that the tactile qualities and sensations of touching can never coincide, or in Legrand and Ravn the dancers' attention to "the experience of the body in movement" (2009,398). Now, in relation to the two questions from the introduction, I think it is possible to see the two forms of the opaque body, or of reflecting/attending to the body, as comparable with Montero's cognition-in-action. A dancer often scrutinises her body, her movements and performance, but she is also aware in a more embodied way of her performance while performing. Montero (2016, 213 f.) distinguishes attention -focusing on what your are doing -from deliberation -thinking about what you are doing -and perhaps attention can roughly be compared to Legrand and Ravn's embodied reflection, while deliberation may to some extent be related to the scrutinising form (although deliberation also includes calculative forms of reasoning). I assume both Legrand and Montero would agree that scrutinising reflection can hinder optimal performance, if misdirected. But what about embodied reflection in Legrand and Ravn's sense -can it impede skilled action? For Montero, both attention and deliberation can impede everyday skills, while for the most part they foster expert performance -except when they are misdirected or take the form of selfconsciousness. Neither Legrand nor Ravn discuss this issue more than in passing -as when Legrand mentions in (2007b) that observation of one's body can be counterproductive at the level of the performative body. They have another aim: to demonstrate that reflection on subjecticity is not necessarily reifiying, but can present our bodily self in an embodied way.

Letting the body find its way: Bodily Reflection
The notion of embodied reflection put forward by Legrand and Ravn seems insufficient, however, to fully capture what is at stake in experiences of letting "the body function on its own" 19 , in the words of cellist Fredrik Sjölin, described by expert performers: dancer and choreographer Simone Forti talks about the "dance state", when one's whole system is "geared to performing … the musical centers of the mind are in focus, in operation, and all you motor intelligence is blossoming" (quoted in Banes 1987, 34-35). These states are explicitly aimed at -Sjölin states that he tries to navigate away from "a high level of conscious control" (Høffding, 2019, 65) and Camille Buttingsrud reports that dancers she interviewed describe their aspiration to let "'go of (conceptual) thinking' while working" that she calls embodied reflection or reflecting through one's body (2021, 7537) -and from what I have seen in my own work as a dancer and with dancers, an inherent part of practice. Although the analyses of Legrand, Ravn and colleagues considerably enhance our understanding of the integration of body and mind in skilled performance, there is, I believe, a form of bodily reflection described by dancers and other performers that is not fully captured by their notions, and that is crucial not least in artistic contexts. Legrand and Ravn's embodied reflection revealed "a form of reflective [rather than reifying] consciousness of one's subjectivity" (Legrand and Ravn 2009, 404; italics in text). The performative body, in contrast, with its system of dispositions and capacities that has been called the body schema, is enhanced in expert performers on their view, but given pre-reflectively.
Legrand and Ravn also describe an aim in dancers towards "'greater selfawareness'" (Legrand & Ravn 2009, 399), at internalising external perspectives or uncovering sensations "that normally stay unnoticed" (398). 20 Dancers report the importance of "'listen[ing]'" to the "'kinaesthetic logic'" of their bodies when they investigate their own corporeality (403), an idea that can be found also in dancer and researcher Caroline Potter's discussion of kinaesthesia, as "a dynamic sense of constantly shifting one's body in space and time in order to achieve a desired end" (Potter, 2008, 449). Potter emphasises that this sense of motion is heightened by training and becomes a vital part of the dancers' social and professional selves. Proficient improvisers report similar experiences of the body leading the exploration: dancer Kent De Spain, for example, speaks of the particular relation between the moving I and the improvisational material, the movements, that the "improvisational awareness" alters between (2003,27). He relates how during improvisation, movements sometimes appear "unbidden", and asks: "Was that just how my 'body' … wanted to move that day?" (33). His essay is published in an anthology on dance improvisation, congenially entitled Taken by Surprise. This idea is expressed also by jazz saxophone player Torben Snekkestad, who has various techniques for surprising himself and comusicians during improvisation (Høffding et al. 2021, S437). 21 In my view, these descriptions point to an experience different both from the prereflective state of absorption or flow, 22 and from embodied reflection in the sense just discussed, and to a profoundly reflective capacity of the body itself that I call "bodily reflection", to distinguish it from the former, that is based on the habitualised structures of the body and of the expert performer's enhanced capacity to sense her movements. Adding a level to Legrand's analysis in (2010), we could imagine a dancer working on a choreography where she is supposed to imagine touching a dog rose, reactivating kinaesthetic and tactile memories as well as habitual movement patterns, and reinvest them with new meaning. Is such a "bodily reflection", then, to be found within phenomenology? 20 In Ravn 2009 andRavn 2009, quotes from dancers are italicised; for reasons of clarity, I have removed the italics here. 21 Snekkestad even puts it in terms of "pulling the rug" from under the feet of the musicians playing (Høffding et al. 2021, S437). 22 Described by Høffding, who calls attention to a rare state of "absorbed not-being-there", described by the musicians as a "blackout", losing awareness, "complete disappear[ing]", "losing oneself in the music", during musical performance (Høffding, 2019, e.g. 66, 67). I would assume that this is an intense aspect of the performative body.
