The Metaphysics of the Mind The Problem of Representation

autumn term , University of Lucerne Philipp Blum

draft version: October , Contents

Introduction . Methodological Preliminaries ...... .. The Metaphysics of the Mind ...... .. Metaphysics and its Siblings ...... .. Bottom-up ...... . Representation as Co-formality ...... .. TheOverallProject ...... .. Two Types of Perspectivality ...... .. Aboutness and Content ...... . Learning from the History of Philosophy ...... .. The relativistic challenge ...... .. Tracing the problem of representation back to its sources ...... .. Perspectivality, Grounding and Fundamentality ......

Perceptual Intake . Seeing Things ...... .. Perceptual Reports ...... .. The Argument from Illusion ...... .. The Intentionality of Perception ...... . Aristotle on Perception and Colours ...... .. UptakeofForm ...... .. Aristotelian Colours ...... .. Representation by Exemplification ...... . ModeandContent ...... .. Relations and Representations ...... .. Hedging Russell’s Retreat ...... .. Perception as intake of form ......

Intentionality and Representation . The Emergence of Content ...... .. Appearancesand phantasiai ...... .. Seeables and Sayables ...... .. Living by Appearances ......

I . Against Reification ...... .. Speaking of the Looks of Things ...... .. Acts and their Objects ...... .. Modes and their Contents ...... . Veridicality Conditions ...... .. .. ...... .. .. ...... .. .. ......

The Good and the Pleasant . Emotion and Valence ...... .. FeelingtheDanger ...... .. Emotions and their Correctness Conditions ...... .. Emotions and Values ...... . Desire ...... .. The Aboutness of Desire ...... .. TheGuiseoftheGood ...... .. The Value of Emotions ...... . Correctness Conditions ...... .. Correctness Conditions and Intentionality ...... .. Intrinsicness and Relationality ...... .. The Category of the Intrinsic-Relational ......

From Intelligible Species to Ideas . The medieval problem: intentionality ...... . Cogitationes: Material for Doubt ...... .. The Cartesian Project ...... .. The Cogito ...... .. Ideas as contents ...... . Ideas ...... .. Lockean ideas ...... .. Berkeleyan radicalisation ...... .. Humean reconciliation? ......

Colours and Values . Conflict ...... .. Different types of conflict ...... .. Resisting Counterevaluatives ...... .. The Negation Problem ...... . Primary and Secondary Qualities ...... .. The Posteriority of Sensible Qualities ...... .. The Molyneux Problem and Common Sensibles ...... .. Distinguishing “red” from “funny” ......

II . The Question of Objectivity ...... .. Faultless Disagreement ...... .. Arguing about taste ...... .. Aspectival reference ......

Belief . Content Attribution ...... .. Correctness and Truth ...... .. Indeterminacy of Content ...... .. The Determination of Aboutness ...... . Content ...... .. The Attribution of Representational Content ...... .. Thinkingthat x is F ...... .. Representation-as ......

III Chapter

Introduction

. Methodological Preliminaries

.. The Metaphysics of the Mind

Cartesian or not, many philosophers have a tendency to think of mind as being ‘outside’ the world, as separating the world, the world in itself, from our thinking about it. Whether or not we are Cartesian egos, immaterial souls, noumenal selves or non-spatiotemporal monads and are thus not just parts of the world, certainly the world, the focus of cognition, object of enquiry and ontological basis of the phenomena does also contain cognizers, thinking things, representations, words and thoughts. Among the things there are, some are special in that their correct description requires us to talk about other things as well, things ‘outside’ the things described and standing in no regular causal or spatio-temporal relations to them. Not only do such things ‘involve’ others, they also do so in a particular way: they are perspectives on them, ‘views’ or ‘takes’ on them. What it is to be one of these special things will thus require not just talking about other things, but about how these things are taken to be, construed or ‘represented’ by the special things. “Representation” is a multi-faceted word: that some item, or mental state, or person is ‘representing’ or ‘represen- tational’ does not, strictly speaking, tell us more than that it is among the special things; we should not be misled into thinking that there is something about it that we understand, or even that there is something we unequivocally speak about, just by using this word. We should not assume at the outset that there is a well-defined topic of our study – the representation relation; for us, representation rather is a problem.

.. Metaphysics and its Siblings

Theoretical philosophy aims to articulate a view of ourselves and the world that is systematic, comprehensive, beautiful and makes sense both of them and of our desire to understand them. Its most general articulation is within the ‘subdiscipline’ (if there is such a thing) of metaphysics, which asks questions about the natures and determinations of things, about the ontological categories into which they fall, their metaphysical status and their modalities. Metaphysics, so conceived, not only concerns the world as opposed to what we think about it, but also us as parts of that world and our relations to it, themselves a part of the world, connecting two worldly items. The philosophy of mind is more narrowly concerned with such relations, in particular their foundations ‘on our side’, as it were – mental phenomena, such as feelings, intentions, imaginings, suppositions, beliefs, desires – and with the nature of these relations, in particular the question whether they relate things that belong to the same or to different metaphysical categories.

Among such relations between us and the world, we may distinguish two different ‘directions of fit’. Some mental phenomena, such as wishes, impose conditions on the world: I determine what wish you have by asking how the world would have to be for that wish to be fulfilled. Others, such as beliefs, have conditions imposed on them by the world: asking what you believe is asking how the world is according to you and whether or not the world really is that way. For mental states of both directions, we distinguish, in the determination of their content, between structural and thematic constraints. In the case of wishes, a structural constraint is that, in wishing it to be the case that p,I represent the world’s being such that p as good. This is not a second, additional thing I do: by the very having of the wish, I represent its fulfillment as good. A thematic constraint on possible contents of wishes might be that I can only wish what is metaphysically possible: when I wish to draw a square circle, for example, I may be criticisable not just for pragmatic, moral or instrumental reasons, but also for having a wish (if it is a wish at all) that is internally defective. The philosophy of mind, as it is nowadays standardly conceived, lacks both thematic and methodological unity, containing parts that are of the “philosophy of X” type (the philosophy of psychology, of the science of percep- tion, of cognitive science, analogously to the philosophy of physics, of mathematics, of economics), parts that use analysis in the armchair (e.g. whether desire is always under the guise of the good), parts that are at least partly phenomenological (e.g. classification of emotions in terms of their formal objects) and parts that greatly overlap with or even coincide with traditional concerns of metaphysics (the mind/body problem, free will, whether there is a soul). That I will be concerned mostly, or even exclusively, with the metaphysics of the mind does not mean that I will treat only of part of the concerns of traditional philosophy of mind, but rather that I approach mental phenomena such that I ask about their metaphysics first. Epistemology, concerned with those mental phenomena that represent the world, correctly or incorrectly, as being in a certain way, and with the ties of justification among these, examines such structural and thematic constraints. With respect to belief, for example, epistemology is concerned to explain why believing that p is representing it as true that p, and to account for the specific type of – “epistemic” – norms and values that characterise a true belief as epistemically better than a false one and ground a specific – “epistemic” – practice of attributing praise and blame to cognitive subjets. Translating “epistemology” as “theory of knowledge” (“théorie de la connaissance”, “Erkenntnistheorie”) is a theo- retical choice, quite apart from the fact that “knowledge” is just one of many acceptable translations of “epistēme”. While nowadays “knowledge” certainly is the key term, earlier philosophers were more concerned about wisdom, certainty or what it is rational to believe. In this introduction, however, we will follow contemporary orthodoxy and take epistemology to be centrally concerned with knowledge. The key question of epistemology,understood as theory of knowledge, is: what can we know? It is both descriptive, aiming at giving an accurate description of what it is in virtue of which we know when we do, normative, exhibiting grounds for praise and blame in these cases and articulating a sui generis way in which intellectual enquiry is shaped by the goals that are characteristic for it, and also, thirdly, self-reflective, being a theoretical enterprise itself, thus aiming at knowledge, or truth at least.

.. Bottom-up

The present study starts with an investigation of perception and then aims to find its way ‘up’, from perception to the ‘higher’ mental faculties such as emotion, cognition and reasoning. Such directionality, and the metaphors that accompany it, are easily misunderstood: I do not think, for example, that ‘higher’ functions are somehow constructed out of, or reductively explainable in terms of, or metaphysically grounded in the ‘lower’ functions. For present purposes, starting with perception and proceeding to emotions, then to the conative and then to cognition and language is just a convenient order of exposition, allowing to introduce concepts and distinctions in comparatively simpler cases before applying them in more complex circumstances.

. Some, e.g. Aquinas (Sum.c.Gent. III , , , ) would even go as far as claiming that for natural wishes at least, it even has to be nomologically possible for them to be fulfilled.

Generally,my aims are not reconstructive: though they are certainly special, I do not think that mental phenomena are somehow ‘weird’, that we need to build, or rebuild, or reconstruct them from supposedly more basic or less problematic material, that we have to ‘find room for them’

. Representation as Co-formality

.. The Overall Project

The plan of these lectures is to start with an Aristotelian slogan – perception is intake of form without matter – and work my way ‘upwards’ the ‘hierarchy’ from perception to thought. The central idea, to be found in De Anima, may be cashed out as follows: Perception naturally occurs, as a biological phenomenon. Its veridicality is primitive: we can never be deceived by it with respect to its proper objects. These proper sensibles are prior to their perception in the order of explanation: we explain sight as the sense that gives us colours, not colours as the things that are the objects of sight. In all types of perception, the medium accounts for the transmission of form that is taken in by the alteration of the sense (where this alteration is the perception). Seeing perception as emerging in this way as the joint outcome of three factors – the form to be seen, the form transmitted and seen by the perceiver, and the intervening medium – makes understandable how it can be both intrinsic and relational. Representational states being intrinsic, they ‘mirror’ the world as it appears to some perceiver, i.e. as perspectival and perceiver-oriented. Being relational at the same time, they connect perceivers to things outside of them, and make themselves evaluable in terms of external objects. Because an understanding of representation as self-location places the ‘subjective’ and ‘phenomenal’ elements in the mode, rather than the content of the attitude, allowing for an adverbialist, rather than propositionalist analysis of it, it makes an objectivist notion of content possible. Such a notion can be found in Aristotle, and the present project aims to defend his central idea: representation is coformality, the sharing of ‘form’ between what is represented and what represents. As the medievals realised, the account of perception as intake of form may be generalised to representation more generally if, for more complex mental states, we draw an act/content distinction – as Duns Scotus did between the ‘subjectively’ present (accident of) thinking and the ‘objectively’ present external form. On such a generalised conception, the form taken in from a is what a is represented as; and to represent a as F is to have F as a’s form in mind. Such having forms in mind as the forms of a is a separate mental operation:it has its natural home in perception, and is only subsequently, and on this basis, performed autonomously. Intentionality is thus explained as the having in mind of a qualitative content (form) in a certain way. To account for states that exhibit primarily not only representational or intentional features (as do, respectively, perception and intention/desire/wishing), but both in an equally fundamental way, a third primitive element is needed: an attitude the subjects takes on what it represents. Such an ‘active’ component explains how the represented form is taken to be and relates it to the formal object of the mental state in question, thus generating its correctness conditions. In fear, for example, the subject both represents something as dangerous and has a negative, flight-inducing attitude towards it; in belief, the subject entertains a thought, and assents to it. In both cases, it is the attitude that ‘generates’ the content, as it were, as it takes the form (itself taken in a way) to relate to what it is a form of. With linguistic representation, finally, the form is externalised – it is the linguistic items themselves, taken in a way, that may be said to reach out to what they are about.

.. Two Types of Perspectivality

What we see, what we feel and what we represent have their perspectivality in common. This does not by itself entail that their perspectivality has to be explained in the same way, nor that it is of the same type. The perspec- tivality of perception is partly physiological, partly a question about the perceptual milieu and partly a matter of the relative orientation of perceiver and perceived: what I see when I see the stick is how half-immersed sticks, in this situation, look to me from where I am. The perspectivality of emotions is at least in part explained by their being reactions by specific emoters to specific situations: what is dangerous to me is perhaps not dangerous to you. This may justify an account of the formal objects as being itself perspectival: the correctness of my fear could

then be said to turn not on danger tout court, but on danger-for-me-now-here. With representation generally, many different factors may matter for perspectivality: while some of them may also be attributed to the formal object (e.g. assertability-in-the-philosophy-seminar), or even to the direct object (e.g. Hesperus rather than Phosphorus), others are types of medium- or message-specificity (e.g. ‘analog’ representations exploiting my contingent discrim- ination thresholds, de-se or de-nunc representation that makes the content itself dependent on the representer or, even more specifically, the act of representing itself). I hope some clarification may be achieved by distinguishing two closely related, but distinct features of at least some mental states: intentionality and representation. Intentionality A mental state is intentional iff it is ‘directed’ towards something outside itself, where this directed- ness exhibits the ‘presence in absence’ feature, i.e. a ‘pointer’ ‘standing in’ for something (at least potentially) absent. Representation A mental state is representational iff it is either correct or incorrect, i.e. can be assessed in terms of its ‘fit’ to the world, i.e. in terms of what (if any) information its occurrence has. According to Brentano, intentionality is the mark of the mental: it is the feature whereby some internal state reaches out to the world, directing the mental subject towards such (potentially absent) features. Correctness conditions specify such intentional content, but – being conditions – do not themselves require this content to be satisfied. If I am looking for the Holy Grail, for example, my activity is directed towards, and rationalisable only with respect to the Holy Grail, which, or so we shall assume, does not exist. I am intentionally directed towards the Holy Grail, without standing in a relation to it: there is nothing, after all, for me to stand in a relation to. While searching, it is in my search that the Holy Grail is ‘present’ to me, as the thing I am looking for; that it is ‘present’ to me here just means that there are conditions, if it is non-pathological, that determine when my search would be successful. For this, I do not have to be able to describe the Holy Grail, and even less to uniquely identify it; it is enough if ‘I know it when I see it’ and would then regard my search as successful. According to Dretske, it is transmission of information, in particular along causal links, that explains how things like us succeed in representing the world. While natural signs, such as smoke indicating fire, may be (and, ac- cording to the orthodox Shannon-Weaver theory of information, are) information about their causes, they do not encode this information in a format suitable for its transmission. Information transmission in full-blown human communication, on the other hand, is incredibly complicated: not only do people lie and mislead, neither mean what they say nor say what they mean, but a number of complicated ‘uptake’ conditions on the side of the hearer must be fulfilled for information to be transmitted (communicative acts have to be understood). Both intentionality and representation ‘involve’, in some way, the world, or something outside their media; both give us a sense in which mental states are ‘about’ something else. They do so, however, in slighly different ways: they differ in how they cross-cut two distinctions commonly identified with each other: while representation is intrinsic, but also relational, intentionality is non-relational, but also extrinsic.

