A Phenomenological Account of Embodied Understanding

A dissertation submitted to the

Graduate School

of the University of Cincinnati

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

in the Department of Philosophy

of the College of Arts and Sciences

Alexander Albert Jeuk

February 2016

Magister Artium, Goethe Universität, Frankfurt am Main

Dissertation Committee:

Professor Anthony Chemero, Ph.D., Chair.

Assistant Professor Peter Langland-Hassan, Ph.D.

Professor Thomas Polger, Ph.D.

Abstract

This dissertation is a phenomenological account of embodied understanding that is located in the theoretical context of contemporary phenomenology (Dreyfus 1972, 2002, 2007a, 2007b; Kelly

2002; Ratcliffe 2002, 2008, 2015) and phenomenologically-inspired embodied cognition (Varela et al. 1991; Noë 2004, 2012, 2013, 2015; Thompson 2007; Chemero 2009; Rietveld and Kiverstein

2014). This dissertation is a phenomenological account in that I apply the phenomenological method as it has been in particular developed by Martin Heidegger in Being and Time (Heidegger

1962). This means that I provide a careful analysis of phenomena which I then analyze in terms of the conditions of their possibility. This phenomenological account is an account of embodied understanding in that it is only about those forms of understanding that are body-relational. This means that I am concerned here only with those forms of understanding that are responsive to the world in relation to the bodily structures of an agent, her bodily needs and her ability to sense and move.

My dissertation identifies and analyzes two central structures of embodied understanding. Embod- ied understanding is body-relationally spatiotemporally schematic and integrated with affective concern. That embodied understanding is body-relationally spatiotemporally schematic means that embodied understanding exhibits a structure that allows it to be responsive to the experience of a world in space and time that is itself reflective of one’s embodied ability to move and sense; i.e. that is reflective of an agent’s embodiment (Husserl 1989; Noë 2004; Merleau-Ponty 2012). This responsiveness is possible, since embodied understanding shares characteristics with embodied experience by means of what Kant (1998) called ‘schemata’; a priori space and time determinations that allow embodied understanding to respond to the spatial and temporal characteristics of

ii experience. That embodied understanding is integrated with affective concerns means that understanding exhibits a structure that enables it to be responsive to a world that is not, as Dreyfus and

Taylor (2015) would call it, ‘neutral’, but that is ‘always already familiar’, to use Heidegger’s

(1962) famous phrase (Dreyfus 2007b; Ratcliffe 2010; Noë 2015). This responsiveness is possible, since embodied understanding shares characteristics with embodied experience that are in both cases co-constituted by affective concern.

iii

Acknowledgments

I would like to thank my dissertation advisor, Anthony Chemero, for the extraordinary freedom of research that he has granted me, for his valuable feedback and for his support in professional matters. I would also like to thank my dissertation committee members, Peter Langland-Hassan and

Thomas Polger, for the feedback and the very helpful advice that they have given me on my dissertation research.

I would also like to thank the Charles Phelps Taft Research Center that has endowed me with a dissertation fellowship that helped me significantly to finish my dissertation. I would in particular like to thank Adrian Parr and Sean Keating for their support in professional and personal matters.

In that context, I also want to thank the Department of Philosophy of the University of Cincinnati for continued professional and financial support.

My biggest thanks go to my wife, Friderike Spang, who I want to thank for her patience, love and countless discussions that helped me progress with my dissertation.

Table of Contents Introduction ...... 1 Chapter 1. A Phenomenological Account of Understanding. Body-Relationality and the Conditions of the Possibility of Skillful Action ...... 24 Abstract ...... 24 1.1. Introduction ...... 25 1.2. Phenomenology as a Transcendental Project ...... 32 1.2.1 Individuating Phenomena of Explanatory Relevance Descriptively ...... 32 1.2.2 Analyzing the Conditions of the Possibility of Phenomena ...... 34 1.3. Dreyfus and Kelly on Skillful Action and Understanding: Introspectivism and Pseudo- Explanations ...... 38 1.3.1. The Five-Stage Model and Introspectivism ...... 38 1.3.2. Pseudo-Explanations and Reifications ...... 42 1.4. Establishing the Conditions of the Possibility of Skillful Action ...... 47 1.5. Structures of Understanding...... 52 1.5.1 Body-Relational Spatiotemporal Structure ...... 53 1.5.2 Understanding and Affect ...... 59 1.6. Krakauer, Stanley and Williamson on Skillful Action and Understanding: Lack of Descriptive Precision and Criteria ...... 63 1.7. Conclusion ...... 68 Chapter 2. Concern and the Structure of Action: The Integration of Affect and Understanding ...... 70 Abstract ...... 70 2.1. Introduction ...... 71 2.2. The Four-Fold Structure of Action and Embodied Cognition ...... 74 2.3. Motivated Behavior: The Direction and Manner of an Action ...... 80 2.4. The World as the Horizon for Action: Concernful Understanding ...... 86 2.5. Conclusion ...... 90 Chapter 3. Constitution Embodiment ...... 92 Abstract ...... 92 3.1. Introduction ...... 93 3.2. Constitution Embodiment ...... 97 3.2.1 PEC and Constitution Embodiment ...... 97 3.2.2 Husserl and Merleau-Ponty on Embodied Spatiality ...... 102 3.2.3 Heidegger on Care ...... 109 3.3. Conceptual Problems with PEC’s Use of ‘Constitution Embodiment’ ...... 113

3.3.1 Constitution Embodiment is a Transcendental Claim — not a Scientific Claim ...... 113 3.3.2 Problems with Claims about the Location of the Embodied Mind ...... 116 3.3.3 Problems with Treating Affect as a Psychological Phenomenon ...... 118 3.3.4 Philosophical and Practical Consequences of these Inconsistencies ...... 120 3.4. Conclusion ...... 126 Chapter 4. Overcoming the Disunity of Understanding...... 128 Abstract ...... 128 4.1. Introduction ...... 128 4.2. The Separation between Embodied and Conceptual-Representational Understanding ...... 131 4.2.1 The Separation is Strict ...... 135 4.2.2 Problems with the Separation ...... 137 4.3. Bridging the Separation with Schematic Structure ...... 139 4.3.1 Kantian Schemata ...... 140 4.3.2 Schemata as a Bridging Structure ...... 144 4.4. Different Interactions: Ways to Connect both Kinds of Understanding ...... 148 4.4.1 Grounding Conceptual-Representational Understanding in Embodied Understanding ...... 149 4.4.2 Preserving Autonomy through Minimal Interaction ...... 150 4.5. Conclusion ...... 151 5. Critical Conclusion ...... 153 6. References ...... 155

vii

Introduction

This dissertation is a phenomenological account of embodied understanding. It is a phenomenological account in that I apply the phenomenological method as it has been in particular developed by Martin Heidegger in Being and Time (Heidegger 1962). This means that I provide a careful analysis of phenomena and then analyze these phenomena in terms of the conditions that make them possible. This phenomenological account is an account of embodied understanding in that it is only about those forms of understanding that are body-relational. This means, I am concerned here only with those forms of understanding that present the world in relation to the bodily structures of an agent, her bodily needs and her ability to sense and move.

This analysis establishes the following two central characteristics of the structure of embodied understanding. Embodied understanding is body-relationally spatiotemporally schematic (chapter

1; also chapter 4) and integrated with affective concern (chapters 1 and 2). To be body-relationally spatiotemporally schematic means as much as that embodied understanding exhibits a structure that allows it to be responsive to the experience of a world that is itself reflective of one’s embodied ability to move and sense; i.e. that is reflective of an agent’s embodiment (chapter 3; see also chapter 1) (Husserl 1989; Noë 2004). To be integrated with affective concern1 means that understanding exhibits a structure that enables it to be responsive to a world that is not, as Dreyfus and

Taylor (2015) would call it, ‘neutral’, but that is ‘always already familiar’ to me, to use

Heidegger’s (1962) famous phrase (chapters 2 and 3) (Dreyfus 2007b; Ratcliffe 2010; Noë 2015).

1 Here the question might emerge why concern is affective or why affect contributes to the motivational aspect of concern? I take this in this dissertation to be a phenomenological and conceptual datum that has been established by Martin Heidegger’s extensive work on concern in Being and Time (1962). For a more detailed discussion see chapter 2 and Ratcliffe (2010).

My dissertation provides a unique perspective on embodied understanding within the landscape of other accounts of embodied understanding. Embodied understanding has increasingly received attention in Anglo-American academic philosophy through the work of Hubert Dreyfus (1972,

1991), his collaborators (Taylor 1982; McDowell 2007) and pupils such as John Haugeland (1998),

Sean Kelly (2000, 2002) and Mark Wrathall (2005), who introduced the work of classic phenomenologists to cognitive science and philosophy of mind. The emergence of the various branches of embodied cognition, some of which also strongly draw on the work of classic phenomenologists

(Varela et al. 1991; Gallagher 2005; Wheeler 2005; Thompson 2007; Chemero 2009; Rowlands

2010; Noë 2012; Rietveld and Kiverstein 2014) has further contributed to the prominence of embodied understanding in the philosophical literature.

This dissertation is unique in that it understands phenomenology, mostly in opposition to the work of contemporary Anglo-American authors mentioned above, as a transcendental discipline. Ac- cordingly, I treat experience and embodied understanding as transcendental phenomena and analyze them as such. This entails, obviously, also a conception of embodied understanding that is different from those conceptions of the authors mentioned above. The reasons for this are the following.

Dreyfus and his disciples conceive of phenomenology as a descriptive discipline that establishes the structures of our experience and that conceives of the structures that explain the structures of our experience nearly in an isomorphic relationship to the structures of experience. For instance, according to Dreyfus, an accurate description of our everyday experience establishes that we primarily do not represent the world in language or mental representations—as for instance certain branches of classic cognitive science or analytic philosophy have claimed—but that our primary dealings with the world are comprised by embodied, engaged and non-deliberative action.

The important point here is that Dreyfus claims that an adequate description of the phenomena that we encounter in experience is also basically exhaustive of an explanation of these experiential phenomena. For instance, Dreyfus (2007a) states that we encounter the world in thought linguistically, disembodied and detached from our environment. Therefore, according to Dreyfus, the understanding that contributes to thought has to be explained with detachable and disembodied linguistic representations. Contrary to that, we encounter the world in action in an embodied, engaged and non-deliberative fashion. Therefore, Dreyfus explains the understanding that contributes to action as being embodied, engaged and non-deliberative.2 As we can see, Dreyfus understands phenomenology in the following way. We provide a proper analysis of the relevant phenomena of our experience. Subsequently, we explain these phenomena with entities that correspond structurally to what we have described before. For instance, our actions are most of the time performed non-deliberatively. Accordingly, the understanding that governs action needs to be of a non-deliberative kind too (see chapter 1).

Many proponents of embodied cognition follow Dreyfus in this interpretation of phenomenology.

In particular, various proponents of embodied cognition seem to conceive of phenomenology as a proper observational, descriptive practice that provides data for scientific explanation. For instance, Wheeler (2005), Thompson (2007), Chemero (2009) and Rowlands (2010) seem at times to conceive of phenomenology in this way.3 And Dreyfus has done so too, particularly in Dreyfus

2 It needs to be said here that Dreyfus is not following this procedure coherently. Particularly in his more scholarly work on Heidegger (Dreyfus 1991) he is not following this procedure, but is rather endorsing a method that comes close to the transcendental method used in this dissertation. However, in his own work on understanding and action (Dreyfus 1972, 1988, 2002, 2007a, 2007b; Dreyfus and Dreyfus 1986), that he deems explicitly phenomenological (Dreyfus 1988, 2002), he proceeds like described in the main text of the introduction and in chapter 1. 3 These authors have at times a very ‘thick’ conception of phenomenology that comes with strong ontological and epistemological assumptions about experience. For instance, Chemero has a neutral monist conception of experience (Silberstein and Chemero 2015) and conceives of it as intrinsically meaningful (Chemero 2009), in accord with the phenomenological tradition. At other times, these authors seem to give explanatory importance rather to the sciences, which explain for them phenomena that have been merely correctly individuated by a merely descriptive phenomenological practice.

(2007b), where he treats phenomenology as a descriptive practice that is complemented by a neu- rodynamic model developed by Walter Freeman.

Even though this understanding of phenomenology deserves significant merit in that it brought criticality and accordingly a commitment to the phenomena with it back to contemporary Anglo-

American philosophy, it has its limitations within the context of its own tradition and the basic assumptions that it makes. First, from a merely historic perspective phenomenology is a transcendental endeavor—an endeavor that analyses the conditions of the possibility of phenomena—as phenomenology’s major proponents made abundantly clear (Heidegger 1962; Husserl 1989; Mer- leau-Ponty 2012).4 For instance Heidegger:

Being and the structure of Being lie beyond every entity and every possible char-

acter which an entity may possess. Being is the transcendens pure and simple. (…)

Every disclosure of Being as the transcendens is transcendental knowledge. Phe-

nomenological truth (the disclosedness of Being) is veritas transcendentalis.

(Heidegger 1962, p. 62)

Second, and significantly more important than an exegetical issue is that an understanding of experience and understanding requires a transcendental method, given the basic assumptions that phenomenologists and the traditions from which their work emerged—British empiricism (Locke

1996; Hume 2007) and German idealism (Kant 1998)—make about experience and understanding.5 As Locke states:

4 I provide ample textual evidence that phenomenology is a transcendental endeavor, in particular in chapters 1 and 3, based on the work of Husserl (1989), Heidegger (1962) and Merleau-Ponty (2012). 5 Obviously, empiricists like Locke and Hume were not ‘full blown’ transcendentalists. Their transcendentalism goes not farther as claiming that experience cannot be transcended by understanding (see the quote from Locke in the main text below). Further, when I discuss Kant and the phenomenologists in the following as part of the transcendental tradition, I focus on two major commonalities in their work. First, a commitment that only from an analysis of experience further understanding of philosophical phenomena can be achieved. Second, a commitment to the idea that

(…) from experience: in that (experience), all our knowledge is founded; and from

that it ultimately derives itself. Our observation either employed about external,

sensible objects; or about the internal operations of our minds, perceived and re-

flected on by ourselves, is that, which supplies our understanding with all the ma-

terials of thinking. (Locke 1996, p. 33, italics in original and brackets mine)

This basic assumption is, that experience is the most basic access that humans have to the world and that experience is the foundation of meaning—a statement that is often echoed in contemporary Anglo-American phenomenology and embodied cognition.6 This has the following consequence. We do not have access to things that we cannot experience and things that we cannot in principle experience are meaningless.7 Differently put, we cannot transcend our perspectival experience of the world and the understanding that accompanies it.8 This is the basic datum, around which the phenomenological method revolves and from which a theory of embodied understanding has to be developed.

experience and understanding have to be analyzed by an analysis of the conditions of their possibility. Obviously, Kant, and particular Heidegger and Merleau-Ponty deviate in their analyses of experience and understanding significantly. However, that is the result of diverging analyses of experience and not a consequence of a major disagreement about method or about the basic role that experience has in any philosophical analysis. 6 Many proponents of embodied cognition draw heavily on the work of the transcendental tradition and its core assumption about experience. The clearest case is Alva Noë, who identifies with the work of Kant (Noë 2004, 2015) and that of Heidegger and Merleau-Ponty (Noë 2012, 2015). Noë makes clear that experience is our basic access to the world and that our embodied engagements with the world allow for understanding. Other proponents of embodied cognition that follow the transcendental tradition in particular in that respect that experience is that what generates meaning are Thompson (2007), Chemero (2009) and Rowlands (2010). We can see a similar commitment to the idea that experience grounds meaning in the work of Barsalou (1999) and Prinz (2004), who follow the transcendental tradition in a looser fashion. 7 There is an important exception to this as we will see later. This exception is that we can analyze the conditions of the possibility of experiencing and understanding. Even if we cannot understand or experience these conditions like positive phenomena, we still can conceptually analyze them as mere logical conditions that have to hold so that something else can come about. 8 Accordingly, all justified philosophical posits have to derive from experience and cannot be arrived at based on opinion, intuition or in general by a high-level belief into a metaphysical idea, be that semantic externalism, naturalism or the idea that God exists, etc. That is what makes phenomenology, to use a Kantian term, critical.

What is important about this brief historical excursus is that it reminds us that experience has a special role in the phenomenological tradition, that particularly proponents of a more naturalistic philosophical tradition or of the analytic tradition might not be familiar with. A role, that requires a particular method to analyze experience and understanding.

As we said, experience and understanding are basic world-disclosing structures for the phenomenological tradition. They do not possess the same structure as positive phenomena. Rather, they grant ‘access’ to the world, as Alva Noë (2012, 2015) would state. To be more precise, experience has a transcendental role in the phenomenological tradition. Mark Rowlands puts it in the following way:

In general, experiences are not just items of which we are aware. They are also

items in virtue of which we are aware, both of non-mental objects and their prop-

erties and also of other experiences. As acts of awareness, experiences are items

that reveal objects to subjects. Thus, experiences are not simply empirical things:

they are also transcendental. As transcendental, experiences are not objects of

awareness but that in virtue of which objects—whether mundane physical objects

or other experiences—are revealed to a subject precisely as objects of that subject’s

experience. (Rowlands 2010, p. 169)

Concretely, understanding and experience are the conditions that make it possible to have a world in the first place, because they allow to experience and understand the world.9 Accordingly, they cannot have the same ontological status as the things that we experience and understand. Under- standing cannot be understood the way objects of understanding can be understood. Experience

9 This might sound circular, but it expresses the circumstance that ‘experiencing’ and ‘understanding’ are the most basic terms we can use to express our basic access to the world. Differently put, there is no more basic terminology— other than rephrasing—to express our access to the world.

6 cannot be experienced in the way objects of experience can be experienced. We can experience objects, and we can be aware of particular aspects of our experience. But we can never experience experience. In the same way, our (embodied) understanding has objects, that, as we will see, reflect the basic structure of experience, but our (embodied) understanding cannot be itself the object of its own understanding.

Accordingly, we require a method that does not analyze embodied understanding and experience as objects of embodied understanding and experience itself. And this is where the transcendental method comes into play. The transcendental method consists in an analysis of the conditions of the possibility of phenomena. Accordingly, I analyze in the following which conditions have to hold so that we can understand things as we understand them. Given my focus on the transcendental method, a central contribution of this dissertation is the analysis of the conditions of the possibility of the structure of embodied understanding in the theoretical context of contemporary Anglo-

American academic phenomenology and phenomenologically-inspired embodied cognition.

Chapter 1

Chapter 1, ‘A Phenomenological Account of Understanding. Body-Relationality and the Condi- tions of the Possibility of Skillful Action’, explains in detail why embodied understanding has to possess body-relational spatiotemporal schematic structure and why it has to be integrated with affective concern. As I said above, phenomenologists have established that embodied, skillful action is our major engagement with the world. Since this is the case, embodied understanding has the major function to govern skillful action. Therefore, chapter 1 provides a careful description of

7 the structure of skillful action, since, in order to determine the structure of the embodied understanding that governs skillful action, we need to be first clear—as stated above—about the structure of skillful action itself.

My descriptive analysis of skillful action focuses on two aspects of skillful action. First, the pace with which an action is performed. Second, the effort that goes into an action and the circumstance that behavior follows an agent’s affective concerns.10 After I have provided analyses of skillful actions such as soccer playing and cooking, I go on to analyze those structures that understanding has to exhibit so that it is possible that understanding can govern skillful action; i.e. I apply the transcendental method to analyze the structure of the embodied understanding that governs skillful action.

My analysis identifies two major structures that characterize embodied understanding: body-relational spatiotemporal schematic structure and affective concern. Embodied understanding has to possess spatiotemporal schematic structure, because it has to exhibit a structure that allows it to be responsive to the spatiotemporal structure of our experience of objects (Heidegger 1990; Kant

1998).

Such a structure is provided by spatiotemporal schemata. Spatiotemporal schemata are, according to Kant, a priori time-, and we might add, space-determinations. That means, schemata present an object as an object by means of the possible spatiotemporal instantiation forms that it could potentially possess. For instance, the schema of dog can re-present all possible spatiotemporal forms

10 These two aspects of skillful action are certainly not sufficient to account for skillful action. Both are of central importance for skillful action, but other aspects might be equally important. Yet, an analysis of those aspects is beyond the scope of this dissertation. A short overview of the contemporary phenomenological work on skillful action is provided in chapter 1.

8 that dogs could take in different situations. Thereby schemata do not only allow embodied understanding to exhibit the spatiotemporal meaning aspect of objects independent of the actual experience of these objects, but they further allow that an actual, perspectival experience of an object can match the spatiotemporal structure of a schema; i.e. they allow embodied understanding to be responsive to our experience of an object through the non-arbitrary match of a homogenous spatiotemporal structure.

If we locate this point in the context of the transcendental tradition, it becomes clearer. For Kant

(1998), appearances of the world and conceptual understanding need to exhibit the same structure.11 Only if they exhibit the same structure, in Kant’s case, spatiotemporal structure, do they possess something significant in common which allows for their non-arbitrary, reliable, covariant interaction. For instance, we can enjoy certain experiences of dogs that present them as particular spatiotemporal entities. According to Kant, conceptual understanding possesses spatiotemporal schematic structure that re-presents possible spatiotemporal configurations that dogs can exhibit.

Accordingly, my conceptual understanding of dogs can be spontaneously directed to my experience of dogs and my experience of dogs can refer to my conceptual understanding through a match of the spatiotemporal structure of experience and understanding.

Now, why is a match of a homogenous structure necessary so that embodied understanding can covariantly interact with experience? As we have stated above, phenomenology is located in a tradition that conceives of experience as the locus of meaning. Since it is experience that is meaningful, understanding should possess a structure similar to that of experience so that itself can be meaningful in the absence of environment-involving experience and even in the presence of the

11 For more detail and textual evidence see chapters 1, 2 and 4.

9 latter, when it contributes spontaneously to environment-involving experience in action and perception. Accordingly, an analysis of experience is necessary to determine the structure of understanding. In that spirit, prominent British empiricists claimed that understanding is comprised by image-like ideas that correspond to their image-like conception of experience. Only this way, could understanding and experience interact. And, principles of association provided the spontaneous mechanisms for the manipulation of these image-like ideas through the understanding (Locke

1996; Hume 2007).

Obviously, the Kantian tradition and with it, phenomenology, has criticized the empiricist conception of experience as pictorial.12 Experience is not image-like, but world-involving, which meant for Kant primarily that our experience is characterized by the pure forms of intuition space and time. And indeed, we directly experience a world in space and time, and not images of entities.

Now, that lead Kant to schemata. If the basic structure of experience is determined by space and time, understanding would have to be centrally characterized by space and time too, to be meaningful and to be able to interact with experience.

Yet, in order to understand how to act, I do not require an understanding of the abstract spatiotemporal properties of the objects of experience, pace Kant. Rather, I require an understanding of the possible ways in which my body can relate to entities in space and time through sensing and moving (Husserl 1989; O’Regan and Noë 2002; Noë 2004; Merleau-Ponty 2012) (see also chapter 3).

This means, I need an understanding of the ways in which I can relate in action with a body like mine—and not as a pure, Kantian consciousness—to objects in space and time. Embodied understanding has accordingly to possess a schematic structure that allows it to not only respond to

12 Alva Noë (2012) is probably the most prominent and adamant opponent of empiricism among proponents of embodied cognition and contemporary proponents of the transcendental tradition.

10 spatiotemporal objects simpliciter, but to objects that appear to me in embodied-space and embodied-time.13 Embodied space and time are determined, among other things, by my ability to move in a certain way, which refers back to my body and its capacities. Accordingly, I argue that embodied understanding has to exhibit body-relational, or ‘embodied’ spatiotemporal schematic structure, so that embodied understanding can interact with the experience of an embodied spatiality and temporality.

Embodied understanding further has to be integrated with affective concern, or so I argue. I provide two reasons why that has to be the case. In chapter 1, I argue that understanding and affective concern have to be integrated with each other, because otherwise we could not make sense of their interaction. In chapter 2 I argue that understanding and affective concern have to be integrated with each other, because only if understanding and affective concern are integrated with each other, can we even want to engage into an ordinary action. Yet, more to the latter point below.

I argue in chapter 1 that affective concern is the regulative ideal of our actions. We act in order to fulfill our concerns. Accordingly, the effort that we put into an action as well as the particular behaviors that we produce are determined by our affective concerns. Therefore, our embodied understanding is required to be responsive to the affective concerns that we have, in order to guide our actions so that they can produce outcomes for which we have affective concern. Now, in the context of the transcendental tradition, the question emerges again, how our embodied understanding, that I have so far primarily characterized as embodied spatiotemporally schematic, can be responsive to affective concern. Methodologically speaking, we have to identify the common

13 See for a detailed conception of such embodied spatiality and temporality chapter 3.

11 ground that is the condition of the possibility for the interaction of affective concern and embodied understanding.

I argue in chapter 1, that we cannot imagine such a common ground. Since we cannot imagine such a common ground, we should conceive of affective concern and embodied understanding as integrated with each other, because only this way can we make sense how understanding and affect can interact. Heidegger comes to a similar conclusion.

Every understanding has its mood. Every state-of-mind (affect) is one in which one

understands. (Heidegger 1962, p. 385, brackets mine)

This integration has not only the advantage of being in accord with the transcendental method, but it also has two other reasons speaking in favor of it. One will be discussed in the next chapter. The other is that our experience of the world and our understanding of objects presents objects as concernfully individuated.14 As the phenomenological tradition has pointed out again and again

(Heidegger 1962; Husserl 1982, 1989; Dreyfus 2007b), we do not experience and understand things merely as abstract spatiotemporal objects with secondary qualities or as bearers of a list of properties; rather, we experience the world that surrounds us as a world of things that matter to us in certain ways so that we can potentially fulfill concerns with them. As Husserl puts it:

This world is there for me not only as a world of mere things, but also with the

same immediacy as a world of objects with values, a world of goods, a practical

world. I simply find the physical things in front of me furnished not only with mere

14 I can only hope that a conception of embodied understanding and experience that individuates the world by means of embodied spatiality and temporality as well as affective concern contributes to an erosion of the barrier that has been drawn between humans and animals based on intellectualist conceptions of human understanding. Even though this unjustified division need not necessarily have moral consequences, depending on one’s ethical framework, it has been used pervasively to justify the ill-treatment of animals. A conception of understanding that focuses in particular on the importance of affective concern, will certainly erode the idea that animals have no world, no interests or no thought and allow for more empathy with animals.

material determinations but also with value-characteristics, as beautiful and ugly,

pleasant and unpleasant, agreeable and disagreeable, and the like. Immediately

physical things stand there as Objects of use, the ‘table’ with its ‘books’, the ‘drink-

ing glass’, the ‘vase’, the ‘piano’, etc. These value-characteristics and practical

characteristics also belong constitutively to the Objects ‘on hand’ as Objects, re-

gardless of whether or not I turn to such characteristics and the Objects. (Husserl

1982, p. 53, italics in original)

Therefore, that our understanding experience presents the world as concernfully to us, is further support for the transcendentally derived statement that our embodied understanding is integrated with affective concern.

Chapter 1 provides more than an account of the structure of skillful action and embodied understanding. It further defines the method for the entire dissertation and introduces it to the reader. As

I have stated already above, this dissertation is a phenomenological project that puts emphasis on the transcendental method. The transcendental method consists in the analysis of the conditions of the possibility of certain phenomena. In chapter 1 I introduce this method and show ways in which it is superior to other methodological approaches.

To differentiate my approach from other phenomenological accounts, chapter 1 provides a critique in particular of the work of Hubert Dreyfus (1972, 1991, 1988, 2002 2007a, 2007b, 2012, 2013;

Dreyfus and Dreyfus 1986; Dreyfus and Kelly 2007; Dreyfus and Taylor 2015), which is arguably providing the most influential understanding of phenomenology in contemporary Anglo-American philosophy. In one from or the other, Dreyfus’s merely descriptive phenomenology has been accepted by most proponents of embodied cognition, by Dreyfus’s influential pupils, such as Sean

Kelly or John Haugeland, and even by influential analytic philosophers such as John McDowell,

Jason Stanley and Timothy Williamson, who are also discussed in this dissertation due to their work on action and understanding.

My critique has roughly the following form. Dreyfus’s merely descriptive conception of phenomenology cannot provide justified explanations of the phenomena it has descriptively identified and accordingly the explanations that Dreyfus and those referring to Dreyfus’s understanding of phenomenology have provided, are not justified. That is the case since Dreyfus does not analyze the conditions of the possibility of phenomena that he has descriptively individuated.

Chapter 1 ends with a short discussion of the work of Stanley and Krakauer (2013) and Stanley and Williamson (2016) who claim that their analytic and naturalistic method entails results with regard to the analysis of skillful action and understanding that is superior to the phenomenological method. I compare my results to theirs, and conclude that the phenomenological method is superior to their own account when it comes to the analysis of embodied understanding and skillful action.

Chapter 2

Chapter 2, ‘Concern and the Structure of Action’, provides an in-depth analysis of action that continues where chapter 1 ended on the relationship between embodied understanding and action.

Here I provide a concern-based account of action, that lends heavily from Heidegger (1962), which is a precondition for a further argument for the integration of embodied understanding and affective concern that I present in this chapter. I argue, that in order to want to perform an ordinary action, we already require a concernful understanding of the world, which itself requires and integration of embodied understanding and affective concern.

As I stated above, our major engagement with the world consists in action. Phenomenologists and proponents of embodied cognition go as far as claiming that all other engagements with the world, among them thought, reduce to a form of action.15

In fact, the whole field of the intellect—thought, concept-use, language—is itself a

sphere of absorbed coping (to use Dreyfus’s phrase). (Noë 2012, p. 9)16

Accordingly, the primary function of understanding, in particular embodied understanding, consists in facilitating action. As Thompson states:

The central idea of the embodied approach is that cognition is the exercise of skill-

ful know-how in situated and embodied action. (Thompson 2007, p. 11)

Therefore, in order to determine the structure of embodied understanding, we need to be clear first about that which it is supposed to govern and facilitate: action.

As I argue in this chapter, action has been treated, despite its importance for contemporary phenomenology and embodied cognition, in surprisingly little detail by these theoretical projects. Fur- thermore, I argue here that in particular embodied cognition lacks an adequate account of action that could explain how agents are motivated to act and how they determine the goals towards which they direct their behavior.

To account for this shortcoming, I provide an account of action in terms of a Heideggerian conception concern (Heidegger 1962). I show with the help of various examples as well as phenomenological and conceptual analyses how concern is structuring our action—in particular, the ways in which it is determining the direction and manner of our actions. For instance, cooking is done

15 Heidegger conceived of thinking as a form of agential, though primarily disinterested perception. See footnote 87. 16 ‘Absorbed coping’ is Dreyfus’s (2002) term for engaged, embodied action. Dreyfus attributes the term to Heidegger and Merleau-Ponty.

15 for a particular purpose. Say, for the purpose of saturating one’s hunger. Accordingly, cooking is directed at one’s concern for saturating one’s hunger. However, cooking can also be done in a variety of different ways. It can be conducted as a social event, it can be wrapped up in minutes and so forth. This manner, in which one conducts an action, depends on a wider set of concerns, such as a concern for time efficiency or a concern for sociality.

As I have already demonstrated in chapter 1, understanding how to govern behavior so that it is in accord with one’s concerns, requires an integration of embodied understanding and affective concern. That is the case, since we can only conceive meaningfully of the interaction of embodied understanding and affective concern if we conceive of them as integrated with each other. In chapter 2 I advance another argument in favor of the integration. I argue that, if embodied understanding and affective concern were not integrated with each other, an agent could not even want to engage into an ordinary action.

As I pointed out above, with an ordinary action I do not only direct my behavior towards a state for which I have concern; I also perform an action in a particular manner. Even if I want to hang a picture in my room, I will only do so, if I can perform the action in a manner that I have concern for. If I have no hammer and nails or comparable equipment at hand, I will not proceed following this goal. For instance, I would not hammer the nails into the wall with my bare hands, even though they afford hammering and hanging. Obviously, I would hurt myself badly, would create a huge mess and above all, the whole procedure would take a serious amount of time; all things I not only do not have concern for—all things I have concern to avoid.

Now, in order to even want to perform an ordinary action, I need to understand that the manner of an action is in accord with my concerns.17 That means, I need to understand the things in the world with which and on which I act as concern-fulfilling. For instance, I need to understand that a hammer allows me to fulfill my actions in a concern-fulfilling manner. And not surprisingly, I do not understand a hammer in this situation as a thing with a wooden handle and a metal top, but as a thing with which I can hang a picture in a comfortable and time-efficient manner; i.e. as a thing that allows me to fulfill my concerns.

If I would not understand a hammer as a thing that fulfills my concerns, I would not even want to engage into the action of hanging a picture—while no similar artifact is present and when I am an ordinary person with concerns for time-efficiency and comfort—because I would not understand how I could hang the picture in a comfortable and time-efficient manner; those conditions, only under which I want to perform the action in the first-place.18 The very wanting to perform an ordinary action—Heidegger’s concernful dealings that are ready-to-hand—is dependent on a concernful embodied understanding that some x enables me to perform a behavior that is in accord with my concerns.19 And this concernful embodied understanding is best explained by an integration of embodied understanding and affective concern, as I suggest in chapter 2 in accord with the work of Martin Heidegger (1962) and Matthew Ratcliffe (2002; 2008; 2010; 2015).