We know that Merleau-Ponty repeatedly refers to Husserl's analysis of the double sensation in his writings. In particular, the French philosopher brings up the remark that when my right hand touches my left hand, my left hand-object at the same time appears as a living body -"es wird Leib, es empfindet", in Husserl's words (1952, 145). He continues: the physical thing comes to life, "or more precisely, it remains what it was, the event does not enrich it, but an exploring power comes … to inhabit it" (Merleau-Ponty 1960, 210). Merleau-Ponty also writes, with a quote from Cartesian Meditations, that "the body carries out a kind of reflection" (1945, 109; italics added) -a citation that comes back in several of his writings and that is found in the first French translation of the text, but not in later editions. 23 The connection between the double sensation and the "kind of reflection" carried out by the body is thus Merleau-Ponty's own.
This reflection is not only a matter of double experiences that can never fully coincide, 24 but also an accomplishment of the body in movement as instituting or creating a first level of meaning (Merleau-Ponty 1995, 279, 289 f.). While the phenomenon of double touch appears when I thematise how the lived body is constituted as a being with two "sides" (Merleau-Ponty 1995, 380), the body is moreover the "scene of a kind of reflection" that opens a first level of symbolism (Merleau-Ponty 1995, 270). 25 We might speak with Rudolf Bernet of "the reflective consciousness of the touching hand" that appears when the right hand touches the left hand that is touching (Bernet, 1994, 173), except that the bodily ("charnelle") reflection on his interpretation implies not only heterogeneity but also fragmentation, whereas Merleau-Ponty sees it as the locus of an integration of the living body with the world and the others (1960,210).
With bodily reflection in a more radical sense, it is the body itself that takes the lead. Not a mindless body of course; it is not a question of pure automaticity, but of letting the limbs with their experience take over and produce the gaze. It is a state where the mind is open and available, where consciousness remains "on the surface" of movement, letting the hand, or the body, explore its inherent possibilities.
In fact, Legrand referred to "'the dancer who moves from a sharp and very present physical state'" and who follows the body's logic (Legrand 2007, 501), 26 and similar descriptions are found in the dancers studied by Ravn, who reported an aim to be present in their bodies and "feel the whole space" (Ravn, 2009, 175). Dancer Cecilia Roos insists that the dancer has to be "razor sharp" in her choices and at the same time have "an expanded state of consciousness" based on her capacity to fine-tune her body (Roos, 2013, 22). This idea of openness or "availability" is echoed in the words of another dancer, Chrysa Parkinson, who relates it to the memories lodged in 23 As Donald Landes points out (in Merleau-Ponty 2012, 515). 24 For Dan Zahavi, this is the reason why this form of self-awareness "contain[s] a dimension of alterity and exteriority" (Zahavi, 1999, 174). 25 Whose relation to the world Merleau-Ponty strived to comprehend during his last years. My aim here is not to deliver an exegesis of these difficult, posthumously published texts, but to suggest that another notion of bodily reflection can be extracted from them. 26 She quotes the dancer Carolien Hermans, whose paper from 2003 is called precisely "When the Body Takes Over" (Legrand 2007b, 515). our bodies, and in particular kinaesthetic memories in the dancer's body (Parkinson and Roos 2013, 101). It is not just that these memories are accidentally awakened through movement, rather the movement itself seems in her description to be active in searching for them: "it's not only the positions or body parts involved in reproducing a movement that make time dilate and events slip across years but actually the movement of the body: motion itself relocates sensory relationships in time" (Parkinson and Roos 2013, 81). 27 When Montero argued against the notion that expert performance can be immersed to the point that it is mindless, she writes at one place, discussing the philosopher and pianist Charles Rosen, that "the mind, it should seem, is on the feeling of movement" (Montero 2016, 183;my emphasis). This is one counterargument to mindlessness among many, but one that points to the experience that I am after here. Giovanna Colombetti also discusses a state of immersion, with reference to a jazz musician who describes how he learns to to improvise. He goes from an observational and reflective stance towards his hands to "a more embodied and nonreflective knowing how", where he is guided by touch and proprioception (Colombetti, 2014, 130). Finally, he learns to "'submit' to the increasing skills of his hands" (ibid.). Here it is both question of a "progressive expansion of bodily self-awareness" and of "'the hand' gradually replac[ing] the 'I' as the improvising agent" (ibid.). 28 Colombetti refers to the musician's awareness as "pre-reflective", but I, preferring to remain in line with the above, would point to a level of bodily reflection here, that relies upon bodily schematic structures that the body itself brings forward in reactualising and transforming movement patterns. 29 Here, it is the body that is active: a reflection is a bodily matter, based on the acquired skills and experience of the performer, with the mind open, "on" the experience, without trying to take hold of this experience. Further, the notion of bodily reflection throws light upon the "just-do-it" principle of Montero, in that it shows how we, rather than simply rejecting this notion, can show what was true in this "myth". The experience described by a professional cricket player in the following way: "You've got to let your body do all those things by itself without letting your mind take control" (quoted in Christensen et al. 2019, 165), appears to substantiate Dreyfus's attitude of absorbed coping. It might, however, also be seen as a state where the body takes the lead, but with a present mind that we saw exemplified above: Colombetti's jazz musician who learned to "submit" to his hands, Buttingsrud's dancers whose bodily attention is directed to the whole performance situation "as an open embrace" (2021, 7542) or else Roos who speaks of letting the body understand and the importance of having "full pitch" for one's body's choices (2013,21). It is dependent on the "performative body" -and thus available to a higher 27 Interestingly, Montero discusses a similar "expertise-induced expansion of time" and suggests that it might be based on the expert's "multiple foci" (Montero 2016, 145), but she sees this experience as yet another form of thinking in action rather than of letting the body find its way -that for Montero is equivalent to "letting the body move automatically" (Montero 2016, 140). 28 An analogous idea is suggested in Buttingsrud's descriptions of dancers' being absorbed and intensely self-aware at the same time, "disclosing experiences through bodily transformation" (2015,118). 29 I discuss this further in Foultier (2021). extent, but not exclusively, to experts -is not yet a thematising reflection, but rather an exploration and rearticulation of the body's own structure.