.. Aboutness and Content

The present project focusses on the concept and relation of representation, central to the philosophies of mind, language and the world. What is it, the central question goes, in both us and the world that makes it possible for some of our states and products to be about other things, both in- and outside of us? Traditionally (cf. e.g. Harman ; Block ; Dretske ; Fodor , a; Devitt ), this question has often been answered in terms of correlation: it is because our states covary with what is outside them, the thought goes, that they may be said to be about these other things. However, this answer is unsatisfactory. For one, we know from other cases (e.g. the discussion of modal accounts of essence, supervenience and grounding) that covariation is not fine-grained enough to do this explanatory work, and recent work in the philosophy of mind has shown how difficult it is for a purely correlational account to account for important features of the representation relation, such as the possibility of (even quite wide-spread) misrepresentation, the normativity of many representational states and, first and foremost, the phenomenal aspect of many, perhaps all of them.

Representation and intentionality, understood in a Brentanian way as the ‘mark of the mental’, are more than lawful correlation. But what are they? How does the representation relation produce states that are both meaning- ful and phenomenal, in the sense of there being something it is to be in them? The crucial idea, to be developed, examined and defended by the present research project is that it is of the very nature of representation that it has a two-fold character, both reaching out to things outside ourselves and locating ourselves with respect to them. A sufficiently broad notion of self-location within egocentric spaces of certain kinds (perceptual, temporal, loca- tional, axiological and phenomenal) will help us explain both why their ‘objective’ representational content does not capture everything that is essential to meaningful states and why it is impossible to characterise their representa- tional content in purely third-personal terms. While self-locating attitudes have been widely studied in connection with the so-called ‘essential indexical’ (Perry ), de se thoughts (Lewis ) and the self-referential guises of Castañeda (; ), the connection to the perspectivality of ‘ordinary’ representations (as of a stick as bent, of a distant tower as rectangular, of a wall as blue) has not yet been explored. Drawing on recent discussions of locational properties (Sider & Hawthorne ; Simons , ; Parsons ), I will characterise Russellian perspectival thoughts as self-locating (cf. Egan (b) and Titelbaum () for some promising ideas): having them, we not only ascribe (subjective, perspective-dependent) properties to (subjective, perspective-dependent) objects, but we locate ourselves in relation to them. A metaphysically informed theory of the representation relation has a lot of work to do. It is hoped that it (i) will shed light on recent debates of relativism, (ii) provide a new account of lesser entities, secondary qualities and Kantian phaenomena, and also (iii) help us understand the ontology of values and the contentfulness of mental states generally. The plan of the project is to approach the question of how the mind connects to the world from three complementary perspectives: an account of our representations of the world as exhibiting perspectivality what is the nature of per- spectival thinking? how can we reconcile objective and subjective factors to account for (apparent) ‘faultless disagreement’? how are perspectives represented, entertained, constituted? what makes our thinking about things perspectival? an ontologically robust account perspectival objects how are perspectives individuated? how can they be objective? in what way are they related to the representing subject? what is the ontological status of so-called ‘lesser entities’ and ‘secondary properties’ – values, colours, odours, tastes, shadows, holes? a cartography of the mental what is the representation relation? what mental states are there? how are they individuated? what role does content play in their individuation? what kind of thing is content and how does it feature in representation? Taking these perspectives from the philosophies of language, mind and from metaphysics together should allow for an unprecedented and ground-breaking approach to the perennial problem of representation, laying the foun- dations for a fruitful research program. To develop a realist theory of perspectives, I plan to bring together two important strands of contemporary philosophical theorising: the re-emergence of relativism and the study of the concept of ontological fundamentality. The theory to be developped combines insights from pertinent recent research into a novel and original form of ontologically robust relativism, according to which the world contains perspectival objects. Relativism and fundamentality have been widely discussed, but in separate subdisciplines of academic philosophy: the former in the philosophy of language and the philosophy of mind, and the latter in metaphysics. No attempt has yet been made to examine their relationship, and their potential to complement each other has not yet been recognized. Drawing on both resources, I aim to articulate a world view that allows for tolerance, i.e. real disagreement that still is – in some sense further to be specified – ‘faultless’. Moreover, a relativist account of the representation relation clarifies a problem in theories of fundamentality: how to account for the problematic status of so-called ‘lesser’ entities, things that are grounded in others and are not part of the fundamental level of reality. Conversely, allowing for grounding ties between the truthmakers of relativist judgments helps to explain faultless disagreement without invoking a problematic relativisation of truth itself.

Such an ontological account of perspectivality and of the representation relation accounts for perspectival objects and facts by taking them to be a special category of so-called ‘lesser entities’, things that are not funda- mental but grounded in things other than themselves. Not all such lesser entities – non-substantial particulars such as boundaries, holes, sounds, as well as ephemeral or ‘non-canonical’ objects of perception such as flames, soap-bubbles, glimmers, highlights, reflections, echoes, shivers, atmospheric phenomena like rainbows and mi- rages, shadows, after-images, constellations and affordances – are perspectival. According to the hypothesis to be explored, however, all perspectival objects are lesser entities – they are grounded, at least in part, in reactive properties of perceiving subjects. Taking the contemporary discussion of grounding and fundamentality as its starting point, this second phase is devoted to a detailed exploration of historical antecedents of this question. I will examine in some detail the (almost always stepmotherly) treatment so-called ‘lesser entities’ (non-substantial particulars) have received in the history of philosophy, and connect it to two much broader topics, the discussions of secondary and response-dependent qualities in both modern philosophy and in the contemporary literature on the one hand, and to the most important and certainly most influential attempt to ontologically explain features of our representations, namely Kant’s transcendental idealism, in particular according to the so-called ‘dual aspect’ interpretation of it. On my way to a unified theory of the representation relation, my starting point and the form of representation I take as paradigmatic is perception. I start with an Aristotelian slogan – perception is intake of form without matter – and work my way ‘upwards’ the cognitive ‘hierarchy’. The central idea, to be found in De Anima, may be cashed out as follows: Perception naturally occurs, as a biological phenomenon. Its veridicality is primitive: we can never be deceived by it with respect to its proper objects. These proper sensibles are prior to their perception in the order of explanation: we explain sight as the sense that gives us colours, not colours as the things that are the objects of sight. In all types of perception, the medium is that of a transmission of form that is taken in by the alteration of the sense (where this alteration is the perception). Seeing perception as emerging in this way as the joint outcome of three factors – the form to be seen, the form transmitted and seen by the perceiver, and the intervening medium – allows, I hope to show, for its most puzzling feature: that representational properties are both intrinsic and relational. Representational states being intrinsic, they ‘mirror’ the world as it appears to some perceiver, i.e. as perspectival and perceiver-oriented. Being relational at the same time, they connect perceivers to things outside of them, and make themselves evaluable in terms of external objects. Because an understanding of representation as self-location places the ‘subjective’ and ‘phenomenal’ elements in the mode, rather than the content of the attitude, allowing for an adverbialist, rather than propositionalist analysis of it, it makes an objectivist notion of content possible. Such a notion can be found in Aristotle, and the present project aims to defend his central idea, which was the almost exclusive focus of almost all of the history of the philosophy of mind, but seems to be largely forgotten today (though cf. its recent excavation in Kalderon ()): representation is coformality, the sharing of ‘form’ between what is represented and what represents. As the medievals realised, the account of perception as intake of form may be generalised to representation more generally if, for more complex mental states, we draw an act/content distinction – as Duns Scotus did between the ‘subjectively’ present (accident of) thinking and the ‘objectively’ present external form. On such a generalised conception, the form taken in from a is what a is represented as; and to represent a as F is to have F as a’s form in mind. Such having forms in mind as the forms of a is a separate mental operation (aka reference), but it not particularly mysterious: it has its natural home in perception, and is only subsequently, and on this basis, performed autonomously. Intentionality is thus explained as the having in mind of a qualitative content (form) in a certain way. To account for states that exhibit primarily not only representational or intentional features (as do, respectively, perception and intention/desire/wishing), but both in an equally fundamental way, a third primitive element is needed: an attitude the subjects takes on what it represents. Such an ‘active’ component explains how the represented form is taken to be and relates it to the formal object of the mental state in question, thus generating its correctness conditions. In fear, for example, the subject both represents something as dangerous and has a negative, flight-inducing attitude towards it; in belief, the subject entertains a thought, and assents to it. In both cases, it is the attitude that ‘generates’ the content, as it were, as it takes the form (itself taken in a way) to relate

to what it is a form of. With linguistic representation, finally, the form is externalised – it is the linguistic items themselves, taken in a way, that may be said to reach out to what they are about. Representational properties, being intrinsic, are not had in virtue of things outside the entities that have them. They are relational, however, in two ways: in virtue of reaching out to their direct objects, if any, and in virtue of being representations of these objects as such and such. It is this second feature, representation-as, which accounts for perspectivality, and the perspectivality is located in the mode of the activity described under (ii) above, in the way it is done. The content-conditions for representation-as have a two-fold structure: To represent something a as F, I must be appropriately related to a item external to the representation (this may be a ‘proposition’, a ‘state of affairs’, a perceptual situation, an object, a plan) and my standing in this relation to the external item must be an appropriate reaction to it (serious, non-lucky, veridical, non-perverse, not practically irrational). We may call the first item the cognitive base, and the second the attitude taken up towards it. Normativity. Under what circumstances is the reaction to the cognitive base appropriate? The answer, I think, exhibits again a double structure: the reaction must be caused by (and justifiable in terms of) its base, and it attitude must match its formal object, i.e. the axiological property it attributes and in terms of which it is individuated. The double structure of representational states, reaching out to something beyond themselves and at the same time situating us with respect to it, allows for two types of failure to meet the characteristic content-conditions for the respective state: either the failure concerns a mislocation of ourselves (wrong direct object), or a constitutive condition on the attitude in question fails to be met (wrong formal object). What this means concretely depends on the form the content-conditions take: . Judgements have success-conditions: My judgement of a to be F is successful iff I succeed in referring to a and in predicating F of it. Two types of failure are possible: (a) Wrong direct object: I do not refer to a (but to something else or to nothing at all) or do not predicate of it to be F (but to be something else, or nothing at all). (b) Wrong formal object: I do not succeed in commiting myself to a’s being F, but perform some other action or no action at all. . Beliefs have truth-conditions: My belief that a is F is true iff Fa. Beliefs are criticisable on two grounds: (a) Wrong direct object: it is not true that a is F. (b) Wrong formal object: my belief falls short of knowledge, i.e. has not the right genesis to be a manifes- tation of my desire for truth. . Perceptual experiences have veridicality-conditions: My perceptual experience of a as F is veridical iff I am visually presented with a and see it as F. Two types of perceptual ‘error’ may be distinguished: (a) Wrong direct object (illusion): While I do have a perceptual experiences as of a being F, I am either not visually presented to a or do not see it as F. (b) Wrong formal object (hallucination): My state cannot be characterised as one of perceiving. . Emotions have appropriateness-conditions: My emotion of type T towards a is appropriate iff a exhibits the formal object of emotions of type T. Two types of error: (a) Wrong direct object: My fear is not directed at the dog or the dog, at which my fear is directed, is not in fact dangerous. (b) Wrong formal object: I see the dangerous dog, but I don’t see it as dangerous. While I am struck by the right value (dangerousness), I don’t react in the right way (by fear). . Desires have satisfaction-conditions: My desire to ϕ is satisfied iff I want to ϕ and I ϕ. Again, two types of error are possible: (a) Wrong direct object (akrasia): I can ϕ, but do not ϕ. (b) Wrong formal object: I ϕ, but I don’t do it as a my plan. While I realise my plan, I don’t react in the right way to my ϕ-ing, i.e. I do not see it as the satisfaction of my desire. Whenever there is a mismatch between attitude and the intended formal object, I dishonour a certain value. Be- cause the attitude itself is voluntary,and can be withhold, the standard of correctness is internal to it and conditional on my having the attitude. This explains why the normativity of the mental, contrary to what is often held, is never a matter of some mental state being obligatory (under some circumstances), but exclusively a question of its being

permissible (or perhaps good) to be in such a state (cf. Blum b). That emotions and beliefs, e.g., are optional in this sense means that they are only internally criticisable, and that the values that justify the (conditional) norms governing them are in some sense ‘internal’ to them. For belief and emotion, the falsity of prescriptivism is demon- strated by the blamelessness of two familiar philosophical characters, the Sceptic and the Stoic respectively: the Sceptic acknowledges things as true, but does not react with belief, ‘stays with appearances’, even though belief would be appropriate. She is blameless, even though perhaps criticisable for not doing as well as she can (though such a criticism certainly requires substantial philosophical argument). The Stoic, on the other hand, acknowl- edges things as offensive/dangerous etc. but does not react with anger/fear etc – he stays with ‘feelings of values’ (cannot avoid being struck), but does not respond with an emotion that would be appropriate. From the point of view of a long and honorable tradition, the Stoic is not only blameless, but indeed praisworthy – certainly not a possibility an account of the nature of emotions should rule out! Disjunctivism. While professing non-sense, wishful thinking, hallucination, psychopathic ‘emotions’ and the ‘desire’ for water of the hydropic are not judgments, believings, perceptions, emotions and desires respectively, they still stand under (and are criticisable in terms of) the same norms. This is a difficulty for disjunctivists who cannot explain it just by the subjective indiscernibility of the two states. Formal objects fill in the lacuna: even though hallucinations are not ‘failed’ perceptions, but something different altogether, they are still subject to the same norms because they have the same formal object as perceptions. In the same way you can be afraid of spiders, but only if you are afraid of them as dangerous, you can hallucinate unicorns, but only if you see them as existing. The metaphysics of emotions. This basic structure of representation-as is perhaps most clearly discernible in the case of emotion: when I am afraid of a dog, I locate myself with respect to it (putting it, so to say,on my mental map of the world), but I also reach out to an axiological aspect of the world, the dog’s being dangerous. It is in virtue of this formal (or, in Brentanian terms, ‘secondary’) object of my fear that I represent the dog as dangerous (Mulligan a, b, ; Teroni ). This representation, however, is not just grounded in the ‘mode’ of my entertaining a certain content, or the ‘attitude’ I have towards the direct object of my state, the dog (as Teroni and Deonna would have it, cf. their ; ; ). It is, rather, a reaction to an independently given feeling of value, which reveals to me the dangerousness of the dog. Only by construing the emotion as a reaction, I argue, can we explain how it contingently has the valence it has, i.e. why, for non-perverted people, fear is not a pleasant emotion to have, and why being afraid feels a certain way. Phenomenology. That the attitude I take towards the cognitive base comes with a feeling explains, I think, the phenomenology of the mental. This ‘inner side’ of our mental states, which enables and sometimes constitutes our awareness of them, is a matter how they feel. To account for how they feel may be one way of cashing out what it means to say that ‘there is something it is like’ to be in them, though the latter notion is so slippery that it is hard to be sure. That my taking up an attitude towards a representational state that relates it to a formal object is quite unmysterious and ubiquitous phenomenon; while it requires, as its ground, that the attitude has a determinate nature, it does not require either phenomenal content nor autonomous phenomenal properties. Intentionality. Taking emotions as my paradigm, I would like to assess to what extent the notion of “formal object”, in terms of which the appropriateness of emotions may be assessed, can be generalised to other mental states. The first group concerns mental actions where the form is produced autonomously, and thus has only intentional existence. The roles of the two factors in perception are here exchanged: the action that reaches ‘out’ is not caused by, but causes the representational state. This constitutes the so-called ‘intentionality’ of the latter. A property of something is intentional iff it is taken to be about something else than itself. It is so taken to be if we attribute to it conditions under which it may be said to be correct. Correctness conditions specify the intentional content, but – being conditions – do not themselves require this content to be satisfied. If I am looking for the Holy Grail, for example, my activity is directed towards, and rationalisable only with respect to the Holy Grail, which, or so we shall assume, does not exist. I am intentionally directed towards the Holy Grail, without standing in a relation to it: there is nothing, after all, for me to stand in a relation to. Intentionality is the flip-side of representation, as it were: whereas representational properties are intrinsic, but relational, intentional properties are extrinsic, but non-relational. Taking up an attitude towards the cognitive base turns the latter’s relatum into the former’s ‘intentional object’. Such a ‘conversion’ of the relatum of a representational state into the intentional