17 I differentiate ordinary actions or Heidegger’s concernful dealings ready-to-hand, from exploratory actions or Heidegger’s concernful dealings unready-to-hand. The latter might exhibit a different structure than the ordinary actions that I describe in this chapter. 18 Mind the difference between wanting to engage into an action—differently put, having an intention to act—and wanting a situation in the world to be the case. For instance, I might want a picture hanging at my wall. I might be able to want this, without having a concernful understanding of those things necessary to perform the action of picture- hanging in a manner that is in accord with my concerns. This wanting of a situation in the world is different from the concrete wanting to act so that I can achieve this situation in the world. The latter requires concernful-understanding. 19 I admit that it is unclear whether action is a transcendental or a psychological phenomenon. Accordingly, I also admit that I am not entirely certain whether my claim above is a necessary claim about ordinary action or whether it is a psychological generalization regarding most ordinary actions.

Chapter 3

I argued in chapters 1 and 2 that embodied understanding has to exhibit a structure that can interact with a world that already matters to us and whose spatiality and temporality reflects our embodiment. So far I have spelled out the latter claims as entailments of a proper embodied conception of understanding. In chapter 3 I spell out in detail, based on the work of Husserl (1989), Heidegger

(1962) and Merleau-Ponty (2012), how our experience of the world is co-constituted by our embodiment so that we experience a world that already matters to us and whose spatiality and temporality reflects our embodiment—a world which possesses itself a structure towards which embodied understanding can be responsive.

Based on the work of Husserl (1989) and Merleau-Ponty (2012), I spell out how our bodies, and in particular our embodied ability to move, determine the spatial and temporal properties of the objects of our experience. For instance, things appear to be of a certain size to me because of my size and my ability to move towards an object in a particular amount of time. Accordingly, as I argued in chapter 1, our embodied understanding has to exhibit embodied spatiotemporal structure to be responsive to such an embodied experience.

Based on the work of Heidegger (1962), I show how our embodiment and our affective concerns determine the significance of the objects of our experience. For instance, a car might appear safe or unsafe, heavy or light with regard to my physical embodiment. These properties of cars come to the fore in the context of an agent’s particular concern structure. Most people will experience cars as a safe means of travel, while others, for instance, particularly anxious people, might experience cars primarily as potential death traps.

In both cases, our experience of the world is directly concernful. Accordingly, as I argued in chapters 1 and 2, our embodied understanding has to be integrated with affective concern so that our understanding and our experience can stand in a covariant interaction relationship. In chapter 3 we see how the world can show up as a place of things that matter to me—something that has been so far taken as a phenomenological datum and as conceptual entailment of my arguments in chapters

1 and 2—the world is itself co-constituted by our embodied concerns.

Yet, chapter 3 does not only complement the transcendental argument about the structure of embodied understanding that has been developed in chapters 1 and 2. It further provides a critique of the accounts of constitution embodiment provided by proponents of phenomenologically-inspired embodied cognition, such as Alva Noë (2004, 2012), Evan Thompson (2007), and Anthony Chem- ero (2009).

I argue that a phenomenological account of constitution embodiment cannot accept the scientific naturalism that these authors accept, that analyzes embodiment in the context of constitution as an entity that can be studied like any other positive phenomenon. Since these authors accept that embodied experience is a condition of the possibility of having a world, they cannot at the same time study this condition of having a world as an entity. Rather, a condition of having a world is a condition of the possibility of entities, not an entity itself. Accordingly, accepting at the same time constitution embodiment and naturalism (or positivism) about the study of embodiment in the context of matters of constitution entails an inconsistency.

Further, I criticize that many proponents of embodied cognition focus on the location of the embodied mind, even when they talk of embodiment in the context of constitution. The problem with this is, as my analysis of Husserl’s and Merleau-Ponty’s conception of embodiment has shown, that in the context of constitution, embodiment is a condition of having space and time as we do.

Accordingly, we cannot study embodiment—again, within the context of matters of constitution— as something that is located anywhere. Rather, constitution embodiment is a condition of having space and time as we have it.

Chapter 3 concludes with a discussion of phenomenologically-inspired embodied cognition’s relationship to rationalism and the proper research alliances that it should seek to achieve.

Chapter 4

Chapter 4 is concerned with the question how embodied understanding and conceptual-representational understanding, as that what underlies our ability to think in a detached way about the world, can interact. Various contemporary proponents of Anglo-American academic phenomenology and embodied cognition (Wheeler 2005, 2008; Dreyfus 2007a; Chemero 2009; Rietveld and

Kiverstein 2014) conceive of conceptual-representational understanding in linguistic or representational terms, not seldom following Wittgenstein (2003) in that they assume that thinking is a social linguistic practice. However, if conceptual-representation understanding is linguistic or symbolically representational, the question emerges how something that is linguistic or symbolically representational can interact with something that is, for instance, embodied spatiotemporally schematic or integrated with affective concern.

Concretely, the question emerges how two forms of understanding, that seem to be radically different from each other, can interact with each other in meaningful ways, even though they have no relevant structure in common—the same kind of problem that we have already encountered in the chapters above with regard to the interaction of embodied understanding and a world of things that

20 matter to us and with regard to the interaction of embodied understanding and a world characterized by spatiotemporal properties.

This problem has not gone unnoticed. In particular McDowell (1994, 2007), Noë (2004, 2012) and

Dreyfus (2007a) have addressed it in various ways. Dreyfus (2007a) and McDowell (2007) have shown that the problem at hand entails the following undesirable consequences. It is difficult to see how embodied understanding and conceptual-representational understanding can interact so that we can make sense of how we can judge about the objects of embodied understanding and how we can make verbal reports about action and perception. Neither is it clear, according to them, how our perceptual experience and agency could inform our conceptual-representational understanding, which would result in a conception of the mind as detached from the world.

The solutions to this problem vary. Noë (2004, 2013, 2015) claims that embodied understanding itself has to exhibit characteristics of conceptual-representational understanding, yet does not spell out in which way embodied understanding concretely exhibits these characteristics. McDowell

(2007) claims that perceptual experience and agency exhibit conceptual-representational structure.

And Dreyfus (2007a) admits that he has no working account of the relationship between embodied understanding and conceptual-representational understanding.

Independent of the merits of these accounts, I propose a different solution, that circumvents the intellectualism at the core of McDowell’s approach and that is responsive to the following transcendental problem that linguistic accounts face. Embodied understanding is transcendental. Yet, linguistic representations, speech, symbols and so on are positive phenomena that are the objects of embodied understanding and experience. We use them, in an action-like fashion to communicate; and as Wittgenstein (2003) has famously pointed out, the use of linguistic entities requires

21 an embodied understanding of its own (see also Noë 2015).20 This should be at least concerning to everybody who otherwise accepts the main posits of phenomenology.

Accordingly, we need to find a conception of conceptual-representational understanding that is different from a linguistic account, that is ontologically at the same level as my account of embodied understanding and that allows that conceptual-representational understanding can stand in meaningful and non-arbitrary relations to embodied understanding.

Concretely, I propose a Kantian conception of conceptual understanding as a solution. Now,

McDowell and Noë are both heavily influenced by Kant too, as both make more than explicit

(McDowell 1994; Noë 2004), and propose some form or another of a synthesis of embodied understanding and conceptual-representational understanding that is supposed to follow Kant. Yet, my Kantian strategy differs from theirs, in that I put emphasis, as the chapters above have done, on a structural synthesis.21 Such a structural synthesis has been elaborated by Kant in the chapter,

On the schematism of pure concepts of the understanding, in the Critique of Pure Reason. Con- cretely, a structural synthesis requires an analysis of the conditions of the possibility of a synthesis of embodied understanding and conceptual-representational understanding.

Following Kant (1998) and Heidegger (1990), I propose a theory of conceptual-representational understanding that grounds conceptual understanding in spatiotemporal schemata, with which we are familiar from chapters 1 and 3. According to such a Kantian theory of conceptual-representational understanding, conceptual-representational understanding can be directed towards the world and can be responsive to the world in that schemata are a general structure that exhibits all possible spatiotemporal forms according to which an entity can appear to me. For instance, the concept of

20 I would like to thank an anonymous referee of the paper version of chapter 2 for this insight. 21 And, obviously in opposition to McDowell, I do not identify conceptual understanding with language.

22 dog is about dogs since the schema that underlies the concept can generate the different possible ways in which dogs can appear to me in experience.

Now, this conception of conceptual-representational understanding allows for the following. If conceptual-representational understanding exhibits (embodied) spatiotemporal schematic structure and if embodied understanding exhibits (embodied) spatiotemporal schematic structure, as I have argued in chapter 1, then both have a structure in common that allows for a covariant, non- arbitrary interaction. Differently put, spatiotemporal schemata correspond structurally to the spatiotemporal structure with which objects and situations are presented as objects and situations in embodied experience and grasped by embodied understanding. Thereby, my embodied understanding of coffee machines and my conceptual-representational understanding of coffee machines can be directed towards the same entities in the world and be responsive to them. If both forms of understanding are governed by schemata, we can see how they could possibly interact. In this chapter I spell out shortly two different conceptions of their interaction. According to the first possible kind of interaction, conceptual-representational understanding is grounded in embodied understanding. That means, whatever the particular function of conceptual-representational understanding is, it receives its meaning in that it stands in a close, non-arbitrary relationship to embodied understanding. According to another possible kind of interaction, the second one that I point out in this dissertation, embodied understanding and conceptual-representational understanding are responding to the same situation qua the same spatiotemporal schematic structure, yet, both perform their functions in the respective situation autonomously.

My dissertation concludes with a critical conclusion that analyzes problems with a transcendental account of embodied understanding that future work in the transcendental tradition will have to address.

Chapter 1. A Phenomenological Account of Understanding. Body-Relationality and the Conditions of the Possibility of Skillful Action

Abstract

In this chapter I develop a proper phenomenological account of skillful action and understanding.

A proper phenomenological account requires a proper phenomenological method. Based on the work of Martin Heidegger (1962), I develop an account that shows that skillful action and understanding have to be analyzed, first, through a careful and detailed analysis of the phenomena that centrally characterize skillful action and understanding, and, second, in terms of the conditions of the possibility of these phenomena. Then I show that contemporary phenomenological accounts such as those by Hubert Dreyfus (2002, 2007a, 2007b), Sean Kelly (2000, 2002), and Dreyfus and

Kelly (2007) fail to adhere to a proper phenomenological method and accordingly entail unjustified and false results. To compensate for this shortcoming, I provide an account of skillful action and understanding that derives from a proper phenomenological method. I argue that skillful action is governed by an understanding that is body-relational, that possesses a structure that is embodied spatiotemporally schematic and that is integrated with affect. The chapter concludes with a methodological criticism of contemporary analytic accounts of skillful action and understanding, which

I discuss using the example of the work of Stanley and Krakauer (2013) and Stanley and William- son (2016).

1.1. Introduction

In this chapter I develop a proper phenomenological account of skillful action and understanding.

I argue that skillful action is governed by an understanding that is body-relational, that possesses a structure that is embodied spatiotemporally schematic and that is integrated with affect. A proper phenomenological account requires a proper phenomenological method. Based on the work of

Martin Heidegger (1962), I develop an account in section 1.2 of this chapter, that shows that skillful action and understanding have to be analyzed, first, through a careful and detailed analysis of the phenomena that centrally characterize skillful action and understanding, and, second, in terms of the conditions of the possibility of these phenomena. In section 1.3 I show that contemporary phenomenological accounts such as those by Hubert Dreyfus (2002, 2007a, 2007b), Sean Kelly

(2000, 2002), and Dreyfus and Kelly (2007) fail to adhere to a proper phenomenological method and accordingly entail unjustified and false results. Based on the method presented in section 1.2,

I provide a proper account of skillful action and understanding in sections 1.4 and 1.5. Section 1.6 concludes the chapter with a methodological criticism of contemporary analytic accounts of skillful action and understanding, which I discuss using the example of the work of Stanley and Kra- kauer (2013) and Stanley and Williamson (2016). In the following I provide a more detailed introduction that commences with a criticism by Stanley and Williamson (2016) marshalled against contemporary phenomenological accounts of skillful action and understanding.

Jason Stanley and Timothy Williamson (2016) have criticized in their paper, Skill, that philosophers working in the phenomenological tradition have not contributed positively to a clear analysis of skillful action and the understanding that governs it.

The connection between knowledge and skill, like that between knowledge and vir-

tue, has been a major topic in the history of philosophy. Over the past century, some

philosophers wrote directly on the nature of skill. But aside from Gilbert Ryle […],

who argued that skill cannot be explained by states of propositional knowledge

alone, most of this work has been in or under the influence of a nonanalytic tradi-

tion. It has been largely negative, saying what skill is not rather than what it is. For

example, the point of Hubert Dreyfus’s work on expertise or Sean Kelly’s work on

skill is to argue for the negative point that genuinely skilled action is not action

guided by propositional attitudes. It is difficult to extract from this work, or even

from the figures who influenced it, such as Heidegger and Merleau-Ponty, a posi-

tive account of the nature of skill. (Stanley and Williamson 2016, p. 1, emphasis

and square brackets mine)

This claim gives reason for consternation. Dreyfus, contrary to what is claimed here, has provided positive contributions on skillful action and understanding for over three decades. For instance,

Dreyfus (2002, 2007a, b, 2012) and Dreyfus and Taylor (2015) have provided explanations of the understanding that informs skillful action in terms of embodied coping, the intentional arc and background understanding. Dreyfus has provided naturalistic interpretations of these explanatory structures in terms of robotics and neuro-dynamics (Dreyfus 1992, 2007b). And Dreyfus has provided a non-rationalist description of skillful action in terms of solicitations and the tendency towards maximal grip (Dreyfus 2002, 2007b) (Dreyfus and Kelly 2007). Further, Dreyfus has worked intensively on skill-acquisition and learning (Dreyfus and Dreyfus 1986), characterized

26 learning as a central feature of the intentional arc (Dreyfus 2002, 2007b), and accounted for skill acquisition in terms of Walter Freeman’s neuro-dynamic model of learning (Dreyfus 2007b).22

Stanley’s and Williamson’s criticism of Sean Kelly’s work is neither appropriate. Kelly (2000,

2002) has argued for a conception of skillful action that requires an intentionality that lies between the purely mechanical and the purely rational; i.e. Merleau-Ponty’s (2012) motor intentionality.

Further, as shown above, he has provided together with Hubert Dreyfus an account of skillful action and understanding in terms of solicitations (Dreyfus and Kelly 2007). And, not unlike Drey- fus, Kelly has contributed to a naturalistic account of embodied intentionality and the background in terms of neuro-dynamics (Borrett, Kelly and Kwan 2000).

A sweeping statement like ‘it is difficult to extract from this work, or even from the figures who influenced it, such as Heidegger and Merleau-Ponty, a positive account of the nature of skill’ merely betrays the lack of familiarity that Stanley and Williamson have with the work of two of the most influential philosophers of the twentieth century, rather than that it would constitute a legitimate criticism of phenomenology or provide support for the claim that only the analytic tradition has provided a positive account of the relationship between skillful action and understanding.23 To the contrary, as we will see in section 1.6 of this chapter, the analytic tradition has failed to provide a precise and analytically sufficient account of the relationship between skillful action and understanding, contrary to the account that I develop that is particularly indebted to the work of Martin Heidegger.

22 That Dreyfus has focused in his work that intensively on learning is particularly delicate, since Stanley and Krakauer (2013) and Stanley and Williamson (2016) purport to have established, nearly 30 years after Dreyfus and Dreyfus (1986) have published their work on skill acquisition, that learning past baseline is a central feature of skillful action. In particular, it is merely one out of two central features of skillful action that Krakauer, Stanley and Williamson establish. 23 Note that Stanley’s and Williamson’s criticism is not even accompanied by the attempt of an engagement with the work of Heidegger or Merleau-Ponty.

Contrary to what my ‘defense’ of Dreyfus’s and Kelly’s work might suggest, there is no reason for phenomenologists to be complacent. Even though Dreyfus and Kelly have provided plenty of work on the nature of skillful action and the understanding that governs it, their accounts fail to produce informative and justified results. This is the case, since Dreyfus and Kelly misapply the phenomenological method. Both misconceive of the phenomenological method as merely descriptive and even as introspectivist (Kelly 2000; Dreyfus and Kelly 2007). As I show in section 1.2, based on the work of Heidegger (1962) in Being & Time, phenomenology is not only a descriptive endeavor, but it is also a transcendentalist discipline and importantly not an introspectivist one.

That means concretely, that phenomenology, properly understood, does not only describe phenomena, even though a detailed and careful analysis of phenomena is the starting point of every phenomenological analysis (section 1.2.1), but it also analyzes the conditions of the possibility of phenomena (section 1.2.2). This means, that phenomenology does not arrive at explanations or ontological posits based on introspective reports, but based on the analysis of the conditions of the possibility of phenomena.

In section 1.3 I show how Dreyfus’s and Kelly’s work fails to adhere to the phenomenological method as described in section 1.2, and how this leads to unjustified ontological claims and pseudo-explanations. In section 1.3.1 I show how Dreyfus’s well-known five-stage model of skill- acquisition (Dreyfus and Dreyfus 1986), that functions as the basic descriptive evidence in Drey- fus’s work on skillful action and understanding, is not properly phenomenological. Rather, the model has been established through the observation of the performance of agents in experiments and introspective reports of instructors and learners. Contrary to what a proper phenomenological description demands (section 1.2.1), this evidence is not drawn from careful and detailed descriptions of relevant and invariant structures of experience. Yet, what is worse than the reliance on an

28 unreliable descriptive method, is that Dreyfus and Kelly make ontological inferences from (unreliable) descriptive evidence in that they state that rules and representations are not governing skillful action, because rules and representations are not experientially present during skillful action.

This means, they make ontological statements based on the presence or absence of entities in experience, which, as section 1.2.2 shows, is introspectivist; independent of whether the descriptive evidence is accurate or not. In section 1.3.2 I further show that Dreyfus’s and Kelly’s explanations of skillful action and understanding in terms of the intentional arc, the tendency towards maximal grip, solicitations and the background are pseudo-explanations. Concretely, I show that these terms are merely names for (unreliably) descriptively established phenomena. As such they are merely names, instead of terms that would refer to explanations. Accordingly, section 1.3 demonstrates that at least the phenomenological accounts of skillful action and understanding by Dreyfus and

Kelly are wanting.

In order to compensate for the lack of an adequate phenomenological explanation of skillful action and understanding, I provide a proper account in sections 1.4 and 1.5. In section 1.4 I commence with a careful and detailed description of ordinary skillful actions, such as soccer playing and cooking, identify central characteristics of skillful action and then analyze these central characteristics in terms of the conditions that make them possible—though I focus here only on those conditions of possibility that pertain to understanding. I concentrate on two central—though not exhaustive—characteristics of skillful action in terms of the understanding that governs it. First, the understanding that is necessary to determine the pace with which an agent has to act. Second, the understanding that is necessary for an agent to evaluate which of the behaviors that she could possibly produce in a situation is in accord with her concerns. The latter form of understanding

29 enables an agent to determine which objects in the environment and which behavioral patterns are relevant in a situation to reach an agent’s goals and which behaviors are ‘worth the effort’.

The central finding of section 1.4 is that both forms of understanding are body-relational. For instance, the understanding of the pace with which an agent has to perform an action, requires, among other things, an understanding of the agent’s bodily position and an understanding of her sensorimotor competence to reach a particular object in a particular place in a particular time. The understanding of an embodied agent’s concerns requires an understanding of the particular embodied affect that renders certain outcomes of an action significant and desirable to an agent.

Section 1.5 analyzes the structure that understanding has to possess so that it can inform skillful action. This analysis step provides an account of the structure of understanding and thereby circumvents pseudo-explanations such as described in section 1.3.2. Section 1.5.1 shows, following

Kant (1998), that the understanding that governs skillful action has to exhibit body-relationally spatiotemporal structure. That is the case, since understanding can only be about the spatiotemporal world in which skillful action unfolds, if it possesses a structure that is responsive to the spatiotemporal structure of the world. Yet, spatiotemporal structure alone is not enough. This structure has also to be responsive to how an embodied agent’s body can behave in a spatiotemporal world. Accordingly, understanding has to possess embodied spatiotemporal schematic structure so that it can guide an embodied agent’s skillful actions in a spatiotemporal world. Section

1.5.2 demonstrates, following Heidegger (1962), that understanding has to be integrated with affect, so that understanding can be about the concern-relating aspects of skillful action. That is the case, since, according to an analysis of the conditions of the possibility of the structure of understanding interacting with affect, we can only properly conceive of their interaction, if they are

30 integrated with each other. Accordingly, section 1.5 establishes that understanding possesses embodied spatiotemporally schematic structure and is integrated with affect.

Sections 1.4 and 1.5 allow us to evaluate contemporary work from analytic philosophy on skillful action and understanding in section 1.6. This introduction commenced with a depiction of Stan- ley’s and Williamson’s derogatory evaluation of the worth of phenomenological explanations, such as provided by Heidegger. Section 1.6 will show that a proper phenomenological approach, that follows Heidegger methodologically (section 1.2) as well as content-wise (section 1.5.2) is explanatorily superior to Stanley’s and Williamson’s account. First, a phenomenological account provides an analytically more precise account of the structure of skillful action than Stanley’s and

Williamson’s account. Second, a phenomenological account provides analytically more informative and precise means to evaluate the structure of understanding, compared to Stanley’s and Wil- liamson’s account.24

In the following I commence with a delineation of a proper phenomenological method, that will be consistently used throughout this chapter to reject Dreyfus’s and Kelly’s account of skillful action and understanding, that will be used to develop a proper account of skillful action and understanding and that will be used to show the explanatory insufficiency of Stanley’s and William- son’s work.

24 I follow Noë (2005, 2007) in that I criticize contemporary phenomenological and analytic accounts of skillful action and understanding based on methodological considerations. However, contrary to Noë I do not argue for a naturalistic methodology, but for a proper phenomenological methodology.

1.2. Phenomenology as a Transcendental Project

I have claimed in section 1.1 that contemporary phenomenologists, such as Dreyfus and Kelly, cannot explain skillful action and its relation to understanding sufficiently, because they misconceive of the phenomenological method and accordingly misapply it. Instead of providing a phenomenological account, Dreyfus and Kelly have developed a model of action and understanding that is based on introspection and pseudo-explanations, which I demonstrate in section 1.3. Yet, before I can show that Dreyfus and Kelly misapply the phenomenological method, we need to gain clarity about what the phenomenological method is. Not only will this allow to appropriately criticize Dreyfus and Kelly, but it will also allow me to develop a conception of skillful action and understanding in sections 1.4 and 1.5 that is phenomenologically adequate. In the following I present a Heideggerian model of the phenomenological method that is comprised by a) the descriptive individuation of phenomena of explanatory relevance and b) the analysis of the conditions of the possibility of these phenomena.

1.2.1 Individuating Phenomena of Explanatory Relevance Descriptively

Kelly (2000, p. 162) is correct when he states that ‘phenomenology is essentially descriptive.’ Yet, from this it does neither follow that phenomenology is interested in ‘completely … describing the phenomena of human experience’ (Kelly 2000, p. 162), nor that phenomenology is able to draw negative ontological inferences from such descriptions, nor that phenomenology does not engage into further explanation beyond mere description.

Rather, phenomenology describes only certain phenomena of human experience, i.e. those that are central of and invariant to human experience and thereby of explanatory relevance.

And this means that it [Dasein] is to be shown as it is proximally and for the most

part—in its average everydayness. In this everydayness there are certain structures

which we shall exhibit—not just any accidental structures, but essential ones which,

in every kind of Being that factical Dasein may possess, persist as determinative

for the character of its Being. (Heidegger 1962, pp. 37-38, emphasis in original,

square brackets mine)

Heidegger is acutely aware of the circumstance that a mere description of experience in itself is not explanatory nor that it licenses ontological inferences about what governs and constitutes experience and the phenomena with which we are presented experientially.

The idea of grasping and explicating phenomena in a way which is ‘original’ and

‘intuitive’ is directly opposed to the naïveté of a haphazard, ‘immediate’, and un-

reflective ‘beholding’. (Heidegger 1962, p. 61, emphasis mine)

A mere description of our experience, how precise and acute it might be, is merely unreflective observing. This is the case, because the appearances, i.e. the aspects and contents of experience that we describe, are merely ‘surface’ phenomena that require an explanation that goes beyond them. Heidegger compares phenomenology in this respect to medicine.

This is what one is talking about when one speaks of the ‘symptoms of a disease’

[‘Krankheitserscheinungen']. Here one has in mind certain occurrences in the body

which show themselves and which, in showing themselves as thus showing them-

selves, ‘indicate’ something which does not show itself. The emergence of such

occurrences, their showing-themselves, goes together with the Being-present-at-

hand of disturbances which does not show themselves. Thus appearance, as the

appearance ‘of something’, does not mean showing-itself; it means rather the an-

nouncing-itself by something which does not show itself, but which announces it-

self through something which does show itself. (Heidegger 1962, p. 52, translation

in original)

Exactly because the structures that govern our experience are not directly given in experience, phenomenology is required. As Heidegger states,

And just because the phenomena are proximally and for the most part not given,

there is need for phenomenology. (Heidegger 1962, p. 60)

1.2.2 Analyzing the Conditions of the Possibility of Phenomena

Now, how does phenomenology analyze the structures that are governing experience and are constitutive of it? It does so in the tradition of transcendental philosophy, by analyzing the conditions of possibility of the phenomena that characterize our experience centrally. Heidegger is concerned, following Kant, with establishing the ‘existential conditions of (the) possibility’ of phenomena

(Heidegger 1962, 304). In accord with this, Heidegger conceives of the meaning (Sinn) of a phenomenon as that what makes it possible.

What does ‘meaning’ signify? (…) Meaning is that wherein the understandability

[Verstehbarkeit] of something maintains itself—even that of something which does

not come into view explicitly and thematically. ‘Meaning’ signifies the ‘upon-

which’ [das Woraufhin] of a primary projection in terms of which something can

be conceived of in its possibility as that which it is. Projecting discloses possibili-

ties—that is to say, it discloses the sort of thing that makes possible. (Heidegger

1962, pp. 370-371, translation and emphasis in original)

This means, to understand the meaning of a phenomenon is to understand the conditions, for instance, certain fundamental structures like spatiality or temporality, that make this phenomenon possible.25 To show what it means to analyze the conditions of the possibility of a phenomenon must wait till sections 1.4 and 1.5 of this chapter. Yet, one of Kant’s arguments that establishes that space is a pure form of intuition rather than an object of intuition nicely illustrates how to conceive of such an analysis.

One can never represent [rather: imagine/vorstellen] that there is no space, although

one can very well think that there are no objects to be encountered in it. It is there-

fore to be regarded as the condition of the possibility of appearances, not as a de-

termination dependent on them, and is an a priori representation [rather: imagina-

tion/Vorstellung] that necessarily grounds outer appearances.26 (Kant 1998,

A24/B38, 39, square brackets mine)

Kant shows here that we cannot imagine or otherwise conceive of phenomena other than conceive of them or imagine them as spatial entities. We can subtract properties like color or density from appearances, but not spatiality itself. Accordingly, space is the condition of the possibility of imagining or conceiving of appearances.

25 ‘Phenomenon’ is a term for a stable object of experience. As an object of experience, a phenomenon has meaning. 26 In German Kant uses ‘Vorstellung machen’ and ‘Vorstellung’ in the passage quoted here (Kant 1956, A24, B38), which we should translate with ‘imagine’ and ‘imagination’.

These transcendental methodological considerations are persistently reflected in the structure of

Being & Time. Heidegger is not only explicit about them methodologically, but his method also determines the structure of the book itself. First, phenomena are described as being such and such as we encounter them in everyday life. For instance, most encounters that we have in the world are ready-to-hand, which means as much as that we encounter the world primarily non-deliberatively as for action. Yet, contrary to what Dreyfus (2002, 2007a, 2007b) seems to suggest,

Heidegger’s analysis does not end here. Heidegger’s analysis does not consist in an ontological inference from what we are presented with in experience to the existence of the phenomena with which we are presented, as Dreyfus purports. That would be introspectivist.

Heidegger merely commences—not unlike the ordinary language philosopher—his analysis with everyday (alltäglich) phenomena, such as those with which we are presented when we are encoun- tering the world, for instance, ready-to-hand.

Thus the term ‘phenomenology’ expresses a maxim which can be formulated as

‘To the things themselves!’ It is opposed to all free-floating constructions and ac-

cidental findings; it is opposed to taking over any conceptions which only seem to

have been demonstrated; it is opposed to those pseudo-questions which parade

themselves as ‘problems’, often for generations at a time. (Heidegger 1962, p. 50)

This initial analysis has the function to establish the space of phenomena that need to be explained and to differentiate genuine phenomena from philosophical artifacts. As we will see in section 1.3, contemporary phenomenologists such as Dreyfus and Kelly are failing at this task too. Rather than critically evaluating our experience of phenomena by providing detailed and complex descriptions of action, they often make claims without such descriptions or are not critical about the example cases that they discuss. For instance, it is difficult to see how chess playing, Dreyfus’s foremost

36 example for the analysis of the relationship between skillful action and understanding, is characteristic of everyday skillful actions, given that it is strictly rule based and involves only minimal bodily behavior.

Be that as it may, importantly, a description of our everyday experience is merely a ‘preparatory

Interpretation of the fundamental structures of Dasein’ (Heidegger 1962, p. 42). Those structures must be established as the conditions of the possibility of certain phenomena of our experience.

Accordingly, Heidegger goes on to analyze the genuine phenomena of our everyday experience in terms of the conditions of their possibility. As such, Heidegger identifies the care structure, which is comprised by ‘understanding, state-of-mind, falling, and discourse’ (Heidegger 1962, p. 384-

385). Then, finally, the care structure is ultimately explained in terms of temporality—differently put, Heidegger shows how temporality is the condition of the possibility of care. This is obviously an oversimplification of the structure of Being & Time, yet it is a clear expression of how the transcendental method is reflected in Heidegger’s approach, and how Heidegger’s own account differs from how his work is presented by authors like Dreyfus and Kelly as merely descriptive.27

The difference between a phenomenological account that establishes the conditions of the possibility of the phenomena that it has prior to that critically identified through complex and detailed descriptions of experience and a merely descriptive and introspective phenomenology is significant. If understood correctly, particularly if we abstract away from Heidegger’s style with which many contemporary philosophers are not familiar, the Heideggerian project is more precise, informative and criteriologically justified than contemporary analytic approaches, as I will show in

27 Dreyfus is aware of this complex, transcendental structure that characterizes Being & Time, which we can clearly see in his commentary on Being & Time, Being-in-the-World (Dreyfus 1991). Unfortunately, this awareness does not seem to have influenced Dreyfus’s own work as a phenomenologist. In general, I do not try here to undermine Drey- fus’s and Kelly’s status as phenomenologists and particularly their scholarly work on phenomenology. Yet, the work on understanding and action that they have proposed by themselves is nevertheless exhibiting the deficiencies that I analyze in sections 1.2 and 1.3.

37 sections 1.4, 1.5 and 1.6. Compared to this Heideggerian approach, Dreyfus’s and Kelly’s introspectivist approach generates unreliable and unjustified results, as we will see in the next section.

1.3. Dreyfus and Kelly on Skillful Action and Understanding: Introspectivism and Pseudo-

Explanations

Section 1.2 demonstrated that Dreyfus’s and Kelly’s conception of the phenomenological method as a merely descriptive endeavor is misguided. In the following I show that this misconception of the phenomenological method leads to considerable problems with their conception of understanding and skillful action. I show in section 1.3.1 that Dreyfus’s well-known five stage model of skill acquisition is not phenomenological, but introspectivist. This has the consequence that naïve analyses of experience are used to justify ontological claims about the existence of structures that govern skillful action. Section 1.3.2 makes the case that Dreyfus’s and Kelly’s explanations of skillful action and understanding in terms of the intentional arc, the tendency towards maximal grip, solicitations and the background not only derive from a misconceived descriptive account, as shown in 1.3.1, but that they are also only pseudo-explanations that confuse naming with explaining.

1.3.1. The Five-Stage Model and Introspectivism

I want to begin my discussion with Hubert and Stuart Dreyfus’s five-stage model of skill acquisition (Dreyfus and Dreyfus 1986) to which Hubert Dreyfus regularly refers in his work on under-

38 standing and skillful action (Dreyfus 2002, 2007a, b)—at times explicitly under the name ‘a phenomenology of skilled behavior’ or the ‘phenomenology of skill acquisition’ (Dreyfus 1988 p.

102, 2002, fn. 1). Contrary to what these titles suggest, this model has little to do with phenomenology.

The basic assumption that the five-stage model of skill acquisition makes is that beginners learn a skill by explicit linguistic instruction and conscious rehearsal of rules. The more agents advance in a domain of skill, the less will they rely on rules and representations and at a certain stage, rule rehearsal merely diminishes performance.

It seems that a beginner makes inferences using strict rules and features just like a

heuristically programmed computer, but that with talent and a great deal of involved

experience the beginner develops into an expert who sees intuitively what to do

without applying rules and making inferences at all. (Dreyfus 1988, 106)

Importantly, Dreyfus and Dreyfus (1986), and particularly Dreyfus (2002, 2007b) do not merely claim that this is how agents experience skillful everyday action—they make the stronger claim that skillful action is not governed by an understanding that is comprised by rules and representations. As we will see in sections 1.5 and 1.6, Dreyfus and Dreyfus are right about this, but unfortunately for the wrong reasons.