Conclusion
I have discussed several prominent theories on the relation between consciousness and skilled action, in order to clarify the questions mentioned in the introduction to this paper: (1) does conscious awareness in general conflict with skilled performance? and (2) what forms of consciousness pertaining to the body in action can we distinguish in order to answer that question? Are there forms of focus or reflection that actually hinder skilled action while other forms on the contrary foster such performance, especially in cases of highly skilled, expert and/or artistic performance?
Firstly (1), I presented Dreyfus's wellknown theory of absorbed coping, according to reflection interfers with skilled performance when used at this level. I then discussed (2) Montero's account of "cognition-in-action", where conscious effort and control are on the contrary seen as crucial for expert performance, and argued that Montero takes us a long way in showing that the "just-do-it-principle" is often at odds with both the phenomenology and the empirical evidence of skilled action, in particular when the performer aims to improve. Meanwhile (3), her account tends to subscribe to the same top-down view of conscious processes as Dreyfus, and relies, as I see it, on too firm an opposition between expertise and everyday skills, where the conflict between thinking and performance is seen as inherent to everyday skills while misdirected or confidence-undermining thoughts may occasionally hinder expert performance. I further argued that Montero, at least in her work up til and including Thought in Action (2016), does not fully acknowledge performers' experience of "letting the body lead" and aspiration to let go of top-down conscious control in order to improve performance.
In (4), I argued that the immersion of the football player in his environment described by Merleau-Ponty does not exclude conscious control, but rather characterises a form of bodily intelligence that is the result of practice, and then proceeded to briefly describe some recent theories of skilled action: the "Mesh" theory of Christensen et al. which makes a distinction between skills in terms of degree of complexity and difficulty rather than years of practice and the "Arch" account of Høffding and Satne presented an extension of this theory, supposed to explain highly integrated shared artistic activity, as in joint music playing. While Christensen et al. are in line with Montero in claiming that attention can disrupt automated aspects of performance, Høffding et al. stress the fact that overthinking can deteriorate performance aesthetically, but not attention in general.
I pointed to the recent, phenomenologically inspired, account presented by Montero with Toner et al. of experts' aim to improve, that elucidates the earlier cognitionin-action story with levels of pre-reflective bodily awareness and aesthetic enjoyment. A more fine-grained account of the bodily experience of dancers, that Montero et al. draws upon, is made by Legrand, Ravn and collagues. In (5) I examined the notions Legrand et al. propose of pre-reflective and reflective awareness of the body in action, arguing that the scrutinising, "reifying" observation of one's body can be compared with "deliberation" in Montero, a form of cognition that, if misdirected or used in a situation where it is not called for, might hinder skilled performance. As I discuss in (6), Legrand and Ravn further argues that a "non-reifying" reflective experience of the body is possible, that they call embodied reflection, and that gives access to the body in its physicality at the same time as its subjectivity.
Finally (7), I outlined my own account of what I call a bodily reflection, intended to supplement the former analyses of an experience where conscious and bodily activity are highly integrated. In bodily reflection, the performative body is itself reflective, while exploring movements not as an automatism but as a reenactment and transformation of habitual structures with an open and even expanded consciousness of the whole situation.
In other words, in response to question 1., thoughts may hinder performance, or at least render it less than optimal, if they are misdirected, mixed with feelings of selfdoubt, or of a deliberative kind that is external to or not yet integrated into the practice. The solution is not to stop thinking and rely on automatic processes -as we have seen, various forms of attention are needed at all stages of skilled performance, and performers often aim to resist strongly habitualised patterns. While consciousness in skilled action can be both pre-reflective and reflective, many expert performers describe strategies for finding a state of mind where the body leads the exploration and consciousness remains, as it were, on the surface, that I have called bodily reflection. It is an active state explicitly aimed at, that can give way to a state of full absorption, but also shift to forms of conscious control and deliberation, and that will be explored in further work. 30 Funding Open access funding provided by Umea University.

Conflict of interest
The author reports no potential conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/.