object on which an intentional state depends for its existence without being related to it, is what happens in Kantian ‘synthesis’: when I see a thing as red and white, redness and whiteness hang together by being aspects of the one thing my perception relates me to; when I, however, only imagine a red and white thing, the link can not come from the object alone – it must be ‘constructed’ by my faculty of imagination, and it is so constructed by my imagining one thing as both red and white. This intentional object will therefore be extrinsic, depending for its existence on my act of taking my representational state in a certain way. Intentions, desires, imaginings. In perception, the intentional object is the representational object as per- ceived a certain way, the object, as it were, internal to the perceptual state, but extrinsically constituted by this state being a reaction to the world (its form, in this sense, has only “intentional existence”, as the medievals said). In contrast to perceptions, intentions, desires and imaginings have their intentional properties in a non-derivative way. They are extrinsic, because they are signs, but non-relational, because they are characterised by “intentional inexistence”: such psychological states may exist even in the absence of what they are about. With intentions, desires and imaginings this is even the point of such states. They consist in an attitude we take up towards some representational state as something to-be-done, as something good and as something possible respectively. Reference and singular thought. A mental state peculiar in many respects I call (for lack of a better word) ‘thinker’s reference’ – it is the ‘having in mind’ of a definite individual and subsumes not only demonstrative thought, but also thinking of oneself and whatever goes on when one uses a description ‘referentially’ or stipulates, as in a dream, that some description is true of something in particular. Even though reference is intimately con- nected with our use of names, it is not explainable in terms of it; rather, in the spirit of the ‘authoring-tool model’ (cf. below), it is names that are explained in terms of reference. According to Travis, a name is “what functions to make it recognisable that the thought expressed in the whole of which it is a part is a singular thought of a certain object, namely, the one which, in this sense, that name names” (Travis ). Names, being tools to achieve reference, are of a special nature. Emphasis on the question how cognitive significance is mediated by modes of presentation has made it appear mysterious how we can have genuine de re thought: what is it that allows us to reach out to an object, even in the absence of any descriptive information about it? In the same way as colours make us see things as extended, it is names allow us to reach back to their referents. It is its name that ties us to the object, through a chain of entia successiva from the original baptism to our use of it. This picture, I think, can help us with some problems in the theory of rigidity. Singular terms, it may plausibly be held, are temporally rigid, they refer to the same thing at all times at which this thing exists. But then how can it be that the sentence “Plato is dead” now expresses a truth? What is there, now, to serve as a referent of “Plato” and as a thing of which we can truly say thatit is dead? These problems reappear in the case of modal rigidity,for we want to count “Plato does not exist” as possibly true. To talk of names ‘holding fixed’ their reference across a (perhaps contextually restricted) range of possible worlds does not explain these phenomena, for they classify descriptions encoding modally and temporally stable conditions like “the positive square root of ” and “the offspring of gametes G” on a par with proper names. This is why, in his foreword, Kripke (: , fn. ) distinguishes de jure rigidity “where the reference of a designator is stipulated to be a single object, whether we are speaking of the actual world or of a counterfactual circumstance” from mere de facto rigidity of a modally stable description. This distinction has given rise to a lot of discussion and it seems fair to say that no clear account of the difference has yet been given. In my account, the distinction is as clear-cut as it could be: names, being processes, depend on their causal origins in ways no description could ever do. Rigidity is existential dependence. A rigid designator is essentially and existentially determined by its referent: the name is a mere tag, without descriptive content, nothing but a name for its bearer: two rigid designators for the same thing differ only notationally. Belief and supposition. Already with wishes (that may be directed towards the past), but definitely with beliefs and suppositions, the ‘representation is coformality’ model may seem to reach its limits. Certainly with such paradigmatically ‘propositional’ attitudes, it may be thought, we have to reintroduce contents as mental particulars. Such pessimism, I hope to show, would be premature. Like perception and emotion, belief is a reaction: the representational state is an entertaining of a thought, and the mental act is a type of Stoic assent (in the case of belief; dissent in the case of disbelief). To believe, Augustine says, is ‘cum assensione cogitare’ (de praed. sanct. ) (cf. Mulligan : ). In the same way that thinking, by the ‘affirmative’ mode, gives beliefs, it may, by the

‘interrogative’ mode, give questions, and by the ‘mode of conjecture’, suppositions. What is belief a reaction to? Brentano may help us here, drawing our attention to the possibility that inner representation is containment (i.e., works roughly in the way in which a quotation name represents what it quotes by surrounding it by quotation marks). Quoting De Anim III. , p. , b, , he remarks that a doubling of intentional objects is only needed if we individuate them by the representational states, but not if we individuate them (correctly) by the mental acts directed at it: “the presentation [Vorstellung] of the sound is connected with the presentation of the presentation of the sound in such a peculiarly intimate way that the former, by obtaining [indem sie besteht], contributes inwardly to the being of the other [innerlich zum Sein der anderen beiträgt].” (bk. , ch. , §, : ) The resulting picture is the following: There is thinking going on, and some of it is partitioned into a belief of a certain content by an attitude of assent the subject directs at its own mental state. Such a belief, however, is not a representational vehicle in the traditional sense, that allows the subject to represent something. Rather, the belief, as a mental particular (and thus distinguished from the state of believing), is itself produced by (and thus posterior to) the representing of form. The attitude of assent directed at the immanent object is at the same time, and by the same act, a representation of itself as a psychological state, i.e. a belief. It thus projects its own form onto an amorphous mental process of thinking. There is a peculiar interweaving [“eigentümliche Verwebung”, : ] of the object of inner presentation (the belief) and the presentation itself (the thinking), and both belong to the same act, the believing (cf. also Burge : ). Judgment. This account of belief as projection of form onto thinking is ‘externalised’ in Russell’s so-called ‘mul- tiple relation theory of judgment’ (MRTJ), where the ‘inner’ commitment to truth is expanded into an ‘outer’ commitment to knowledge. The main attraction of the MRTJ, in my view, is not it’s avoidance of mysterious (and arguably incoherent) abstract objects such as propositions (though this is certainly a plus), but rather its emphasis on how, in an act of judging, the different objects the judgment is about are brought together. The MRTJ describes an attractive middle position between some higher-order self-attribution of the judgement and the introduction of egocentric modes of presentation of some hyperintensional judgement content. Contrary to Stout’s criticism that it requires that judgment entails “fixing our attention on our own minds and on the act of judging or believ- ing” (: ), only the ordering it effects, not the judgement relation itself must be transparent to the judger: while I may be unsure whether I really am judging, I do judge as long as I am certain that I am entertaining the thought that bRc rather than the thought that cRb. Neither does it require the ‘content’ to be hyperintensional. It is perspectival, directed from the point of view of the judging subject, and ordered relative to it, without thereby entailing that the subject (or a representation of it) is a part, or a constituent, of the judged content. As I show in “Egocentric judgments: multiple, but perspectival”, that “the act of judging itself actually arrange[s] the term in the way they [are] judged to be arranged” (Griffin : ) is, contrary to what Griffin thinks, an asset rather than a liability: it explains why explicitly egocentric judgments are shielded from a certain type of reference failure explains their ‘immunity from error through misidentification’. The perspectivality of the content judged is in these cases automatically endorsed as one’s own: the anaphoric link between (part of) what the judgment is about and the judger makes misidentification impossible. Linguistic representation. As with names, I start with a broadly Travisian picture of linguistic representa- tion. On his so-called ‘authoring-tool model’ (cf. also Kaplan b), words are taken to be means to express thoughts (or to contribute to the expression of a thought), not themselves (at least not primarily) things that express thoughts (because they may express very different thoughts on different occasions, even if their meaning(s) is/are held constant). This picture, as sketched, is silent on what it is to express a thought in words, and this is where the representation-as-coformality model may prove helpful. Suppose I produce the shape depicted as follows: “Zed is dead.” By fixing your gaze on it, you learn that Zed is dead. How is this possible? How is it that by fixing our gaze on some marks of chalk we learn something about people and things in no (obvious) connection to us? This, I take it, is the primordial philosophical question about linguistic representation. It is ordinarily assumed that the representation question divides into two sub-questions, i.e. (i) how is it possible that the marks of chalk mean what they do? and (ii) how is it possible that by understanding what they mean, we learn something about the distant past? It is only if you accept this sub-division, that the notion of a proposition starts playing an (apparently) explanatory role: in terms of it, two allegedly helpful answers may be given: (i∗) the marks of chalk express the proposition that p; and (ii∗) the proposition that p is true iff (or: in virtue of the fact that) Zed is dead. Once

we accept the division, deflationism about truth becomes attractive: it becomes almost irresistibly plausible to say that (ii*) holds just because the proposition in question is what it is – saying that it is true that Zed is dead is just saying that Zed is dead. (i*), though it encodes a contingent fact about language, seems utterly trivial – all you need to see its truth is some competence of English. Our initial puzzlement evaporates into two trivialities and we wonder what made us think our question was interesting in the first place. Just a little below the surface of the two trivialities lie two deep mysteries, which seem completely intractable if approached in these terms: (i+) How is it possible that our activities as language-using intentional agents bestow mind-independent and abstract entities with powers of representation? (ii+) How is it possible that relations of aboutness and of truthmaking hold between these abstract entities and things in the world? I think the sub-division of our original question was a mistake and that we should cut the middle-man out: propositions, understood as abstract objects of belief, do not serve a useful explanatory function. We should not construe belief in terms of content, but rather explain content in terms of belief. The sentence means what it may be used to express (under certain conditions). Such an expressivist account of linguistic meaning, I hope, not only does away with propositions, but shows how sentences mean what they do in virtue of sharing their form with the beliefs they express, i.e. the forms of things out there that thinking subjects project on their own representational states. Cratylos was right: representation is a form of mimesis after all.

. Learning from the History of Philosophy

.. The relativistic challenge

Various forms of relativism have formed part and parcel of philosophical theoretizing since Socrates took on the Sophists. Relativism has often played the rôle of a backdrop, in contrast to which realist systems have been de- veloped. In its common-sensical or so-called ‘naïve’ form, it has often been taken to have been decisively refuted (Gowans ; Baghramian ; Engel & Rorty ; Boghossian , ; Kölbel ). In recent years, however, there has been a remarkable surge of interest in relativistic theories of semantic content, examining attributions of truth to a diverse array of ‘subjective’ judgments, such as judgments of taste, of epistemic possibility, of future contingents and of knowledge attributions. While semantic relativism was first presented as aiming at accounting for “a new kind of linguistic context-sensitivity” (MacFarlane ) or for forms of context-dependence that cannot be (at least, not easily) handled using the existing models (cf. e.g. Scott () on locational and agential operators, and the double-index theory of Kaplan () and of Lewis ()), it has soon become clear that more radical forms are possible. One particular point of crystalisation is the notion of ‘faultless disagreement’ (Kölbel ; MacFarlane ): if two of us can disagree by uttering “liquorice is delicious” and “liquorice is not delicious” respectively without anyone of us being ‘at fault’, some novel account of either disagreement or faultlessness is needed (cf. Egan ). Such accounts will need to say something about truth, as it has seemed obvious to many (cf. e.g. Wright ) that for disagreement you need truth and for truth you need objectivity, which is plausibly taken to be a matter of representing how things are in a way independent of one’s personal perspective on the world. In contrast to early relativist theories, which were presented as a continuation of multiple-indexing, the notion of faultless disagreement raises questions of a more metaphysical nature: about the nature of content, the metaphysical status of truth, and about what makes our utterances true and what determines their truth. In its first part, the present project addresses these three questions, aiming at making progress beyond the state of the art. The nature of content has been at the center of much of recent theorising in the philosophies of language and of mind. A powerful argument in favour of more radical forms of relativism, which go well beyond the resources of contextualist theories to account for context-sensitivity, index-relativity and the logic of context-shifting oper- ators (cf. the articles collected in Preyer & Peter () for different assessments of the potential of contextualist approaches), stems from the intuition that claims like “liquorice is delicious” are not, neither overtly nor implicitly, about the speaker:

“The contextualist takes the subjectivity of a discourse to consist in the fact that it is covertly about the speaker […]. Thus in saying that apples are “delicious”, the speaker says, in effect, that apples taste good to her (or to those in her group).” (MacFarlane : )

“On this [contextualist] view, “It’s wrong to cheat” involves ellipsis, or a place holder indicating a set of standards, a code, whatever. What [its] use says depends on what has been elided or what is being assigned to the place holder.” (Richard : )