Now, whether the five-stage model of skill acquisition is an adequate model of performance en- hancement, problem solving or whether it has other instrumental merits shall not concern us here.

What is important for our purposes is that this model does not derive from a proper phenomenological method and therefore does not provide an explanation of skillful action and its relationship to understanding. First, the model is not based on careful descriptive analyses of one’s own experience. Rather, it came about by a mixture of different forms of evidence, such as experiments in

39 which the performance of agents has been observed, reports by instructors of agents of various skill level and the intuitive hunches of skillful agents about the nature of the understanding that governs their skillful actions.28 That might be already enough to discredit this model.

Second, what is worse though, is the way how this model is used to arrive at explanatory and ontological posits. Dreyfus makes ontological claims about the existence of entities and their explanatory role (or lack thereof) based on the presence (or lack thereof) of these entities in experience. Yet, just because rules and representations are not ‘present’ in experience when I engage into skillful action does not necessarily entail that they are not governing skillful action beyond our awareness. But exactly this introspective inference from experience to ontological structures that govern experience is made by Dreyfus and Dreyfus.

In defense of Dreyfus and Dreyfus, it needs to be said that they are right about pointing out against the analytic tradition in philosophy, that our experience does not suggest an explanation of the understanding that governs skillful action in terms of rules and representations and that accordingly such an explanation is highly unmotivated—and as we will see in sections 1.4, 1.5 and 1.6, it is not only not motivated but as little justified as the suggestions made by Dreyfus, Dreyfus and

Kelly. But, this does not justify the latter authors’ negative ontological inference.

As we said in section 1.2, phenomenology does not arrive at explanations and ontological statements through introspective ontological inferences, even though Dreyfus and Kelly believe that phenomenology trades in ‘introspective reports’ (Dreyfus and Kelly 2007, p. 45). Experience in

28 Dreyfus and Dreyfus (1986, p. 20) state, ‘we studied the skill-acquisition process of airplane pilots, chess players, automobile drivers, and adult learners of a second language and observed a common pattern in all cases, which we call the five stages of skill acquisition.’ This makes already clear that the model does not derive from a proper phenomenological method that commences with a careful and detailed description of one’s own experience. Rather, Drey- fus and Dreyfus cite the introspective reports of novice agents (Dreyfus and Dreyfus 1986, p. 22), cite reports by instructors, such as those by an expert nurse (Dreyfus and Dreyfus 1986, p. 23), and cite reports of chess masters, such as Jan Donner and Susan Polgar, who refer to their intuitions and instincts about how they perform their actions (Dreyfus and Dreyfus 1986, p. 25).

40 itself does not directly provide us with the structure of things. To the contrary, Heidegger made clear that basic ontological structures are often among the things that are farthest remote from our everyday experience.

In demonstrating that Dasein is ontico-ontologically prior, we may have misled the

reader into supposing that this entity must also be what is given as ontico-ontolog-

ically primary not only in the sense that it can itself be grasped ‘immediately’, but

also in that the kind of Being which it possesses is presented just as ‘immediately’.

Ontically, of course, Dasein is not only close to us—even that which is closest: we

are it, each of us, we ourselves. In spite of this, or rather for just this reason, it is

ontologically that which is farthest. (Heidegger 1962, p. 36)

Instead of providing ontological and explanatory statements founded on experience directly, phenomenology arrives at such statements through an analysis of the conditions of the possibility of phenomena. Accordingly, Heidegger does not claim that rules or representations are not governing our everyday action because they are not present in our experience when we engage into skillful, everyday action. Rather, his claim is that they cannot possibly function as an explanation of the understanding that governs skillful action, because they do not possess the right structure; i.e. rules and representations do not possess the right structure so that they can function as the conditions of the possibility of skillful everyday action. I will demonstrate this claim in sections 1.4, 1.5 and 1.6.

1.3.2. Pseudo-Explanations and Reifications

It seems that Dreyfus und Kelly have not only provided negative ontological statements about understanding, but that they have also delivered positive explanations of the relationship of understanding and skillful action in terms of the intentional arc, the tendency towards maximal grip, solicitations, and the background. Unfortunately, these explanations are as unjustified as their negative ontological characterizations of understanding and skillful action. The reason for this is not only that most of their explanations derive from the five-stage model of skill acquisition and other doubtful analyses of experience. Rather, the primary reason for this is that these accounts are either pseudo-explanations or unwarranted introspectivist reifications of objects of our experience; i.e. positive introspective ontological inferences. In the following I provide a discussion of Dreyfus’s and Kelly’s explanation of understanding and skillful action in terms of the intentional arc, the tendency towards maximal grip, solicitations, and the background and show why these explanations are unjustified.

Dreyfus claims that (2002, 2007b) skillful behavior is governed by the intentional arc, a term that

Dreyfus adopts from Merleau-Ponty’s Phenomenology of Perception.

The intentional arc names the tight connection between the agent and the world,

viz. that, as the agent acquires skills, those skills are ‘stored,’ not as representations

in the mind, but as dispositions to respond to the solicitations of situations in the

world. (Dreyfus 2002, p. 367)

Even though the intentional arc is denoted here as a connection, other statements by Dreyfus suggest that we should conceive of it as a dynamic structure that is similar to a ‘feedback loop between the embodied agent and the perceptual world’ (Dreyfus 2007b, p. 250). This feedback structure is

42 a dynamic learning structure, in which learning is not constituted by an accumulation of mental representations, but by a refinement of dispositions to act.

Dreyfus makes this point more explicit by adopting Merleau-Ponty’s concept of the maximal grip.

‘Maximal grip’ denotes the disposition of an agent to literally maximize her ‘grip’ on the world, i.e. to refine her ‘dispositions to respond to solicitations’.

Maximal grip names the body’s tendency to respond to these solicitations in such a

way as to bring the current situation closer to the agent’s sense of an optimal gestalt

(Dreyfus 2002, p. 367).

As we can see from Dreyfus’s account of the intentional arc and the maximal grip, solicitations play a central role in the explanation of skillful action. Dreyfus and Kelly describe solicitations in the following way.

We sense the world’s solicitations and respond to their call all the time. In backing

away from the ‘close talker,’ in stepping skillfully over the obstacle, in reaching

‘automatically’ for the proffered handshake, we find ourselves acting in definite

ways without ever having decided to do so. In responding to the environment this

way we feel ourselves giving in to its demands. (Dreyfus and Kelly 2007, p. 52)

Solicitations are situation-specific affective ‘invitations’ of the environment to an agent to act, that are made possible by the intentional arc. As such, solicitations constitute Dreyfus’s non-cognitivist alternative to goals—the goals of skillful action.

Heidegger and Merleau-Ponty argue that, thanks to our embodied coping and the

intentional arc it makes possible, our skill in directly sensing and responding to

relevant changes in the world is constantly improved. In coping in a particular con-

text, say a classroom, we learn to ignore most of what is in the room, but, if it gets

too warm, the windows solicit us to open them. We ignore the chalk dust in the

corners and chalk marks on the desks but we attend to the chalk marks on the black-

board.29 (Dreyfus 2007b, p. 263)

I stated earlier that Dreyfus does merely provide pseudo-explanations of skillful action and understanding instead of actual explanations. Why is that the case? Are, for instance, the intentional arc and the tendency towards maximal grip not genuine explanations? The reason why they are not genuine explanations is the following. The tendency towards the maximal grip is a descriptive term. It expresses the circumstance that most agents become more skilled over time in the manipulation of objects and situations that are relevant for them. This is a correct, even though trivial description of an aspect of skillful action. To name this circumstance the ‘tendency towards maximal grip’, however, is not an explanation. It is merely a new name for the phenomenon that agents become more skilled over time in the manipulation of objects and situations that are relevant for them. Therefore, the tendency towards maximal grip is, if we interpret Dreyfus charitable, at best the attempt at a new vocabulary for a particular phenomenon that pertains to understanding and skillful action and in the worst case a pseudo-explanation.

The case is even more problematic for the intentional arc. To repeat Dreyfus’s claim about the intentional arc:

The intentional arc names the tight connection between the agent and the world,

viz. that, as the agent acquires skills, those skills are ‘stored,’ not as representations

29 Importantly, Heidegger uses nowhere the terms ‘embodied coping’ or ‘the intentional arc’. As we will see below, neither does he support anything remotely similar to what Dreyfus solicits here as an explanation.

in the mind, but as dispositions to respond to the solicitations of situations in the

world. (Dreyfus 2002, p. 367, my emphasis)

Dreyfus uses, whether intentionally or not, the right vocabulary here, i.e. ‘names’. This means,

‘intentional arc’ is merely a name for a descriptive phenomenon, not an explanation. Yet, ‘intentional arc’, contrary to ‘tendency towards maximal grip’ names not only a descriptive phenomenon, but it also presupposes an ontological assumption that is based on the unjustified negative introspective ontological inference that representations do not govern skillful action: ‘those skills are ‘stored,’ not as representations in the mind, but as dispositions to respond to the solicitations of situations in the world.’ For this reason, the intentional arc is not merely a pseudo-explanation, but also based on an unwarranted introspectivist inference about the existence of representations.

Similar considerations apply to solicitations. A solicitation is, according to Dreyfus, an entity that invites actions non-deliberatively. In that it seems at first glance that ‘solicitation’ is just another name for the phenomenon that we encounter the world as soliciting actions that are relevant to us; i.e. it seems that solicitations are another pseudo-explanation for the phenomenon that we often act non-deliberatively towards objects and situations that are relevant for us. For instance, when I am thirsty, the can of diet coke in front of me looms larger than other objects in my visual field and has a particular character that invites, differently put, motivates grasping and drinking—without me consciously deliberating to grasp the can and to drink from it.

Yet, importantly, we do not encounter solicitations in experience. Rather, we encounter a world that solicits options for action. Therefore, Dreyfus does not only provide a pseudo-explanation, but he further reifies the soliciting aspects of the environment into solicitations. Concretely, Drey- fus provides this time not a negative introspective inference with ontological ramifications, but a positive introspective inference the result of which are solicitations.

‘Background’ is another term that Dreyfus uses to characterize the understanding that governs skillful action (Dreyfus 2007a, 2012) (Dreyfus and Taylor 2015). Dreyfus claims that understanding of the world and skillful action are characterized by a pre-conceptual background that provides us with an understanding of a plethora of non-deliberatively and consciously graspable information about the world. This background is for Dreyfus further a condition of conceptual understanding.

So absorbed bodily coping, its motor intentional content, and the world’s intercon-

nected solicitations to act provide the background on the basis of which it becomes

possible for the mind with its conceptual content to think about and act upon a

categorially unified world. (Dreyfus 2007a, p. 360)

Another statement suggests that the background is nothing else other than the world.

To sum up, human beings when performing at their best are open to and absorbed

in a non-propositional, non-intentional, background field of forces that Heidegger

and Merleau-Ponty call the phenomenon of the world. (Dreyfus 2012, p. 9)

As we can see from only these two examples, the concept of background is a conceptual mess. But again, and more importantly for our purposes, is that ‘background’ is merely a name for whatever x that grants us a non-propositional understanding of the world that is complex, holistic, and action-enabling. Background is again no explanation, but merely a name for a phenomenon that requires an explanation.30 And the plethora of the different characterizations that Dreyfus has given of this phenomenon do not seem to help to provide such an explanation.

30 Analytic philosophers should not rejoice about this lack of explanation. Searle’s (1978) conception of background has basically the same lack of explanatory value as Dreyfus’s conception of background and the term ‘context’ that has become so famous among analytic philosophers of language and epistemology (Stanley 2005) is as much a stand- in for whatever x that explains context phenomena as Dreyfus’s background is a stand-in for whatever x that explains background phenomena.

We can draw the following conclusion from section 1.3. We are in desperate need of a precise and explanatory informative phenomenological account of skillful action, understanding and their relationship. Accounts by Dreyfus and Kelly are not only based on a misconception of the descriptive aspect of phenomenology—which both confuse with introspection—but also the explanatory part of their project is characterized by pseudo-explanations and ontological reifications of descriptive phenomena. In order to provide a sufficient phenomenological account of skillful action and understanding I will describe in the following the complex, body-relational structure of skillful action in terms of understanding being the condition of the possibility of skillful action.

1.4. Establishing the Conditions of the Possibility of Skillful Action

As I have shown in section 1.2, a proper phenomenological method demands a careful and detailed description of phenomena that are central to human being. It further requires an analysis of these phenomena in terms of the conditions that make them possible. In the following I provide a careful and detailed description of ordinary skillful actions, such as soccer playing and cooking. I analyze these ordinary skillful actions in terms of the conditions that make them possible. Yet, I do not focus here on all the conditions that make ordinary skillful action possible, as for instance gravity.

I confine myself to those conditions of possibility that we would be disposed to consider forms of understanding.31

31 The attentive reader might find some of the following analyses trivial. Yet, it is important to point out that they go beyond what has so far provided by both phenomenological and analytic approaches. Phenomenological approaches have primarily focused on the circumstance that skillful action requires situation-specific, sensorimotor learning (see section 1.3). Analytic approaches have primarily focused on learning too, and on the circumstance that the understanding that governs skillful action requires a grasp of the start and finish conditions of an action (this section; section 1.6). My analysis establishes further, that we require a particular spatiotemporal understanding of action that allows, among other things, to determine the pace of a skillful action. Additional to that, I show that skillful action requires

As we will see, the understanding that guides skillful action is body-relational. Body-relationality means in this context that understanding has to account for the behavior relevant relations that hold between situations and objects in the world on the one hand, and the morphological structure, affective concerns and the sensorimotor abilities of an embodied agent on the other hand. It is a condition of the possibility of action that understanding can account for these relations. Skillful action requires me to understand the objects that I use in my actions as for action which means as much as for my body. This means that understanding needs to be able to disclose the particular behaviors that objects afford to an embodied agent; an agent with a certain body, with particular hands that can perform certain movements and so forth. For instance, I have to understand that the structure of my body does not allow me to grasp water as a firm object, but rather as something in which I can swim; among many other things. I need to understand that a cup of coffee allows me to grasp it in a particular way—given, for instance, the size of my fingers. Accordingly, in one sense of an affordance (Chemero 2009), I need to understand things as co-relational; as co-relational with regard to my gross-morphological bodily structure, affective concerns and sensorimotor abilities. An object is only intelligible for action, so to speak, if I understand its properties relationally to mine. Or, as Merleau-Ponty (2012, 140) expressed it, ‘to move toward an object, it needs to exist for the body’. In this sense, understanding has to be body-relational.

In the following I focus on two body-relational dimensions of the understanding that governs skillful action. First, the body-relational understanding of spatial and temporal aspects of the environment and the agent’s body that is necessary to determine the pace with which an agent has to conduct skillful actions. Second, the body-relational understanding of affective concerns that is

an understanding of one’s affective concerns to determine the relevance of certain behaviors and to determine the effort that goes into an action.

48 necessary to evaluate whether a possible action is in accord with the goals of an agent, that determines the effort that an agent should exert in a situation and that assesses the relevance of a particular behavior of an agent in a situation.

First, an agent has to understand the spatial and temporal aspects of objects relationally to the spatial and temporal dimensions of her embodiment—such as the location and extension of objects and their temporal changes in relation to her effector space and movability—in order to determine the pace with which she has to act. Her pace is a function of her bodily, sensorimotor abilities and of the spatiotemporal structure of the environment. Accordingly, she needs to understand spatial and temporal aspects of the environment relative to her own. In order to understand whether she can catch a ball—whether she is fast enough to do so—she needs to understand how long it will take her to stride through the space in front of her, which presupposes that she understands how fast she, as the particular embodied agent that she is, can move on the particular surface structure of the space that she has to traverse. She needs to understand how—for instance—wind and rain, affect her sensorimotor ability to move and she needs to understand the movements that she has to make, given the particular environmental conditions, such as a wet and slippery track, to preserve the pace that she is required to exert to perform the action at hand.

Into the category of spatial and temporal aspects of the understanding that governs skillful action fall also what Stanley and Krakauer (2013) have identified as initiation conditions.

In the case of virtually any activity ϕ, having skill at ϕ-ing requires knowing what

to do to initiate actions of ϕ-ing. (Stanley and Krakauer 2013, p. 4)

Importantly—contrary to what Stanley and Krakauer fail to assert—the understanding of initiation conditions is always relative to an agent’s body and requires an intimate understanding of the spatiality and temporality of the situation in which she acts. In order to make coffee, she needs to

49 understand when to open the coffee machine’s lid, when to pour coffee into it, where the coffee is and so forth. This ‘when’ and ‘where’ is relational to her position, her effector reach and her sensorimotor abilities. Opening the lid of a coffee machine requires her to understand how much force to exert and where to stand, given the position and weight of the coffee machine.

I have asserted above, that skillful action necessitates, second, an understanding of my affective and thereby embodied concerns (Heidegger 1962; Taylor 1985). This means that an agent has to understand that the action that she is about to perform is an action that she wants to perform, given the complex hierarchy of goals that structures not only her everyday living, but even the most minor activities into which she engages. She has to understand that the particular action that she wants to perform belongs to the set of actions that she actually should want to perform, given her interests and goals. If she plays soccer, she needs to understand whether it is more beneficial to run along the sideline, to tackle the player in front of her or to hit the ball wide. Not only does this require an overall understanding of her goals in that situation—say, winning the game in grace or prohibiting a crushing defeat—but also an understanding of her sensorimotor ability level in that situation, with respect to her opponent, and an understanding of her situational fitness and level of exhaustion. The latter means, that I need to understand how effortful the action should be, which relates to the relevance of this single action episode for my overall goal of either winning or prohibiting a crushing defeat (Bower and Gallagher 2013). A novice soccer player will charge after the ball no matter what, without an understanding that this will exhaust her too soon. And obviously, an understanding of one’s endurance relative to the game relevant movements is an important skill, imperative for successful action.

Cooking is another action that can be performed with more or less skill. Importantly, my cooking action requires a concernful understanding of the entire chain of behaviors that I have to engage

50 into in order to produce a result I have concern for, say, a lentil stew of a particular quality and make—the work towards-which my concernful dealings, i.e. actions, aim (Heidegger 1962). In that sense, concernful understanding is the regulative ideal that anticipates the goal of an agent’s actions and guides their coordination. Not only that, concernful understanding is a condition of the possibility to finish an action, because concernful understanding enables an agent to realize when the chain of behaviors in which she engages should come to an end: then, when the object of her concern—the lentil stew—is completed to her personal satisfaction. And a situational level of personal satisfaction is not an abstract finish condition that would correspond to a rule, as Stanley and

Krakauer (2013) purport, but a concernful, affective state of mind.

The cooking example also shows us how the most mundane action requires an understanding of the effort and relevance that should be put into an action or assigned to it, in relation to an agent’s bodily ability level and the temporality of an action. An action requires an understanding of the level of, for instance, diligence and detail an agent should put in chopping carrots, given the overall complexity of the task, the time at her disposal relative to it, and the other obligations and concerns that determine how much time she should spend on cooking in the first place. Concretely, she is required to understand her concerns in relation to her bodily, sensorimotor competence level at the task at hand and the overall temporal course of time that her cooking action should reasonably take.

From the circumstance that we are disposed to call these action-enabling structures ‘understanding’—or ‘knowledge’ for that—it certainly does not follow that this understanding—or knowledge—possesses propositional or representational structure. The claim that understanding possesses propositional or representational structure is not a trivial claim—pace Krakauer, Stanley and Williamson—but presupposes a rationalist or intellectualist understanding of understanding

51 that is not further critically questioned (Heidegger 1962; Noë 2004, 2012). Neither does it follow from the circumstance that this understanding is not directly consciously present or deliberative, that we should not call it understanding—or knowledge—pace Dreyfus. I will spell out in the next section how to properly determine the structure of understanding so that it can body-relationally guide skillful action.

Based on the careful description of ordinary skillful action and an analysis of skillful action in terms of the conditions of its possibility, I have shown in this section that the understanding that governs skillful action is body-relational. I have demonstrated this for the spatiotemporal understanding that determines the pace with which we perform an action and the concernful understanding that determines the goals and relevance of our actions as well as the effort that we put into them. I will provide in the next section an analysis of the structure that understanding has to exhibit so that it can determine the pace of our skillful actions as well as the concernful dimension that it possesses.

1.5. Structures of Understanding

In the following I establish the structural conditions that make it possible that understanding can be body-relationally about the spatiotemporal and concernful aspects of skillful action. First, I demonstrate based on Kant’s synthesis of conceptual understanding and experience that understanding has to exhibit spatiotemporally schematic structure so that it can be about the spatiotemporal aspects of skillful action. Second, I argue that we need to conceive of understanding in embodied terms, for instance, in sensorimotor terms, so that we can make sense of the circumstance

52 that the spatiotemporal aspect of understanding is body-relational. Third, and finally, I argue with

Heidegger that we need to conceive of understanding and affect as co-constitutive of each other, to make sense of how understanding can be about the concernful aspects of skillful action.

1.5.1 Body-Relational Spatiotemporal Structure

As is well known, Kant argues in the Critique of Pure Reason that conceptual understanding and experience have to exhibit commonalities so that experience can be put under concepts like ‘substance’ and so that conceptual understanding can receive an object (Gegenstand) as a referent through experience. Now, what is regularly overlooked in contemporary literature, foremost in

John McDowell’s (1994) interpretation of Kant’s synthesis of conceptual understanding and experience, is that Kant provides a structural synthesis of concepts and experience, because Kant is aware that we have to establish the conditions of the possibility of the interaction between conceptual understanding and experience in order to determine which structure concepts and experience have to exhibit so that they can actually be synthesized.

Without establishing the conditions of the possibility of the interaction of conceptual understanding and experience, we have no criteria to evaluate whether the particular conception of understanding and experience that we adhere to actually allows that both stand in a close, interacting relationship, and accordingly whether we are justified in adhering to the particular conception of understanding or experience respectively. It is this aspect of an analysis of the conditions of the possibility of phenomena that makes it epistemically superior to contemporary phenomenological and analytic accounts. An analysis of the conditions of the possibility of phenomena gives us not

53 only an insight into the structure of phenomena, but it also shows us whether we are justified in conceiving of the phenomena that we analyze as we do.

Kant developed a conception of experience and conceptual understanding that allows for their interaction, based on an analysis of which structure both have to exhibit so that they can possibly interact. As we will see below, Kant argues that concepts are governed by spatiotemporal schemata and that experience is characterized basically by spatiality and temporality. Accordingly, both share spatiotemporal structure and can interact.

Concretely, Kant argues that the common structural ground that understanding and experience share are schemata. A schema is a mediating structure, that preserves the unique nature of experience and understanding, while it at the same time acts as the common ground between both.

Now it is clear that there must be a third thing, which must stand in homogeneity

[in Gleichartigkeit stehen] with the category on the one hand and the appearance

on the other, and makes possible the application of the former to the latter. (Kant

1998, A138/B177, emphasis and translation in original)

Importantly, conceptual understanding (here: categories) and appearances (here: experience) require something that they have in common so that both can interact. If experience were structurally unintelligible for concepts, concepts had no way to refer to them. And if concepts were structurally unintelligible for experience, experience could not be ordered by concepts. For instance, if a concept were a word, and if an experiential episode were an image, then there were no structural means of their connection, because a concept and an image do not have a ‘third thing’ in common—or at least not a third thing that could function as a principle of interaction.32 Accordingly, both could

32 The point here is not that I argue for a particular conception of reference or aboutness, such as causal covariance, causal history and so forth. Rather, I argue that whatever aboutness and reference are, we need to account for the

54 not possibly interact as we conceive of them as interacting. Therefore, we have to determine a condition that allows for their interaction: schemata.

Schemata themselves are ‘a priori time-determinations’ (Kant 1998, A145/B184)—and we might add ‘a priori space-determinations’. What that means becomes clearer if we focus on Kant’s conception of experience. Kant establishes in the Transcendental Aesthetic of the Critique of Pure

Reason that space and time are the conditions of possibility of experience. That is the case, among other things, because we cannot experience an object without it being located in space and time.

Take for instance Kant’s transcendental argument that I have already introduced in section 1.2:

One can never represent [rather: imagine/vorstellen] that there is no space, although

one can very well think that there are no objects to be encountered in it. It is there-

fore to be regarded as the condition of the possibility of appearances, not as a de-

termination dependent on them, and is an a priori representation [rather: imagina-

tion/Vorstellung] that necessarily grounds outer appearances. (Kant 1998,

A24/B38, square brackets mine)

This makes spatiality and temporality essential properties of experience.33 That spatiality and temporality are essential properties of experience allows schemata, as a priori time- and space-determinations to relate to experience, because both have spatial and temporal structure in common.

For instance, I experience objects as substances because they are spatially persistent in time. The concept of substance can be about substances, because it is determined by a schema that determines something as a substance if it is spatially persistent in time.

conditions that allow that they can come about, given the particular conception of understanding and world that one adheres to. 33 Space and time are for Kant not only essential properties of experience, they are the basic, pure forms of intuition for him. However, for the purpose of this chapter this does not need to concern us.

As I have shown in section 1.4, skillful action requires, among other things, an understanding of pace. Pace requires an understanding of various actual and potential spatial and temporal relationships that hold between an embodied agent and her environment. The Kantian considerations above allow us to partially determine the structure of the understanding that provides such spatial and temporal information; i.e. they allow us to conceive of understanding as exhibiting spatiotemporally schematic structure that is the condition of the possibility of understanding being about the spatial and temporal relationships that hold between an embodied agent and her environment.

We do not necessarily have to accept this Kantian analysis. Depending on how we characterize, for instance, experience as that what presents environmental information, we can develop a different analysis of the conditions of the possibility of the interaction between understanding and the environment in which skillful action occurs. Yet, what is important is that we have to provide an analysis of the structural conditions of the possibility of understanding being about the world in which skillful action unfolds. Because if we do not do so, we have no criteria whatsoever that would allow us to assess whether the particular conception of understanding that we have descriptively individuated is actually able to govern skillful action.

For instance, I have argued in section 1.2.2 that Dreyfus’s intentional arc is merely as pseudo- explanation of understanding that is just a name for a descriptive circumstance. Differently put, it cannot even be criteriologically evaluated whether the intentional arc is governing skillful action because the intentional arc is not characterized in terms of a structure based on which we could assess whether the intentional arc would be able to function as the condition of the possibility of skillful action. Compared to this, my analysis of understanding in Kantian terms allows at least for the following characterization. Whatever else the structure of understanding is, it has to exhibit

56 spatiotemporal schematic structure. What is so important about this, is that this explanation provides strong criteria for what the understanding that governs skillful action is. For instance, it is not a structure that could be about something that is not spatiotemporally imaginable or a structure that is primarily definable by its syntactical properties.34 Such strong structural criteria do not only prohibit pseudo-explanations à la Dreyfus and Kelly as described in section 1.3, but they also undermine the inflationary wide and uninformative explanations of skillful action provided by

Krakauer, Stanley and Williamson in terms of facts, representations and knowledge-that, as I will show in section 1.6.35

Even though a Kantian analysis leads into the right direction, it is not sufficient to explain the spatiotemporal structure of understanding, since it does not account for how spatiality and temporality are understood with regard to an agent’s bodily structure and sensorimotor abilities. Even though Kant’s analysis is methodologically brilliant, Kant unfortunately still conceives of the agent descriptively as a disinterested, disembodied observer and reasoner. What is required for a sufficient analysis of understanding is an explanation that does not only show that understanding has schematic structure, but that it has body schematic structure so that it can relate to how an agent understands spatiality and temporality in the context of her particular embodiment.36

34 Mind that these points about the understanding are merely analytic entailments that are arrived at through an analysis of the conditions of the possibility of understanding. 35 Hurley (1998) and Noë (2004) have recently argued for a similar point. Hurley has argued that the connection between the mechanisms that produce action and the mechanisms that produce perception cannot be arbitrary, i.e. that we have to account for the structural connection between both of them. Noë has argued that a functional description of perception and action cannot be autonomous from the particular embodiment of the agent that perceives and acts. However, both authors, though they rely heavily on Kant or Heidegger, do not argue transcendentally for these claims, but naturalistically, which is why I do not further discuss them here. 36 It seems that Merleau-Ponty’s (2012) account of the body schema might give us insight here. Unfortunately, Mer- leau-Ponty’s account of the body schema is less accessible than Kant’s considerations on schemata. Though I believe that Merleau-Ponty’s thoughts on the body schema—qua an analysis of Husserl’s (1989) considerations on the body and constitution—are a logical extension of Kant’s conception of schemata that include embodiment, I will not attempt to reconstruct his account here and rather provide a logical extension of Kant’s account of schemata in body-relational terms.

In order to make sense of how understanding can exhibit body schematic structure we have to conceive of the structure of understanding and the structure of experience with which we perceive skillful action unfolding, in embodied terms. Therefore, I suggest that we conceive of the understanding that governs skillful action as intrinsically embodied, for instance, in a sensorimotor format, i.e. in a way in which objects are conceived of according to the behavioral and perceptual patterns that they afford in relation to our embodiment (Noë 2004). Further, I suggest that we conceive of experience not as Kant did, in disembodied terms, but with Husserl and Merleau-Ponty as reflective of our embodiment and conceive therefore accordingly also of the spatiality and temporality of our experience in body-relational terms (Husserl 1989; Merleau-Ponty 2012) (for a detailed account of this embodied version of experience see chapter 3, in particular section 3.2.2).

That experience is embodied means as much as that our embodiment is co-constitutive of the contents of experience so that we do not experience things as part of a disembodied space, but as graspable, close, easily-reachable, out of reach, far away and so forth. This embodied conception of our experience is in accord with our experience. We indeed do not experience a world of objects for which we do not care in objective space and time. Rather, we experience objects as directly relevant for us in certain ways.

I do not claim that experience is embodied, because we experience the world as reflective of our embodiment, even though this motivates my claim. However, if I would infer the claim that experience is embodied from a mere descriptive analysis of experience, I would commit the introspectivist fallacy that I accused Dreyfus and Kelly (2007) of when they analyze how we encounter the world in terms of solicitations. Rather, my point here is that only if we conceive of understanding and our experience as characterized by our embodiment, can we make sense of how they could possibly interact. That means, only if our understanding exhibits information about the spatiality

58 and temporality of our environment in body-relational terms and only if our experience itself is presented to us body-relationally, is there a possible means of their interaction: spatiotemporal schematic structure that is embodied. This means from an explanatory point of view, that the understanding that governs skillful action is not only spatiotemporally schematic, but it is embodied spatiotemporally schematic.

1.5.2 Understanding and Affect

Similar considerations apply to our understanding of concernful aspects of our skillful action. Sec- tion 1.4 made clear that an agent requires an understanding of the goals of her actions, an understanding of the effort that she should put into a behavior and an understanding of the factors that determine the relevance that she should assign to a particular behavioral outcome. Heidegger’s

(1962) considerations on the care structure, which suggest that understanding and affect are co- constitutive of each other, allows us to understand how the structure of the understanding can be about goals, effort and relevance.

As I have said in section 1.4, goals, effort and relevance pertain to an agent’s concerns. That I do not eat a muffin for breakfast is dependent on my concern for losing weight—my goal to lose weight. That I put particular effort into my work from 9-5 depends on my concern for having a relaxing evening thereafter. That I perceive lettuce as a particularly relevant stimulus in my visual field depends on my concern for healthy eating. Accordingly, we can say, that our understanding of goals, effort and relevance in relation to a skillful action, pertains to my concerns, or as

Heidegger (1962) would say, to an agent’s ‘care’ structure. As I said in section 1.2, Heidegger argues that the care structure is comprised by ‘understanding, state-of-mind, falling, and discourse’

(Heidegger 1962, pp. 384-385), where ‘state-of-mind’ is the translators’ term for the German ‘Be- findlichkeit’, which I translate in the following with ‘affect’.

In particular, the relationship between understanding and affect shall concern us in the following, since it seems that the interplay of understanding and affect is the basis of our concerns. I want to live healthy because I am afraid of being sick and because I enjoy being in good shape. I temper my efforts, because I am threatened by exhaustion and rejoice being rested. I see certain things as relevant because I am happy about reaching my goals and sad about not achieving them. Focusing in the following only on the interplay between understanding and affect further simplifies the subsequent analysis in terms of the structural conditions of a possible interaction. The reader is wel- come to further incorporate falling and discourse into her analysis, however, should keep in mind that this will render an explanation of the structure of understanding significantly more complex.

Now, if we assume that concerns require affect in one form or another, we need to explain how understanding and affect can stand in a relationship towards each other. Heidegger’s answer to this is that both, understanding and affect, are co-constitutive of each other. ‘Every understanding has its mood. Every state-of-mind (affect) is one in which one understands’ (Heidegger 1962, p. 385, brackets mine). Further, as Heidegger states, ‘to any state-of-mind (affect) or mood, understanding belongs equiprimordially’ (Heidegger 1962, p. 315, brackets mine) and ‘understanding always has its mood’ (Heidegger 1962, p. 182).

As I already stated above, contrary to Dreyfus and Kelly, the point here is not that understanding and affect are co-constitutive—for instance, for solicitations to come about—because things show up experientially in a way so that I understand them affectively as things that matter to me. That again would be an introspectivist inference that would not be justified. Rather, the point is, like in the case of body-relational spatiotemporal schematic structure for understanding and experience,

60 that we can only make sense of the interaction between understanding and affect, if they are co- constitutive. Differently put, we can only make sense of the interaction of affect and understanding if they are integrated with each other, because we cannot imagine another principle of their interaction. Why is that? Take for instance the idea that understanding is sensorimotor or representational and that affect is, simply that, affective. What could be the structure that brings them possibly together, as for instance spatiotemporal structure in the case of schematic understanding and spatiotemporal aspects of the environment?

Indeed, it seems that we cannot conceive of a structure that can function as a means of their interaction. If understanding were housed in a sensorimotor format, as for instance Noë (2004) argues, there is no common structure of interaction that affect would share with sensorimotor understanding. If understanding were housed in a representational format, as for instance Stanley and Kra- kauer (2013) argue, there is no common structure of interaction that affect would share with, for instance, a symbolic representation.37

Now, one might claim that affect and understanding interact through a transduction process—and a similar argument like this might be advanced in the context of my claim that understanding and experience have to be bridged through spatiotemporally embodied schematic structure. Yet, to claim that a transduction process turns, for instance, affect into a representational desire state or an objective spatiotemporal representation into an embodied one, is neither an answer to the transcendentalist challenge that I present here, nor constituting a satisfactory explanation on its own.

‘Transduction process’ is rather a filler word for ‘whatever x that makes a theory of understanding

37 I want to thank Anthony Chemero (in conversation) for pointing out that the integration of embodied spatiotemporal understanding and affective concern might be the basis for an explanation of the experiential phenomenon that we experience spatial and temporal distances often according to our moods.

61 coherent that cannot provide a structural explanation of the interconnection of phenomena in terms of the conditions that make this interconnection possible’.

Differently put, ‘transudation process’ has become a pseudo-explanatory term that is supposed to overshadow an explanatory gap in one’s theory. This matters in particular for phenomenologists or classic ordinary language philosophers, like Stanley and Williamson, who formulate their theories in personal level terms. ‘Transduction process’ is, besides being pseudo-explanatory, not a personal level term and should therefore be shunned.38 Likewise, to claim that affect is a disembodied representational state, that is not primarily characterized by its qualitative nature and which can directly interact with understanding qua being a disembodied representational desire state is descriptively unjustified and seems to have, too, primarily the function to save a theory of understanding that cannot explain the interaction of its parts in terms of the conditions that make this interaction possible, from explanatory collapse.

To conclude this section, our analysis of the structural conditions of the possibility of understanding that informs action has established that the understanding that governs skillful action exhibits the following structure. Understanding is integrated with affect and possesses embodied spatiotemporal schematic structure.

38 For phenomenologists, it is important that we use personal level terms, because they are about experiential phenomena. For ordinary language philosophers, it is obviously important to use personal level terms, because only personal level terms are part of ordinary language.

1.6. Krakauer, Stanley and Williamson on Skillful Action and Understanding: Lack of De- scriptive Precision and Criteria

I began this chapter with Stanley’s and Williamson’s ‘criticism’—or rather, handwavy and unin- formed dismissal—of phenomenological accounts of understanding, skillful action and their relationship. I hope that sections 1.2, 1.4 and 1.5 have shown that this criticism is without substance.

Instead, I hope that the reader familiar with analytic accounts has already noticed that the phenomenological account provided here has significant explanatory advantages over analytic ones.

In the following I swiftly spell out two of these explanatory advantages, taking Krakauer’s, Stan- ley’s and Williamson’s account of skillful action and understanding as an example for analytic accounts. The first explanatory advantage is the analytic precision of a proper phenomenological account when it comes to determining the structure of skillful action (section 1.4). I will not say much about this advantage here other than that the reader is invited to make her own picture based on the complex characterization of only a few features of skillful action that I have provided in section 1.4, compared to Krakauer’s, Stanley’s and Williamson’s analysis in the following.

The ﬁrst [criterion for an x being a skillful action] is that having skill at ϕ-ing re-

quires being trained past baseline. The second is that there is a large class of skills,

the ones the manifestations of which are intentional actions, such that having skill

at ϕ-ing requires knowing what to do to initiate an act of ϕ-ing, or more colloquially,

how to start ϕ-ing (knowledge that may be different in different circumstances).

(Stanley and Krakauer 2013, p. 7, square brackets mine)

First, learning is a trivial aspect of skillful action that has been identified as a central feature of skillful action for nearly four decades by Dreyfus (section 1.3)—and that has been accounted for by Heidegger (1962) in particular through the social aspects of the care structure. Second, initiation conditions are, as I have shown in section 1.4, merely a species of a more complex understanding of the spatiotemporal and concern-related aspects of skillful action. Accordingly, analyses by ordinary language philosophers falls short of the complex descriptive phenomenological accounts that analyze skillful action in terms of its conditions of possibility. This is in particular the case since analytic approaches seem to reduce the complex procedure that goes into an analysis of actions to analyses of action verbs which themselves might even be reduced to ‘phi-ings’, as the quote above shows.

What is more concerning though than the lack of analytic precision when it comes to the description of the complexity of the structure of skillful action, is the lack of analytic precision in determining the structure of understanding.

Stanley and Williamson assert that skillful action requires an explanation in terms of knowledge of facts.

So, in a very clear sense, skilled action is action guided by propositional knowledge,

the propositional knowledge that is revelatory of that agent’s skill. (Stanley and

Williamson 2016, p. 6)

Further,

In our view, skilled action is action guided by ongoing accrual and improving ap-

plication of knowledge of facts about an activity (…). (Stanley and Krakauer 2013,

p. 2)

The problem is that Krakauer, Stanley and Williamson are evasive about specifying the structure of knowledge-that and facts. Whereas the phenomenological account presented in this chapter, determines the structure of understanding as integrated with affect and as possessing embodied spatiotemporal schematic structure, ‘facts’ and ‘knowledge-that’ seem criteriologically unbound in their application. The problem here is not even that ‘fact’ and ‘knowledge-that’ are used so liberally that they are basically meaningless, even though this is reason for concern too. Every situation of skillful action, co-constituted by the environment and the embodied agent, can be called a ‘fact’. Whatever can be called ‘understanding’ is allegedly a form of ‘knowledge-that’, because it can be called so, independent of the structural constraints that an analysis of the conditions of the possibility of understanding has determined. Rather, the actual problem is that ontological posits are implicitly made here.

If the structure of knowledge-that is methodologically best explained by an analysis of language, in particular declarative sentences, then it seems that the structure of declarative sentences is exhaustive of or at least characteristic of understanding. Yet, it is unclear in which sense something that is primarily structurally characterized by its symbolic and syntactic properties could exhibit something like affective or embodied spatiotemporally schematic structure or any other structure that understanding has to exhibit so that it can guide skillful action.

Similarly, if the paradigm for the analysis of understanding—or knowledge-that—are declarative sentences, we should expect that it makes sense, that understanding is exhaustively expressible in language. Yet, this is obviously not the case. Stanley and Krakauer defend their claim against such an allegation in the following way.

But there is simply no reason to conclude that concept possession generally requires

the capacity to linguistically articulate. There is no reason to deny that at least some

animals without a capacity to verbally articulate their content nevertheless possess

the same concepts we do. The capacity to have attitudes to propositions generally

does not depend on the capacity for verbal articulation. (Stanley and Krakauer

2013, p. 7)

What is problematic, if not bizarre, about such a defense, is that the case of animals seems to exactly speak against an account as theirs and seems to rather support accounts of skillful action and understanding, as the one proposed in this chapter, that argues that understanding, i.e. that what performs functions Stanley and Krakauer ascribe to concepts, is not housed in a structure that is linguistic or quasi-linguistic. Such a phenomenological account makes sense of the idea that animals and humans not able to use language still have a rich understanding that allows them to generate skillful actions. The problem here is again that analytic accounts such as Stanley’s and

Krakauer’s have no structural constraints on claims about understanding. Accordingly, implausible post hoc additions are made to their account, such as that even though their paradigm for understanding are declarative sentences and even though their method consists primarily in linguistic analysis, understanding has supposedly not to be linguistic itself or verbally articulable.

Why have so many philosophers and scientists wrongly thought that knowing a fact

requires the ability to verbally articulate it? This is because some examples of

knowledge do seem to be ones that characteristically manifest themselves in ver-

balization. But to take this as a requirement on factual knowledge generally is to

mistake a property of some instances of a type with an essential property of the

type. (Stanley and Krakauer 2013, p. 7)

The reason why ‘so many philosophers and scientists’ thought that understanding requires verbal articulation is that they were probably concerned about what we might want to call ‘structural

66 coherence’. If we assume that the paradigm for the analysis of the structure of understanding is linguistic, as Krakauer, Stanley and Williamson do, it makes only sense to expect that the structure of understanding is linguistic and accordingly it makes a lot of sense too, to expect that understanding is verbally articulable. Accordingly, it seems that Stanley and Krakauer confuse ‘a property of some instances of a type with an essential property of the type’ (Stanley and Krakauer 2013, p.7). It is ‘a property of some instances’ of understanding that they can be expressed—or perhaps merely be described—with declarative sentences. They infer from this, that the structure of understanding, the ‘essential property of the type’, is structurally equivalent to declarative sentences, even though not necessarily expressible through them. But as I have shown above, if we do not analyze the structural conditions of the possibility of understanding, we have no actual criteria to make explanatory or ontological claims about the structure of understanding. And this leads exactly to the situation described here, where it seems that arbitrary, post hoc claims about the structure of understanding are the only thing at one’s disposal—where even though one states that understanding is structurally equivalent to declarative sentences, it is for some arbitrary post hoc reason not linguistically expressible.

It is this uncritical and structurally unbound use of ontological posits and the explanatory vague- ness of rationalist terminology, that motivated Kant and his phenomenological successors to analyze the conditions of the possibility of phenomena to determine precise and informative criteria for the use of concepts. And I hope the reader will see too, that an analysis of the conditions of the possibility of skillful action, understanding and their relationship beyond the present analysis, is desperately needed.

1.7. Conclusion

A proper understanding of the phenomenological method allows us to analyze skillful action and the understanding that governs it in the following way. First, we have to carefully and in a detailed fashion identify the central aspects of our experience of skillful action and understanding descriptively. Second, we have to analyze the central aspects—the basic phenomena that comprise our experience of skillful action and understanding—in terms of the conditions of their possibility.

This proper understanding of the phenomenological method allowed us to see, that contemporary phenomenological accounts, such as those by Dreyfus and Kelly, are methodologically flawed and provide therefore unjustified and mostly false results.

As an alternative to these faulty accounts, I have argued, based on a proper phenomenological method, that the structure of skillful action necessitates us to conceive of the understanding that governs skillful action in body-relational terms. I have then provided an analysis of the structure of understanding that established the conditions that have to hold, so that understanding can govern skillful action body-relationally. This analysis established that understanding possesses embodied spatiotemporal schematic structure and that understanding is integrated with affect. Only if understanding exhibits this structure can we explain in terms of the conditions of the possibility of understanding governing skillful action, how understanding can determine the pace of a skillful agent’s behavior and how understanding can monitor that the behavior that an agent produces is in accord with her concerns.

Finally, I have shown that contemporary analytic accounts, such as those by Krakauer, Stanley and

Williamson fall methodologically short of a proper phenomenological account too. In particular, analytic accounts lack precision in establishing the structure of skillful action descriptively and

68 they fail to provide informative criteria that would allow to determine the structure of the understanding that governs skillful action.

The account presented in this chapter invites refinement. The major theoretical contribution of this chapter is to show how a proper phenomenological account allows to establish the structure of skillful action and understanding more appropriately than contemporary phenomenological and contemporary analytic accounts. Therefore, I have merely established the structure of skillful action and understanding with regard to two aspects of both of them. The pace of skillful action and the concern-related aspects of skillful action—and accordingly the understanding that is necessary to determine the appropriate pace and the understanding that is necessary to govern behavior so that an agent’s skillful action is in accord with her concerns. There is obviously a plethora of further functional and structural dimensions that characterize skillful action and understanding.

These will have to be identified and established in future phenomenological work on the nature of skillful action and understanding.

Chapter 2. Concern and the Structure of Action: The Integration of Affect and

Understanding

Abstract

In this paper, I propose a theory of action based on a Heideggerian conception of concern, that integrates (embodied) understanding and affect. I proceed in three steps. First, I provide an analysis that identifies four central aspects of action that a satisfying theory of action needs to explain: i) the spontaneous aspect of behavior, which pertains to how action is motivated by situational interests. ii) The receptive aspect of behavior, which is about how a background of life-projects and long-term interests constrains action. iii) The spontaneous and iv) the receptive aspects of understanding, which pertain to how understanding provides a non-representational background for possible actions. Second, I provide an analysis of the i) spontaneous and ii) receptive aspects of behavior in terms of concern. Thereby I establish that concern structures the direction and manner of an action, which explains a) autonomous and purposeful behavior without recourse to representational vocabulary (e.g. goal, desire, intention) and b) how situational and long-term interests can interact through the same structure. Third, I show that we understand the world concernfully by iii) spontaneously not only contributing embodied abilities to our understanding of the world but also affect, which generates iv) a passive-receptive background for action. This passive-receptive background allows to explain a) how most of our actions can be non-deliberative and direct, without recourse to belief-desire psychological reasoning, b) how we can want to engage in an action the first place, and c) it finally reveals how all aspects of action are united through concern.

2.1. Introduction

Action is central to human being; a circumstance that has been in particular emphasized by existential phenomenologists (Heidegger 1962; Merleau-Ponty 2012). In accord with their work and inspired by their insight, embodied cognition has put action center stage and made it central to explanations of cognition and perception.

Claims about the centrality of action range from statements like ‘cognition is for action’ (Glenberg

1997; Prinz and Clark 2004) to ‘action, perception and cognition are integrated with each other’

(Hurley 1998; Thompson 2007) to ‘cognition is a form of action’ (Noë 2012). Action-oriented representations, skilled coping, and action affordances are central concepts of the paradigm

(Chemero 2003; Anderson and Rosenberg 2008; Gallagher 2008; Rietveld 2008; Wheeler 2008).

However, despite the efforts of proponents of embodied cognition to explain the environmental and bodily constituents of action, such as motor control, the body schema, skill, affordances, and sensorimotor knowledge, the research program is surprisingly quiet about the phenomenon of action itself. McGann goes as far as claiming that embodied cognition lacks a theory of action:

Though exalting action, researchers and writers within these new ways of thinking

(embodied cognition) have tended to gloss over just what they mean by the term.

(McGann 2007, p. 464, brackets mine)39

Independent of whether McGann’s evaluation is correct, I doubt that embodied cognition possesses an adequate depiction of the phenomenon of action and an adequate theory of action in embodied terms. Therefore, I propose in the following, first, an analysis of the structural components of

39 Similar concerns have been voiced by Dotov and Chemero (2014) with regard to ecological psychology.

71 action that identifies and establishes the basic structure of action. Second, I propose an explanation of action based on a Heidegger-inspired account of concern, that is based on an integration of affect and embodied understanding.

An adequate conception of action has to conceive of action, as Dreyfus and Taylor remind us,

As that of an engaged agent, determining the significances from out of its aims,

needs, purposes, desires. These significances arise out of a combination of sponta-

neity and receptivity, constraint and striving, they are the ways the world must be

taken in for a being, defined by certain goals or needs to make sense of it. (Dreyfus

and Taylor 2015, p. 69)

Dreyfus and Taylor identify correctly what is missing from an adequate conception of action in embodied terms: An account that shows a) how the understanding that informs action is both spontaneous and receptive and that further demonstrates b) how our behavior itself is both spontaneously (‘striving’) directed at the world and at the same time receptively-passively ‘constrained’ by the world, our own life-projects and long-term interests.

Suppose I want to hammer a nail in order to hang a picture. In order to do so I need to understand what a hammer does and how to use it. This understanding has to be both spontaneous and receptive. Without my embodied abilities40 and my interest in the results of potential actions that I can achieve with hammers—the ‘work’, in Heidegger’s words—I could not understand what a hammer

40 Embodied abilities are the abilities and skills that are necessary to perform various forms of sensory and motor behavior, which renders embodied abilities instrumental for action. Embodied abilities range from the ability to move one’s legs in rhythm in order to enable walking, to the ability to move one’s eyes so that one can read, to the motoric and auditory abilities that are necessary to produce speech. But embodied abilities also scale up, as Rietveld and Kiverstein (2014) have convincingly argued, to complex social abilities, as those of architects and musicians—even though they remain properly grounded in more basic embodied abilities.

72 is and how to use it.41 Yet, without being receptive to the physical and spatiotemporal properties of hammers—their material structure—I could not understand what a hammer is and how to use it either. Further, without the world showing up as a passive-receptive horizon for the realization of the concerns that motivate my behavior I could not want to hammer in the first place.42

So far I have merely listed aspects of action that allow me to understand how to use a hammer.

This understanding, in combination with the presence of, say, a hammer, a nail, a wall and a picture is not yet sufficient to account for my action of hanging a picture.43 The action requires that I want to hammer a nail to hang a picture—it requires me to have concern in a situation to act. Addition- ally, this situational concern is embedded into a background of life-projects and long-term interests. I will only actualize the action of hammering a nail if I am not constrained by a wider set of long-term concerns of mine. If I am giving a lecture about Heidegger’s conception of readiness- to-hand, and have brought a hammer to my lecture to exemplify this existential mode of engagement, I will not engage in hammering just because a hammer is present. The context of the lecture, in which I pursue one of my life-projects, being a professor, does not permit the action to occur.

41 Accordingly, people with bodies different from ours or bodily abilities different from ours as well as animals might have a significantly different understanding from ours. 42 I use ‘receptive’ with regard to understanding in two different senses in this chapter. The first sense is about how the mere physical environment contributes to our understanding of the world. The second sense, is about how understanding functions as a passive background or horizon for action. If I use it in the first sense, I merely write ‘receptive’. If use it in the latter sense, I write ‘passive-receptive’. Importantly, the passive-receptive aspect of understanding does not correspond symmetrically to the spontaneous aspect of understanding that I also describe in this chapter. This spontaneous aspect is about what we contribute to our understanding of the world and it is only important for this chapter in so far it is about how concern, as that what is a necessary condition for having interests and motivation, is a co-originary contribution of ours to the passive-receptive background for action, i.e. understanding. 43 Understanding comes in different degrees. Somebody who does not understand how to hammer might have a certain understanding of hammers, but it is different from that of somebody who understands how to hammer.

2.2. The Four-Fold Structure of Action and Embodied Cognition

Accordingly, an adequate conception of action needs to account for at least the following four aspects of action that I have identified above: 1) The spontaneous aspect of behavior (striving), 2) the receptive aspect of behavior (constraint), 3) the spontaneous aspect of understanding, and, finally, 4) the receptive-passive aspect of understanding. In the following I describe these aspects in more detail and show how embodied cognition accounts for them. I further introduce the conception of concern that is used in this chapter to explain the four-fold structure of action.

1) The spontaneous aspect of behavior (striving). This aspect determines the direction of an action and is about the ways in which I situationally direct my behavior towards the world. If I were not motivated to behave in a certain way, my embodied abilities and the presence of objects would not dispose me to act. I am not merely receptive to what is given to me by the material structure of the environment in relation to my embodied abilities—rather I need to want to act in accord with my concerns.

This spontaneous aspect of behavior has been stressed in particular by Rietveld (2008) and to a lesser degree by Noë (2016). Both authors emphasize the importance of motivational factors for action and suggest that concern grounds these motivational factors. But, Rietveld and Noë leave the exact nature of concern open—in particular, they do not provide an analysis of the conditions of the possibility of action in terms of concern. To circumvent this shortcoming, I will show in section 2.3 how affect, as part of a Heidegger-inspired concern structure, determines the direction and manner of an action. And in section 2.4, I will discuss how the integration of affect and understanding makes it possible to want to engage into an action in the first place.

Importantly, embodied understanding, merely understood as a set of embodied abilities or sensorimotor knowledge, is not sufficient to account for the spontaneous aspect of behavior, as Noë

(2004) and Chemero (2009) seem to claim. Even though I understand, for instance, an apple relationally through the structure of the environment, thereby receptively, and through my embodied abilities, thereby spontaneously, the mere presence of an apple is not sufficient to account for the spontaneous action of apple eating. Rather, in order to act I further need to spontaneously behave in accordance with my interests—I need to have concern for eating apples or not eating apples.44

At this point, it might seem unlikely that so many proponents of embodied cognition have overlooked the spontaneous aspect of behavior in the form of interest or motivation. But, it seems that their ‘oversight’ is motivated by the following two reasons. First, most proponents of embodied cognition are anti-representationalists when it comes to everyday action and experience. Accord- ingly, they seek to shun reference to goals, desires, beliefs, and intentions.45 That is laudable, but we still require a non-representational explanation of the spontaneous aspect of our actions that pertains to motivation and interests.

Many proponents of embodied cognition seek to provide such an explanation, secondly, via recourse to phenomenology and claim that the world itself is intrinsically significant and soliciting

44 Importantly, we do not have to be consciously aware of these interests when we act spontaneously; we merely have to act according to them. Consequently, spontaneous action must not be confused with deliberative action, but rather is a form of autonomous action, where autonomy is determined by acting according to one’s interests. 45 The spontaneous aspect of behavior has also been emphasized by Hurley (1998), Rowlands (2006) and McGann (2007) and they explain this aspect with recourse to intentions or goals. Though it is correct to emphasize the spontaneous aspect of our actions, I am skeptical that goals or intentions are the best way to do so for the following three reasons. First, most of our actions are not making any explicit reference to goals (Dreyfus 2007a; Wheeler 2008). Second, goals, as representational states, do not explain at the sub-personal level. Rather, goal-based explanations run into the frame problem, since ‘it is completely unclear how it (a particular goal) was selected out of the many possible ‘goals’ or ‘desires’’ (Bruineberg and Rietveld 2014, p. 12, brackets mine). Third, and most importantly, it is not clear how goals or intentions, in particular if we conceive of them as representational entities, are related to the embodied abilities which enable and disclose opportunities for action for us. This means, it is not clear how goals, if they would constitute the spontaneous aspect of behavior, are interacting with the spontaneous aspect of our understanding; i.e. our embodied abilities that grant us an understanding of environmental affordances for action.

75 options for action. Yet, the physical world—or the animal-environment system for that—is not intrinsically significant. It is the ‘human world’ (Dreyfus 2007b, p. 255), we might also say, ‘the living world’ or ‘the possible horizon for action’, that is intrinsically significant. And we will see in section 2.4 of this chapter, that this possible horizon for action requires concern.

2) A further aspect of action for which a sufficient theory of action has to account is the receptive aspect of behavior (constraint). This aspect pertains to the manner of an action and is about the ways in which my wider set of concerns constrains my disposition to behave in a situation. I do not just engage in any kind of action, but primarily in those actions that are in accord with my overall life-projects and interests. For instance, if I am very hungry and surrounded by a table with delicious foods, I might still not eat any of these foods, if I am interested in my health and if I am therefore fasting at the moment. Section 2.3 of this chapter describes how concerns are the basis for this receptive aspect of behavior.

3) The spontaneous aspect of understanding is about what I bring to bear on the world so that the world can be, 4), a horizon of possible actions for me—the receptive-passive aspect of understanding (section 2.4). For instance, if my understanding of hammers were not structured spontaneously by my embodied abilities that allow me to grasp and move a hammer, I could not understand a hammer as a thing that I could use for certain actions that require certain bodily behaviors such as moving and grasping. And accordingly I could not want to engage in the action of hammering in the first place, since in order to act I presuppose an understanding of the movements necessary to achieve what I aim at. In particular, Noë (2004) and Chemero (2009) have accounted for this spontaneous aspect of action. Both authors have argued that our understanding of possibilities for action is co-constituted by embodied abilities of an agent and the material structure of the environment.

Yet, proponents of embodied cognition have widely neglected that my interests have to be part of what I spontaneously bring to bear on the world so that the world can show up as a passive-receptive horizon for action. As Ratcliffe correctly states:

The world can only show up as it does in so far as things matter to us. (Ratcliffe

2010, p. 128)

However, embodied abilities do not constitute what matters to us—teaching a lecture does not matter to me, because I have the ability to do so, but because I have concern to do so. A conception of interest or relevance that goes beyond embodied abilities is needed here. We will see in section

2.4 of this chapter, that I cannot want to engage in an action if I do not already receptively-passively understand the world as a place of things that realize my interests. I will argue that concern is necessary for this receptive-passive understanding too.

Even though different embodied cognition approaches account for one or the other of the aspects of action described above, we lack an overarching explanation of how these aspects are integrated with each other so that action can come about. As an overarching background structure in this sense, I propose a conception of concern that is inspired by Martin Heidegger’s work in Being and

Time.46 Obviously, Heidegger’s analysis of Dasein and the role that concern (‘care’) plays in it is highly complex. Importantly, I draw on certain of Heidegger’s insights, in so far as they are relevant for the present analysis of action. The argument in the following does not depend on the conclusiveness of Heidegger’s analysis, nor does it draw the same conclusions. Yet, it agrees with

46 I translate in the following Heidegger’s concept ‘Sorge’ (Heidegger 2006) with ‘concern’ rather than with the term used in the ‘standard’ English translation, ‘care’ (Heidegger 1962). ‘Concern’ seems to be the more fitting term to capture the positive as well as negative connotation of the German term as well as the motivational force that ‘Sorge’ expresses. Further, the standard English translation uses for the terms ‘besorgt’ and ‘besorgen’, which derive from Sorge, the terms ‘concernful’ and ‘concerned’. For reasons of conceptual consistency, it seems therefore more advis- able to use the term ‘concern’ instead of ‘care’.

Heidegger on the centrality of the integration of affect and understanding, in the form of concern, for the explanation of action.

Heidegger identifies four structural parts (Strukturmomente) that constitute concern equiprimordially (gleichursprünglich). Affect (Befindlichkeit), which is commonly translated as affectedness, attunement or state-of-mind; understanding (Verstehen); falling (Verfallen); and discourse (Rede)

(Dreyfus 1991; Ratcliffe 2002; Crowell 2007).

Care has been characterized with regard to its temporal meaning, but only in its

basic features (…): understanding, state-of-mind, falling, and discourse.

(Heidegger 1962, pp. 384-385, italics mine)

In the following I will only focus on the first two of these four structural parts of concern, understanding and affect, which I deem sufficient for the constitution of action.47

Some readers might be surprised that I do not also account for discourse, given its social implications. This might seem particularly surprising, given the popularity of social practices in the explanation of cognition in embodied cognition. My reasons for omitting discourse or social practices from my discussion of action are the following. This chapter is concerned with the conditions that

47 It seems as if autopoietic enactivism (Thompson and Stapleton 2009) provides a treatment of action that is similar to mine in two senses. First, it is similar in that the autonomy of organisms from their environment is emphasized, which is comparable to my emphasis on the spontaneous aspect of behavior. It is further similar, second, in that sense- making, a major theoretical theme of autopoietic enactivism, is an affective and understanding-based process, which is comparable to my claim that concern contributes spontaneously to an understanding of the world. Yet, there are also deep discontinuities between my approach and those of autopoietic enactivists, which are the reason why I do not discuss autopoietic enactivism in this chapter. First, instead of conceptually characterizing the relations between the various aspects of action, autopoietic enactivists assert that all these aspects are interacting in complex and dynamic ways. This emphasis on complexity might be desirable, if we want to explain action biologically, but it also leaves the exact nature of the relationship between, among others, affect and understanding open. Second, and more important, whereas my approach is primarily phenomenological, autopoietic enactivists use simultaneously phenomenological, psychological and biological terms. Whereas I seek to establish the conditions that make action possible, autopoietic enactivists treat action and purposes as positive phenomena. I do not want to engage into the meta-philosophical and methodological ramifications of this difference here, but rather want to confine myself to the statement that these differences make it difficult to discuss both approaches at the same level of analysis.

78 make action possible. This means that the exclusive focus of this chapter is only on the fundamental, invariant conditions of action. And it seems that we can make a good point that sociality and discourse are not such fundamental, invariant conditions. For instance, we can easily imagine a person being born in a non-social situation and not being able to produce speech, but who is still perfectly able to act. That is so, or so I argue in the following, because action is at a fundamental level co-constituted by affect and understanding, i.e. concern. Affect and understanding are invariant features that people possess even in a non-social situation as the one imagined above.

Importantly, social practices and discourse are not at the same ontological level as affect and understanding. Social practices and speech are actions themselves (Austin 1962). Even though a certain speech act of mine might have a significant effect on somebody else’s action as well as on her concerns, my speech act is itself an action that requires an explanation in more foundational terms.

For this reason, the subsequent considerations confine themselves to affect and understanding only.

In the following, the concept of concern will be developed functionally, by an analysis of ordinary action—or, as Heidegger would say, by establishing the conditions of possibility of action by an analysis of the ‘everydayness’ (Alltäglichkeit) of Dasein (here: the embodied human agent). Ac- cordingly, I will discuss concern, its structural parts, and how the here presented conception of concern relates to Heidegger’s conception of concern, as the argument unfolds.

Further aspects of behavior and understanding are important to understand action, but are not discussed here. For instance, the sense of agency, the body schema, skill acquisition and learning. I do not engage with these phenomena for two reasons. First, they have been well-explained already elsewhere (Dreyfus 2002; Gallagher 2005). Second, as Dreyfus (2002) correctly points out, most of these phenomena relate to the ‘phenomenological’— or rather experiential—but not ‘logical’—

79 or rather ontological—requirement for a behavior counting as an action. And it is this logical, or rather, in Heideggerian terms, ontological or phenomenological structure of action—the structural conditions that make action possible—which I seek to identify and explain.48

The next section, section 2.3, provides an analysis of the spontaneous and receptive aspects of behavior. I will establish that concern is that which specifies both aspects of behavior in the form of the direction and the manner of an action. I will further discuss the affective nature of concern in this section. In section four I will show that our understanding of the world has to be concernful so that the world can show up as a receptive-passive horizon for our actions, which is the precondition for the possibility of engaging in an action in the first place.

2.3. Motivated Behavior: The Direction and Manner of an Action

In this section, I show that the spontaneous and receptive aspect of behavior requires concern. As we will see, concern allows us to coherently explain how situational interests and long-term life- projects motivate, generate and constrain our actions.

Let us begin our analysis of action with the example of the action of cleaning a carpet. If I perform the action of cleaning a carpet, my behavior is structured by concern in two ways. I) My behavior is directed at transforming the environment into a state that I have concern for, e.g. transforming

48 In particular Heidegger would not use, like Dreyfus, ‘phenomenological’ to merely describe an ‘experiential episode’ compared to a logical or ontological phenomenon. Rather, phenomenology establishes the conditions of the possibility of phenomena (see the introduction, chapter 1 and 3), which is a form of a logical analytic exercise and as Heidegger made clear, phenomenology is first and foremost ‘fundamental ontology’ (Heidegger 1962). As Heidegger (1962, p. 60, italics in original) states: ‘Only as phenomenology, is ontology possible.’ Or take for example: ‘With regard to its subject-matter, phenomenology is the science of the Being of entities—ontology’ (Heidegger 1962, p. 61).

80 a dusty floor into a tidy floor, because I have concern for inhabiting a tidy home. Differently put,

I spontaneously strive for such a home in the particular situation in which I act.49 II) My behavior is also structured in a manner for which I have concern. Concerns that structure the manner of my behavior are concerns like time-efficiency and comfort. This is why I perform the action with a vacuum cleaner rather than with a toothbrush or my bare hands, which would be more uncomfort- able and time-consuming. In this sense, my action is receptively constrained by my concerns.

An analysis of the action of cooking reveals the same background structure. If I want to perform the action called ‘cooking’, I) I direct my behavior towards a state in the world for which I have concern. This can be a healthy meal or a quick snack. The choice depends on whether I have concern for my health, or whether I have concern for being successful at work.

I will further engage in cooking behavior in, II) a particular manner, which is also structured by my concerns. For instance, I might want to cook energy efficiently because I might have concern for the environment. Accordingly, I will cook with a stove that offers me to act in a particular manner, such as cooking my meal on a low flame. Alternatively, I might rather have concern for the efficient use of my time because I have concern for achieving important life-projects swiftly.

This is why I might either use a slow cooker or a microwave oven, which are an integral material part of the structure of my cooking action, depending on my concerns.

One might object, that contrary to the actions that I analyzed in these examples, it seems that certain actions might not be subject to an analysis in terms of concern; for instance, ‘actions’ like sitting, grasping, touching, seeing, running, walking, lying or standing. When I walk, not every

49 As it is apparent from this description of the structure of action and the examples above, the direction of an action is directed towards something that is not present yet (a similar future-directedness is hinted at with the term ‘horizon for action’). Therefore, a complete explanation of the structure of action will require an account of anticipation, which is beyond the scope of this chapter. This future directedness of action and concern is apparent in Heidegger’s (1962) own work, where he seeks to ground concern in future-oriented temporality.

81 aspect of my walking seems to be determined by a particular concern, but rather by my embodied ability to walk and the walkability offered by the material structure of the path that I walk. It seems as if these types of actions can be exclusively analyzed in terms of embodied abilities in relation to the structure of the environment, without reference to spontaneous concern of an agent (Hauge- land 1998; Chemero 2009).