If content itself is relative in this way (as in the early theories of Cappelen, cf. e.g. a b), there is no com- mon content for the speakers to disagree about: Cappelen & Hawthorne () call this the ‘problem of shared content’. To answer this charge, contextualists can draw on the rich literature on the semantics/pragmatics in- terface and invoke the recently much studied mechanisms of ‘free enrichment’, whereby contextual information is made available for the hearer without thereby becoming part of the semantic content of the utterance. Even though much progress in this debate has been made recently, in particular by members of the Marie Curie Initial Training Network Perspectival Thoughts and Facts (PETAF, grant agreement no. FP-), of which I am a mem- ber, and the Consolider-Ingenio program PERSP equally based in Barcelona (cf. in particular García-Carpintero & Kölbel () and the papers therein, the two recent books by François Recanati (; ) and the recent exchange between Isidora Stojanović () and John Lasersohn ()), there remains much to do. Particularly for relativism concerning epistemic modals such as ‘must’ and ‘might’ (Egan et al. ; MacFarlane a, a; Egan & Weatherson ), it is very difficult to neatly separate semantic from pragmatic factors (cf. von Fintel & Gillies ; Wright ; von Fintel & Gillies ; Dietz ; Egan ; Bach ) and the alleged retraction ‘data’ themselves have been challenged (cf. e.g. von Fintel & Gillies ). Alternative contextualist (Dowell ) and presuppositional accounts (López de Sa , ) have been proposed. On the question whether there may be faultless disagreement, the jury is still out (cf. also Iacona (), Goldman ()). In particular, the lessons learned within the philosophies of mind and language have not yet fully been brought to fruition (the important work of (Kölbel , , ) notwithstanding) with respect to moral relativism, which is usually presented as a form of contextualism (cf. Gowans () and the articles in Hales () for an overview). Rationally irresolvable moral disagreements, at the heart of the moral relativists’ motivation (cf. e.g. Harman () and Wong (, )), are dissolved by appeal to hidden indexicals (Kölbel : ). The present project will innovate on this state of the discussion by presenting a genuinely metaphysical form of relativistic content, where each party states a perspec- tival fact, and is thus epistemically faultless. Because the different perspectival facts are still grounded in the same item of reality, there nevertheless is something for the contestants to disagree about. Truth is another perennial topic of philosophy that has been put into sharper focus by the relativism debate. Recent discussions, in particular of MacFarlane’s notion of “assessment sensitivity” (MacFarlane b), have focussed on the question whether an utterance being true or ‘accurate’ involves its standing in some relation to a context of assessment (cf. in particular MacFarlane (b: ) and Cappelen & Hawthorne (b: )). In this context, MacFarlane () has sparked a lot of discussion. Cappelen & Hawthorne () have mounted a concentrated attack against the very intelligibility of relative truth, to which relativists are responding in quite different ways (cf. the book symposia in Analysis (Cappelen & Hawthorne d; Glanzberg ; Soames ; Weatherson ; Cappelen & Hawthorne c) and in Philosophical Studies (Cappelen & Hawthorne a; Lasersohn ; MacFar- lane b; Richard ; Cappelen & Hawthorne b), cf. also the reviews by Brogaard (b) and Almér & Westerståhl ()). It is not yet clear whether the very idea of relative truth is coherent (cf. also Heck ; Ein- heuser ; Wright ; MacFarlane ; Glanzberg , ), nor how these discussions relate to the more familiar question about whether truth is correspondence with reality (cf. e.g. Künne () for an excellent survey of the traditional conceptions, and Wright () and Brogaard (a) for critical assessments of their compatibil- ity with relativism). In particular, it would be interesting to find out whether assessment sensitivity might privilege the truth-as-operator view over the more familiar truth-predicate account (cf. Mulligan (), Künne (, ) and Blum (a) for more discussion). The present project defends the view that a relativisation of truth should be avoided. What is relative is what makes truths absolutely true – i.e. some truthmakers are perspectival, i.e. relative to a perspective, in the sense of being grounded on other things than themselves. Truth is not itself a relative thing like deliciousness – what is relative, at best, is rather what truths there are. It is useful to compare truth-from-a-perspective to truth-at-a-world in this respect. Neither is a kind of truth; they both rather involve the notion of representation, and are a species of truth-according-to-some-view-of-the-matter: in the same way merely possible worlds represent things differently from how they really are, perspectives are perspectives on something existing independently of them. Even though

they often are not, such perspectives may be wrong, poor, misguided or ill-informed. This leaves conceptual room for there being experts, and training, also in taste-related matters like viticulture. Wine experts, and wine tasting courses, may help us to sharpen, broaden and in other ways improve our perspectives on tastes, and we may thereby get a better grip of perspectival facts existing independently of our perceiving them. A third important metaphysical problem in the background of recent discussions of relativism does not concern truth itself, but rather what is responsible for the truth of relativistically interpreted utterances: granted that their truth-value varies with contexts of assessment, what is it in the world that is responsible for their having the truth- value they do? Kölbel (: ), for instance, frames the issue of relativism as the question “whether there are novel truth-determining factors, such as standards of taste and states of knowledge, and how exactly such a determination relation should be construed”. If the truth of “this is delicious” depends on the standards of taste of the speaker (Kölbel ; Lasersohn ; Récanati ; MacFarlane ), then we may legitimately ask what such standards of taste are. If standards of taste really exist, and determine the truth-value of judgments of taste, we may wonder, why should it not then be possible to describe them in an objective and non-relative way? The issue is especially pressing in the case of moral relativism: if moral judgments are relative to moral codes, as both Copp () and Wong () have it, then this is explanatory only insofar as moral codes are something over and above (sets of) moral judgments – but what else could they be (cf. Mackie () and Kölbel () on this issue)? The present project avoids the postulation of mysterious standards and codes, and takes the truthmakers of rela- tive judgments to be secondary qualities – real, but non-fundamental properties of the external world. Faultless disagreement is not so much about truth itself (otherwise it could not be faultless), but about the truthmakers of our ascriptions of secondary qualities. When you and I disagree about whether liquorice is delicious, we represent different perspectival facts – in this sense, our disagreement is faultless. It is, however, still a disagreement about some real properties of liquorice, those responsible for you finding it tasteful and me finding it disgusting, and it is in these independently and objectively existing properties that our respective perspectival objects are grounded. Recognising this difference, we may reasonably opt for tolerance.

.. Tracing the problem of representation back to its sources

Apart from Kalderon () and Glauser (), there is relatively little previous work on the metaphysics of the representation relation. Even though the history of the philosophy of mind has recently received some attention – good overviews are the collections published by the “History of Mind” research unit of the Academy of Finland, centred around Simo Knuuttila (Saarinen ; Knuuttila & Niiniluoto ; Alanen ; Knuuttila ; Knu- uttila & Kärkkäinen ), by Henrik Lagerlund (Lagerlund & Yrjönsuuri ; Lagerlund b, a), by Dominik Perler (,,,, Perler & Haag b; a and Perler & Wild ), as well as Brunschwig & Nussbaum (), Easton () and Ebbersmeyer () – and there are a couple of important monographs, either on specific philosophers or periods – I will just mention Annas (),Aquila (), Brook (),Clarke (), Della Rocca (), Grüne (), Hausman & Hausman (), Kukla & Walmsley (), Long (), Marshall (b), Need- ham (), Oaklander (), O’Daly (), Perler (), Reed (), Rollinger (), Schlutz (), Sorabji (), van der Burg (), Waxman (), Wedin () and the papers in Jolley () and in in Nightingale & Sedley () – as well as a couple of studies specifically on the theory of ideas – Watson (),Perler (), Wild () –, its metaphysical aspects, as well as the metaphysics of the mind more generally, are still quite neglected. Although some important work has been done on the ontology of mind (esp. by Helen Steward, cf. her and some recent articles (; ), Mike Martin (, , , , ) and Rowland Stout ()) and Peter Geach () is of course an important influence, only the monographs French (), Lowe (, ), Matthews (), Senderowicz (), Tye () and the papers in Haugeland () and in Kim () seem to deal with the metaphysics of the mind directly (Jaworski () looks interesting, though I have not yet had the chance to have a look at it). Given this scarcity of resources, a fresh attempt is needed. The present project attempts to do exactly this, starting by retracing the history of the problem. Tracing the problem back to its sources. Perhaps surprisingly, the methodology I plan to adopt for the proposed monograph is partly historical. Only such an approach that is informed by the history of the subject, I believe, will allow for the direly needed enrichment of the quite restricted conceptual repertoire that is nowadays

brought to bear on these issues. Studying the classic problems in their original settings thus may, I hope, open up new avenues towards their solution, occluded, or at least much less perspicuous in their textbook presentations. I thus plan to organise the proposed monograph along historical lines, ‘reaping’, as it were, insights (or rather what I take to be insights) into contemporary problems along the way: Aristotle: perception is a mode of a perceiver Aristotle’s starting point, in De Anima is certainly not one we today would find natural. Perception, for Aristotle, is a natural phenomenon and as such does not require an explanation: it occurs when and because a sense-organ is altered, and thus a potentiality is actualised. There is no other story to tell: perception just is this alteration of the sense-organ. Perception (aisthēsis) is thus an alteration of the perceiving object (De An., II.), the effect of being affected by an external object, which is actually so as the perceiving subject, before being affected by it, is potentially. We will discuss in more detail later how to understand that alteration (cf. p. below). For the time being, it is important to note that Aristotle’s question is not how perception can come about, or how it can provide us with a route to the outside world. We start with an eye, which is already a part of the natural world, and note that its function, which it has essentially, is sight, i.e. the uptake of colours. That (veridical!) perception occurs is part of the data. Lesson to be learned: This theory thus combines two very attractive features: a naturalistic account of the experience we undergo when and because we perceive, and an explanation how perception and its objects may be mutually dependent, without making the objects ‘subjective’ or dependent on their being perceived. Aristotle: colours are potentialities of sight The essence of each sense, which is a capacity, is “naturally relative” to its objects, the proper sensibles of this sense, which (may) exist before the sense in question exists (Cat. a-), for “perception is hardly of itself – but there must be something else beyond the perception, and this must be prior to the perception; since what brings about change is naturally prior to what is changed, even if they are spoken of in relation to each other” (Met. Γ. , b-a). Sight is thus posterior to its proper sensibles, colours and ‘shininess’ (to produce perception in the dark) – categorical properties which ground the disposition of things to be seen and are thus the ‘agents’ of perception. The proper sensibles are prior in the order of explanation: we explain sight as the sense that gives us colours, not colours as what are the objects of sight. In this sense, the proper sensibles are the only things seen. Lesson: This understanding of Aristotle solves, I think, the old conundrum where to ‘locate’ the way things look: when I have an illusion as of a red object, which is really white, it is sight, not the eye, that becomes red. It also gives, I think, a plausible model of how names represent their referents (cf. below). the Stoics’ reifying of contents It is only with the Stoics that contents, in our sense, are reified and attributed to perception. The Stoic theory, however, avoids the false and misleading ‘conceptual vs. non-conceptual’ dichotomy, and is best understood as simply ‘enlarging’, as it were, the subject of affection in the Aristotelian theory: it is no longer just sight, but the whole subject, including its conceptual faculties, which may be said to be ‘affected’ by red. They thus allow for ‘cognitive penetration’, as it is now called, without, however, giving up the characteristic Aristotelian thesis of the posteriority of the perceiving act to its object. The crucial new ingredient is the distinction between sensory appearance (phantasia) and assent (sunkatathesis). It is assent to the so-called rational appearances that explains us humans being in the grip of the problem of illusion. Lesson: To what extent the received form is conceptually structured will thus receive highly specific answers, both depending on the form and on the subject that receives it. Perception-as, in its cognitively demanding sense, does not have to be seen as a kind of ‘projection’, but may instead by subsumed under the humdrum phenomenon that perception depends on its milieu (where this is broadly construed, so as to include the conceptual repertoire of the perceiving subject). The Stoic notion of assent provides us with what is needed to account for the ‘active’ element in imagination, emotion and belief (cf. below). the medievals’ notion of ‘intentional being’ Albert the Great claims that the form (ie. a colour, a sound) is taken in not in its ‘material being’ (esse materiale), but in its ‘spiritual being’ (esse spirituale), or rather that what is really taken in is just its intentio – a thought which, with Aquinas, became the claim that the form has ‘intentional being’ in the soul. This species theory, most clearly in Aquinas, even though it takes itself to side with the ‘spiritualist’ interpretation of Aristotle, still allows, as I hope to show, for a ‘literalist’ interpretation,

. . We will return to this important point below, on p. ,

with ‘form’ reinterpreted as a structural property of the perceiving act. This, I think, is needed to make sense of the otherwise quite mysterious claim by Aquinas that not just the subject, but also the intervening perceptual medium (normally air) is ‘perceptive of’ the colour. Lesson: This elaboration of the ‘perception is intake of form’ doctrine allows to conceive of representation-dependent properties of objects (e.g. colours) as existentially-independent of perceiving objects, while not importing ineliminable modality: lava was red, even before the first eye evolved, not because it would have appeared so and so to us if we would have been there, but rather because the form was actually taken in, not by a perceiver but by a perceptual medium. That the form has intentional being, I think, helps understand how it can be a form of sight, but not of the eye. Descartes: objective and material reality of ideas It was with Descartes’ reinterpretation of the species doctrine that conceptual room was made for intrinsic representation, which allowed himto claim that it is a change in me to see the sun, but no real change in the sun to be seen by me. Even though representation is a relation, we are thus not forced to conclude that the property representing a to be F is extrinsic. Intrinsic repre- sentation allows for a crucial, but often neglected distinction between two types of falsity, non-representation (material falsity) and misrepresentation (formal falsity), most clearly expressed in AT VII - (AT IX/I -). With a materially false idea (such as, in Descartes’ view, the idea cold), there is a difference between what it purports to represent (a real property in objects) and what it relates us to (a mere privation). Lesson: The distinction between formal and material falsity,I think, is in an interesting way tantamount to (but more plausible than) the disjunctivist reply to the argument from illusion. The materially false idea is not of what it purports to be; it is an internal state in some ways subjectively, or more broadly speaking functionally, indistinguishable from an idea, but does not have any content at all. It is a form, however, and taken by the subject in some way (ie. for an idea), but it is not the form of what it is taken to be of. Locke and Hume’s causalism These subtle distinctions were lost, and the contemporary form of the problem of perception created, by the retrograde and reductive understanding of ‘idea’ and ‘impression’ as mental particulars many commentators thought to find, respectively,in Locke and Hume (though this interpretation is itself very questionable). From (something like) the notorious argument from illusion, Locke seems to have quite carelessly inferred the essential independence of our ideas from their existential independence (II.xxii,)This conception of ideas, perhaps to be found in Locke and Hume but at any rate still predominantly employed by philosophers of perception, that ideas are the objects of sensation, and the associated conception of perception, has been successively reduced ad absurdum by Berkeley. If whether or not an idea represents something external to it is just a matter of how it feels to entertain it, as it is in Hume, then Berkeley is right that even “figure and extension are not patterns or resemblances of qualities existing in matter” (§ ), because our perception of them may change without any change in an external object. Lesson: Though this point is familiar, its repercussions have not, I think, been properly appreciated. The lesson to learn from the Berkeleyian reductio, I argue, is that an idea is not a mental particular, but rather – as Descartes had it – the “form of my thinking by the direct perception of which I have knowledge of it” (AT VII - / IX/I ), i.e., on my Aristotelian picture, what a certain form in my mind is taken to be. Rather than being mental particulars, the form(s) I have in mind are ‘stuffy’ – they are particularised and made countable only by the attitude I have towards them. Reid’s denial of the act/object distinction Reid correctly observed that Berkeleyianism is a reductio not of empiricism, but of what Reid calls ‘the way of ideas’: if “we can think, and speak, and reason about [spirits] and about their attributes, without having any ideas of them” (IHM (Reid : )), as Berkeley allows, then why not allow for the same in the case of bodies? Doing away with the intermediaries, he instead proposed a rather interesting form of adverbialism: ‘The form of expression, I feel pain, might seem to imply that the feeling is something distinct from the pain felt; yet, in reality, there is no distinction. As thinking a thought is an expression which could signify no more than thinking, so feeling a pain signifies no more than being pained. What we have said of pain is applicable to every other mere sensation. […] [T]he sensation by itself […] appears to be something which can have no existence but in a sentient mind, no distinction from the act of the mind by which it is felt.” (IHM ., /a) Lesson: This adverbialist picture, applies not just to mere sensations, but to representations more generally, giving us what is needed, I think, to legitimatise talk of “contents”: rather than hypostatized ‘meanings’ or ‘thinkables’, they are the ‘internal accusatives’ of