Yet, embodied abilities are not sufficient to explain action as long as we do not conceive of ‘mere sensings’ and ‘mere movings’, i.e. mere behaviors, as actions. An ability is not sufficient to count as an action—an ability is literally an ability to do something; not the doing of the something itself.

For instance, eating an apple might require under normal circumstances the embodied abilities to be able to focus on an apple, to grasp it, to chew it and to swallow it. But as long as I do not have concern for apple eating—e.g. satisfying hunger, appreciation of apple taste, being on an apple diet—an apple’s presence and my sensorimotor ability to eat it, are not sufficient to account for the action of apple eating.

Another reason for reservation about my analysis might be that not every aspect of, say, my cooking seems to be guided by my concerns. Many different ways of behaving might fulfill my concerns. And the variations that fulfill my concerns might be primarily explainable by the material structure of the environment and my embodied abilities. For instance, I can cook with pans of different sizes, sit or stand while cooking, use my left or my right hand for stirring and so forth.50

But that does not change the fact that the varying behavioral patterns are part of my cooking action

50 Anthony Chemero (in conversation) pointed out that he sees the explanation of the plethora of varying ways in which environmental, behaviorally fine-grained and non-experiential factors structure our actions, as one (among many) important contribution of an affordance-based approach to the explanation of action. And indeed, these aspects cannot be explained by a phenomenological or conceptual approach and are best explained by models from ecological psychology. Yet, as much as such an account enriches our understanding of the behavioral aspects of action, it is not sufficient to explain action, because it does not account for concern.

82 because my behavior is directed towards a state for which I have concern, in a manner for which I have concern—a particular meal that I create in a particular fashion.51

Importantly, simple behavioral movements are, with the exception of reflex behaviors and basic biological processes, never done for their own sake. Rather, walking is done to think, to relax, to exercise, or to get from place A to B, all of which are things for which I have concern. Similarly,

I do not sit simpliciter, but I sit to rest, to meditate, to look out of the window, to participate in a meeting or to work with my laptop. ‘Sitting’ denotes in these cases the state change of my body, brought about by movement and sensing, that is necessary to engage in something for which I have concern. But this bodily state change does not exhaust my action.52 Actions, other than mere behavior, can therefore not be understood just in terms of embodied abilities, exploratory behavior and the structure of the environment, as many proponents of embodied cognition claim (Haugeland

1998; Noë 2004; Chemero 2009), but require reference to concern.

Heidegger puts this point in the following way:

51 There is a strong linguistic arbitrariness in at least English and German as to what counts as an action and what does not count as an action or rather counts as part of another action. For instance, we would not say that sleeping is comprised by two actions: lying and something we could call ‘dormating’. Rather we would just say that we sleep, independent of whether we lie, sit or stand. In a similar fashion it is unclear what constitutes drinking. Is grabbing the water bottle, from which I drink, part of the action of drinking or is only bringing the bottle directly to my mouth, combined with the synchronized swallowing, part of my drinking? What about the walking that I have to do to get the bottle; does this walking fall under the action of ‘getting something to drink’ or should it rather be part of the drinking action itself? It seems as if there would not be an interesting answer to these questions other than that much of this is arbitrary. What ontologically matters though in all these cases is that concern unites these different behaviors to meaningful actions. 52 The question might emerge whether any behavior that we perform is a non-action, given that most of our daily dealings are part of some concern or other. A general answer to this question is that every aspect of our behavior that is not fulfilling a concern is a non-action. Therefore, for instance, reflex based behavior will count in most cases as a non-action.

That with which our everyday dealings proximally dwell is not the tools themselves

[die Werkzeuge selbst]. On the contrary, that with which we concern ourselves pri-

marily is the work - that which is to be produced at the time (…). (Heidegger 1962;

p. 99, italics mine)

And the work, at which our behavior is directed, is something for which we spontaneously have concern in a situation.

The previous considerations can be summarized in the following way. An action A is a bodily behavior that is directed towards a state for which the agent of A has concern and is performed in a particular manner for which the agent of A also has concern. The way in which my behavior is directed is spontaneous in that it is not solely enforced by the environment, but also generated by my autonomous concerns. Still, my action is receptive in that I normally engage only in behaviors that I non-deliberatively understand to be achievable by my embodied abilities and to be in accord with the concerns that structure the manner of my action.

The manner of an action, is normally structured by ‘wide’ concerns, concerns we might call ‘existential’. These are the concerns on which Heidegger (1962) focused in Being and Time in his analysis of human being in terms of concern (care). Such concerns are, for instance, the concern for the efficient use of the time at one’s disposal, concern for comfort, concern for survival (fear of death), or concern for the self-realization of one’s individual life-projects. As much as wide concerns constrain our actions receptively, they allow us at the same time to spontaneously have concern for particular actions in the first place. For instance, if I did not have wide concern for being healthy, as part of my wider concern ‘fear of death’, I would probably not be disposed to direct a concrete action spontaneously at broccoli eating rather than on an infinite number of other possible actions that I could engage in at a particular moment.

These wide concerns structure nearly all actions to a certain degree. From this we see that we are seldom aware of our concerns; they structure our actions in a holistic background fashion of which we often cannot be aware.53 Contrary to that, the direction of an action towards a worldly state is characterized by more situational, fine-grained concerns, which are often derivative of wide concerns. These fine-grained concerns show up against the backdrop of concrete environmental situations. And in these situations we sometimes consciously present these concerns deliberatively as

‘goals’ or other intentional states to ourselves.

I stated above that concern is constituted by the integration of affect and understanding. The affective aspect of concern renders the objects and outcomes of actions as attractive or repelling, as valenced in a particular way; i.e. as something I have concern for.

For instance, drinking lemonade is in many situations attractive, though it can be repelling if I had too much of it and my concern for the taste of lemonade stands back behind my concern for feeling

‘not too full’. I agree with Heidegger that this attraction or repulsion is best explained by the affective component of concern. As Heidegger states,

Mood has already disclosed, in every case, Being-in-the-world as a whole, and

makes it possible first of all makes it possible first of all to direct oneself towards

something. (1962, p. 176, italics in original)54

53 I admit that this claim is not based on a strong argument, but that it is an intuition-based statement, that might elude philosophical argumentation. For psychological support of this intuition, please compare Wilson (2002). 54 Importantly, ‘Stimmung’ (‘mood’) does not mean the same as the English term ‘mood’. Rather, ‘Stimmung’ has a wider, affective component. As Dreyfus (1991, p. 169) states, ‘Stimmung has a broader range than ‘mood’. Fear, for example, is a Stimmung for Heidegger, but it is clearly an affect not a mood. Stimmung seems to name any of the ways Dasein can be affected’. There are plenty other examples of Stimmung that Heidegger discusses next to fear, which are indicative of Heidegger’s wide conception of Stimmung, such as Erschrecken (alarm), Grauen (dread), Entsetzen (terror), Schüchternheit (timidity), Scheu (shyness), Bangigkeit (misgiving), Stutzigwerden (becoming startled) (Heidegger 2006, p. 142, Heidegger 1962, p. 181). Further, Heidegger uses the German term ‘Affektion’ (‘affection’ or ‘affect’) in the same sense as ‘Befindlichkeit’ (‘affect’) (Heidegger 2006, p. 137). And finally, Heidegger uses ‘Stimmung’ and ‘Befindlichkeit’ interchangeably, as the following example shows: ‘Jedes Verstehen hat eine Stim- mung. Jede Befindlichkeit ist verstehend’ (Heidegger 2006, p. 335), which we can translate as ‘Every understanding

But, affect has not only a directive, motivational role, as I suggested in this section. Rather, it is the precondition for things showing up as the objects with which I perform an action or for which

I perform an action. Affect makes this possible, by being integrated with understanding. As

Heidegger states,

Every understanding has its mood. Every state-of-mind (affect) is one in which one

understands. (Heidegger 1962, p. 385, brackets mine)

In the following I will spell this claim out and show how the integration of affect and understanding is a precondition for the possibility of action.

2.4. The World as the Horizon for Action: Concernful Understanding

In order to act, I require an understanding of a horizon of possible actions, i.e. a living world that is populated with opportunities for action. Concretely, in order to act, I need to passively-receptively understand the world as a place in which I can realize my concerns at which my actions are directed.55

has its mood, every affect has its understanding’. Given all this evidence, we can make a good case that Heidegger uses ‘Stimmung’ (‘mood’) and ‘Befindlichkeit’ (‘affect’) in a wide, theoretically underdetermined (i.e. ‘everyday’), affective sense and not merely in a narrow sense, as the English translation of ‘Stimmung’ seemingly suggests to some authors (Ratcliffe 2008; Guignon 2009). 55The point that the world needs to show up as concernful or significant has also been stressed by a variety of contemporary philosophers, most of them phenomenologists, that are working in proximity to cognitive science and embodied cognition (Taylor 1982; Kelly 2000; Dreyfus 2002, 2007b; Wrathall 2005). Yet, these authors do not focus on the role that affect, especially integrated with understanding, has for concern or significance, nor do they spell out how our understanding has to be concernful so that we can want to engage into an action.

In the following I argue that understanding, at least the kind that is guiding ordinary action—

Heidegger’s concernful dealings ready-to-hand, has to be intrinsically affective.56 Affective understanding, or understanding affect, if you will, is a condition for the possibility of ordinary action. Importantly, the relationship between affect and understanding is not merely causal, but co- originary. As Heidegger states,

To any state-of-mind (affect) or mood, understanding belongs equiprimordially

(Heidegger 1962, p. 315, brackets mine),57 and

Understanding always has its mood. (Heidegger 1962, p. 182)

This means for embodied cognition and cognate approaches, that embodied abilities (Chemero

2009), sensorimotor knowledge (Noë 2004) or motor intentionality (Kelly 2000) are not sufficient to account for the understanding component of action, since they do not integrate affect with understanding.

If I want to perform an ordinary action, I need to understand in advance that the objects and situations, on which or with which I perform an action, will produce not just any result, but exactly the result that I want to achieve, i.e. the result for which I have concern. To engage in an ordinary action, already requires an understanding of the affect-satisfying properties that are offered by objects or situations because objects or situations are co-constitutive for having particular concerns

56 I differentiate this non-deliberative form of action from, for instance, exploratory actions, Heidegger’s concernful dealings unready-to-hand. I do not claim that actions other than ordinary actions might not be characterized by the same structure that I propose here. Rather, I merely want to confine my claims to ordinary actions. That has partly to do with the circumstance, that it is not entirely clear whether action belongs into a transcendental or a psychological category. If action belongs into the latter, my claims here have merely the character of a psychological hypothesis and I accordingly do not want to make necessity claims about all kinds of action, if the former would hold true. 57 The German original is: ‚Zur Befindlichkeit (Stimmung) gehört gleichursprünglich das Verstehen‘ (Heidegger 2006, p. 270). Importantly, Heidegger uses, contrary to the translation, ‘Stimmung’ in brackets after ‘Befindlichkeit’, which is again an indicator for the circumstance that Heidegger uses the terms interchangeably.

87 and are thereby co-constitutive for the generation of a concernful action. Concern, as the integration of understanding and affect, functions as a passive-receptive horizon that enables me to understand which things allow me to fulfill my concerns—the aims of my action—which enables me to want to act in the first place. As Ratcliffe states:

Care (concern) is the condition for the possibility of apprehending the world as a

significant whole, as an arena of possible projects, goals and purposes. (Ratcliffe

2002, p. 289, brackets mine)

Understanding concernfully what a vacuum cleaner is, is co-constitutive of the state I have concern for, i.e. the state that directs and motivates my ordinary action of carpet cleaning. I want a carpet that is as clean as only a vacuum cleaner can achieve it in a time-efficient and comfortable manner—compared to removing dust from my carpet with my bare hands, for instance. Therefore, in order to even want to perform the ordinary action of carpet cleaning, I have to understand that a vacuum cleaner is a thing with which I can satisfy this concern.58 Therefore, I need to understand the vacuum cleaner concernfully; not merely qua its sensorimotor affordances.

Dreyfus, in his interpretation of Heidegger’s work, makes this point clear too:

Things are always encountered in some specific way, as attractive, threatening, in-

teresting, boring frustrating, etc. Possible actions are always enticing, frightening,

intriguing, etc. We care when a piece of equipment breaks down and whether or

58 Please mind the difference between wanting to engage into an action and wanting a situation in the world that is the consequence of an action. I might want a situation in the world that might require an action of mine so that it can be achieved, independent of my concernful understanding how to achieve this situation in accord with my concerns. However, I cannot want to engage into an ordinary action—an action ready-to-hand—without a concernful understanding of how to achieve the result of my action in accord with my concerns.

not we achieve our goals. Affectedness is the condition of the possibility of specific

things showing up as mattering. (Dreyfus 1991, p. 175)

Yet, importantly, we do not merely encounter things as attractive, frightening and so forth. Rather, we understand things as significant for the achievement of our concerns and thereby our actions.

Mind that understanding has to be intrinsically affective—and affect has to be intrinsically understanding. Both stand in a co-constituting relationship. For instance, vacuum cleaning requires an understanding of the concern fulfilling aspects of the object. If I would understand a vacuum cleaner merely relative to my embodied abilities, i.e. based on those abilities that allow me to grasp, move, turn off and on a vacuum cleaner and to perceive the effects of its use, I would understand a vacuum cleaner as a plethora of possibilities for mere moving and mere sensing. If merely embodied abilities, motor intentionality or sensorimotor knowledge would govern action,

I would not understand a vacuum cleaner as a particular cleaning thing, but merely as a noisy thing, a thing that is located in a particular place, or a thing of a particular color.59

Yet, if that were the case, I could not want to engage into an action, because, as we said, engaging into an action already requires to understand the objects of my action concernfully. I need to be able to understand that an object on and with which I act can satisfy the concern that motivates my action. But, if affect stood merely in a causal relationship to understanding, this would not be possible. A similar point has been made by Gallagher and Bower:

Schemata of sensorimotor contingencies give an agent the how of perception, a tacit

knowledge of potential sensorimotor engagements, without giving its why, which

depends on latent valences that push or pull in one direction or another for attention

59 That is the case, since, according to most conceptions of sensorimotor understanding, affect is not a sensorimotor phenomenon.

and for potential sensory-motor engagement, reflecting, for example, a degree of

desirability. (Gallagher and Bower 2014, p. 234)

Accordingly, both, affect and understanding need to be understood as integrated with each other, to make sense of action. As Heidegger states, understanding is always already based on a ‘towards- which’ or ‘what-for’ that is directly based on concern.

The work which we chiefly encounter in our concernful dealings (…) has a usabil-

ity which belongs to it essentially; in this usability it lets us encounter already the

‘towards-which’ for which it is usable. (Heidegger 1962, p. 99, italics mine)

This usability is not ‘added’ to understanding after an affective interpretation—rather, it is intrinsic to it—it ‘belongs to it essentially’. Therefore, concernful understanding is the very ground for the generation of action.60

2.5. Conclusion

In this chapter I have argued, that concern is what unites the various aspects of action and explains them in embodied terms. It spontaneously directs behavior towards something that we want to achieve—towards which we strive. It also constrains our behavior against the backdrop of our long-term projects and interests, and determines the manner of our actions receptively. Further,

60 Matthew Ratcliffe reviews interesting cases from psychiatric research that seem in accord with my claims about the centrality of affect for understanding. He describes cases in which schizophrenic patients lose an understanding of the significance of objects (of the what-ness of objects) (Ratcliffe 2015, pp. 166-167), which might even yield the result that certain objects are not usable for certain patients anymore (Ratcliffe 2015, p. 54). Importantly, he suggests that these symptoms are not merely based on a loss of motivation, but on a loss of practical significance of the patients’ understanding of objects (Ratcliffe 2015, pp. 170-171); a result of the disturbance of the affective part of concern, which he calls ‘existential feelings’.

90 concern provides a receptive-passive horizon for possible actions that confines the scope of actions at our disposal, while it is at the same time the ground for the possibility of action.

The account that I have provided certainly allows for refinement. In particular, it will be important to address the question how concern for particular things and long-term projects is acquired, given the virtuously circular nature of concern that oscillates between understanding and affect.

Kiverstein and Rietveld (Kiverstein and Rietveld 2015), based on the work of Merleau-Ponty, suggest to ground concern in a more basic form of bodily equilibrium and disequilibrium. Such an account might provide an important insight into the acquisition of particular concerns.

It will be further interesting to explore the possibilities for mutual enrichment between the here presented account of concern, that is entirely based on phenomenological and conceptual grounds, on the one hand, and naturalistic approaches, on the other hand. In particular, the work of Luiz

Pessoa and Lisa Feldman Barrett might be of interest in this respect. Both argue for integrative accounts of understanding and affect. For instance, Pessoa:

(…) emotion and cognition are functionally integrated systems (…) they continu-

ously impact each other’s operations. (Pessoa 2015, p. 18, italics mine)

Similarly, Feldman Barrett:

Emotion-cognition-perception distinctions are phenomenological and are not re-

spected by the brain. (Barret et al. 2015, p. 85)

Finally, we can conclude with Heidegger that,

The totality of Being-in-the-world as a structural whole has revealed itself as care.

In care the Being of Dasein is included. (Heidegger 1962, p. 275)

Concern—care—is central to the constitution of action. And action is most central to human being.

Chapter 3. Constitution Embodiment

Abstract

In this chapter I analyze a particular conception of embodiment, constitution embodiment. Propo- nents of constitution embodiment claim that the body is a condition of the constitution of entities.

Constitution embodiment is popular with phenomenologically-inspired embodied cognition

(PEC), including research projects such as enactivism (Noë 2004, 2012; Thompson 2007) and radical embodied cognitive science (Chemero 2009). Unfortunately, PEC’s use of constitution embodiment is neither clear nor coherent; in particular, PEC uses the concept of constitution embodiment so that a major inconsistency is entailed. PEC conceives of the body in a transcendental sense as a condition of the constitution of entities and in an ontic sense, as a scientifically describ- able entity. Yet, a condition of the constitution of entities cannot be itself an entity—rather, it is the very condition of the possibility of an entity. This inconsistency entails further problems, among them PEC’s misguided focus on the location of the embodied mind. In order to correct these mistakes, I develop a conception of constitution embodiment based on the work of Heidegger

(1962), Husserl (1989) and Merleau-Ponty (2012). This has two purposes. First, it provides the conceptual groundwork to secure the status of PEC as a consistent and coherent research project and to clarify PEC’s conception of the relationship between phenomenology and the sciences.

Second, my account provides an elaborate concept of constitution embodiment that can function as the basis for more sophisticated work in the future.

3.1. Introduction

As it is well known by now, ‘embodiment’ can mean quite different, at times opposing things

(Haugeland 1998; Wilson 2002; Gallese and Lakoff 2005; Clark 2008a; Barsalou 2008; Goldman and de Vignemont 2009; Kiverstein and Clark 2009; Gallagher 2011; Alsmith and de Vignemont

2012; Kiverstein 2012). In the following I examine one particular kind of embodiment, that I call for the lack of a better term ‘constitution embodiment’. Proponents of constitution embodiment state that the body is constitutive for the objects of experience and thought.61 As Husserl states:

We need only consider how a thing exhibits itself as such, according to its essence,

in order to recognize that such an apprehension must contain, at the very outset,

components which refer back to the subject, specifically the human (or, better: an-

imal) subject in a fixed sense. (Husserl 1989, p. 61)

The concept of constitution embodiment has been most prominently developed by Husserl (1989) and by Merleau-Ponty (2012) in the context of their discussion of the Leib (lived body) and phenomenological body respectively.62 As we will see, Heidegger (1962) has developed his own conception of the term in the context of his treatment of care.

Constitution embodiment is the opposite of what we might want to refer to as ‘body neutrality’.

Body neutrality is a particular version of what Husserl and Merleau-Ponty call the ‘natural’ or

61 Sometimes I state in this paper, in the context of constitution embodiment, that the body is a condition of the possibility of entities. Please mind, with recourse to the introduction, that experience has a basic world-disclosing role for the transcendental tradition. Accordingly, it makes little difference to claim that the body is a condition of the possibility of entities, compared to the claim that the body is a condition of the experience of objects, because we cannot understand what an entity is beyond how it is presented to us in the basic structure of experience, that is, or so I argue in this chapter, co-constituted by embodiment. 62 Some contemporary writings (Zahavi 2005; Fuchs 2011) that seek to introduce phenomenology to a primarily analytically educated audience, treat the Leib or lived body primarily as the body as it is experienced in ‘qualitative experience’ or as something that can be researched through the ‘first-person perspective’. This approach should be considered with caution, since these categories that are applied to the Leib here are those of analytic philosophy and for good reasons alien to the very idea of phenomenology. If I use the terms ‘Leib’ or ‘lived body’ in the following, I do not make any such connotations, but primarily treat the Leib as a condition of the constitution of entities.

‘dogmatic attitude’ respectively.63 According to the natural attitude, the world is as it appears, independent of the invariant structures that organize our experience; i.e. the world is constitutively independent of a conscious agent. By extension, body neutrality states that the body is not a constitutive condition of our experience and understanding of entities.

As I show in section 3.2.1, constitution embodiment has become popular again due to the work of enactivists (Noë 2004, 2012; Thompson 2007; Thompson and Stapleton 2009; Stapleton 2013) and proponents of radical embodied cognitive science, henceforth RECS (Chemero 2009; Rietveld and Kiverstein 2014). But, as I suggest, constitution embodiment should be also of interest to all other proponents of phenomenologically-inspired embodied cognition (henceforth PEC); particularly to those that have stressed the importance of Husserl’s concept of ‘Leib’ or Merleau-Ponty’s concept of ‘body schema’ in their writing.

Despite its regained popularity, neither the concept of constitution embodiment nor its entailments are particularly well spelled out by proponents of PEC. My discussion of Husserl, Merleau-Ponty and Heidegger in sections 3.2.2 and 3.2.3 demonstrates this lack of clarity in contemporary work.

I provide in these sections a detailed discussion of these classic authors to provide a sufficiently complex, consistent and explicit conception of constitution embodiment. I focus hereby on two aspects of constitution embodiment. First, in accord with Husserl’s and Merleau-Ponty’s work, I show how our embodiment is co-constitutive of the spatiality of the objects that we can understand and experience. Second, I examine, following Heidegger’s work on the care structure, how our embodiment constitutes the significance of the objects—their ‘whatness’—that we encounter in the world.

63 For a recent discussion of the natural attitude see also Dreyfus’s and Taylor’s (2015) conception of the ‘neutrality condition’ of the ‘mediational picture’.

Developing a sufficient concept of constitution embodiment has two purposes. First, it provides

PEC with an elaborate concept of constitution embodiment that can function as the basis for more sophisticated work. Second, it helps to identify conceptual problems with the use of the term in

PEC. These conceptual problems are manifold and severe, as my discussion in section 3.3 demonstrates. As such, a proper concept of constitution embodiment is necessary conceptual groundwork to secure the status of PEC as a consistent and coherent research project.

Section 3.3.1 provides a demonstration of how proponents of PEC use the concept of constitution embodiment so that a major inconsistency arises. I show that many authors confuse what is a transcendental claim about the structures that constitute a world of entities with a positive claim about entities themselves. Constitution embodiment is a condition of having a world for a person. As such, it cannot be examined as an entity in the world—a physical body—as the scientifically inspired work of various contemporary champions of constitution embodiment suggests, because it is co-constitutive of a scientifically examinable world and of the entities in it. Here, a proper concept of constitution embodiment allows us to demarcate clearly between philosophical and scientific claims. Whatever abstract reasons have been given in the methodological debates about the relationship between science and philosophy (Petitot et al. 1999; Dreyfus 2007; Noë 2005, 2011;

Wheeler and Kiverstein 2012; Carel and Meacham 2013; Fernandez 2016), the present discussion shows concretely how the relationship has to be conceived of given a concrete philosophical problem.

Further, I show in section 3.3.2 that the focus on the location of the embodied mind in PEC is a result of confusing constitution embodiment as a constitutive condition of entities with a spatially extended entity. In particular, if, as I will show in section 3.2.2, constitution embodiment is itself

95 a co-constituting condition of the spatiality of objects, the embodied mind cannot be located in space itself—as claims about the extendedness of the mind suggest.

In order to highlight the significance of these conceptual inconsistencies, I show in section 3.3.3, that claims about the location of the embodied mind are empty conceptual claims. If claims about the location of the embodied mind were meaningful, we could derive conceptual entailments about the structure of the mind from them. As I show with reference to the work of Thompson (2007) and Hutto and Myin (2013) respectively, this is not the case.

Finally, section 3.3.4 contains a short discussion of desirable philosophical and scientific research alliances within PEC, i.e. a short discussion of the work that should be included and discussed within the boundaries of the paradigm. I suggest that productive research alliances should be founded, if they are concerned with matters of constitution, not on an examination of the alleged location of the embodied mind, but on investigations into how the embodied mind constitutes or at least structures the world of an organism. The unfruitful focus on location has lead PEC astray.

The proper argumentative opponent of PEC has to be body neutrality, the claim that the body does not constitute the objects of our experience and understanding. Contrary to that, the emphasis on a rejection of internalism, even though correct, often enough leaves body neutrality untouched or might even support it. Accordingly, PEC should overcome its opposition to grounded cognition

(Barsalou 1999; Gallese and Lakoff 2005), metaphor theory (Johnson and Lakoff 1999; Borodit- sky and Ramscar 2002) and other projects that reject body neutrality, and reconsider how much it actually has in common with projects such as radical enactivism (Hutto and Myin 2013) that leave body neutrality untouched.

There are plenty reasons to accept constitution embodiment as a condition of the possibility of things showing up as they do. They range from general, complex considerations that characterize

96 the work of Husserl, Heidegger and Merleau-Ponty, to more specific claims, such as that constitution embodiment is a condition of the possibility of action (Heidegger 1962; Noë 2004; Chemero

2009; Merleau-Ponty 2012). I will not further discuss these claims, though I mention reasons for accepting constitution embodiment when it fits the structure of the argument. Rather, I take it for granted that constitution embodiment is a condition of the possibility of objects and spell out what is conceptually entailed if we accept it.

3.2. Constitution Embodiment

3.2.1 PEC and Constitution Embodiment

Various proponents of PEC, in particular enactivists as well as proponents of RECS have provided their own take on constitution embodiment in terms of enaction, sense-making and affordances

(Varela et al. 1991; Noë 2004, 2012; Thompson 2007; Chemero 2009; Thompson and Stapleton

2009; Kiverstein 2012; Rietveld and Kiverstein 2014). Particularly enactivists have stressed the constitutive role of the embodied organism for experience and understanding.

A cognitive being’s world is not a prespecified, external realm, represented inter-

nally by its brain, but a relational domain enacted or brought forth by that being’s

autonomous agency and mode of coupling with the environment. (Thompson 2007,

p. 13).

Similarly Kiverstein:

The body enactivist understands embodiment in terms of bodily skills we draw on

all the time when we act unreﬂectively, and in virtue of which we can encounter

situations that are meaningful (Kiverstein 2012, p. 13).

For enactivists, the world is not passively given, but brought forth by an embodied organism. The main case to which enactivists constantly refer to exemplify enaction or sense-making, is that of a simple organism, like an amoeba or bacterium, that enacts its world. Among the objects of the organism’s world, sugar is the most significant one.

Sugar is signiﬁcant to these organisms and more of it is better than less because of

the way their metabolism chemically realizes their autonomous organization. The

signiﬁcance and valence of sugar are not intrinsic to the sugar molecules; they are

relational features, tied to the bacteria as autonomous unities. (Thompson and Sta-

pleton 2009, pp. 24-25)

According to Thompson and Stapleton, sugar, as sugar-as-food, is at least co-constituted by an embodied organism in that sugar-as-food is only significant for an embodied being that uses sugar in a particular way—in this case as food.

Alva Noë (2004, 2012) has made similar claims about how the body constitutes the objects of experience and understanding. However, he is less explicit about this than Thompson and Staple- ton. Yet, we can clearly see a conception of constitution embodiment at play in his enactivist account of experience and understanding.

Like other enactivists, Noë identifies himself with ‘existential phenomenology’—Heidegger and

Merleau-Ponty in particular—in that he rejects the idea that humans are ‘detached representer(s) and interpreter(s)’ of their environment (Noë 2012, p. 8). As we will see later, this means for

98 existential phenomenologists not only that we are primarily active agents, but that we also co- constitute the world we live in based on our embodiment.

Noë argues that we need to achieve presence of the world. We need to make the objects of our experience present to us; they are not merely given to us. Differently put, we spontaneously contribute something to the world so that it can show up for us. In many ways following Kant and the phenomenological tradition, Noë (2012, p. 20) claims that ‘knowledge, understanding, skills’ are that what we contribute to the world to make it present.

Importantly, Noë argues that knowledge, understanding and skill are dependent on one’s embodiment in a twofold sense. Understanding, knowledge and skill are primarily sensorimotor understanding, sensorimotor knowledge and sensorimotor skill (Noë 2004). Yet, sensorimotor skill is not an abstract form of skill; it always refers back to a particular body. As Noë states:

If perception is in part constituted by our possession and exercise of bodily skills

(…) then it may also depend on our possession of the sort of bodies that can en-

compass those skills, for only a creature with such a body could have those skills.

To perceive like us, it follows, you must have a body like ours. (Noë 2004, p. 25)

As we will see below, this idea resembles closely Merleau-Ponty’s considerations that the world has to be reflective of our embodied skills so that it can show up in the first place. In accord with

Noë’s statements about the world-disclosing role of experience and perception as well as his statements about constitution embodiment, Noë makes clear that we ‘enact content’ based on our embodied skill (Noë 2004, p. 100) and that this embodied skill is not merely a causal perceptual ability that represents a given, but the spontaneous precondition of having ‘world-presenting sensory experience’ (Noë 2004, p. 117).

More recently Noë has specified his claims about constitution embodiment. In older work (Noë

2004) embodiment had ‘only’ been reflected in the nature of skill, understanding and knowledge.

In more recent work he has added the idea of a somatic field.

The organic, somatic field of sensations (is that) that forms the ever present back-

ground to our lived achievement of the environment's presence. (Noë 2012, p. 12,

brackets mine)

The somatic field is the set of all bodily abilities and states that presents the world in a particular way. In that it equals the conception of the body schema in Merleau-Ponty quite closely and the function it plays for constitution embodiment. Bower and Gallagher come to a similar conclusion about Noë’s work.

An enactive account of perceptual presence integrates bodily factors into the per-

ceptual event as an essential, constitutive ingredient. The body here is understood

as what phenomenologists call the ‘lived body’, which includes the related notion

of a ‘body schema’ and the full ensemble of bodily factors prenoetically governing

conscious life (…). (Bower and Gallagher 2013, p. 110)

RECS (Chemero 2009; Rietveld and Kiverstein 2014) is another member of PEC that defends a version of constitution embodiment. A central term of RECS that exhibits the idea of constitution embodiment is that of an affordance. For proponents of RECS, we experience and understand not objects in the world, but affordances. An affordance itself is a relation between the embodied abilities of an agent and the agent’s environment (Chemero 2003). Accordingly, we always experience or understand something that is at least co-constituted by embodied abilities and thereby our embodiment.

100

Importantly, embodied abilities need to be understood in the context of affordances as structural features of the body that determine the structure of the world that an agent encounters. That is, embodied abilities must not be merely understood as an ability, that belongs to a body, that is the causal precondition of perceiving or understanding the world as it is. This becomes clear if we exemplify what an affordance is. For instance, if I perceive a chair affordance, I do not experience a chair as a chair because I have the perceptual ability to perceive chairs. Rather, I experience a relation between me and an object that affords me sitting, because I have a body that can sit in the particular environmental situation. In accord with these deliberations, Silberstein and Chemero

(2015) defend a form of neutral monism, which they also ascribe to Husserl, Heidegger and Mer- leau-Ponty, according to which the world is constituted by relations that are co-constitutively dependent on embodied human beings.

As we can see, various contemporary philosophical projects, that are strongly inspired by phenomenological thought, are committed to one form or another of constitution embodiment. Even though the particular conceptions vary, there is a commitment to the idea that the body is in some sense co-constitutive of the world as it is experience-able or understand-able. Despite this popularity of constitution embodiment, the idea itself is not as well spelled out as we might want it to be. In the following I provide a detailed discussion of the work of Husserl, Merleau-Ponty and

Heidegger on constitution embodiment to remedy this circumstance. This will not only provide

PEC with a coherent and detailed conception of constitution embodiment, but it will also unveil major inconsistencies in the work of proponents of PEC on constitution embodiment that we can thereby remedy.

101

3.2.2 Husserl and Merleau-Ponty on Embodied Spatiality

Husserl and Merleau-Ponty both claim that the body, as lived body (Leib) or phenomenological body—as an organizational structure of experience—is constitutive of the objects of our experience (Dingkonstitution) (Husserl 1989; Merleau-Ponty 2012).

The qualities of a material thing as aestheta, such as they present themselves to me

intuitively, prove to be dependent on my qualities, the make-up of the experiencing

subject, and to be related to my Body and my ‘normal sensibility’. (Husserl 1989,

p. 61, emphasis in original)

Similarly, Merleau-Ponty claims that ‘perceptual consciousness’, which is dependent on the body schema, ‘constitutes its objects’ (Merleau-Ponty 2012, p. 28); that ‘my body is that by which there are objects’ (Merleau-Ponty 2012, p. 94); and that the body schema not only unites my senses, but also unites the objects of my perception (Merleau-Ponty 2012, p. 244).