mental acts. In the same way the equivalence of ‘dancing’ and ‘dancing a dance’ does not commit us to a transitive construal of dancing and the acceptance of objects that are dancings, so ‘thinking that a is F’ is nothing more than a way of thinking, even though it may be individuated by its being a thinking that a is F. Kantianism about experience as a radicalisation of Reid While Reid did not adequately explain how ex- trinsically representational sensations bestow thoughts with their intentional objects, Kant tried to fill this lacuna by a theory of how objects are, in a somewhat mysterious way, ‘constituted’ in experience, which has as its proper object just an “unbestimmter Gegenstand” (KrV A/B, B). On this reading, Kant does not just, or not even, try to debunk some ‘myth of the given’, but rather tries to explain how experience is partitioned, by the rather mysterious operation of ‘synthesis’, into discrete ‘units’, ie. contents. This expla- nation allows him to explain how, in a judgment of experience (Erfahrungsurteil), concepts are predicated as combined in the object, and how such objective combination carries epistemic necessity. Lesson: Such a ‘realist’ interpretation of Kant sheds light, as I hope to show, on his objectivist understanding of “appear- ances”. Appearances are not putative, or postulated, or conceptually constructed entities that somehow stand between us and the real world, but rather objective, mind-independent objects of empirical inquiry and knowledge. We cognize them through concept-formation; the predication of such a concept of an ap- pearance represents it as falling under it. Kant’s intricate theory of how understanding and sensibility work together to allow for synthetic a priori knowledge is, so far, the most detailled account of representation-as. Brentano’s and Meinong’s reintroduction of contents What he saw as Kant’s misapplication of the act/object distinction and his hypostasizing of ideas on the side of the world was corrected by Brentano who charac- terised psychological phenomena in terms of their ‘immanent objects’, rather than the objects of perception in terms of how we perceive them. A sound, in Brentano, is “that which is as an appearance [Erscheinung] the immanent object of our hearing, that is different from it and according to whether or not we think that the appearance has outside us a matching cause [eine ihre entsprechende Ursache] do we believe that there exists a sound in the outer world as well” (bk. , ch. , §, : ). While Brentano was right about intention- ality, revealed in the inner awareness we have of our own mental states, he was wrong about representation, not allowing, as Meinong observed, for a different, roughly Stoic notion of content. Lesson: Combining Brentano with a Stoic notion of content, we can do justice to what Brentano himself (bk. , ch. , §, : ) sees as the Aristotelian roots of this notion of “immanent object” and “intentional non-existence”. Rather than departing from Aristotle, he extends the ‘proper sensibles’ account to inner pereption, explaining our awareness of our mental states by their being the secondary objects of our sensory acts (bk. , ch. , §, : ). There is thus no need, I will argue, to follow Brentano in taking the represented object to be immanent, as long as we grant an ontological status to our taking some internal form to be a certain way (and thus have a ‘secondary’ object). Sense-data and the perspectival objects of neutral monism Proper sensibles made their re-appearence on the modern stage as sense-data, distinguished by both Russell and Moore from their sensations. On the interpretation they receive by the New Realists (and in Russell’s neutral monism), not just the sense- data, but the sensations as well are “out there”: “If, as Aristotle said, ‘thought and its object are one,’ so are sensations and perceptions one with their ‘objects.’ In fact, there are not sensations or perceptions and their objects. There are objects, and when these are included in the manifold called consciousness they are called sensations and perceptions.” (Holt : ) Such a holistic Aristotelian account takes a “perception” to be an essentially perspectival object, “the appearance of the object from a place where there is a brain (or, in lower animals, some suitable nervous structure), with sense-organs and nerves forming part of the intervening medium” (Russell : ) The neutral monist picture is the most radical and wide-reaching of the Aristotelian idea that perception is grounded in co-formality. I would like to defend a version of it where not just Reidian sensations, but Kantian appearances are ‘out there’, where this may, in a Brentanian way, include the inside of our mind. The argumentative strategy to be realised in what follows, roughly, is to take the two Aristotelian insights described above very seriously, and develop a theory of perceived things as perspectival, grounded entities which share their form with the sense-organs by which they are perceived, thus giving concrete sense to the objective/material distinction we find in Descartes’ reinterpretation of the species theory, and vindicating the new realists’ insistence

on the perceived things being ‘out there’, while still leaving room for a Brentanian notion of perceptual content and a Kantian notion of representation-as. This not only gives an interpretation of the history of the philosophy of mind that stresses continuity rather than often quite superficial or terminological differences, but it also ‘synthetically’ justifies the Aristotelian point of departure, as least by showing that it is not quite as crazy as Burnyeat thinks it is, claming, as a reason not to take Aristotle’s metaphysics seriously, “[o]ne might say that the physical material of animal bodies in Aristotle’s world is already pregnant with consciousness, needing only to be awakened to red or warmth” (: ). While Burnyeat takes this to be a reductio, I think this is exactly right.

.. Perspectivality, Grounding and Fundamentality

The perspectivality of our representations is not confined to the cases widely discussed in the relativism literature, such as judgments of taste, but of much wider importance. Perspectivality has often been considered as essential to intentionality (Searle ): to refer to something is always to refer to it from a certain standpoint, under a certain aspectual shape (Moore ). To a first approximation, the connection between intentionality and perspectivality may be illustrated by considering the case of visual perception. Visual perception is always perspectival: while I do not normally see myself, or the time or place where I am located, what I see tells me where I am; I see the table and the cup from a point of view, at a certain distance from me, and my perspectival perception informs me not just about the shape of the cup and the table, but also about my own location with respect to them. Perception, like many intentional episodes involving perspectivality, is Janus-faced: it informs us both about its object and about our location with respect to it. How is it even possible to simultaneously get these two kinds of information? I cannot, after all, simultaneously see the cup both as round (informing me on its shape) and as oval (informing me about my location). Two main theories of intentionality have been distinguished: mental episodes are directly related to their object; mental episodes refer to their object through the mediation of some intentional content. The second theory splits in turn into two, depending on whether the intentional content is understood as conceptual or non-conceptual. We thus have three general ways of conceiving the intentionality of mental episodes: () as a direct relation, () as relation mediated by conceptual content, () as a relation mediated by non-conceptual content. These three kinds of aboutness have all been taken to be paradigms of representation: Russell famously held that my thinking of the Mont Blanc is to be accounted for in terms of my being related to a representing object (a Russellian propo- sition) containing the mountain itself, whereas Frege defended a theory of representational devices as abstract objects, containing concepts and grasped by thinkers mastering these concepts (see e.g. Dokic (, ), Travis (, , ) and Martin (a, b) for recent reassessments of the Russellian approach to intentionality as applied to perception, and Sellars () or Brewer () for reassessments of the Fregean view). Recently, a third line of thought, originating in the works of Twardoswki, Meinong and Husserl (revived by authors such as Evans (, ) or Peacocke (, , )), holds that the representational powers mental states have must be ac- counted for in terms of their non-conceptual content. To these three ways of characterising content correspond three different ways of individuating mental phenomena: perspectival thoughts and attitudes are individuated in terms of their genesis, by when, where and by whom they are had, while propositional thought is intrinsically and essentially characterised by what it expresses; emotional attitudes introduce a third component: in order to distinguish two different emotions with respect to the same object, we need to appeal to their criteria of adequacy. An ontologically robust (‘Russellian’) account of representation endeavours to explain differences of perspectives or points of view in term of differences among the facts perceived. Perspectives are there to be seen (and pho- tographed as the late Russell repeatedly noticed in defence of his full-blooded perspectival realism). What we see, the intentional object of our perception, is: ‘A table from here’. According to the rival Fregean account, by con- trast, the intentional object of our perception is only the table. The ‘from here’ (or ‘oval’) should not be included in the intentional object but in the conceptual content of our perception. On this alternative account, perspectival properties are semantic, not ontological: what it is for the cup to look oval to me is for that concept to apply to the cup. Epistemic (propositional) seeing, however, is problematic in its own right: do I really have to apply the concept distant to the cup I see in order to see it as being one meter from me? Could not some non-linguistic infant or animal see exactly the same thing? The third, Husserlian, approach to intentionality, takes yet another

line on this: rather than asking about the direct objects or propositional vehicles of my perception, it urges us to ask about its non-conceptual modes of presentation. We can have many different standpoints with respect to the table. While those standpoints are not intrinsic to the table or any other objects of our perception (contra Rus- sell), they are neither to be defined in terms of concepts or propositions (contra Frege). What is really essential to perspectival perception are the non-conceptual, non-propositional or non-epistemic modes of presentation under which it presents its objects (Mulligan , ; Bermúdez ). The present project aims to defend a Russellian account of perspectivality and apply it to the problem of per- ception, thereby providing a new and original answer to the so-called ‘argument from perspective’. The central thesis defended is that perspectives are real: perspectival facts are mind-independent. As a consequence, they are intentional objects, not intentional contents. Perspectival facts account for the perspectivality of many of our thoughts: if my thinking “I’m hungry” consists in my being related to a Russellian proposition containing me and the property of being hungry, then so does your thinking “he’s hungry” – and still our two thoughts seem very different, motivating me but not you to reach for the ham sandwich in front of us. There is something perspectival about my thinking about me, which is shared not by your thinking about me but rather by your thinking about you. Neo-Russellian theories of this “subjective” or “indexical” component of thoughts have postulated an extra ingredient to the semantics of such thoughts, which they call “character” (Kaplan , , a; Perry , ). The study of character has led to the development of two-dimensional semantics, and a revival of Neo-Fregean theories. It is, however, normally taken to be uncontroversial in the intricate and complex debate between Neo- Russellians and Neo-Fregeans that reality itself is not perspectival. Only under this assumption does it make sense for Neo-Fregeans to press Neo-Russellians on the question of how to account for the perspectival “presentation” of non-perspectival facts. If the facts themselves are perspectival, this is a non-issue: the perspectivality of the facts themselves explains the perspectivality of our thinking about them. Philosophers distinguish between essential and non-essential (or accidental) features of things. A feature is said to be essential to a thing if it pertains to what the thing is, in a specifically metaphysical sense, i.e. if it helps constitute its nature; a non-essential feature is a feature the thing has but which merely characterises how the thing is, without pertaining to what it is. The concept of essence or of an essential feature is central to metaphysics. There are two fairly standard accounts of what it is for an object to have a property essentially, both formulated in terms of the notion of metaphysical necessity. On the first account, for an object to be essentially F is for it to be metaphysically necessary that the object is F; and on the second account, for an object to be essentially F is for it to be metaphysically necessary that the object is F if it exists. In an important series of works, Kit Fine (, a, b, , c) has argued that essence, and related concepts such as dependence, cannot in this way be understood in modal terms (cf. also Dunn (b), Almog (), Gorman (), Zalta ()). Such non-modal notions of essence have been widely studied in recent years (Almog ; Mulligan ; Correia , , ) and recent work, including Blum (, ), has broadened this perspective to include other forms of non-causal determination and to examine, in general terms, what it means to say that some things are fundamental, or more fundamental than others. According to the thesis defended, the connection between essence and fundamentality is much tighter than is ordinarily assumed. Grounding is explained in terms of essential dependence: if a is grounded in b, and b therefore more fundamental than a, this is because the very nature of a requires b to exist, or involves b being such-and-such (or requires things such as b to exist, or being such-and-such, in cases of multiple realisability). The relation between the less and the more fundamental, in virtue of which the less fundamental is somehow ‘derivative’ or ‘nothing over and above’ the fundamental, has recently come into focus within the discussion of the grounding relation (Audi (a, b, ); Betti (); Cameron (c); Correia (); Daly (); deRosset (, ); Liggins (); Morganti (); Nieland & Weber (); Raven (); Rosen (); Schaffer (, ); Schnieder (, b, a, , , ); Sider (); Weber et al. (); Williams ()). Grounding is widely taken to be at the heart of a number of important philosophical claims: according to physicalism, mental facts obtain in virtue of neurophysiological facts; according to categoricalism, dispositional properties are grounded in categorical properties; according to ethical naturalism, morally wrong acts are wrong because they have certain non-moral properties. Much progress has been made recently on how to understand such grounding claims, but the discussion has often been at a relative high level of abstraction, barely touching on concrete examples. The