These are strong claims that suggest that the body plays a significant role in the constitution of how we experience and understand objects—or even in the constitution of objects themselves. In the following I discuss examples provided by Husserl and Merleau-Ponty to receive a better understanding of constitution embodiment.

Husserl discusses the role of the Leib for object constitution as an organ of perception

(Wahrnehmungsorgan). In order to do so, he describes the dependence of perception on the body.

For instance, touch is dependent on the hands, hearing on the ears.

The Body is, in the first place, the medium of all perception; it is the organ of per-

ception and is necessarily involved in all perception. In seeing, the eyes are directed

102

upon the seen and run over its edges, surfaces, etc. When it touches objects, the

hand slides over them. Moving myself, I bring my ear closer in order to hear. (Hus-

serl 1989, p. 61, emphasis in original)

This characterization of constitution embodiment is admittedly weak. Husserl does not state more here than that the sense modalities that present us objects refer back to the body, for instance, touch to the hands. Obviously, Husserl presents this characterization of constitution embodiment in the context of his general considerations about constitution, i.e. Husserl has already established object constitution as dependent on a subject as part of his transcendental-phenomenological project. This subject has been primarily characterized, resembling Kant, as a form of pure consciousness. And accordingly, Husserl seeks to introduce with the considerations above the idea that this constituting subject is embodied. Yet, the passage above does so only in the form of a trivial entailment of his transcendental project and does not provide a sufficient characterization of constitution embodiment on its own.

Similar concerns apply to other examples that Husserl provides to elucidate constitution embodiment. For instance, Husserl discusses how disturbances of and changes to the sense organs cause changes in the perception and appearance of objects. He describes how blisters on one’s fingers cause changes in the perception of the textual structure of objects and how taking drugs or wearing colored glasses changes the perceived color properties of objects (Husserl 1989, p. 66).

Again, these considerations provide only a case for constitution embodiment within the wider framework of Husserl’s transcendental project. Without this context, these examples could be taken as descriptions of, for instance, breakdowns of the proper functioning of sense organs, which create disturbed perceptions, rather than differently constituted objects. This means, the examples

103 of constitution embodiment that Husserl has provided so far could even be cited by proponents of body neutrality, if taken out of context of Husserl’s wider project.

Different from these weaker considerations is the following case. Husserl (1989, p. 56) presents the lived body and its role in object constitution as ‘the bearer of the zero point of orientation’, or more mundanely expressed, as the necessary perspective on the world.

Furthermore, obviously connected with this is the distinction the Body acquires as

the bearer of the zero point of orientation, the bearer of the here and the now, out

of which the pure ego intuits space and the whole world of the senses. Thus, each

thing that appears has eo ipso an orientation relation to the Body, and this refers not

only to what actually appears but to each thing that is supposed to be able to appear.

If I am imagining a centaur I cannot help but imagine it as in a certain orientation

and a particular relation to my sense organs: it is ‘to the right’ of me; it is ‘approach-

ing’ me or ‘moving away’; it is ‘revolving’, turning toward or away from ‘me’—

i.e., from my Body, from my eye, which is directed at it. (Husserl 1989, pp. 61-62)

Similar points about the perspectival nature of perception, relative to our embodiment, can be found in Merleau-Ponty’s work. Merleau-Ponty claims that the body is our ‘point of view on the world’ (Merleau-Ponty 2012, p. 73) and that the body is functioning as ‘pivot of the world’ (Mer- leau-Ponty 2012, p. 84).

104

Even though not fully explicated, Husserl describes here a particularly strong condition of constitution embodiment. Like Kant and the transcendental tradition, Husserl argues that space is intuited: ‘the pure ego intuits space and the whole world of the senses’ (Husserl 1989, p. 61).64 Im- portantly, space is not intuited as one thing among others; it has an excellent role as an invariant structure of experience, because, as already Kant argued, it is the precondition of objecthood in the first place:

One can never represent (rather: imagine/vorstellen) that there is no space, although

one can very well think that there are no objects to be encountered in it. It is there-

fore to be regarded as the condition of the possibility of appearances, not as a de-

termination dependent on them, and is an a priori representation (rather: imagina-

tion/Vorstellung) that necessarily grounds outer appearances.65 (Kant 1998, A24,

B38, brackets mine)

Each object that I can meaningfully conceive of—imagine—either in thought or experience, requires that I understand it as a spatial object.

Where Husserl seems to deviate from Kant, is that the former argues that bodily perspective characterizes the spatiality of our experience. Accordingly, we do not experience the world in a disembodied space in which all parts are not differentiable from each other, as Kant argued (Kant 1998,

B40, A25). Rather, the spatiality of our experience is relative to me and my perspective—parts of

64 Thompson (2011) follows Kant and Husserl here—even though in an inconsistent way as we will see in section 3.3.2—in that he claims that spatiality and temporality are created by the body. Noë (2004) also makes clear that embodied beings enact spatial content. 65 Kant uses the German terms ‘Vorstellung machen’ and ‘Vorstelleung’ in the passage quoted here (Kant 1956, A24, B38). To ‘vorstellen’ or ‘Vorstellung machen’ means in German clearly to imagine something—and might be technically translated as ‘intuiting something’. A ‘Vorstellung’ is an imagination—and might be technically translated as an ‘intuition’. To translate these terms with ‘representation’ and ‘represent’ is the expression of an unwarranted rationalist re-interpretation of Kant’s work that severs the clear connection between Kant’s and Husserl’s work shown above, which helps us to understand constitution embodiment more clearly as part of transcendental philosophy.

105 space, or rather, places, are closer to me, others far away, and still others show merely up as a hardly differentiable horizon. In that sense, Husserl provides an important correction of Kant’s disembodied conception of space.66

Yet, it seems that we can draw an even stronger conclusion from Husserl’s considerations. We can argue that the size of objects is experienced relative to our morphological embodiment. Objects appear as tall and small relative to my own size and effector space. In that sense, embodiment is in a twofold sense constitutive for space. First, it determines the spatiality of our ordinary, or as

Heidegger would state, ‘everyday’ conception of spatiality in which we dwell. Second, the possibility of acting depends on this form of embodied spatiality. One important condition for the possibility of acting on objects is the size of objects relative to my body and relative to, for instance, the reach of my arms and grasp. Understanding the spatiality of objects as relative to my own embodiment is the condition that makes it possible to utilize objects for action. Similarly, Merleau-

Ponty (2012, p. 140, emphasis mine) states: ‘to move toward an object, it needs to exist for the body’. And an object exists only for the body, if the object reflects the abilities and properties of the body—bodily spatiality among them—so that the body can be receptive to the object.

If we interpret Husserl in this sense—that embodiment is co-constitutive of the spatiality of objects— we can gain even more insight from his work.

66 Please mind that this Husserlian point does not imply that we have to think of every object in an experiential relation to us or another agent. Rather, it means that, even if we think about things abstractly, that they reflect the structure of our embodiment, because our thinking, which is governed by our understanding, reflects the structure of our experience. And this experience is co-constituted by our embodiment. Mind that this argument is transcendental. We cannot directly evaluate whether we can think about things independent of our experiential structures or not. Rather, if follows from the basic assumption of the transcendental tradition that we cannot understand anything meaningfully without being able to experience it potentially.

106

To the possibility of experience there pertains, however, the spontaneity of the

courses of presenting acts of sensation, which are accompanied by a series of kin-

esthetic sensations and are dependent on them as motivated: given with the locali-

zation of the kinesthetic series in the relevant moving member of the Body is the

fact that in all perception and perceptual exhibition (experience) the Body is in-

volved as freely moved sense organ, as freely moved totality of sense organs, and

hence there is also given the fact that, on this original foundation, all that is thingly-

real in the surrounding world of the Ego has it relation to the Body. (Husserl 1989,

p. 61, emphasis in original)

Now it might seem at first glance that Husserl argues here for a position similar to the sensorimotor view of experience that has been proposed by O’Regan and Noë (2002) and Noë (2004). This position states that the perceptual experience of objects is dependent on the movement and potential movability of our bodies. Take visual perception as an example. Our retina is at best receptive to information about a two-dimensional world. Yet, we experience a world in three dimensions.

Accordingly, we need to account for what adds to perception so that we can perceive the world as we do. O’Regan and Noë argue that full-blown perception comes about, if the perceiving agent utilizes her sensorimotor, bodily understanding. This understanding accounts for the possible movements that the agent could perform with regard to the objects of her perception, which entails an understanding of the possible perceptual vantage points that the agent could have with regard to the objects of perception. For instance, through my sensorimotor knowledge I understand how the backside of the tomato in front of me could look like if I would sense it directly from a position opposite to mine. This perspectival understanding then allows for perceptual understanding of full- blown objects, rather than of an understanding of two-dimensional objects only.

107

As insightful as this account might be, it is only an account of the perception of objects. However, it does not account for the constitution of objects and accordingly falls short of constitution embodiment.

Yet, if we understand Husserl along the lines that I have suggested above—that our embodiment co-constitutes the spatiality of our experience, we can see how movement and movability are constitutive for the spatiality of objects. As I suggested above, the body of an agent determines the experience of the spatial properties of the objects surrounding her. Yet, these properties are not only determined by static factors, i.e. the particular size and morphological form of my body. Ra- ther, movement, or, my ability to move in particular ways, further specifies the spatiality of objects.

Movement and movability add a dynamic factor to the morphological one. The spatial attributes of objects are dependent on the pace with which I can move relative to them. For instance, the apple in front of me is not only close, because I can reach it fast, but it is also small, since I can walk around it in a matter of a ‘stride’.

According to Husserl and Merleau-Ponty, embodiment constitutes the spatiality of objects in a two-fold sense. First, it determines their everyday appearance—the appearance that makes objects objects for us and that creates the familiarity that we have with them. An apple is partially an apple for me, because it has the particular spatiality that it has in most cases in which I encounter it and that it can potentially take. Second, constitution embodiment is a condition for the possibility of action. If objects would not reflect the properties and abilities of the agent’s body that acts with and on them, they would not be there for the body, as Merleau-Ponty argues.

As central and important as spatiality is for the constitution of objects, it is only one among various constitutive elements of objects. Importantly, the spatiality of objects primarily seems to have significance for us in so far as it is relevant for action. Yet, action is not only not motivated by the

108 spatiality of objects—I do not want to sit on a chair, because a chair has a particular spatial profile, even though this spatial profile allows me to sit in the first place. Rather, I am motivated to sit on a chair, if I have concern to do so. But, action further requires concern so that we experience and understand things as things for action, i.e. concern is the condition of the possibility that things show up as things that matter to us.67 Importantly, this concern is of an embodied nature. In the following I present Heidegger’s considerations on concern (‘care’) and show how they lend themselves to a comprehensive version of constitution embodiment in connection with the deliberations of Husserl and Merleau-Ponty discussed above.

3.2.3 Heidegger on Care

That Heidegger has not explicitly written about the body, with a few exceptions (Heidegger 1962, p. 368, 2001), is ‘well-known’ (Guignon 2009; Zahavi and Overgaard 2012). Yet, we should not confuse Heidegger’s silence on the morphological, let alone biological body, with silence on the body itself. Despite generally circumventing the word ‘body’, Heidegger has provided a particularly sophisticated idea of what I denote with the term ‘constitution embodiment’. What is of particular relevance here is Heidegger’s idea of care (Sorge), or, as I will call it in the following,

‘concern’.

Heidegger argues in Being and Time that concern is the being of Dasein, the ontologically basic way of being-in-the-world (Heidegger 1962, p. 329, brackets mine): ‘Dasein’s being is care (concern)’. This means that concern is a condition of the possibility of being-in-the-world. Differently

67 Please mind that this statement does not mean the circular, ‘concern is necessary for concern’, but rather, ‘concern is the necessary condition for things showing up as concernful’.

109 put, concern is an invariant structure of our experience and understanding. Concern itself has the following four structural components (Strukturmomente): Affect, understanding, falling and discourse (Heidegger 1962, p. 335).68 In the following I will focus on affect to show how concern provides us with a convincing conception of constitution embodiment.

The considerations by Husserl and Merleau-Ponty that I have discussed above provide a conception of constitution embodiment that suggests that our embodiment determines the spatiality of objects. But as such, objects are not sufficiently individuated. The objects that surround us are not just points in space-time. Neither are they merely persistent (beharrlich) substances, as Kant argued. Rather, objects have significance to an agent that enacts them.

This is quite obvious, when we think about artifacts, such as a car.69 A car ordinarily appears as a car, in that it affords fast, safe and comfortable travelling from point A to B. Being fast, safe and comfortable are properties of cars that are co-constituted by our embodiment. A car is fast compared to how swift we can move. A car is safe compared to the physical robustness of our bodies.

And a car is comfortable, given that embodied beings such as us sense sitting in cushioned seats in a particular way. Yet, importantly, these properties show up, because we have, as Heidegger would state, concern for time-efficiency, safety and comfort—in particular, these things have affective appeal to us.70

68 Affect is my translation of the German term ‘Befindlichkeit’ (Heidegger 2006). It is also translated with ‘state-of- mind’, such as in the standard English translation by Macquarrie and Robinson (Heidegger 1962) or with ‘affectedness’ (Dreyfus 1991). 69 The phenomenological or transcendental method does not provide definitions of entities, but establishes the conditions of the possibility of these entities. Before we discuss the example of how our experience and understanding of cars is co-constituted by concern, it is important to keep in mind that concerns vary from person to person and are not only based on a person’s embodiment, but also on a history of individual and social interactions with one’s world. Therefore, the following is not attempting to give a definition of a kind (car), but to exemplify how one possible among many ways in which concern as a condition of the possibility of understanding cars, might present cars to a person. 70 Please mind that I describe here only one possible way according to which an object can appear to me as concernful. Importantly, these ways can vary from person to person, from animal to animal, from organism to organism. Im- portantly, phenomenology does not define objects, but analyzes the conditions under which objects can show up as

110

If my concern structure had no place for time-efficient travel, say, in case I were an immortal being that does not have to care about time, a car would accordingly not show up as a fast means of travel, even if it were objectively ‘faster’ than me walking from place A to B. Importantly, we are not immortal beings, or beings that are so robust that they do not have to care about safety. And, we are such beings, because we have a particular embodiment. We are mortal and vulnerable. Even more importantly, we are not only such beings, but we care for our survival and goals, since we are beings with embodied interests, i.e. we have affect. I fear death and accordingly want to travel safe. I feel joy if I make ‘the sale’, and accordingly need to be on time for the meeting. I appreciate comfort, and dislike the strain in my legs from long walks. Given these very embodied concerns, things show up in a certain way for me. A car as safe. A bed as comfortable. A computer as a time- efficient means of working.

One might reply that this is the case for artifacts, but is this also the case for non-artifacts?

Heidegger’s answer to this is clear.

Here, however, ‘Nature’ is not to be understood as that which is just present-at-

hand, nor as the power of Nature. The wood is a forest of timber, the mountain a

quarry of rock; the river is water-power, the wind is wind ‘in the sails’. (Heidegger

1962, p. 100, emphasis in original)

Let us imagine to walk along a forest path in the woods. If I am not a dedicated lumberjack, I will not primarily conceive of the trees around me as commercial forest—pace Heidegger’s example.

Rather, trees might appear as objects of contemplation and might be part of my concern to recreate

objects. I merely use an example here, that could have been described differently, given different concerns. What is important is not under which particular concerns things show up as things, which, again, can differ from person to person. What is important is that things have to show up as things that are concernfully individuated.

111 in nature. Alternatively, I might conceive of them as obstacles, that hinder me from moving com- fortably and safely. For instance, if I leave the path I might fall over the trees’ roots that cover the forest floor and I might break my leg or might cut myself with branches.

That I can cut myself is relative to the particular density of the branches relative to my skin and movement. My skin is vulnerable, because it can be cut quite easily by something mundane like a tree branch and the tree branch is sharp and dense, because a creature with my skin and my movement speed will cut herself with the branches. Importantly, among all the possible relations that hold between me and trees, these things show up, because they are significant.71 They are significant to me, because my body can get easily hurt relative to this forest structure. But what ultimately grounds this significance is not my morphological or biological body, but the circumstance that I have affective concern for not hurting myself and for averting pain. Accordingly, the same considerations that apply with regard to constitution embodiment towards artifacts also apply to non- artifacts.

Here we can clearly see how Heidegger provides a conception of constitution embodiment. Affect is embodied and affect underlies concerns. Thereby, concerns are not mere concerns, but concerns of embodied beings, immersed in a world of objects that matter to them.

71 Please mind that concern is not about wishful-thinking or wanting things to show up in a particular way. Usually we have no control over nor any awareness of our most basic concerns, that determine the structure of how things appear to us.

112

3.3. Conceptual Problems with PEC’s Use of ‘Constitution Embodiment’

I have focused in my analysis of constitution embodiment on two factors. How the body constitutes the spatiality of objects and how the body renders objects as significant in the first place. In the following I will analyze how PEC’s conceptions of constitution embodiment fare with regard to these considerations. I will show that PEC’s versions of constitution embodiment are inconsistent in that they confuse constitution embodiment, a transcendental claim, with a positive claim. I will further show, that the focus of PEC on the location of the body is an entailment of this inconsistency, that yields unproductive philosophical and scientific debates and research alliances.

3.3.1 Constitution Embodiment is a Transcendental Claim — not a Scientific Claim

Constitution embodiment is what Heidegger calls an ‘ontological’ claim; a claim about structures that determine the possible ways the world can appear to us and how we can understand it. Differ- ently put, constitution embodiment is a transcendental claim that makes the existential conditions—in this case, bodily conditions—explicit that are the possibility of experience and understanding. As we have seen above, the body co-constitutes the spatiality and significance of objects that we understand and experience. As such, constitution embodiment is a genuinely philosophical claim that is established by means of the phenomenological method. Importantly, constitution embodiment cannot be established or formulated as a claim that ‘explain(s) Reality ontically by Real connections of interaction between things that are Real’ (Heidegger 1962, p. 251), i.e. ‘Being cannot be explained through entities’ (Heidegger 1962, p. 251).

Unfortunately, many approaches by proponents of PEC defend a version of constitution embodiment that does not treat the body as a constitutive condition that gives us entities, but as a physical

113 thing among others, i.e. as an entity. Take for instance enactivism. The bacterium as well as the world of sugar that it inhabits—see the example in section 3.2.1—is treated exactly as what

Thompson (2007, p. 13) has called a ‘prespecified world’. A world that can be accounted for by providing a description of how the organism is physically, chemically and biologically coupled to its environment—a world that can be ‘explain(ed) by Real connections of interaction between things that are Real’ (Heidegger 1962, p. 251). Differently put, Thompson and Stapleton treat sugar as a prespecified given that can be represented in different ways, for instance as food, instead of as something that is itself constituted or ‘enacted’ by the body.

This obviously entails an inconsistency. It is analytically impossible to argue for constitution embodiment from a perspective that is different from our own, since constitution embodiment states that an embodied agent constitutes her own world. Accordingly, we can only describe the world from our own perspective, by describing it as reflective of our embodiment. And this is importantly a phenomenological task, and not a scientific one that describes the body, as a physical given, as coupled to a pre-given world.72

A similar criticism applies to RECS. If affordances are co-constituted by our embodiment, if doing science is itself dependent on embodied abilities (Rietveld and Kiverstein 2014), and if we encounter the world by experiencing affordances, then we cannot treat the body as a physical thing that is located in a world that is not itself co-constituted by our own embodiment. But this is exactly what RECS does when it treats claims about affordances and the relationship between agents and

72 Similar considerations apply to Noë’s work. Noë (2005) explicitly asks for an intermeshing of science and philosophical method and a rejection of pure conceptual analysis. Noë (2007) does the same, only with regard to phenomenological analysis rather than conceptual analysis. Here, as in the case of Thompson, the problem emerges that the boundaries between ontological or transcendental and positive or empirical claims vanish.

114 their environment as scientific claims that can be primarily explained by means of dynamic systems theory and the explanatory apparatus of ecological psychology (Chemero 2009; Favela and

Chemero 2016). The body is treated here as an entity, not as a condition of the constitution of entities.73

Like enactivism, RECS commits the fallacy to understand the Leib not as a constitutive condition of understanding and experience, but as a real, physical thing that is placed inside the world that it is supposed to co-constitute. Like enactivism, RECS confuses an ontological or transcendental claim with a positive or ontic claim.

This is not only inconsistent, but it also leads to the problem that enactivism and RECS risk that their claims receive uncharitable re-interpretations. For instance, enactivism can be, in its current form, consistently reduced to selective representationalism (Mandik and Clark 2002) which claims that we merely represent selectively features of a prespecified world that are important for action.

And RECS is at the risk that affordance theory is reduced to the idea that affordances are functional predicates for action, not unlike the post-hoc value predicates criticized by Heidegger (1962) and

Dreyfus (2007). Obviously, this is not in accord with the phenomenological foundations of enactivism and neither consistent with the phenomenological and neutral monist foundations of RECS that clearly attribute an ontological role to affordances and our experience of them (Chemero 2009;

Silberstein and Chemero 2015).

Here we see clearly that PEC might not be as clear about the relationship that is supposed to hold between phenomenology, or philosophical method in general, and scientific research, as the plethora of literature on the topic otherwise seems to suggest.74 If, for instance, the body is treated

73 I am not claiming that the body should never be treated as a biological or physical entity. It should merely not be treated that way in the context of the constitution of experience and its objects. 74 See the literature referred to in the introduction.

115 undifferentiatedly as a condition of the constitution of entities and as an entity, inconsistency is entailed.

Obviously, we can treat the body as a condition of the constitution of entities and as an entity— dependent on the context of research. For instance, if we research into the nature of psychological, positive phenomena such as concrete movement or problem solving, we have to treat the body as an entity. But if we deal with ontological phenomena, we need to treat the body as a condition of the constitution of entities and not as an entity.

3.3.2 Problems with Claims about the Location of the Embodied Mind

A conceptual entailment of the inconsistency demonstrated above is how the body is treated by proponents of PEC in relation to space. The body is treated as an entity located in space. Indeed, it seems as if claims about the location of the ‘embodied mind’ are the major argumentative domain of PEC. The basic claim is that the embodied mind is not in the brain or body alone, but that it is in the world (Wheeler 2005; Rowlands 2010; Gallagher 2015), ‘out of our heads’ (Noë 2009), spanning the entire animal-environment system (Favela and Chemero 2016), or that it is an extended phenomenological-cognitive system (Silberstein and Chemero 2012). The embodied mind cannot be a brain in a vat but requires its environment (Thompson and Cosmelli 2011) and the embodied agent is constantly coupled to its environment (Thompson 2007; Chemero 2009)—so the arguments go.

But, as we have seen in section 3.2.2, the body itself is a constitutive condition of the possibility of how things show up to me in space. As Heidegger states:

116

Dasein is essentially not a Being-present-at-hand; and its ‘spatiality’ cannot signify

anything like occurrence at a position in ‘world-space’, nor can it signify Being-

ready-to-hand at some place. Both of these are kinds of Being which belong to

entities encountered within-the-world. Dasein, however, is ‘in’ the world in the

sense that it deals with entities encountered within-the-world, and does so concern-

fully and with familiarity. (Heidegger, 1962, p. 138)

If that is the case, the body, at least as conceived of according to constitution embodiment—as an ontological individuation condition—cannot be located in space itself, because the way we conceive of space is itself constitutively dependent on the body.75 As Merleau-Ponty (2012, p. 40) makes clear, ‘the thing is in place, but perception is nowhere, for if it were situated it could not make other things exist for itself, since it would remain in itself in the manner of things.’76

Constitution embodiment is a condition of the constitution of entities, not an entity itself and can therefore not be located anywhere. To repeat this point; it is clear that there are interesting psychological and biological claims about positive phenomena such as intelligence, motor control or problem solving, which pertain to the location and potential extendedness of these phenomena.

But we need to clearly pry apart what pertains to this positive, psychological project and what pertains to the ontological project surrounding constitution embodiment. And PEC is at the very least not clear about making the distinction between the body as a condition of the constitution of entities and the body as an entity that is the subject of psychological processes that might or might

75 One might claim that the way that we experience things in space is dependent on the brain. Since the brain is located in space, it seems that we can locate that on which spatial experience depends in space. What is fallacious about this is that our conception of the brain and how it interacts with cognitive phenomena is itself only meaningful because we conceive of it in experience; differently put, our access to the brain depends constitutively on experience. 76 My point here is obviously not to claim that the we should accept brain or body internalism about the embodied mind. Rather, my point is that we should not make claims about the location of the embodied mind in the first place.

117 not extend into the environment. We can see this in Silberstein’s and Chemero’s work on neutral monism and the temporality of experience.

The experience of time, for example, is neither in the head (the subject) nor the

external world (the object), the experience is fundamentally relational or extended.

It is the self-consistency relation between subject and object that allows for the

experience of time. This relation or structure is not in anything nor located any-

where. (Silberstein and Chemero 2015, p. 192, emphasis mine)

Silberstein and Chemero state correctly—following the phenomenological tradition—that the experience of time, or perhaps rather the temporality of experience, has no location. At the same time, though, they claim that the structure that is the condition of the possibility of the temporality of experience is extended. But, as soon as something is extended, as already Descartes (1986) made clear, something is taking place in space and has thereby a location. And this then, again, leads to inconsistency.77

3.3.3 Problems with Treating Affect as a Psychological Phenomenon

As Heidegger (1962) has argued—see section 3.2.3—objects that I encounter in the world are not spatiotemporal objects, nor substances with properties. They are first and foremost things that mat-

77 We encounter a similar mistake in the literature quite often. On the one hand ‘localism’ (de Oliveira and Chemero 2015) or the ‘the spatial containment language of internal/external or inside/outside’ (Thompson and Stapleton 2009, p. 26) are correctly criticized. But then on the other hand, Oliveira and Chemero argue for extended cognition based on empirical evidence and Thompson and Stapleton provide an account of the embodied mind as a neurological system that is continuously coupled to its environment; i.e. they defend claims about the extendedness of what they identify with the physical realization basis of the embodied mind. Yet, as I have pointed out, if something is extended, it is localizable in space.

118 ter to me. Yet, objects do not matter to me based on their physical properties. Objects, as I experience and understand them, matter to me, because they are co-constituted by concern; importantly, by affect.78

This Heideggerian insight has been widely ignored by most proponents of PEC—with the laudable exception of the so called ‘autopoietic’ branch of enactivism (Thompson 2007; Thompson and

Stapleton 2009). Independent of the inconsistency that I have demonstrated in section 3.3.1,

Thompson and Stapleton (2009) go at least in the right direction, when they claim that that, for instance, sugar shows up for the bacterium as sugar-as-food because it has significance for the organism. Unfortunately, Thompson and Stapleton treat this significance as the result of chemical coupling, but other work by Thompson (2007; 2011) suggests that he correctly identifies this generation of significance with purpose, which he grounds in affect. Again, this insight is tainted, since affect is treated by Thompson as a positive phenomenon, which entails that affect plays merely a selective representational role in the generation of significance. But, if PEC becomes clear about the treatment of affect as a constitutive phenomenon that pertains to constitution embodiment, this problem can be circumvented.

Enactivists, and with them all proponents of PEC, need to be clear about whether they treat affect transcendentally or psychologically. Only a psychological conception of affect is also a biological phenomenon that is locatable in the ‘inner body’ (Colombetti and Thompson 2008; Kiverstein

2012; Stapleton 2013; Colombetti 2014; Kiverstein and Miller 2015). Only a psychological conception of affect determines the effort that is necessary to perform an action (Bower and Gallagher

2013). Only a psychological conception of affect pertains to direct relevance detection, motivation or the guidance of attention (Stapleton 2013; Colombetti 2014; Kiverstein and Miller 2015). All

78 See section 3.2.3.

119 these aspects of affect are psychological, and in a certain sense, trivially embodied. They have to be clearly differentiated from a conception of affect that pertains to constitution embodiment; a conception of affect that co-individuates the objects that we can possibly experience and understand.

To be sure, a proper conception of constitution embodiment might entail an account of the psychological phenomena named above. Since objects are co-constituted through concern, they have, for instance, intrinsically motivational power and solicit options for action that show up relevant in a context. Yet, if we make use of a transcendental conception of affect for the explanation of psychological phenomena, we still need to be clear about what of the explanation of these phenomena requires a transcendental investigation, because it is about the invariant structures that constitute entities, and which part of it pertains to a positive, psychological project, that is about

‘variant’ structures of experience that vary from person to person and from culture to culture (Sar- tre 1962).

3.3.4 Philosophical and Practical Consequences of these Inconsistencies

So far my criticism has primarily focused on conceptual inconsistencies. But, there is also a different way to highlight what is problematic about confusing embodiment as a constitutive, ontological condition of entities with claims about psychological and biological entities. The problem is that claims about the physical location of the embodied mind—in the context of constitution— are usually conceptually empty. To be sure, they are conceptually empty because they are based on the conceptual mistakes pointed out above. But phrasing it this way makes it more apparent what is problematic about these inconsistencies.

120

Now, it is obviously difficult to show in concreto how claims like these are conceptually empty.

Yet, I think the following deliberations support my claim. Various philosophers, among them proponents of PEC, but also proponents of the extended mind (Clark 2008b), radical enactivists (Hutto and Myin 2013) and semantic externalists (Putnam 1975) all claim that the (embodied) mind is spreading into the world or part of the world. This should already give us pause, given the different conceptions of the mind to which these authors adhere. Take, for instance, the cases of Thompson on the one hand and Hutto and Myin on the other hand. Thompson is a phenomenologist, whereas

Hutto and Myin accept behaviorist, Davidsonian and Wittgensteinian conceptions of the mind.

Yet, both claim that the embodied organism is constantly, dynamically coupled to its environment and that the embodied mind spans the organism and the environment to which it is coupled.79

It seems, that if such claims about the location of the embodied mind were meaningful, i.e. not empty, that there were conceptual entailments that would either derive or at least suggest a phenomenological or a behaviorist-Davidsonian-Wittgensteinian account of the mind. But, it seems that there are none such entailments or criteria. Rather, both, the claims about the location of the mind and claims about the specific nature of the mind seem to stand in no relationship whatsoever; which is dubious, if claims about the location of the mind should be meaningful. The same does not hold for constitution embodiment proper. Constitution embodiment is inconsistent with both a behaviorist or Davidsonian-Wittgensteinian theory of the mind. For instance, constitution embodiment determines the significance of objects as deriving from the spatiality and concern structure

79 The coupling terminology shows us another problem with the interaction between phenomenology and a scientific approach. Thompson claims that ‘an autonomous system is always structurally coupled to its environment. Two or more systems are coupled when the conduct of each is a function of the conduct of the other’ (Thompson 2007, p. 45). Yet, this is a causal connection between entities. Contrary to that, Heidegger (1962) and other proponents of a classic conception of constitution embodiment argued that the objects of our experience and understanding are always objects of our world and that our world always reflects our (constitutive) embodiment. Here we see how a causal relationship, coupling, is confused with an intentional and co-constitutive relationship.

121 of an agent, which is at odds with a theory of meaning that primarily grounds meaningfulness in a linguistic speaker community, let alone with behaviorist notions that reject the idea that non-linguistic structures generate experience, understanding and meaningfulness.

In a similar fashion, and connected to the point above, is a pragmatic concern about proper scientific and philosophical alliances among research projects, which also underlines the problems with the conceptual inconsistencies pointed out earlier. We find constantly strong opposition between externalists and internalists about the embodied mind. For instance, Anderson and Chemero (2013) criticize Clark’s (2013) brain-based account of predictive coding as ‘threaten(ing) to return us to the bad old days of epistemic internalism’ (Anderson and Chemero 2013, p. 24). Similarly, Gal- lagher (2015, p. 99) criticizes proponents of grounded cognition—such as Barsalou, Gallese, John- son, and Lakoff—for being ‘internalists’ and thereby close to ‘classic cognitivism’. At the same time, for instance, Hutto and Myin (2013), are seen as part of either enactivism or RECS (Chemero

2009), because they are externalists about the embodied mind.

What is concerning about this, is that these research alliances are based on—at least with regard to constitution and specification—inconsistent and empty claims such as that about the location of the embodied mind. For instance, it seems dubious that Hutto and Myin’s ‘radical enactivism’ belongs to the same research project as Chemero’s or Thompson’s work, while the former claim that animals have no meaningful thought, because they lack language; where at least for phenomenologists meaningful thought arises from more basic structures, like concern. Further, language, as that what constitutes meaningful thought for Hutto and Myin, is understood by them in a disembodied, classic sense, according to which the world can, at least in principle, be mirrored as it is. This means, Hutto and Myin are close to the classic conception of body neutrality towards which constitution embodiment is radically opposed. And not surprisingly, they mock enactivist

122 claims—claims at the heart of constitution embodiment— about how embodied beings enact their world as ‘extravagant claims’ (Hutto and Myin 2013, p. 5).

At the same time, researchers that are committed to the idea that the experience and understanding of the world reflect our embodiment—such as Lakoff and Johnson (1999) and Gallese and Lakoff

(2005)—are criticized for being classic cognitivists. And other researchers (Berthoz and Christen

2009) that make strong claims about how our experience and understanding of the world are reflective of our embodiment are widely ignored by proponents of PEC, because they focus in their research on the brain. Importantly, these researchers make the same mistake as the proponents of

PEC; they are treating embodiment as a condition of the constitution of entities and as an entity.