present projects will make progress beyond the state of the art by considering three categories of non-fundamental, i.e. grounded objects, and analysing in some detail their perspectival character: • so-called ‘lesser entities’ such as boundaries, holes, sounds, as well as ephemeral or ‘non-canonical’ objects of perception; • secondary qualities of external objects, such as colours, odours, tastes and textures, and so-called ‘response- dependent’ properties, such as normative, chromatic and locational properties; • ‘phenomenal’ objects (phaenomena) in Kant’s philosophy, under the dual-aspect interpretation of it. Lesser entities. Even though so-called ‘lesser entities’ have rarely been considered in their own right (but cf. Fine ; Chisholm ), important special cases have been studied in some detail. Inspired by the pioneering studies of Roderick Chisholm (, , c, a, , b, b, b, , , ) and Avrum Stroll (, , ), there have been a number of individual studies on shadows (Casati , , ), holes (Lewis & Lewis ; Casati & Varzi , , , ), boundaries (Hestevold ; Smith , ; Smith & Varzi ; Smith ), sound (Casati & Dokic (, ); Hamilton (); Nudds (, ); O’Callaghan (); O’Shaughnessy (); Pasnau (); Scruton () and the articles in Nudds & O’Callaghan ()) and a number of studies of non-substantial individuals in Aristotle (Corkum ) and in medieval philosophy (Erismann & Schniewind ). No attempt, however, has yet been made to characterise lesser entities in general and in terms of grounding. Understanding lesser entities in non-modal terms will help solve the conundrum of what Armstrong (: ,) calls “the doctrine of the ontological free lunch”: that lesser (in Armstrong’s terminology: supervenient) entities are grounded in (in Armstrong: necessitated by) their basis, and therefore, in one sense at least, are ‘nothing over and above’ them, while still being numerically different from them (Lewis (a), Campbell (), Oliver (), David (), Melia () and Schaffer () have voiced the obvious objection to Armstrong, that nothing can be both a free lunch and not exist; cf. also Cameron (b: ) and Cameron (a: )). A particularly promising route to the clarification of the ontological status of lesser entities has recently been sketched by Kit Fine (, b, a, b, , ) (cf. also Horwich ()andFine’sreply()), distinguishing between two concepts of ‘reality’, as “what is “objective” or “factual”” on the one side, and “what is “irreducible” or “fundamental”” on the other (Fine : ). This distinction not only allows for forms of reductionism which are not eliminativist, but also for the category of the ‘unreal’ (i.e. non-objective, non-factual) and irreducible, which allows for disagreement about ‘subjective’ qualities and objects which is not really about their objective and factual bases. Taking the distinction by Kit Fine () between two concepts of reality as my starting-point, I will provide a general characterisation of lesser entities, building on my earlier work on ‘qua-objects’. Lesser entities, I will argue, are constituted by others, they are manifestations or aspects of the latter, and in this sense ‘less substantial’ than them. This does not, contrary to what many philosophers seem to think, make them ‘less real’ or even non-existent. They do exist, and are the objects of disagreements about taste, epistemic possibility and future contingents. Secondary and response-dependent qualities. Another important distinction, rarely discussed nowadays (Nolan (b) is a notable exception), is the distinction between primary and secondary qualities. The impor- tance of this contrast for modern philosophy, in particular the thinking of Descartes and of Locke, can hardly be over-estimated (cf. Hüttemann ). It is at the heart of the scientific world-view, and motivated its first incor- poration, the mechanist philosophies of the th century. Despite its huge importance, the characterisation itself has proved surprisingly elusive (cf. Nolan a: ). What lies at the bottom of the distinction between properties such as extension, size, shape, motion and position on the one hand and colour, sound, taste, odour, heat, cold etc. on the other? It is often said that primary qualities are primary in the order of explanation (Smith ). The notion of ‘explanation’ in play here is of a distinctively ontological kind. Building on García-Carpintero (), the present project aims at answering this question in terms of grounding. Secondary qualities, the hypothesis to be explored says, are grounded in primary qualities – they are perspectival manifestations of the latter, part of reality, but not themselves fundamental. In particular, the present proposal aims at elucidating further the connections between secondary qualities and representation (Brooks ), in particular perspectival self-location (Egan a), perceptual content (Kulvicki ) and moral judgments (Wright ). The notion of response-dependence was introduced by Mark Johnston () as a generalisation of the notion of secondary qualities (cf. Menzies (); Casati & Tappolet (); Yates () for overviews of the debate). Johnston

(: ) characterises a predicate F as response-dependent “iff there are ‘substantial’ specifications of S, C and R such that it is necessary and a priori that, for all x,”x is F” is true iff x has the disposition to produce in all of S a mental response R under conditions C” (the requirement that the biconditional is necessary is not present in Johnston’s original criterion, but clearly needed, cf. also Pettit () and Wright () for similar characterisations). While much work has been done on the modal and epistemological status of such ‘basic equations’ (Smith & Stoljar ), often in connection with the so-called ‘missing explanation argument’ (cf. Johnston (: –), (: – ) and : – and the discussion in Blackburn (), López de Sa (), McFarland & Miller (), Menzies & Pettit (), Miller (, , ), cf. also the papers in Menzies () and Casati & Tappolet ()) and their rigidifications (cf. Haukioja : ), they have rarely been discussed in the context of metaphysical explanation (cf. Ásta [Ásta Kristjana Sveinsdóttir] () for an interesting first step). The present project attempts to do so, characterising response-dependent properties as a special kind of secondary, i.e. dependent qualities, namely those which are grounded in reactive properties of perceiving subjects. Noumenal and phenomenal objects. According to the so-called “dual aspect” interpretation of Kant’s tran- scendental idealism (cf. e.g. Abela (); Allais (); Allison (, ); Bird (); Collins (); Friebe (); Prauss (); Quarfood (); Walker (); Willaschek (); Rosefeldt () cites Schultz (, ) as a his- torical antecedent, Nicholas Stang cites Adickes ()), the Kantian dichotomy of phaenomena and noumena should be interpreted in terms of aspects, privileging one strand in Kant’s thinking over another (cf. e.g. Kant (: xxvii) vs. Kant (: , , )). While it has led some to talk of the whole Kantian system as a ‘system of perspectives’ (Palmquist ) and certain shed new light on Kant’s notion of “thing-in-itself” (cf. e.g. Dalbosco () for a thor- ough survey), the exact nature of Kantian aspects remains elusive. Recent work has emphasised the metaphysical nature of the distinction, has considered restricting the interpretation to objects, as opposed to subjects (Hanna ) and has applied it to Kant’s theory of the self (Di Maria ). The dual-aspect theory has been defended against its critics (cf. e.g. Guyer (); Falkenstein (); van Cleve (); Hogan (b, c, a)), distinguished from a dualism of descriptions, as in Buchdahl (, , , , ), and stripped from the epistemological interpre- tation it has in Bird, Prauss, Allison and Collins. But much work needs to be done (cf. Schulting () for a helpful overview). I would like to develop, in more detail, a variant of the dual-aspect theory, in its two-properties, rather than two-objects version. The property version of the dual aspect theory has been put forward in two versions, situating the crucial contrast either in the intrinsic/extrinsic (Langton ) or in the relational/non-relational di- vide (which, according to Rosefeldt (: ) may be found in a nutshell in Paton (: ff.) and Dryer (: ch. .)). Building on my earlier work on how the intrinsic/extrinsic and the relational/non-relational distinctions cross-cut, I will side with Bird (), Rosefeldt () and Allais () against Langton, taking into account not just the historical antecedents quoted above, but also very recent work by ‘systematic’ metaphysicians (cf. McDaniel ; Stang , ; Marshall a). In particular, I would like to connect the discussion about the ontological status of phaenomena with the theories of secondary and response-dependent qualities mentioned above, explicit links which have rarely been made (but cf. Pettit (), Allais (), Rosefeldt (), Goldberg (), Groff () and Hare (, )).

Chapter

Perceptual Intake

. Seeing Things

.. Perceptual Reports

From its very beginning, and very explicitly so from Descartes on, the philosophy of mind has been intertwined with epistemological concerns. Perception is important to us mostly because it is a source of knowledge, and the ‘inner’ is marked out by what we have some special type of access to (introspection, perhaps). From the perspective of a metaphysics of the mind, epistemological properties are just some of the properties we believe our minds, and their states, to have, and there is nothing especially important about them. Both historically and systematically, however, epistemology starts with perception, or at least the supposedly paradigm case of visual perception of visible properties of opaque, middle-sized, relatively immobile material things. “How do you know the stick is straight? – I have seen it, by my own eyes.” It does and did not take long to realise how problematic even this paradigm case is. Here are some of the questions that may be raised. What does the “it” refer to in the example given? What is the object of perception, what is it that we see? Many have thought it necessary to distinguish between at least the following two: (A) S sees that p. (B) S sees x. The supposed contrast between (A) and (B) has been used to introduce, motivate or explicate a certain number of theoretical distinctions: . intensional / extensional: While we may substitute coferential terms for each other within “x” in (B) salva veritate, we cannot do so normally within “p”: if S sees Hesperus, S thereby sees Phosphorus, while seeing that Hesperus is the brightest heavenly body in the sky is not the same as seeing that Phosphorus is the brightest heavenly body in the sky. . propositional / objectual: While (B) relates S (at least paradigmatically) to an object / individual / particular, (A), at least on some readings, relates S to the referent of “that p”, usually taken to be a proposition. . conceptual / non-conceptual: While propositions are normally taken to contain concepts and mastery of these may be taken to be a prerequisite for the truth of (A), a corresponding requirement seems absent for (B). . epistemic / non-epistemic: While (B), at least prima facie, seems direct and unmediated, (A) seems to involve some exercise of our epistemic faculties, involving something like a belief in a way (B) does not.

. Unless, of course, these introduce intensional contexts of their own, as would, e.g., a substitution of “that p” for “x”.

None of these uses of the contrast between (A) and (B) is unproblematic, however. What is more, the binary dichotomy itself is too simplistic. In addition to (A) and (B), we also have (C), (D), (E), (F), (G) and (H), and perhaps others as well: (B) I see the stick and it is straight. (A) I see that the stick is straight. (C) I see the straight stick. (D) I see the stick and its straightness. (E) I see the stick as straight. (F) I see the stick as if (it were) straight. (G) I see the stick being straight. (H) I see the stick’s being straight. At least prima facie, it is not clear that all these are analysable just in terms of either the ‘purely objectual’ (B) and the ‘purely propositional’ (A). Neither is it clear how the proposed dimensions of contrast classify these cases. Moreover, these different grammatical construals of perceptual reports leave it surprisingly open what we may legitimately take to be the object of the perception so reported. ‘Propositional’ seeing, as in (A), is not, or at least not only, a seeing of propositions. Even if propositions are perceptible, I also see the cat when I see that it is on the mat. If the proposition involved in (A) is not the primary object of the perceptual act, in what way is it involved at all? Saying that it is the content of that act is not explaining anything, but labelling the phenomenon we would like to have explained. The objectual sense also raises further questions: do we ever see the stick? Perceptually given to us, it might seem, is at best its surface, or rather the parts of its surface visible from a certain point. Do we see its back-side too, the other parts of its surface, its inside? After all, sticks with different backsides would look just the same to us. If we perceive what is not given to us, how come we do not acquire information about it? If we do not, on the other hand, then how do we know, on the basis of our perception, that sticks, e.g., are three-dimensional? As this brief and cursory survey of just some of the problems of perception reports makes clear, one of the main problem of contemporary theories of perception is to reconcile the relationality of perception, its being a way of making contact with the world, with its representational character, its being a certain take or perspective on the world.

.. The Argument from Illusion

When we see, we undergo certain physiological processes and we also have a visual experience. But we also have visual experiences when we are hallucinating (though, conventionally, we would not call them “seeings” or “per- ceptions”). Macbeth did not see or perceive a dagger before him, but instead he merely seemed to see one; he had an experience as of one or, perhaps, an experience that represented one. We can have visual experiences when we are not seeing. These are of two types:

. With respect to (), both parts of the claim may be disputed: does Oedipus, when he sees the stranger he then kills, really see his father? does the dog, when he sees that the cat ate its food, see the cat under some mode of presentation, in a way that would, in principle (if only it could speak and reason …), allow it to describe the food-eater as a cat? With respect to (), we should keep in mind that the proposition that p may be de-se, as in “S sees that he himself∗ is making a mess” using Castañeda’s ‘quasi-indicator’ as applied to Perry’s ‘messy shopper’. In these cases, the proposition does not have autonomous semantic content, and is not detachable – in the sense that “T saw that too” is made true by T (not S!) realising that s/he is making a mess. We also should not forget that “x” in (B) may refer to a fact, or another worldly item of propositional form. With respect to (), it may be questioned whether concept-mastery is required for all so-called ‘propositional attitudes’ (for example, it seems that you can regret that p without ‘having’ the concepts needed to understand that p); if they are not, then why require them for (A)? Conversely, it may be maintained that we need concepts to see artefacts, beginnings and endings, social events (the crash of the stockmarket) and things such as Maria’s regret, the apple’s being edible and the sky looking as if it will soon rain. With respect to (), we should not forget that in persistent visual illusions, we normally continue to have the visual impression that p (e.g. that one of the lines is shorter than the other in the Muller-Lyer illusion) even when we not only lack the belief but even know it to be false. On the other hand, seeing x may be taken to be itself an epistemic relation (‘acquaintance’). . (E), for example, seems to side with (A) on criteria () and (), being epistemic and conceptual, but rather with (B) on () and (), being extensional (in the position of “stick”, at least) and objectual.