Accordingly, their internalism needs to be revoked. But, the focus on constitution and specification in their work itself can be maintained and used for fruitful research alliances.

Obviously, there are strong differences in the projects of ‘externalists’ and ‘internalists’ when it comes to their research of psychological phenomena. In this sense it is sensible to stress the divide between these different frameworks. Yet, from a philosophical, ontological point of view it seems that the commonality researchers express towards how the body constitutes or at least specifies the objects of our experience and understanding, is significantly more important than the commonality that is based on conceptually empty claims about the location of the embodied mind.

Constitution embodiment is basically a rejection of the natural attitude, in particular of body neutrality. Yet, if we focus on the alleged location of the embodied mind, we not only overlook this circumstance, but run into the danger that we accept positions that are clearly expressions of the natural attitude, such as Hutto’s and Myin’s linguistic conception of thought.

123

Another example of the infelicitous consequences of focusing on location, is PEC’s relationship to the father of rationalism, Descartes. Cartesianism is the argumentative ‘opponent’ of many proponents of PEC (Wheeler 2005; Rowlands 2010; Gallagher and Zahavi 2012). The major reason for this hostility towards Cartesianism is the belief that Descartes is the founder of internalism.

Yet, contrary to this popular belief, Descartes (1986) made clear that the mind, as a res cogitans, is the opposite of a res extensa—a thing with extension—which entails by definition that the mind has no place or location in space. It is true that Descartes claimed that the pineal gland, as part of the brain, is the locus of mind-body interactions. But, since Descartes did obviously not assert that the mind is in any form reducible to the pineal gland or other physical entities, this should be taken more as a speculative, and probably rather incoherent claim of Cartesian philosophy than something that needs to concern proponents of PEC any further.

Descartes’ (1986) claim, that we can imagine ourselves without a physical body, is not problematic for proponents of constitution embodiment either, as long as we do not draw negative ontological conclusions from it, as Descartes did, pace Gallagher and Zahavi (2012). From a merely descriptive point of view it seems that we can imagine ourselves without a physical body—and, for the sake of the argument we can concede to Descartes that this is possible—again, as long as we do not draw, like Descartes, a negative ontological inference from it. Importantly, the objects that I imagine when I imagine that I do not possess a physical body, are still constituted by constitution embodiment and therefore reflective of constitution embodiment. If I withdraw meditatively from the world and my body and imagine a car, then this imagined car still exhibits those properties that are constituted by my embodiment—being a safe means of travel that has particular spatial properties that refer back to my size and pace. That Descartes would have accepted this claim is unlikely. But, it still shows that Descartes’ thought experiment is in itself not problematic. More

124 importantly, Descartes’ thought experiment clearly shows that it is not the physical body that matters for constitution embodiment, but the body as a condition of the constitution of entities.80

The problem with Descartes’s rationalism is not internalism or the idea that we can imagine ourselves without a physical body. Rather, the problem with Descartes and the rationalist tradition— as Dreyfus and Taylor (2015) point out correctly and as Husserl and Heidegger have done before them—is that they conceive of the world as independent of us and our embodiment.81 This means, rationalism supports body neutrality and thereby the opposite of constitution embodiment. And exactly this aspect of rationalism is not only overlooked due to the infelicitous focus on location, but it is even accepted by various proponents of PEC that focus on location, but leave rationalist conceptions of representation and thought untouched (Wheeler 2005, 2008; Rowlands 2010).

Constitution embodiment allows us to overcome body neutrality and with it the ‘intellectualism’

(Noë 2012) and ‘rationalism’ (Dreyfus 1971, 1988, 2007) at the heart of contemporary analytic philosophy and cognitive science. Instead of focusing on futile debates about the location of the embodied mind—at least when it comes to constitution—we should make use of fruitful alliances between theoretical frameworks that emphasize the constitutive role of the body for experience and understanding.

80 The case is different for ‘brain in a vat’ thought experiments, such as criticized by Thompson and Cosmelli (2011), which are commonly seen as an extension of Descartes’ thought experiment. Wheras Descartes’ thought experiment departs from a descriptive analysis of an experiential episode that we can have, a brain in a vat thought experiment starts from the very beginning with an ontological assumption or statement—that brains in vats could realize or are identical to the mind (or not)—but does not include any experiential analysis whatsoever. Wheras Descartes’ thought experiment is partially epistemically dependent on an analysis of an experiential episode, a brain in a vat thought experiment is merely justified by an intuition about whether internalism or externalism about semantic content is the case. Therefore, my partial defense of Descartes’ ‘thought experiment’, or rather, phenomenological analysis, does not apply to brain in a vat thought experiments, because I defend Descartes’ experiential analysis that we can have experience that does not refer to our physical body, not the negative ontological conclusion that he draws from it, which depends on his epistemology and the resultant idea that whatever we can imagine clearly and distinctly is also of ontological import. 81 Varela et al. (1991) are also clear about this. Their famous term ‘Cartesian Anxiety’ refers to the epistemic aspects of Cartesian philosophy, not to the ontological ones; let alone to those that pertain to the location of the mind.

125

3.4. Conclusion

Based on the work of Husserl, Heidegger and Merleau-Ponty, I have spelled out a consistent and sufficiently complex conception of constitution embodiment that can inform further philosophical research into the nature of constitution in general and into how the body constitutes the objects of experience and understanding in particular. I have further shown how PEC fails to use the concept of constitution embodiment consistently—in particular, PEC confuses a transcendental claim about the constitution of entities through the body with a claim about the body as an entity. As I have demonstrated, this is not a harmless inconsistency that can be easily removed, but a deeper- rooted problem that will require PEC to be more specific about the boundaries between philosophy and science. Further, PEC will be forced to move its research focus from the location of the body to how the body constitutes and specifies the objects of experience and understanding. PEC will also have to consider with which research projects it will have to work in the future. I have suggested that PEC should engage into research alliances with other projects that focus on constitution and specification, rather than on the location of the embodied mind.

An obvious problem is that PEC seems to be torn apart between being a psychological project that investigates the nature of positive phenomena, such as problem solving, and being a philosophical project that provides an ontological analysis of the structure of the world. Whereas the former project might rightfully focus on the extendedness of, say problem solving—or at least on how problem solving abilities require structures that are not only located in mechanisms in an organism’s brain—the latter project will have to cede its claims about the location of the embodied mind altogether. PEC has to be very clear about which project it is concerned with and which particular conception of embodiment it accordingly has to use, depending on the respective research context

126 with which it engages. Accordingly, PEC has to be clearer about the relationship between philosophy and science.

I hope my deliberations provide further insight into the truly ‘revolutionary’ or ‘radical’ potential of PEC. Instead of confining itself to claims about the realization-base of the embodied mind, in which PEC incoherently follows research questions defined by the metaphysical presuppositions of analytic philosophy and classic cognitive science, that themselves derive from rationalist philosophy (Dreyfus 1971; Dreyfus 1988; Varela et al. 1991; Dreyfus 2007; Dreyfus and Taylor

2015), PEC should embrace the opportunity to see itself as a successor of the phenomenological tradition, as Thompson (2007) and Noë (2012) already seem to suggest. As such, PEC should form a research program with which it can establish itself as a clear competitor to the rationalist agenda, that defends the natural attitude and body neutrality, which still guides wide sectors of contemporary philosophy and cognitive science. Clearly, constitution embodiment should be at the forefront of this research program.

127

Chapter 4. Overcoming the Disunity of Understanding

Abstract

I argue that embodied understanding and conceptual-representational understanding interact through schematic structure. I demonstrate that common conceptions of these two kinds of understanding, such as developed by Wheeler (2005, 2008) and Dreyfus (2007a, b, 2013), entail a separation between them that gives rise to significant problems. Notably, it becomes unclear how they could interact; a problem that has been pointed out by Dreyfus (2007a, b, 2013) and McDowell

(2007) in particular. I propose a Kantian strategy to close the gap between them. I argue that embodied and conceptual-representational understanding are governed by schemata. Since they are governed by schemata, they can interact through a structure that they have in common. Finally, I spell out two different ways to conceive of the schematic interaction between them—a close, grounding relationship and a looser relationship that allows for a minimal interaction, but preserves the autonomy of both forms of understanding.

4.1. Introduction

Proponents of embodied cognition as well as contemporary phenomenologists usually separate between embodied understanding on the one hand and conceptual-representational understanding on the other hand (Wheeler 2005; Dreyfus 2007a, b, 2013; Chemero 2009; Hutto and Myin 2013).

Whereas embodied understanding is supposed to constitute what we can call a practically-engaged

128 mode of dealing with the world, conceptual-representational understanding is supposed to constitute what we can call a theoretically-detached mode of thinking (section 4.2).82

The separation between these kinds of understanding is strict. This is apparent from how embodied and conceptual-representational understanding are characterized. Whereas embodied understanding is supposed to be governed by embodied and sensorimotor abilities, conceptual-representational understanding is linguistic or symbolic. Embodied understanding is context-sensitive, action-oriented and pragmatic; conceptual-representational understanding is general, abstract and disembodied. Accordingly, it seems as if the separation renders these kinds of understanding as autonomous and structurally unrelated to each other (section 4.2.1).

I argue in section 4.2.2 that we receive a segmented conception of the mind, in which two autonomous kinds of understanding pull the embodied agent into different directions, if we cannot account for the relationship between these two kinds of understanding. Given such a segmented conception of the mind, it is unclear how conceptual thought can relate to embodied understanding, as is the case when we report an action or make a judgment about a perceived situation (Dreyfus

2007a, b; McDowell 2007). Similarly, if conceptual-representational understanding is not structurally connected to embodied understanding, which is the ground of concernful interactions with the world, it is mysterious how conceptual-representational understanding exhibits practical meaning and significance (Heidegger 1962; Dreyfus 1991). Even though the problem of a segmented mind has been addressed by Dreyfus (2007a, b) and McDowell (1994, 2007), we are left without

82 The exception to the rule is Alva Noë (2004, 2012) who has rejected the separation as intellectualist.

129 a sufficient account to overcome it. Dreyfus (2007a) admits that he has no explanation of the relationship between these two kinds of understanding (section 4.2.1), and McDowell’s (1994, 2007) account is unacceptably intellectualist (section 4.3.1).

In section 4.3 I propose the following positive solution to overcome the separation. Like McDow- ell, I draw on Kant to do so. Yet, unlike McDowell, I refer to Kant’s thoughts on spatiotemporal schemata. For Kant, the basic structure of experience is that of spatiality and temporality. Kant connects conceptual thought with experience, by arguing that concepts are governed by schemata that determine the possible spatial and temporal forms that objects of experience can have. For instance, the schema of the concept of dog determines all possible forms that dogs can have in space and time and thereby allows the concept to be about the experience of things in space and time. Conceptual thought and experience are able to interact, since both exhibit a common structure: spatiality and temporality (section 4.3.1).

I argue in section 4.3.2 that spatiotemporal schemata govern both—embodied and conceptual- representational understanding. Schemata can be conceived of as ontologically and cognitively modest. They are ontologically modest, because most objects of the understanding can be minimally characterized by their spatial and temporal properties. They are cognitively modest, because understanding has to be minimally responsive to the spatial and temporal structure of the world surrounding us; i.e. the spatiotemporal world that contains objects and situations about which embodied and conceptual-representational understanding are.

If both kinds of understanding exhibit at a basic level spatiotemporally schematic structure, this schematic ground allows for a structural connection between both of them. In section 4.4 I describe two ways to conceive of this structural connection. We can conceive of it, first, as a grounding relationship, where embodied understanding grounds conceptual-representational understanding

130

(section 4.4.1). Or we can conceive of it, second, as a looser relationship, where schemata allow for ‘cross-talk’ between the two kinds of understanding, yet where the autonomy of each is pre- served to a certain degree (section 4.4.2).

4.2. The Separation between Embodied and Conceptual-Representational Understanding

In contemporary philosophy, a distinction is frequently made between two kinds of understanding—embodied understanding and conceptual-representational understanding. These two kinds of understanding are supposed to roughly correspond to two ways of engaging with the world that we can call ‘practically-engaged’ and ‘theoretically-detached’. In the following I discuss how these distinctions are drawn, what motivates them and which problems they entail.

Let us first focus on the phenomenologically inspired work of Michael Wheeler (2005, 2008) and

Hubert Dreyfus (2007a, 2007b, 2013). Both differentiate between different modes of engagement based on phenomenological analyses that are inspired by Martin Heidegger’s (1962) and Maurice

Merleau-Ponty’s (2012) work. The first mode of engagement that Dreyfus and Wheeler identify is reflective of our experience of most of our interactions with the world. They call it in accord with

Heidegger and Merleau-Ponty ‘readiness-to-hand’ or ‘smooth coping’ respectively. It is characterized by environmentally immersed, non-reflective and contextual action.

This mode of engagement exhibits no experienced distinction between subject and object and is structured by the body. As Wheeler states:

(…) smooth coping in the domain of the ready-to-hand has a non-representational

phenomenology. Smooth coping involves a form of awareness in which there are

131

no subjects and no objects, only the experience of the ongoing task (e.g. typing).

(Wheeler 2008, p. 338)

For instance, if I hammer a nail, I am engaging in a skillful activity in which my awareness of myself is lost in the activity. I am often not only not aware of myself in these engagements, I am also not aware of operating on determinate objects with decontextualized properties.83 To the contrary, I focus on the activity that is directed towards the end result of my action and the work that is to be achieved. As Heidegger states:

That with which our everyday dealings proximally dwell is not the tools themselves

[die Werkzeuge selbst]. On the contrary, that with which we concern ourselves pri-

marily is the work—that which is to be produced at the time; and this is accordingly

ready-to-hand too. (Heidegger 1962, p. 99)

In this mode of engagement, readiness-to-hand, things show up to me only as things-for-the-sake- of-the-work that I seek to achieve. In our case, hammer and nail show up contextually as things with which I, for instance, hang up a picture.

Dreyfus and Wheeler distinguish this mode of engagement, from deliberative, reflective, decontextualized and detached modes of engagement. We are involved in these detached modes of engagement, for instance, when we encounter practical problems, reason or do science. Both authors identify these modes of engagement with what Heidegger (1962) called ‘un-readiness-to-hand’ and ‘presence-at-hand’. In the former, we are disturbed in our smooth coping and search for solutions to the problems that caused the disturbance. In the latter, we take an observer stance towards the world and conceive of it in terms of objects with decontextualized properties.

83 Decontextualized properties are properties that objects have independent on the concerns and embodiment of an agent—if there are any such properties.

132

In both of these cases, a ‘cognitive distance’ is introduced between subject and object which is not present in skillful coping (Wheeler 2008, p. 383). This introduction of an experiential distinction between object and subject seems to entail for Dreyfus and Wheeler that we conceive of objects in detached modes of engagement in a literally ‘objective’ way. As Wheeler states:

When revealed as present-at-hand (e.g. by detached theoretical reflection) an entity

will be experienced in terms of properties that are action-neutral, specifiable with-

out essential reference to the representing agent, and context-independent. Moreo-

ver, according to Heidegger, this group of properties will also characterize the con-

tents of the agent’s related representational states. (Wheeler 2008, p. 339, empha-

sis mine)

Wheeler and Dreyfus argue that readiness-to-hand is governed by embodied understanding and that un-readiness-to-hand and presence-at-hand are governed by (action-oriented or classical) representations (Wheeler) or linguistic concepts (Dreyfus), which exhibit the characteristics of ‘conceptual content, mindedness, and rationality’ (Dreyfus 2013: 29). The assumption here is that embodied understanding enables engaged-practical modes of engagement with the world and that conceptual-representational understanding enables theoretical-detached modes of engagement with the world, in particular, decontextualized, body-neutral conceptual content.

We find similar claims about different modes of engagement and corresponding differences in the underlying kinds of understanding in other embodied conceptions of understanding. For instance, various authors argue for the existence of something like a pre-conceptual background that characterizes most of our engagements with the world. This background is conceived of as a holistic and contextual background structure that allows us to act and interact with our living world. This

133 pre-conceptual background is then contrasted with explicit concept use (Dreyfus 2007a, 2007b,

Wheeler 2008; Hutto 2012; Dreyfus and Taylor 2015).

Another similar separation is made in radical embodied cognitive science and closely related versions of enactivism (Chemero 2009; Kiverstein and Rietveld 2015). One central concept of this project is ‘affordance’ which is defined by Chemero (2009) as a relation between agents and en- vironments and, in accord with J.J. Gibson, as the embodied, pragmatic meaning of objects. In this sense, affordances grant an understanding of objects for engaged-practical purposes. The explanatory scope of affordances is yet unclear for Chemero (2009). In particular, he deems it an open question to what extent radical embodied cognitive science will be able to explain, what he calls in accord with Clark and Toribio (1994), ‘representation hungry tasks’.

As we have seen above, many authors differentiate between different ways of relating to the world.

On the one hand, we have a practical-engaged mode that pertains to action and perception. This practical-engaged mode is contextual, holistic, value-laden, body-centric and is governed by embodied understanding (Dreyfus 2007b; Wheeler 2008). On the other hand is a theoretical-detached mode, which is thought to include engaging in decontextualized, general propositional thought

(Hurley 1998), distinctly cognitive intentionality (Kelly 2002), doing science, theorizing and using language (Wheeler 2005; Chemero 2009), engaging with the world in a detached, observational fashion (Dreyfus 2007b), deliberative, reflective rationality (Dreyfus 2013), judging, believing or planning (Hutto and Myin 2013), or participating in the space of reasons (Dreyfus and Taylor

2015). This detached-theoretical mode is supposedly governed by conceptual-representational understanding.

134

4.2.1 The Separation is Strict

Now we need to ask how strict the separation between embodied and conceptual-representational understanding is. Is there continuity between the different kinds of understanding or are they radically separate from each other? The answers to this question differ. For instance, Hutto and Myin

(2013) see no connection between the embodied abilities that generate action and perception, and contentful conceptual-representational understanding. Contrary to that, Dreyfus in particular has stressed the continuity between the two kinds of understanding. As Dreyfus claims:

Intelligence is founded on and presupposes the more basic way of coping we share

with animals. (Dreyfus 2007b, p. 250)

Another expression of Dreyfus’s commitment to the continuity between both kinds of understanding is the following.

Absorbed bodily coping, its motor intentional content, and the world’s intercon-

nected solicitations to act provide the background on the basis of which it becomes

possible for the mind with its conceptual content to think about and act upon a

categorially unified world. (Dreyfus 2007a, pp. 360–361)

Importantly, Dreyfus deems conceptual-representational understanding constitutively dependent upon embodied understanding.

Similar remarks have been made recently from a Neo-Pragmatist perspective by Gallagher:

Pragmatists and neo-pragmatists would treat the intentionality of propositional at-

titudes as derived from a more original form of embodied intentionality, what phe-

nomenologists like Husserl and Merleau-Ponty call ‘motor intentionality’. (Gal-

lagher 2014, p. 121)

135

Further, in particular proponents of embodied cognition support the claim that cognition, action, and perception are integrated with each other (Thompson and Stapleton 2009) or that they are non- separable, interaction-dominant components of one dynamic system (Chemero 2014). If that is the case, i.e. if cognition, action, and perception are integrated and inseparable from each other, it should be entailed that both kinds of understanding are also integrated with each other and inseparable from each other, which again entails a strong form of continuity.

And not surprisingly, a standard response by proponents of embodied cognition to the question about the relationship between these kinds of understanding is that they are ‘somehow’ connected and continuous with each other. But there is no concrete philosophical explanation of what the relationship between embodied and conceptual-representational understanding actually is.84

We can see this at the example of Dreyfus’s (2007a) account. As we have seen above, Dreyfus claims, with recourse to Heidegger and in particular Merleau-Ponty, that motor intentionality and other embodied abilities enable conceptual-representational understanding. Yet, Dreyfus does not provide an explanation of the relationship between the two kinds of understanding. He merely claims that they are connected with each other, without providing an analysis of how our theoretical-detached engagements with the world characteristically exhibit signs of motor intentionality or embodiment.

To the contrary, Dreyfus (2007a, p. 364) claims that we experience ‘context-free, self-sufficient substances with detachable properties’ when we engage with the world as theoretically-detached, which he (Dreyfus 2007a, p. 364) identifies with ‘McDowell’s world of facts, features and data’.

However, it is not clear what it means that decontextualized, self-sufficient substances exhibit signs of embodiment or motor intentionality.

84Accounts by McDowell (1994, 2007) and Noë (2004, 2012) are the exceptions to this explanatory shortcoming. I discuss McDowell briefly in section 4.3.1. A discussion of Noë unfortunately goes beyond the scope of this chapter.

136

Further, Dreyfus states in another passage that ‘motor intentional content’ cannot in any ‘’form’’ be ‘’suitable to constitute the contents of conceptual capacities’’ (Dreyfus 2007a, p. 360), which seems to contradict the claims Dreyfus makes about how embodied understanding, in particular motor intentionality, is the basis of conceptual-representational understanding.85 And to further heighten the confusion, Dreyfus concludes that neither he nor Heidegger nor Merleau-Ponty would have been able to provide an account of the relationship between embodied and conceptual-representational understanding: ‘It seems that (…) the phenomenologists can’t account for what makes it possible for us to step back and observe [the world]’ (Dreyfus 2007a, p. 364, brackets mine).

Concretely, Dreyfus admits that he cannot account for the relationship between conceptual-representational understanding and embodied understanding.

4.2.2 Problems with the Separation

The lack of an account that explains the relationship between two kinds of understanding is deeply concerning. What is in particular concerning is that the two kinds of understanding, according to how they are standardly described, seem structurally disparate—so disparate even that they are seemingly autonomous from each other. If that were the case, it is deeply mysterious how they could interact.86

85 Dreyfus understandably has to make this claim, given his own theoretical presuppositions, to thwart John McDow- ell’s (1994, 2007) claim that experience exhibits conceptuality. However, Dreyfus’s claim entails unfortunately a contradiction to his own claims about the continuity between embodied and conceptual-representational understanding. ‘To focus on the motor intentional content, then, is not to make some implicit conceptual content explicit—that’s the myth—but rather to transform the motor intentional content into conceptual content, thereby making it available for rational analysis but no longer capable of directly motivating action’ (Dreyfus 2007a, p. 360). Worse, Dreyfus’s statement, cited in the main text, clearly cuts off embodied understanding from conceptual-representational understanding. 86 The real problem here is that embodied understanding is transcendental and that language is not transcendental. Rather, language is an object of experience and understanding, a structure that we use to communicate and memorize things. Accordingly, embodied understanding and linguistic structures are not comparable with each other. Therefore,

137

Yet, it is obvious that they do interact. For instance, as McDowell (2007) has repeatedly pointed out, we need to account for how our embodied experience can be the object of verbal reports and how it can inform conceptual judgments. And as Alva Noë (2004, 2012) has pointed out, our embodied understanding itself exhibits such cognitive complexity that it requires a close relationship to conceptual-representational understanding.

Further, if these kinds of understanding were separate and autonomous from each other, then it would be puzzling how human thinking and action are so synchronized in everyday behavior. If the two kinds of understanding were separately operating in an embodied agent, it would seem as if she would have to be torn in different directions, by autonomously operating kinds of understanding.

Even more, if embodied understanding is that which generates meaning and significance for an agent, as at least phenomenologists and phenomenologically-inspired philosophers argue

(Heidegger 1962; Merleau-Ponty 2012; Thompson 2007; Ratcliffe 2008; Noë 2012), then we need to address the question how conceptual-representational understanding itself receives meaning, in particular if we do not want to accept classic intellectualist conceptions of a disembodied intellect.87 Accordingly, we need to account for the relationship between embodied understanding and conceptual-representational understanding.

those who seek for an interaction of embodied understanding and an understanding that is supposedly governed by language, commit a category mistake. In the following elaborations, in which I focus on Kant’s conception of understanding, I do not criticize authors for this category mistake. What is important, is that I myself do not commit it, since I propose a Kantian conception of conceptual-representational understanding. Still, I need to admit though, that this leaves the interaction between linguistic entities and conceptual-representational understanding puzzling. 87 Importantly, Heidegger, who inspires Dreyfus and Wheeler to sharply separate between both kinds of understanding, does not make the same separation for two reasons. First, Heidegger does not differentiate between embodied understanding and conceptual-representational understanding, or rather, between readiness-to-hand and presence-at-hand, the way Dreyfus and Wheeler do. Heidegger nowhere claims that we are experiencing ‘context-free, self-sufficient substances with detachable properties—McDowell’s world of facts, features, and data’ when we engage with the world as present-at-hand, as Dreyfus (2007a, p. 364) suggests. Rather, he makes clear that presence-at-hand is grounded in readiness-to-hand. As he states, ‘readiness-to-hand is the way in which entities as they are ‘in themselves’ are defined ontologico-categorically’ (Heidegger 1962, p. 101). Further, ‘when we merely stare at something (presence-at-hand), our just-having-it-before-us lies before us as a failure to understand it any more. This grasping which

138

In what follows I seek to account for this relationship with the following strategy. First, I suggest that spatiotemporal schemata can function as the ground for an interaction between these kinds of understanding. Second, I describe different ways to conceive of the relationship between the kinds of understanding, given that they are connected through schematic structure.

4.3. Bridging the Separation with Schematic Structure

In the following I suggest that schemata are the means by which embodied understanding and conceptual-representational understanding interact. A schema, as I will argue, is an ontologically and cognitively minimal structure that preserves a certain degree of autonomy for both kinds of understanding, yet, allows for their close interaction. The following considerations draw heavily on Immanuel Kant’s conception of schemata, developed in the Critique of Pure Reason. I will introduce the conception of a schema based on his work and show how we can make the basic idea

is free of the ‘as’, is a privation of the kind of seeing in which one merely understands (readiness-to-hand). It (presence- at-hand) is not more primordial than that kind of seeing (readiness-to-hand), but is derived from it’ (Heidegger 1962, p. 190, brackets mine). And Heidegger makes elsewhere plainly clear that presence-at-hand is this deprived form of merely looking at things. The clearest expression of this might be, ‘theoretical behavior is just looking, without cir- cumspection’ (Heidegger 1962, p. 99, emphasis mine). This means, presence-at-hand is a ‘deficient mode of concern’ (Heidegger 1962, p. 103)—in which something that was formerly ready-to-hand is only just present (‘Nur-noch- vorhandensein eines Zuhandenen’) (Heidegger 2006, p. 73). Here we can also see the second reason, why Heidegger would reject Dreyfus’s and Wheeler’s separation between both kinds of understanding. For Heidegger, the difference between readiness-to-hand and presence-at-hand is not a division between conceptuality or representational thought on the one hand and sensorimotor or embodied understanding on the other hand. Heidegger nowhere claims that conceptual or representational understanding governs presence- at-hand. As we have seen above, Heidegger uses perceptual and agential vocabulary to describe how we encounter the world as present-at-hand. This makes further sense, if we consider the role of language in Heidegger’s care structure in the form of discourse. The care structure, Heidegger’s most basic ontological structure after temporality (Heidegger 1962, p. 329), is co- constituted by understanding, state-of-mind (affect), falling, and discourse (Heidegger 1962, p. 384-385). Understand- ing is characterized by its disclosing ability, which presents an object or a situation, together with affect, as mattering to an agent by showing up for which purpose something can be used (Heidegger 1962, p. 182). Discourse, which is the condition for speaking a language, is not conceived of as detached theoretical linguistic thinking, but as an existential structure that contributes to the structure of understanding and allows for its articulation (Heidegger 1962, p. 203-204). Discourse, as part of the care structure, is characteristic of both, readiness-to-hand and presence-at-hand. Yet, in both cases it is a co-constitutive part of the care structure, but never a self-standing intellectualist device for judgment and representation, as both Dreyfus and Wheeler suggest.

139 work in different ways, without having to accept the wider ramifications of Kant’s epistemological and ontological project. In order to do so, I spell out two different ways in which schemata can connect embodied and conceptual-representational understanding.

4.3.1 Kantian Schemata

Schemata are a central, though often overlooked aspect of Kantian philosophy (Sherover 1971;

Heidegger 1990; Carman 1999; Hanna 2005; van Mazijk 2016). They are the structures that explain, for Kant, how experience and conceptual thought are synthesized. In the following I will discuss schemata in the context of Kant’s considerations on the synthesis of experience and conceptual thought.

As is well known, Kant insisted that concepts and experience have to interact in order to make sense of either of them. Yet it is less well known that Kant is not satisfied with claiming that concepts and experience interact or that they are synthesized. Kant is concerned with the conditions that make it possible that concepts and experience can interact; i.e. Kant does not merely describe the necessity of the interaction between them, but he seeks to explain it further. This means that

Kant is concerned with the structure that is the ground for the interaction between concepts and experience, and accordingly with the conditions that make it possible that concepts are about objects of experience.

Kant argues that there has to be a common structural ground, a ‘homogenous’ ‘third thing’, that has to be definitive of concepts and experience so that they can interact.

Now it is clear that there must be a third thing, which must stand in homogeneity

[in Gleichartigkeit stehen] with the category on the one hand and the appearance

140

on the other, and makes possible the application of the former to the latter. (Kant

1998, A138 / B177, emphasis and translation in original)

Kant’s point is that concepts and experience have to exhibit a structural commonality, so that they can interact; or differently put, so that each can non-arbitrarily match with each other. Kant identifies this matching structure, the ‘third thing’, with schemata.

According to Kant, schemata can perform their function as the common ground of both experience and concepts by being ‘a priori time-determinations’ (Kant 1998, A145 / B184).88 Whether Kant then actually conceives of them merely as ‘a priori time-determinations’, i.e. as exhibiting temporal structure, is doubtful. It seems as if schemata not only exhibit temporal structure, but also spatial structure, as we can see from his discussion of the concept of substance.

The schema of substance is the persistence of the real in time, i.e., the representation

of the real as a substratum of empirical time-determination in general, which there-

fore endures while everything else changes. (Kant 1998, A144 / B183)

What allows for the concept of substance and the experience of a substance to interact is that both—or rather schemata in the case of concepts—have spatial and temporal structure in common.

For instance, the experience of a substance is that of an entity that is a spatially stable thing that does not change in time. The schema of a substance directs the concept of substance to a substance, since it exhibits spatial and temporal structure based on which it determines an experience of x as a substance if x has spatially stable structure that does not change in time. The matching consists in the overlap of the same spatial and temporal structure of the experience and the schema—in the case of substance: spatially stable form that does not change in time.

88 Please mind that we cannot have positive understanding of the structure of schemata. We can merely understand them analytically as the conditions of the possibility of, for instance, the interaction between conceptual-representational understanding and embodied understanding.

141

The same considerations apply to empirical concepts too—not only to the concepts of substance, cause, reality, and so forth—Kant’s pure concepts of the understanding. The concept of dog is governed by a schema that determines all possible spatiotemporal forms that dogs can exhibit.

The concept of a dog signifies a rule in accordance with which my imagination can

specify the shape of a four-footed animal in general, without being restricted to any

single particular shape that experience offers me or any possible image that I can

exhibit in concreto. (Kant 1998, B181, 182 / A 142)

At the same time, all the possible ways in which one can experience dogs exhibit a particular spatiotemporal form, i.e. dog-form. Accordingly, my concept of dog is about dogs, and can only be about dogs in the first place, because it is connected through the schema of dog to dogs in the world, since both exhibit matching spatiotemporal form—the particular instantiation of the dog schematas1-t1-particular dog with dog forms1-t1. For instance, the declarative sentence, ‘there is a dog on the mat’, is about a dog in the world because it is governed by schemata that determine the spatiotemporal forms according to which dogs can appear on mats and it can match an actual appearance of a dog on a mat in space and time.89

Importantly, what is required for experience and conceptual thought to interact is that both are at the basic level characterized by spatial and temporal attributes. Experience is basically characterized by the invariant structures spatiality and temporality; in Kant’s case, the pure forms of intuition space and time. Concepts are basically characterized by the invariant structures spatiality and

89 Importantly, I do not claim that spatiotemporal structure is sufficient to account for aboutness. Spatiotemporal structure is merely one condition that has to hold to relate conceptual-representational understanding reliably and covariantly to experience and embodied understanding.

142 temporality; in Kant’s case, through spatiotemporal schemata. Since both experience and conceptual thought exhibit the same basic structure, spatiality and temporality, they can interact and be synthesized.

Kant is quite clear about what is entailed for conceptual understanding if we accept that it is governed by schemata. Concepts cannot constitute an arbitrary, autonomous kind of understanding, but schemata determine the scope, application and meaning of concepts.

Thus the schemata of the concepts of pure understanding are the true and sole con-

ditions for providing them with a relation to objects [Beziehung auf Objekte], thus

with significance [Bedeutung] (…). (Kant 1998, A145-146 / B185, emphasis and

translation in original)

This is the case, since not only the aboutness relationship between conceptual thought and objects is determined by schemata, but also because that what concepts can mean is completely determined by the possible ways in which an object can appear to me in space and time.

As we can see from this, Kant’s own approach is quite different from John McDowell’s Kant- inspired account (McDowell 1994, 2007), according to which concepts reach into experience, without further mediation by schemata; where neither conceptual thought is constrained by schemata, nor experience is primarily characterized by spatiality and temporality.

McDowell breaks with Kant in that he does not provide an analysis of the relationship between experience and conceptual thought by means of a mediating structure, i.e. schemata. Rather, he claims that our experience is readily available for conceptual understanding without providing an account of the mediating structure that could make experience available for conceptual understanding; i.e. without an explanation of the conditions that make it possible that experience and conceptual thought can interact. As McDowell claims:

143

We can equip ourselves with new conceptual capacities, in that sense, by isolating

and focusing on—annexing bits of language to—other aspects of the categorially

unified content of the experience, aspects that were hitherto not within the scope of

our capacities for explicit thought. (McDowell 2007, p. 347, emphasis mine)

For McDowell, conceptual thought is simply linguistic and it can reach into experience qua its linguistic structure, which presupposes that the structure of experience exhibits linguistic structure too—McDowell’s ‘categorically unified content of the experience’.

Yet, as I have shown above, at least for Kant, what provides experience with ‘categorically unified content’ or the Kantian equivalent thereof, is that experience is structured by the pure forms of intuition space and time, i.e. that experience is itself structured into objects that have a particular spatiotemporal form that differentiates them from other objects and makes them available to be the object about which a schema, and thereby a concept, can be.

Since McDowell does not account for such a mediating structure, he not only fails to provide an analysis of the conditions that make the application of concepts to experience possible. He also renders the relationship between experience as well as embodied understanding and conceptual thought intellectualist, since he argues that experience and embodied understanding exhibit conceptual structure.

4.3.2 Schemata as a Bridging Structure

Importantly, we do not have to accept the particular ontological ramifications of Kantian philosophy in order to see in schemata an attractive option for the explanation of at least a minimal inter-

144 action between embodied understanding and conceptual-representational understanding. In the following I will spell out why schemata can be considered to be an ontologically and cognitively non- demanding structure. Then I will show, that, given the ontologically and cognitively non-demanding structure of schemata, we can conceive of embodied understanding as minimally characterized by schematic structure, which gives us the ground for an interaction between the kinds of understanding. Finally, I will spell out different ways of how we can conceive of the schematic relationship between embodied understanding and conceptual-representational understanding—ranging from the strong Kantian (and phenomenological) project, to a minimalist interaction between both kinds of understanding that explains how we can, for instance, make verbal reports about the objects of embodied understanding.

Many readers might be skeptical about accepting a Kantian conception of schemata, because they might worry that such a strategy commits them to an acceptance of wider aspects of the Kantian project that they might find not desirable. Yet, we do not have to accept other aspects of Kant’s system to adopt his conception of schemata; for instance, his claim that space and time are pure forms of intuition or that there are pure concepts of the understanding, i.e. basic categories such as substance or causality.

Indeed, the idea of a schema is ontologically modest, since it is formulated about primitive, invariant spatiotemporal properties of objects that characterize these objects in a minimal and essential fashion. There is nothing ontologically obscure about claiming that the objects of our thoughts, such as dogs, have a particular spatiotemporal form and have it essentially. There is further nothing cognitively obscure or demanding about the claim that understanding has to be responsive to the temporal and spatial aspects of the world; for instance, by exhibiting temporal structure itself.90

90 That understanding exhibits temporal structure means that it exhibits a structure that is responsive to the temporal structure of our experience of the world. We experience a world that constantly changes in time and not a scenery of

145

Quite to the opposite, it seems difficult to make sense of the responsiveness to a spatial world in temporal change without such a conception of schemata—for instance, based on an explanation formulated over disembodied, symbolic representations.

Since there is neither anything cognitively nor ontologically demanding about schemata, we can conceive of embodied understanding as schematic too—which is obviously a necessary condition for the interaction of the kinds of understanding through schemata (see chapter 1).

If I bring my embodied understanding to bear on an action, for instance, the action of making coffee, I need to have a practical understanding of the behaviors and objects involved in the action.

This embodied understanding is itself characterized by an aboutness relation, as in particular Mer- leau-Ponty (2012) has argued, that is neither cognitive nor merely causal, i.e. a motor intentional aboutness relationship (Carman 1999; Kelly 2002).

This intentional relationship is characterized by an interaction of motor abilities (e.g. embodied abilities) of an agent and spatiotemporal objects. As is well known, Merleau-Ponty conceives of the spatiotemporal nature of objects as determined though a body schematic relationship. As Mer- leau-Ponty states:

Each figure appears perspectivally against the double horizon of external space and

bodily space. (Merleau-Ponty 2012, p. 103)

Independent of how we conceive of this (body) schematic structure exactly, whether it is determined through embodiment (Merleau-Ponty) or without embodiment (Kant), it is essentially of a spatiotemporal nature, as the following example illustrates.

atemporal facts and propositions to which an atemporal propositional understanding would correspond. Neither do we experience as movie-like chain of images. Accordingly, we require an understanding that can be responsive to the temporal structure of the world, which is an understanding that is governed by schemata.

146

Take for instance the action of coffee making—a simple, cognitively non-demanding action governed by embodied understanding. In order to make coffee, I need to understand, non-deliberatively and non-thematically, among many other things, what a coffee machine does and I need to understand how to behave towards it, i.e. I have to understand non-conceptually which buttons I have to press, where I have to insert the coffee, when I have to stop inserting it and so forth.

This means, my understanding of which action I have to perform is relative to my understanding of which objects are involved in the action (coffee machine, coffee ground, kitchen cabinet, etc.).

And my understanding has to be about these objects—in order to identify a coffee machine as a coffee machine, i.e. as something that in this situation produces coffee for me. What allows for the application of my embodied understanding to objects is schematic structure.

The most basic, invariant features of a coffee machine are its spatiotemporal properties. I do not have to be aware of them as such, yet they nevertheless allow me to see a particular physical configuration in space and time as a coffee machine. My understanding of the coffee machine itself is not primarily spatiotemporal. It is about what I can do with the coffee machine and how it can fulfill my concerns. But a spatiotemporal schema allows me to pick out an otherwise insignificant object out of space and time as a thing of which I have a particular concern-fulfilling and action- oriented understanding.

Similarly, the structure of my behavior is guided by a spatiotemporal schema in relation to objects.

As Stanley and Krakauer (2013) and Stanley and Williamson (2016) have pointed out, action is governed by an understanding of initiation conditions. This means, I need to understand, in relation to an environmental situation, when I have to initiate or change my course of behavior in order to produce a result I have concern for. For instance, if I open the lid of my coffee machine I need to understand when to start pouring ground coffee into the filter of the machine and when to stop.

147

Krakauer, Stanley and Williamson argue that this understanding of initiation conditions is housed in mental representations that correspond to facts. Yet, we can equally conceive of this understanding as governed by spatiotemporal schemata. The schema cannot only specify which objects are involved in an action, but it has further an intimate connection to the temporality of experience itself, which is necessary to determine stop and initiation conditions of an action.

Now that we see that we can conceive of embodied understanding as spatiotemporally schematic too, we can analyze the possible relationships that can hold between embodied understanding and conceptual-representational understanding, based on their schematic structure.

4.4. Different Interactions: Ways to Connect both Kinds of Understanding

In the following I discuss two different ways how we can conceive of the schematic relationship between embodied understanding and conceptual-representational understanding. First, a tight relationship, that grounds conceptual-representational understanding in embodied understanding through schemata. Second, a loose relationship, that grants autonomy to both kinds of understanding, yet, lets them interact at various levels through schemata. I cannot spell out the specifics of such a relationship in this chapter. Rather, I merely describe which kinds of relationship can in principle follow from the structural connection between the two kinds of understanding that is grounded in schematic structure. The details of such a relationship will have to await further, more detailed deliberations.

148

4.4.1 Grounding Conceptual-Representational Understanding in Embodied Understanding

We can conceive of the relationship between embodied and conceptual-representational understanding, as Kant conceived of the relationship between experience and conceptual thought. In that case, embodied understanding, as that which contributes to our structure of experience or is at least closely connected to it (O’Regan and Noë 2001; Noë 2004; Thompson 2007), is that which governs conceptual-representational understanding insofar as it bestows meaning and significance to it.

Accordingly, we can conceive of the relationship between the two kinds of understanding as a grounding relationship.

Conceiving of the relationship as a close one could be interesting for any author that follows

Heidegger (1962) in arguing that the meaning of thought or linguistic expressions requires a grounding in our practical understanding of the world, that itself is grounded in concern or sensorimotor skill.

For instance, my concept of a coffee machine receives significance through the significance that coffee machines have for me based on the ways I can relate to them based on my embodied understanding of them. This grounding relation allows then that concepts have pragmatic meaning.

Whatever the functional contribution of conceptual understanding is, it is about the world in that it is covariantly connected to embodied understanding.

149

4.4.2 Preserving Autonomy through Minimal Interaction

If we want to grant conceptual-representational understanding a high degree of autonomy and argue for a marked difference between the modes of engagement realized by embodied and conceptual-representational understanding respectively (Wheeler 2005; Dreyfus 2007a), we can obviously also conceive of the relationship between both kinds of understanding in a weaker form; yet still schematically mediated.

If both, embodied understanding and conceptual-representational understanding are exhibiting schematic structure, both kinds of understanding can be elicited or exerted in the same situation.

If I make coffee, my spatiotemporally schematic understanding of the situation not only allows me to bring my embodied understanding to bear on the situation, but through a similar or the same schema, my concept of, say, ‘coffee machine’ can be elicited too. This allows then to make verbal reports about the situation at hand and it allows to form conceptual judgments about actions; such as ‘I am making coffee’—an action that is otherwise governed by embodied understanding.91

I need to admit, that I cannot spell out the concrete interconnection between both forms of understanding in more detail. Such accounts will in particular depend on how one conceives of the function of conceptual-representational understanding—something about which I want to be quite in this chapter and dissertation. In this chapter I have merely pointed out how both forms of understanding can in principle interact, in particular if we assume that something like conceptual-representational understanding exists, and provided in this section short sketches of possible schematic

91 To explain the relationship between both kinds of understanding in terms of schemata has further explanatory advantages, even if we should only accept a loose relationship. For instance, if conceptual thought is governed by spatiotemporal schemata, we might be able to account for the grounding of demonstratives and the identification of spatiotemporal objects in perceptual judgments. Both these properties of thought relate to spatial and temporal properties of objects and schemata are structures that explain how we can relate to objects in space and time.

150 interactions that are supposed to show that the relationship between both forms of understanding could in principle take different forms, which I deem to be a strength of schemata.

4.5. Conclusion

I have argued that spatiotemporal schemata are the condition that makes it possible that embodied understanding and conceptual-representational understanding can interact. I have shown that the separation that is made between both kinds of understanding by contemporary phenomenologists

(Dreyfus 2007a, b) and proponents of embodied cognition (Wheeler 2005, 2008; Chemero 2009) renders the interaction between them puzzling. Concretely, I have argued, in accord with McDow- ell (1994, 2007), Dreyfus (2007a) and Noë (2004, 2012), that the separation between these kinds of understanding cannot account for how we can produce reports about our actions that are governed by embodied understanding, or how we can make judgments about them. Further, I have argued that the separation can further not account for the seeming interaction between both kinds of understanding in everyday action and, at least from a phenomenological point of view, for how conceptual-representational understanding receives meaning and significance through embodied understanding. These problems make it necessary to account for the interaction between both kinds of understanding.

I have then argued that spatiotemporal schemata are well-suited to account for this interaction. I have advanced a conception of schemata that is based on Kant’s own conception of schemata. I have argued that schemata are basic, ontologically and cognitively modest structures that relate to spatial and temporal properties of objects. I then suggested that schemata underlie both, embodied

151 as well as conceptual-representational understanding. If that is the case, both, embodied and conceptual-representational understanding have the same basic structure in common that allows for their interaction.

Finally, I have spelled out two different ways in which the two kinds of understanding could interact through schemata. I have described a close interaction relationship, in which conceptual- representational understanding is grounded in embodied understanding—an idea valuable for phenomenologists and philosophers who conceive of embodied understanding as the primary locus of meaning and significance. I have further depicted another way to conceive of the relationship as a looser one in which both kinds of understanding minimally interact—so that we are able to make reports about our actions or to transform our thoughts into actions—yet, where embodied understanding and conceptual-representational understanding are in many ways autonomous from each other and preserve their unique properties and functions.

152

5. Critical Conclusion

In this conclusion, I want to hint at future directions for research in the phenomenological tradition that employs the transcendental method and shed some critical light on my own dissertation. In particular, I want to point out problems with the transcendental method in the context of an analysis of (embodied) understanding that will need to be clarified and overcome.

My dissertation sits at the border between the transcendental and the psychological. This is the case, since understanding interacts closely with experience.92 Understanding is only meaningful if it responds to experience. But experience can also only be meaningful, or at least be experience- as-we-enjoy-it, if our understanding spontaneously contributes to it. In this sense, understanding is clearly transcendental.

However, understanding is obviously also a psychological phenomenon and it is used in this way in the dissertation too; particularly then when I am concerned with the relationship between understanding and action. The obvious problem that emerges here, is that the transcendental method cannot account for psychological phenomena, yet it seems that we are required to treat understanding, but also action, as both transcendental and psychological phenomena, at times, at one and the same time.93

The theoretical discomfort with this situation can be documented in my own dissertation, as we can see particularly in chapter 2. There I concede that it is unclear whether we should treat action

92 Please consult the introduction for the transcendental role of experience within the transcendental tradition. 93 This psychological and transcendental double-character of action and understanding is also reflected in this dissertation in that it started out as a psychological project that turned over time into a transcendental project, in particular for more paradigmatic reasons that are mentioned in the introduction and in footnotes throughout the dissertation. These paradigmatic reasons pertain to the ontological and epistemological role that the close interaction of experience and embodied understanding plays. I believe we can see a similar situation and transformation in the work of Alva Noë, which also started out as a psychological and naturalistic project (O’Regan and Noë 2001; Noë 2004), but which turned successively into a transcendental project, given that Noë (2012, 2015) realized more and more the world- disclosing role that experience and understanding play in close interaction with each other.

153 as a transcendental or a psychological phenomenon. I am also forced to admit that my conclusion in chapter 2—that we require a concernful understanding of the world in order to even want to engage into an ordinary action—might turn out to be more of a psychological hypothesis rather than the result of an analysis of the conditions of the possibility of acting.

Yet, it is not only a ‘pure’ transcendental account of embodied understanding that is confronted with this problem. As I have shown in chapter 3, phenomenologically-inspired embodied cognition is confronted with very similar problems. Accordingly, future work in the transcendental tradition will have to clearly address the relationship between the interaction of psychological and transcendental phenomena.

154

6. References

Alsmith, A., & De Vignemont, F. (2012). Embodying the mind and representing the body. Review of Philosophy and Psychology, 3(1), 1–13.

Anderson, M., & Rosenberg, G. (2008). Content and action: The guidance theory of representation.

Journal of Mind and Behavior, 29(1-2), 55–86.

Anderson, M., & Chemero, A. (2013). The problem with brain GUTs: Conflation of different senses of ‘prediction’ threatens metaphysical disaster. Behavioral and Brain Sciences, 36(03),

204–205.

Austin, J. (1962). How to do things with words. Oxford: Oxford University Press.

Barrett, L., Wilson-Mendenhall, C., & Barsalou, L. (2015). The conceptual act theory: A road map.

In L. Barrett & J. Russell (Eds.), The psychological construction of emotion (pp. 83–110). New

York: Guilford.

Barsalou, L. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22(04), 637–660.

Barsalou, L. (2008). Grounded cognition. Annual Review of Psychology, 59, 617–645.

Berthoz, A., & Christen, Y. (2009). Neurobiology of ‘Umwelt’: How living beings perceive the world. Berlin: Springer Verlag.

Boroditsky, L., & Ramscar, M. (2002). The roles of body and mind in abstract thought. Psycho- logical Science, 13(2), 185–189.

Borrett, D., Kelly, S., & Kwan, H. (2000). Bridging embodied cognition and brain function: The role of phenomenology. Philosophical Psychology, 13(2), 261–266.

155

Bower, M., & Gallagher, S. (2013). Bodily affects as prenoetic elements in enactive perception.

Phenomenology and Mind, 4(1), 78–93.

Bruineberg, J., & Rietveld, E. (2014). Self-organization, free energy minimization, and optimal grip on a field of affordances. Frontiers in Human Neuroscience, doi: 10.3389/fnhum.2014.00599

Carel, H., & Meacham, D. (2013). Phenomenology and naturalism: Editors' introduction. Royal

Institute of Philosophy Supplement, 72, 1–21.

Carman, T. (1999). The body in Husserl and Merleau-Ponty. Philosophical Topics, 27(2), 205–

226.

Chemero, A. (2003). An outline of a theory of affordances. Ecological Psychology, 15(2), 181–

195.

Chemero, A. (2009). Radical embodied cognitive science. Cambridge, MA: MIT Press.

Chemero, A. (2014). Synergies and systematicity. In P. Calvo & J. Symons (Eds.), The Architec- ture of Cognition: Rethinking Fodor and Pylyshyn's systematicity challenge (pp. 353–370. Cam- bridge, MA: MIT Press.

Clark, A., & Toribio, J. (1994). Doing without representing? Synthese, 101(3), 401–31.

Clark, A. (2008a). Pressing the flesh: A tension in the study of the embodied, embedded mind? Philosophy and Phenomenological Research, 76(1), 37–59.

Clark, A. (2008b). Supersizing the mind: Embodiment, action, and cognitive extension. Oxford:

Oxford University Press.

Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(03), 181–204.

156

Colombetti, G. (2014). The feeling body: Affective science meets the enactive mind. Cambridge,

MA: MIT Press.

Colombetti, G., & Thompson, E. (2008). The feeling body: Towards an enactive approach to emotion. In W. Overton, U. Müller, & J. Newman (Eds.), Developmental perspectives on embodiment and consciousness (pp. 45–68). New York: Lawrence Erlbaum Associates.

Crowell, S. (2007). Sorge or Selbstbewusstsein? Heidegger and Korsgaard on the sources of nor- mativity. European Journal of Philosophy, 15(3), 315–333.

Descartes, R. (1986). Meditations on first philosophy. (J. Cottingham, R. Stoothoff, & D. Murdoch,

Trans. & Eds.). The philosophical writings of Descartes. Volume II. Cambridge: Cambridge Uni- versity Press. (Original work published 1641)

Dotov, D., & Chemero, A. (2014). Breaking the perception-action cycle: Experimental phenomenology of non-sense and its implications for theories of perception and movement science. In M.

Cappuccio & T. Froese (Eds.), Enactive cognition at the edge of sense-making: Making sense of non-sense (pp. 37-60). Basingstoke, UK: Palgrave Macmillan.

Dreyfus, H. (1972). What computers still can’t do: A critique of artificial reason. Cambridge, MA:

MIT Press.

Dreyfus, H. (1988). The Socratic and Platonic basis of cognitivism. AI & Society, 2(2), 99–112.

Dreyfus, H. (1991). Being-in-the-world: A commentary on Heidegger's Being and Time, Division

I. Cambridge, MA: MIT Press.

Dreyfus, H. (1992). What computers still can’t do. A critique of artificial reason. Cambridge, MA:

MIT Press.

157

Dreyfus, H. (2002). Intelligence without representation—Merleau-Ponty's critique of mental representation. The relevance of phenomenology to scientific explanation. Phenomenology and the

Cognitive Sciences, 1(4), 367–383.

Dreyfus, H. (2007a). The return of the myth of the mental. Inquiry, 50(4), 352–365.

Dreyfus, H. (2007b). Why Heideggerian AI Failed and how fixing it would require making it more

Heideggerian. Philosophical Psychology, 20(2), 247–268.

Dreyfus, H. (2012). Introductory essay: The mystery of the background qua background. In Z.

Radman (Ed.), Knowing without thinking: Mind, action, cognition and the phenomenon of the background (pp. 1–10). London: Palgrave McMillan.

Dreyfus, H. (2013). The Myth of the Pervasiveness of the Mental. In J. Schear (Ed.), Mind, reason, and being-in-the-world: The McDowell-Dreyfus debate (pp. 15–40). Abingdon: Routledge.

Dreyfus, H., & Dreyfus, S. (1986). Mind over machine. The power of human intuition and expertise in the era of the computer. New York: The Free Press.

Dreyfus, H., & Kelly, S. (2007). Heterophenomenology: Heavy-handed sleight-of-hand. Phenom- enology and the Cognitive Sciences, 6(1-2), 45–55.

Dreyfus, H., & Taylor, C. (2015). Retrieving realism. Cambridge, MA: Harvard University Press.

Favela, L., & Chemero, A. (2016). The animal-environment system. In Y. Coello & M. Fischer

(Eds.), Foundations of embodied cognition: Volume 1: Perceptual and emotional embodiment (pp.

59–74). Abingdon: Routledge.

Fernandez, A. (2016). The subject matter of phenomenological research: Existentials, modes, and prejudices. Synthese, 1–20.

158

Fuchs, T. (2011). The brain – A mediating organ. Journal of Consciousness Studies, 18(7-8), 196–

221.

Gallagher, S. (2005). How the body shapes the mind. Oxford: Oxford University Press.

Gallagher, S. (2008). Are minimal representations still representations? International Journal of

Philosophical Studies, 16(3), 351–369.

Gallagher, S. (2011). Interpretations of embodied cognition. In W. Tschacher & C. Bergomi (Eds.),

The Implications of embodiment: Cognition and communication (pp. 59–71). Exeter: Imprint Ac- ademic.

Gallagher, S. (2014). Pragmatic interventions into enactive and extended conceptions of cognition.

Philosophical Issues, 24(1) 110–126.

Gallagher, S. (2015). Invasion of the body snatchers: How embodied cognition is being disembodied. The Philosophers’ Magazine, April(2), 96–102.

Gallagher, S., & Zahavi, D. (2012). The phenomenological mind. Abingdon: Routledge.

Gallagher, S., & Bower, M. (2014). Making enactivism even more embodied? AVANT. Trends in

Interdisciplinary Studies, 5(2), 232–247.

Gallese, V., & Lakoff, G. (2005). The brain's concepts: The role of the sensory-motor system in conceptual knowledge. Cognitive Neuropsychology, 22(3-4), 455–479.

Glenberg, A. (1997). What memory is for. Behavioral and Brain Sciences, 20(1), 1–55.

Goldman, A., & de Vignemont, F. (2009). Is social cognition embodied? Trends in Cognitive Sci- ences, 13(4), 154–159.

159

Guignon, C. (2009). The body, bodily feelings, and existential feelings: A Heideggerian perspective. Philosophy, Psychiatry, & Psychology, 16(2), 195–199.

Hanna, R. (2005). Kant and nonconceptual content. European Journal of Philosophy, 13(2), 247–

290.

Haugeland, J. (1998). Mind embodied and embedded. In Having Thought. Essays in the Metaphys- ics of Mind (pp. 207–237). Cambridge, MA: Harvard University Press.

Heidegger, M. (1962). Being and time. (J. Macquarrie & E. Robinson, Trans.). New York: Harper and Row. (original work published 1927)

Heidegger, M. (1990). Kant and the problem of metaphysics. (R. Taft, Trans.). Bloomington: In- diana University Press. (original work published 1929)

Heidegger, M. (2001). Zollikon seminars: Protocols – conversations – letters. (F. Mayr, Trans.).

& (J. McCumber & D. Kolb, Eds.). Evanston, Il.: Northwestern University Press. (original work published 1987)

Heidegger, M. (2006). Sein und Zeit. Tübingen: Max Niemeyer Verlag.

Hume, D. (2007). An enquiry concerning human understanding and other writings. (S. Buckle,

Ed.). Cambridge: Cambridge University Press. (original work published 1748)

Hurley, S. (1998). Consciousness in action. Cambridge, MA: Harvard University Press.

Husserl, E. (1982). Ideas pertaining to a pure phenomenology and to a phenomenological philosophy. First book. General introduction to a pure phenomenology. (F. Kersten, Trans.). Dordrecht:

Kluwer Academic Publishers. (original work published 1913)

160

Husserl, E. (1989). Ideas pertaining to a pure phenomenology and to a phenomenological philosophy. Second book. Studies in the phenomenology of constitution. (R. Rojcewicz & A. Schuwer,

Trans.). Dordrecht: Kluwer Academic Publishers. (original work published 1952)

Hutto, D. (2012). Exposing the background: Deep and local. In Z. Radman (Ed.), Knowing without thinking: Mind, action, cognition and the phenomenon of the background (pp. 37–56). Basing- stoke: Palgrave Macmillan.

Hutto, D., & Myin, E. (2013) Radicalizing enactivism: Basic minds without content. Cambridge,

MA: MIT Press.

Kant, I. (1956). Kritik der reinen Vernunft. W. Weischedel (Ed.). Frankfurt am Main: Suhrkamp

Verlag.

Kant, I. (1998). Critique of pure reason. (P. Guyer & A. Wood, Trans.). Cambridge: Cambridge

University Press. (original work published 1781 and 1787)

Kelly, S. (2000). Grasping at straws: Motor intentionality and the cognitive science of skillful behavior. In M. Wrathall & J. Malpas (Eds.), Heidegger, coping, and cognitive science: Essays in honor of Hubert L. Dreyfus, Volume 2 (pp. 162–177). Cambridge, MA: MIT Press.

Kelly, S. (2002). Merleau-Ponty on the body. Ratio, 15(4), 376–391.

Kiverstein, J. (2012). The meaning of embodiment. Topics in Cognitive Science, 4(4), 740–758.

Kiverstein, J., & Clark, A. (2009). Introduction: Mind embodied, embedded, enacted: One church or many? Topoi, 28(1), 1–7.

Kiverstein, J., & Miller, M. (2015). The embodied brain: Towards a radical embodied cognitive neuroscience. Frontiers in Human Neuroscience, doi:10.3389/fnhum.2015.00237

161

Kiverstein, J. & Rietveld, E. (2015). The primacy of skilled intentionality: On Hutto & Satne’s the natural origins of content. Philosophia, 43(3), 701–721.

Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: The embodied mind and its challenge to western thought. New York: Basic Books.

Locke, J. (1996). An essay concerning human understanding. (K. P. Winkler, Ed.), Indianapolis:

Hackett Publishing Company. (original work published 1689)

Mandik, P., & Clark, A. (2002). Selective representing and world-making. Minds and Ma- chines, 12(3), 383–395. van Mazijk, C. (2016). Kant and Husserl on the contents of perception. The Southern Journal of

Philosophy, 54(2), 267–287.

McDowell, J. (1994). Mind and world. Cambridge, MA: Harvard University Press.

McDowell, J. (2007). What Myth? Inquiry, 50(4), 338–351.

McGann, M. (2007). Enactive theorists do it on purpose: Toward an enactive account of goals and goal-directedness. Phenomenology and the Cognitive Sciences, 6(4), 463–483.

Merleau-Ponty, M. (2012). Phenomenology of perception. (D. A. Landes, Trans.). Abingdon:

Routledge. (original work published 1945)

Noë, A. (2004). Action in perception. Cambridge, MA: MIT Press.

Noë, A. (2005). Against intellectualism. Analysis, 65(4), 278–290.

Noë, A. (2007). The critique of pure phenomenology. Phenomenology and the Cognitive Sci- ences, 6(1-2), 231–245.

162

Noë, A. (2009). Out of our heads: Why you are not your brain, and other lessons from the biology of consciousness. New York: Hill and Wang.

Noë, A. (2012). Varieties of presence. Cambridge, MA: Harvard University Press.

Noë, A. (2013). On overintellectualizing the intellect. In J. Schear (Ed.), Mind, reason, and being- in-the-world: The McDowell-Dreyfus debate (pp. 178–193). Abingdon: Routledge.

Noë, A. (2015). Concept pluralism, direct perception, and the fragility of presence. In T. Metzinger

& J. M. Windt (Eds.), Open MIND (27(T)). Frankfurt am Main: MIND Group. doi:

10.15502/9783958570597

Noë, A. (2016). Sensations and situations: A sensorimotor integrationist approach. Journal of Con- sciousness Studies, 23(5-6), 66–79. de Oliveira, G., & Chemero, A. (2015). Against smallism and localism. Studies in Logic, Grammar and Rhetoric, 41(54), 9–23.

O'Regan, J., & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. Be- havioral and Brain Sciences, 24(05), 939–973.

Pessoa, L. (2015). Précis on the cognitive-emotional brain. Behavioral & Brain Sciences, doi:

10.1017/S0140525X14000120

Petitot, J., Varela, F., Pachoud, B., & Roy, J-M. (Eds.). (1999). Naturalizing phenomenology: issues in contemporary phenomenology and cognitive science. Stanford: Stanford University Press.

Prinz, J. (2004). Furnishing the mind. Concepts and their perceptual basis. Cambridge, MA: MIT

Press.

163

Prinz, J., & Clark, A. (2004). Putting concepts to work: Some thoughts for the twentyfirst century.

Mind & Language, 19(1), 57–69.

Putnam, H. (1975). The meaning of ‘meaning’. Minnesota Studies in the Philosophy of Science, 7,

131–193.

Ratcliffe, M. (2002). Heidegger's attunement and the neuropsychology of emotion. Phenomenol- ogy and the Cognitive Sciences, 1(3), 287–312.

Ratcliffe, M. (2008). Feelings of Being: Phenomenology, Psychiatry and the Sense of Reality. Ox- ford: Oxford University Press.

Ratcliffe, M. (2010). The phenomenology and neurobiology of moods and emotions. In S. Gal- lagher & D. Schmicking (Eds.), Handbook of phenomenology and cognitive science (pp. 123-140).

Dordrecht: Springer.

Ratcliffe, M. (2015). Experiences of depression: A study in phenomenology. Oxford: Oxford Uni- versity Press.

Rietveld, E. (2008). The skillful body as a concernful system of possible actions. Phenomena and neurodynamics. Theory & Psychology, 18(3), 341–363.

Rietveld, E., & Kiverstein, J. (2014). A rich landscape of affordances. Ecological Psychology,

26(4), 325–352.

Rowlands, M. (2006). Body language: Representation in action. Cambridge, MA: MIT Press.

Rowlands, M. (2010). The new science of the mind: From extended mind to embodied phenomenology. Cambridge, MA: MIT Press.

164

Sartre, J.-P. (1962). Sketch for a theory of the emotions. (P. Mairet, Trans.). London: Methuen

Publishing. (original work published 1939)

Searle, J. (1978). Literal meaning. Erkenntnis, 13(1), 207–224.

Sherover, C. (1971). Heidegger, Kant and Time. Bloomington: Indiana University Press.

Silberstein, M., & Chemero, A. (2012). Complexity and extended phenomenological-cognitive systems. Topics in Cognitive Science, 4(1), 35–50.

Silberstein, M., & Chemero, A. (2015). Extending neutral monism to the hard problem. Journal of

Consciousness Studies, 22(3-4), 181–194.

Stanley, J. (2005). Knowledge and practical interests. Oxford: Oxford University Press.

Stanley, J., & Krakauer, J. (2013). Motor skill depends on knowledge of facts. Frontiers in Human

Neuroscience, 7, 1–11.

Stanley, J., & Williamson, T. (2016). Skill. Nous, doi:10.1111/nous.12144 (online first publica- tion).

Stapleton, M. (2013). Steps to a ‘properly embodied’ cognitive science. Cognitive Systems Re- search, 22-23, 1–11.

Taylor, C. (1985). The concept of a person. In Human agency and language. Philosophical papers

I (pp. 97–114). Cambridge: Cambridge University Press.

Thompson, E. (2007). Mind in life: Biology, phenomenology, and the sciences of mind. Cambridge,

MA: Harvard University Press.

Thompson, E. (2011). Living ways of sense making. Philosophy Today, 55(Supplement), 114–

123.

165

Thompson, E., & Stapleton, M. (2009). Making sense of sense-making: Reflections on enactive and extended mind theories. Topoi, 28(1), 23–30.

Thompson, E., & Cosmelli, D. (2011). Brain in a vat or body in a world?: Brainbound versus

Enactive Views of Experience. Philosophical Topics, 39(1), 163–180.

Varela, F., Thompson, E., & Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human

Experience. Cambridge, MA: MIT Press.

Wheeler, M. (2005). Reconstructing the cognitive world: The next step. Cambridge, MA: MIT

Press.

Wheeler, M. (2008). Cognition in context: Phenomenology, situated robotics and the frame problem. International Journal of Philosophical Studies, 16(3), 323–349.

Wheeler, M., & Kiverstein, J. (Eds.). (2012). Heidegger and Cognitive Science. Basingstoke: Pal- grave Macmillan.

Wilson, M. (2002). Six views of embodied cognition. Psychonomic Bulletin & Review, 9(4), 625–

636.

Wilson, T. (2002). Strangers to ourselves: Discovering the adaptive unconsciousness. Cambridge,

MA: Harvard University Press.

Wittgenstein, L. (2003). Philosophische Untersuchungen. Frankfurt am Main: Suhrkamp Verlag.

Wrathall, M. (2005). Motives, reasons, and causes. In T. Carman & M. Hansen (Eds.), The Cam- bridge companion to Merleau-Ponty (pp. 111–128). Cambridge: Cambridge University Press.

Zahavi, D. (2005). Subjectivity and selfhood: Investigating the first-person perspective. Cam- bridge, MA: MIT Press.

166

Zahavi, D., & Overgaard, S. (2012). Time, space and body in Bergson, Heidegger and Husserl. In

R. Baiasu, G. Bird, & A. Moore (Eds.), Contemporary Kantian metaphysics: New essays on space and time (pp. 270–297). Basingstoke: Palgrave Macmillan.

167