• hallucinatory: We do not see anything, we merely seem to see something / have a visual experience as of something: there is nothing to perceive. • illusionary: We do see something, but see it to be in some other way than it really is. Some cases may be hard to qualify one way or other: do I see some region (wrongly) as filled when I hallucinate a pink elephant? Do I hallucinate the white wall’s redness when I see it in red light? The very existence of hallucinations and illusions may be questioned: is there ever nothing to perceive, do I ever see ‘more’ than how things appear to me? While I think these questions are pressing, I will follow orthodoxy in what follows and assume that illusions exist and are roughly as they are standardly described. If it is possible to perceive things differently from how they really are or even to seem to perceive things that are not there at all, how can perception give us “immediate and direct access” to the world? Crane (: ) gives the following version of the argument from illusion: . “When one is subject to an illusion, it seems to one that something has a quality, F, which the real ordinary object supposedly being perceived does not actually have. . When it seems to one that something has a quality, F, then there is something of which one is aware which does have this quality. . Since the real object in question is, by hypothesis, not-F, then it follows that in cases of illusion, either one is not aware of the real object after all, or if one is, one is aware of it only “indirectly” and not in the direct, unmediated way in which we normally take ourselves to be aware of objects. . There is no non-arbitrary way of distinguishing, from the point of view of the subject of an experience, between the phenomenology of perception and illusion […]. . Therefore there is no reason to suppose that even in the case of genuine perception one is directly or imme- diately aware of ordinary objects. . Therefore our normal view about what perceiving is – sometimes called “naïve realism” or “direct realism” – is false. So perception cannot be what we normally think it is.” It goes without saying that all premisses, as well as the inferences from (), (), (), () to () and from () to () have been challenged. Acceptance of the first inference may distinguish between two (families of) views about the relationship between the phenomenology of perceptual experiences and their relationality: Common factor theorists : If veridical experiences and hallucinatory experiences are the same in phenomenal character (what they are like to have), they must have the same type of objects (sense-datum theories and representationalism). Disjunctivists : Two visual experiences may be phenomenally and even introspectively indiscriminable and only one of them be veridical or even have an object at all. Another way of marking the same distinction is in terms of the direction of explanation of the ‘content’ of the non-veridical experience: common factor theorists take this content to be settled independently of the veridicality of the experience, while disjunctivists hold that the determination (and even the having) of the content by the non-veridical experience is irreducibly counterfactual, in terms of the content it would have if it were veridical. Common-factor theorists thus explain the good case as a complex of the bad case + whatever is needed for veridi- cality, while disjunctivists explain the bad case as ‘botched’, as something that would be a good case if only the world cooperated. With respect to perceptual beliefs, i.e. beliefs we acquire and are justified in acquiring in virtue of undergoing some perceptual experience, we distinguish two articulations of the problem of explaining how we sometimes, but only sometimes, acquire justification for (J) on the basis of (I): (I) I have the visual experience that p. (J) I know that p. Common-factor theorists seem committed to the claim that (I), because it relates me to the same object or to the same content as (A), justifies (J) in the same way as would (A). In the case of illusions and hallucinations, this is the wrong answer – how can it be avoided? The only way, it seems, is for them to deny, as in the case of (I)⊢(A), so

equally in the case of (A)⊢(J), that the transmission of justification is direct: some other conditions have to obtain. But what could they be? Sense-data theorists face this objection in the following form: if in visual experiences I am related to sense-data, then knowledge claims about ordinary mind-independent objects such as sticks can only be indirectly justified by them; the sense-data act as a ‘veil’, hiding the world from us. The best thing sense-data theorists can do, it seems, to distinguish (I) from (A) is by imposing accuracy conditions on the sense-data: a sense-datum is accurate if it stands in some representational isomorphism to the scene it is a sense-datum of, i.e. pictures it in the way accurate pictures do. (I) can then be said to transmit justification to (J) iff the sense-datum in question is accurate. Under what conditions, however, is a sense-datum accurate? Suppose I am looking at a tree and would be able, given enough time and patience, to visually discriminate exactly ’ leaves, each one suitably contrasting against its background. Instead of staring at the tree, I could take a high-resolution photograph and examine it in my study: it would be accurate only if it depicted exactly ’ leaves. When I am not counting them, however, and even though I may be said to see that the tree has more than leaves and less than ’, it is just wrong to say that I see that the tree has ’ leaves. What I see is that the tree has many leaves, where this leaves it open how many exactly it has. The problem is that the sense-datum, whatever it is, cannot match this indeterminacy (Armstrong : ). If it is accurate, it has to depict the right number of leaves, but then it can appear to be the way the tree appears to be only if the sense-datum appears to be differently than it is. This is the answer Jackson () gives to Armstrong, but it is problematic. If sense-data have a ‘hidden nature’ in virtue of which they can appear differently from how they are, would not then another argument from illusion (about the properties of sense-data) show that another veil has to be introduced, between us and our own sense-data, thus starting a regress? At root the same objection can be put to representationalists as follows: It may seem an advantage of represen- tationalism that it explains the content of the knowledge claim (J) (i.e. that it is a claim to know that p) on the grounds that (J) ‘inherits’ its content from the visual experience: the visual experience ‘transmits’ its content to the knowledge claim and thereby justifies it. On closer inspection, however, things are not so simple. The lin- guistic context for “p” in (J), as we have seen, is intensional; insofar as (at least some types of) knowledge require concept-possession, it is even hyperintensional: I may know that Sam is an eye-doctor without knowing that he is an ophtalmologist. Even if perception has conceptual content, however, it seems implausible to take it to be so finely grained. Suppose I recognise his profession by some external marks, in what would a perception that he is an eye-doctor differ from a perception that he is an ophtalmologist? If the content is not ‘handed down’ entirely from the perception to the knowledge claim, then the way it changes must be determined by the change in ‘mode’: because the ‘modes’ of (A) and (J) are different, some rules of transition have to be specified, and they will be different in the (I) to (J) and the (A) to (J) cases. Disjunctivists also face a similar problem. While they may hold that only (A), but not (I) transmits justification to (J), they also need a story to back this up. What is it about the subjectively indiscriminable difference between cases where only (I), but not (A) is true and cases where both are true that makes (J) turn on it? The beginning of a plausible answer is to take the veridicality of perception to be ‘primitive’ and instead of explaining (A) as ‘(J)-plus’, explain (J) as ‘(A)-minus’. One way of doing this is in terms of success conditions: a perceptual experience is successful iff it is a perception, and only successful perceptual experiences ground knowledge claims. Requiring truth for success, however, is characteristic not just of perception, but also of belief and, to some extent to be determined later, of emotions as well. The central explanandum is thus the relation that is required to hold between the state reported by (A) (but not every state reportable by (J)) and the state that is claimed to obtain by (I): a relation of grounding and justification I will call the “basing relation” in what follows.

.. The Intentionality of Perception

Let us consider one version of the argument from illusion: A stick in front of me looks bent. A stick in front of you looks straight. There is at most one stick in front of us.

No stick can be both bent and straight. ∴ In at least one case, one of our perceptual experiences is non-veridical. In the non-veridical perceptual experience, one is presented with a mind-dependent intentional object. Both perceptual experiences have the same type of intentional object. ∴ The intentional object of both experiences is mind-dependent. Sticks are not mind-dependent. ∴ In neither case do we see the stick. Hence nothing that may look to have contrary properties is ever seen – a reductio. While most attention has been focussed on denying () or () or both, I would like to focus attention on the description of the situation, i.e. on () and (). My main motivation is that already () is implausible: while I may certainly form false beliefs on the basis of one of my perceptual experiences, they themselves are not necessarily misleading: at least on one sense of “looks”, the straight stick half-immersed in water does look straight, i.e. looks as straight sticks half-immersed in water do. How else should the stick from here? If the stick were to look to me from here like a straight stick in air does, I would rightly conclude that it is bent, and do so on the basis of how it looks to me from here. Perspectivality is not illusionary, but a real fact. But let us first look at the other premisses and transitions. Austin denies that () and () are compossible. Irrealists deny (), and may therefore also deny (). () appears vulnerable, but depending on how we understand ‘intentional object’ and ‘mind-dependent’, it allows of uncontroversial interpretations. () is denied by disjunctivists. The phenomenal principle, or something like it, may be appealed to justify (), by deriving it from (), (), () and (): A stick in front of me looks bent. ∴ There is something of which I am aware that is bent. A stick in front of me looks straight. ∴ There is something of which I am aware that is straight. There is at most one stick in front of me. No stick can be both bent and straight. There is a most one thing of which I am aware in the two cases. ∴ In at least one case, one of our perceptual experiences is non-veridical. This is not a particularly good argument, however: the conclusion () is too strong (as it incorporates already a version of ()), and the additional premise () is hard to justify. Themostpromisingwayout, Ithink, istoblockthedeductionof(), whilepreservingasenseinwhichboth()and() may be true. A first option is to attack (). One may hold that the properties in question (being bent, being straight) are not (parts of) the intentional object of these perceptions, but rather of their contents, conceived of as modes of presentations of the object: the stick is presented in the straight-mode or in the bent-mode. Alternatively, one may hold that being presented in the straight- (or bent-)mode are not representational at all, but rather sensational properties, qualia of the experiences themselves. The problem with these views is that do not address the real problem: even if we grant that bentness may be a way the stick is presented to me or a qualitative property of my experience, it is certainly also a property of the stick: in one of the two cases at least, the roundness present in my experience is ‘matched’ by a roundness of the stick – and we have to find a ground for saying that this is not the case for bentness. Adverbialists reinterpret () and () as i I am appeared to bent-ly. i I am appeared to straight-ly. Even though we do not then have analogues of () and (), () still seems to follow: If we are appeared to bent-ly and straight-ly, then in at least one case, the perceptual experience is non-veridical. We also have trouble explaining why (), or at least some version of it, belongs to a exhaustive description of the situation: to see this, compare the scenario with what changes when we take the stick out of the water: even if we keep our respective perspectives on the stick, an important source of conflict and disagreement disappears. This source, whatever it is, is not captured by (i) and (i). In this second, but not the first respect, representational views fare better:

ii I see the stick as bent. ii I see the stick as straight. Insofar as bentness and straightness may figure in contents (as represented properties, rather than as being exem- plified by things as sticks), we do account for the conflict noted in (). Given () and (), however, it still follows that in at least one case one of us sees the stick as F where the stick is not F – so () still follows. The sense data theorist reinterprets () and () as iii It appears to me: something is bent*. iii It appears to me: something is straight*. Bentness* and straightness* are not properties of sticks, but of sense-data or of experiences: if they can both be had by the same sense-datum, then we do not have an analogue of (); if they cannot, then we do not have an analogue of (). The victory is Pyrrhonic, however, because even if they can perhaps provide a reading of “seeing” on which it is true that I see the stick, I will not on their view see its bentness or straightness. Perspectival realists also deny () by reinterpreting () and (). Their version is: iv The stick appears-bent-to-me. iv The stick appears-straight-to-me. This, I think is part of the right response to the argument. It combines the adverbial account of the (allegedly) contrary properties with the introduction of appearance properties – properties like being bent-to-me, or rather: being-bent-from-here, the relativisation encapsulating the information about the position and perceptual apparatus of the observer and the information about the perceptual milieu we would need to appeal to correctly predict from what other actual or counterfactual experiences the perceptual experience in question would be subjectively indiscriminable. Bent-to-me and straight-to-me are not incompatible, so we do not have an analogue of (); but neither are they proper- ties of experiences, nor of sense-data, nor of points of views or observers. Are they properties of sticks? In a sense yes, but also in another sense no. The property the half-immersed stick appears to have from here, i.e. the one I can correctly describe by painting a inflected line, is a property the stick no longer appears to have once I have taken it out of the water. It is very well possible, however, that the half-immersed stick did not to you appear to have this property, e.g. if you were looking at it from another angle it may have looked straight to you, even when half-immersed in water. When we took the stick out, it started to appear differently to me, but not to you. While bent-to-me certainly was a property the stick appeared (to me) to have, it was not so for you and is now not so for me. When we immersed it in water, there was a look the stick acquired, a look it did not have before: by changing the perceptual milieu and the set of available perspectives on the stick, the stick acquired a new look it did not have before, a look I describe ascribing the perspectival property bent-to-you and you describe ascribing the different, but not incompatible property straight-to-me. Objects of perception are things from a certain point of view, i.e. perspectival facts or ‘looks’. The points of view accounting for their perspectivality are not the perceptual acts, nor are they modes of presentation of the object (if there is one): they are located on the ‘object side’ as it were. Take, to change the example, the round coin lying on the table in front of me, looking, at we would say, oval from here. The same coin, it seems, can look round from certain points of view and elliptical from others. As before, round a from certain point of view and elliptical from another point of view are not incompatible properties, so the coin does not look to have incompatible properties. The cup always appears round from here or there. To appear round from here means that the cup appears to be at a certain distance and orientation from the given point of view. Orientation is reducible to relative distance of parts: the orientation of the cup changes from a point of view to another iff the ratio of distances between each part and the point of view changes. For instance, from some points of view (right above of the cup), all the parts of the edge are at equal distance from the point of view; while for some other points of view (from the side of the cup), there are parts of the edge that are closer to the point of view than other ones. These ratios, however, should not be confused with the ratios among the parts of the edge themselves: they are all at the same distance of each other, and appear to be so in the perceptual experience. Ellipses then are just artifacts from painters that project three-dimensional

oriented objects on two dimensional vertically oriented screens. When we adopt the attitude of the painter, we create an illusion from a perception: we change the orientation and shape of the perceptual objects. We change the relative distances of the edges of the cup to the point of view, thereby creating the illusion of having an elliptical object in view. These perspectival appearance properties are thus not features of perspective-independent things, but rather of appearances, appearances of perspective-independent things, to be sure, but not things that are themselves inde- pendent of perspectives. Kant is right in thinking that the perception of appearance properties is the perception of appearances, where such appearances are grounded in the things they are appearances of. Objects of perception are things-from-a-certain-point-of-view, i.e. perspectival facts. The points of view accounting for their perspecti- vality are not the perceptual acts, nor are they modes of presentation of the object (if there are any such things): they are located on the ‘object side’ as it were. It is the appearance of the stick half-immersed in water that looks bent-from-here and the same appearance that may look straight-from-here with respect to some other location or observer. It is the change in appearance that is responsible for the fact that once the stick is taken out of the water and seen by both of us in clear air, it may no longer be said to appear bent in any sense. Because appearances are grounded in the things appearing, together with their environment, they are appearances of things – things with which we are in perceptual contact when we perceive the appearances. Can we thus say that the things are only indirectly seen, through their appearances as it were? We should not, however, in the way of sense-data theorists, take these appearances to be the direct objects of perception – rather, it is their grounds that are seen through them. SAY MORE ABOUT THIS! This explanation of appearance properties thus takes us to v The stick-as-seen-by-us appears-bent-to-me. v The stick-as-seen-by-us appears-straight-to-you. We thus have two perceptions, but still just one stick appearing to both of us. () does not follow if the two ap- pearances really have their appearance properties. Such a reply to the argument from illusion, I surmise, is not only independently plausible, but also generalises to the analoguous arguments by McTaggart against temporal A-determinations and and by Frank Jackson in favour of perceptual qualia. Perspectival facts are not sense data, but are ‘out there’, full citizens of the mind-independent external world. They are not parts or aspects of the experiencing subject and they exist independently of experiencing subjects. They are perspectival only in that they contain a perspectivally ‘modulated’ property, ‘modulated’ by its being relational with respect to a point of view. A point of view is just a point in space (compare Atkin ()), where a (point-sized) seeing eye can be located, but that can as well remain unoccupied. Actual points of view do not imply actual views. Points of view are not acts of perception, nor part of the act of perception: they do not depend on perceiving objects. Before drawing such morals, however, we have to know more about sticks-as-seen-by-us, i.e. appearances of sticks. When I say that the wine tastes sweet, I am qualifying the taste as sweet, and only secondarily the wine as having a sweet taste. On one reading, the first qualification is not subject to doubt: in this sense, I am qualifying the taste as tasting-sweet-to-me, something which you cannot dispute. Sometimes, however, I make a stronger claim, not only that the taste is sweet-to-me, but that it is sweet tout court, and so that you should taste it as sweet as well, if you are tasting properly. Here we have conceptual space for taste courses and experts, judges and faultless disagreement. Another dimension of possible disagreement opens up with the secondary qualification: I may be wrong that it is the wine that tastes sweet. This suggests that the ‘extension’ of my assertion is via a causal link: I am saying of the taste that it is sweet (and there cannot be doubt about this) and of the wine that it has that taste, i.e. causally produced it (and about this I could be wrong):

. The “primary”/“secondary” contrast is from Martin (: ).

For to say that a certain experience is a seeing that something is the case, is to do more than describe the experience. It is to characterize it as, so to speak, making an assertion of claim, and – which is the point I wish to stress – to endorse that claim. (Sellars : )

Things stand differently with colours, or rather with chromatic profiles:

“What puts vision apart from hearing, smell, and taste is that we do not conceive of the visible world as offering us objects of visual awareness and attention distinct from (but coincident with) the concrete objects that we also see.” (Martin : ), citing O’Shaughnessy (: –)

The chromatic profile of a coloured thing (its visual appearance, its look) is more closely tied to it than its taste or its visual shape: the being-the-appearance-of relation here is not causal, but constitutive. Looks are nothing but looks of things, i.e. things looking a certain way. In other terms: for a to have a visual look that appears red is both, for the look to be red and for the thing itself to appear red. Two possible sources of error present in the case of taste are thereby eliminated: it is not possible that only its look, but not the thing looking a certain way appears red, and the look itself cannot just appear red but fail to be red. What Mary gets newly acquainted with, when she gets out of her black-and-white room, are not new properties, but new things: the looks things have when they appear to you e.g. in clear daylight. Being an eminent colour scientist, she knew everything there is to know about colours; being equally an eminent psychologist of colour perception, she also knew everything there is to know about visual appearance properties, properties such as looking red from here. What she did not know, however, are an important class of things, visual looks – ie. the things that primarily have the colours and in virtue of which coloured things appear the way they do. Instead of a The look-of-the-patch-in-this-light looks-red-to-me. a The look-of-the-patch-in-this-light looks-green-to-you. we thus have the simpler: b The look-of-the-patch-in-this-light is red. b The look-of-the-patch-in-this-light is green. MORE ON APPARENT-COLOUR PROPERTIES WHAT IS IT ABOUT THEM THAT IS INTROSPECTIVELY GIVEN? H is a perfect hallucination and has a phenomenal character and H is subjectively indiscriminable from a percep- tion P, but differing from P in phenomenal character (as per relationalism). You do not have to deny infallibility of introspection in the favourable circumstances (and deny what Locatelli calls ‘superficiality’), ie you do not have to claim how H is ≠ how H introspectively seems for we are in the presence of: how H introspectively seems = how H introspectively is how H is = how H introspectively is () and () can be granted, ie () is fine how H is = how H introspectively seems But what we do not need to grant is that this means that H cannot differ in phenomenal character from a subjec- tively indiscriminable perception. So we have that () and () are compatible in the presence of (): how H is = how H introspectively seems H and P cannot be distinguished introspectively but have different phenomenal characters. a how P introspectively seems = how H introspectively seems

. The switch to chromatic profiles is needed to preserve Aristotle’s claim that they are the proper sensibles of sight: we also see black and white things, transparent things and things that do not have surfaces, such as shadows, holes, holograms and rainbows.

b how P is ≠ how H is how P is ≠ how P introspectively seems So what P is subjectively indiscriminable from does not exhaust its phenomenal character. This is compatible with introspection being infallible with respect to P too: how P introspectively is = how P introspectively seems but it requires that it has a broader nature than what is available through introspection (eg, to what hallucinations it is introspectively indiscriminable from).

. Aristotle on Perception and Colours

.. Uptake of Form

Perception, for Aristotle, is a natural phenomenon and as such its occurrence does not require an explanation: it occurs when and because a sense-organ is altered, and a potentiality is actualised. For Aristotle, there is no additional story to tell: perception just is this alteration of the sense-organ. We may, however, ask about its causes: Perception (aisthēsis), an alteration of the perceiver (De An., II.), is the effect of being affected by an external object, which is actually so as the perceiver (or the respective sense-organ, cf. Hamlyn (: , ad a)), before being affected by it, is potentially. The things initiating the causal process of affection which is the perceiving are the proper sensibles (colours for sight, sounds for audition, flavours for taste, smells for olfaction) and, derivatively, so-called accidental unities (such as Darius-qua-white) involving them. Each sense is a potentiality / capacity / power; its essence is “naturally relative” to its objects, its proper sensibles, which (may) exist before the sense in question exists (Cat. a-), for “perception is hardly of itself – but there must be something else beyond the perception, and this must be prior to the perception; since what brings about change is naturally prior to what is changed, even if they are spoken of in relation to each other” (Met. Γ., b-a). Sight, for example, is thus posterior to its proper sensibles, the colours. In De Anima II., Aristotle says that a sense (aisthēsis: sight, hearing, smelling, taste, touch), as distinct of its sense organ (the eyes, the ears, the nose, the tongue, the skin), receives form without matter:

Καθόλου δὲ περὶ πάσης αἰσθήσεως δεῖ λαβεῖν In general, with regard to all sense- For that which perceives must be a particular ὅτι ἡ μὲν αἴσθησίς ἐστι τὸ δεκτικὸν τῶν αἰ- perceptionA, we must take it that the senseA extended magnitude, while what it is to be σθητῶν εἰδῶν ἄνευ τῆς ὕλης, οἷον ὁ κηρὸς τοῦ is that which can receive perceptible forms able to perceive and the senseA are surely not δακτυλίου ἄνευ τοῦ σιδήρου καὶ τοῦ χρυσοῦ δέ- without their matter, as wax receives the im- magnitudes, but rather a certain principleL χεται τὸ σημεῖον, λαμβάνει δὲ τὸ χρυσοῦν ἢ print of the ring without the iron or gold, and and potentiality of that thing. (Aristotle : τὸ χαλκοῦν σημεῖον, ἀλλ’ οὐχ ᾗ χρυσὸς ἢ χαλ- it takes the imprint which is of gold or bronze, –) κός· ὁμοίως δὲ καὶ ἡ αἴσθησις ἑκάστου ὑπὸ but not qua gold or bronze. Similarly too in τοῦ ἔχοντος χρῶμα ἢ χυμὸν ἢ ψόφον πάσχει, each case the senseA is affected by that which ἀλλ’ οὐχ ᾗ ἕκαστον ἐκείνων λέγεται, ἀλλ’ᾗ has colour or flavour or sound, but by these τοιονδί, καὶ κατὰ τὸν λόγον. αἰσθητήριον δὲ not in so far as they are what each of them is πρῶτον ἐν ᾧ ἡ τοιαύτη δύναμις. ἔστι μὲν οὖν spoken of as being, but in so far as they are ταὐτόν, τὸ δ’ εἶναι ἕτερον· μέγεθος μὲν γὰρ things of a certain kind and in accordance ἄν τι εἴη τὸ αἰσθανόμενον, οὐ μὴν τό γε αἰ- with their principleL. The primary sense- σθητικῷ εἶναι οὐδ’ ἡ αἴσθησις μέγεθός ἐστιν, organ is that in which such a potentiality ἀλλὰ λόγος τις καὶ δύναμις ἐκείνου. (a- resides. These are then the same, although ) what is is for them to be such is not the same.

. Psychology is part of biology because all affections of the soul (psychē) involve the body. Aristotle may be characterised as a physicalist using a much broader notion of ‘physical’ as is customary today (cf. Burnyeat (: ), but cf. Shields (: ) for some reservations). . It is important to note that Aristotle’s question is not how perception can come about, or how it can provide us with a route to the outside world. We start with an eye, which is already a part of the natural world, and note that its function, which it has essentially, is sight, i.e. the uptake of colours. That (veridical!) perception occurs is part of the data. This reductive aspect of Aristotle’s theory is rightly stressed by Everson (: ): “An eye does not see in virtue of undergoing any other change – rather it sees just because its capacity for vision is activated by an object of vision.” . In his “Notes on the Translation”, Hamlyn (: xvii) says that he uses the subscript “A” to flag occurrences of aisthēsis, which “may be translated variously as ‘sense’, ‘perception’, perhaps ‘sensation’, etc., and is sometimes used even to refer to the sense-organ. . According to his “Notes on the Translation”, any word subscripted with “L” translates “logos” (: xvii).

It is necessary to grasp, concerning the whole way, perception is also in each case affected gan and this potentiality are, then, the same, of perception generally, that perception is by what has the colour or taste or sound, but though their being is different. For what does what is capable of receiving perceptible forms not insofar as each of these is said to be some- the perceiving is a certain magnitude; nev- without the matter, as wax receives the seal thing, but rather insofar as each is of a certain ertheless being capable of perception is not; of a signet ring without the iron or gold. It quality, and corresponding to its proportion. nor is perception a magnitude, but is rather a acquires the golden or the metallic seal, but The primary sense organ is that in which certain proportion and a potentiality of that not insofar as it is gold or metal. In a similar this sort of potentiality resides. The sense or- thing. (Aristotle : –)

Two features of the more general Aristotelian context make this claim less surprising that it otherwise could be: First, we have here a straightforward application of the more general hylomorphic account of change. Because the reception of the sensible form (e.g. the colour, the sound) by the animal is an alteration of the animal (alloiōsis)(De An., II., b), it is, like any alteration, the acquiring of a new form by some underlying thing. The underlying thing here is either the sense-organ or the sense, i.e. its capacity to sense, and the acquired form is said to be perceptible and what is ascribed to the perceptual objects kata ton logon, i.e. as a form. Second, that perception and all forms of knowledge more generally are ‘of the like by the like’ is an endoxon, a plausible opinion put forward by some of Aristotle’s predecessors and as such discussed in the first book of the De Anima. When Socrates acquires the capacity to be musical, i.e. by learning to play the flute, Aristotle describes this change thus in the Physics .:

ἔστι γὰρ γίγνεσθαι ἄνθρωπον μουσικόν, ἔστι Wecansaythemanbecomesmusical, orwhat A man can come to be knowing music, and δὲ τὸ μὴ μουσικὸν γίγνεσθαι μουσικὸν ἢ τὸν is not-musical becomes musical, or the not- also the not knowing music can come to be (190a.) μὴ μουσικὸν ἄνθρωπον ἄνθρωπον musical man becomes a musical man. Now knowing music, or the not knowing music μουσικόν. ἁπλοῦν μὲν οὖν λέγω τὸ γιγνόμε- what becomes in the first two cases – man and man a man knowing music. I call the man νον τὸν ἄνθρωπον καὶ τὸ μὴ μουσικόν, καὶ ὃ not-musical – I call simple, and what each and the not knowing music simple coming- γίγνεται ἁπλοῦν, τὸ μουσικόν· συγκείμενον δὲ becomes –musical – simple also. But when to-be things, and the knowing music a sim- καὶ ὃ γίγνεται καὶ τὸ γιγνόμενον, ὅταν τὸν μὴ we say the not-musical man becomes a musi- ple thing which comes to be. When we say μουσικὸν ἄνθρωπον φῶμεν γίγνεσθαι μουσι- cal man, both what becomes and what it be- that the not knowing music man comes to be κὸν ἄνθρωπον. (b-a) comes are complex.(Aristotle : –) a knowing music man, both the coming-to-be thing and that which comes to be are com- pound. (Aristotle : )

When Socrates becomes musical (by, e.g., learning how to play the flute), we have three transitions, which together constitute the change: . matter from anthropon to musikon: a man, who is potentially musical and has the capacity to become a musician (to musikon, becomes a musician; a musician comes to be from a man; a change happens to a man: he becomes a musician; . form from mē musikon to musikon: musicality or musical ability (to musikon) comes to be, replaces non-musicality or absence, mere potentiality, of musical ability; . compound from mē musikon anthropon to musikon anthropon: a musical man (to anthropon musikon) comes to be, a non-musical man goes out of existence. Transitions of type () are changes out of something: musical comes out of non-musical. But changes of type () and () are not: it is the man who becomes a musician and the non-musical man who becomes a musical man, but we cannot say that the musician comes out of the man, nor can we say that the musical man comes out of the non-musical man. () and () are distinguished by the fact that in (), but not in (), the ‘coming-to-be thing’ (i.e. the thing that is changing) ‘remains’. What we designated by “anthropon” at the beginning of the change is still there at the end of it, and can now also be designated by “musikon”; the lack of musical ability in virtue of which we applied “ignorant of music”, however, is no longer there. The change reported in () is thus shown to have two aspects: one of constancy, exhibited in (), where one and the same thing persists through change and acquires a new quality; but also one of variation, exhibited in (), where one thing (ignorance of music) is replaced by something else (musicality) which comes out of it. It is a change not

only of coming to be, but of coming to be-such-and-such, i.e. exhibits not just that the change in question is a substantial change (as the other two), but also that it is a qualitative change. Aristotle continues by saying that this underlying thing is ‘one in number’, but not ‘one in form’:

οὐ γὰρ ταὐτὸν τὸ ἀνθρώπῳ καὶ τὸ ἀμούσῳ εἶ- For to be a man is not the same as to be un- The being of a man is not the same as the ναι. καὶ τὸ μὲν ὑπομένει, τὸ δ' οὐχ ὑπομένει· musical. One part survives, the other does being of ignorant of music[; a]nd the one re- τὸ μὲν μὴ ἀντικείμενον ὑπομένει (ὁ γὰρ ἄν- not: what is not an opposite survives (for the mains and the other does not. That which θρωπος ὑπομένει), τὸ μὴ μουσικὸν δὲ καὶ τὸ man survives), but not-musical or unmusical is not opposed remains – the man remains – ἄμουσον οὐχ ὑπομένει, οὐδὲ τὸ ἐξ ἀμφοῖν συγ- does not survive, nor does the compound of but the not knowing music and the ignorant κείμενον, οἷον ὁ ἄμουσος ἄνθρωπος. (a- the two, namely the unmusical man. (Aristo- of music do not remain, and neither does the ) tle : ) compound of the two, the ignorant of music man. (Aristotle : )

The underlying thing is ‘one in number’ (i.e. numerically one, one in reality) because the change can be completely characterised by (), where nothing goes out of existence. It is ‘two in form’ or (Aristotle says: equivalently) ‘two in account’ because the change is between opposites, as nothing is preserved in (). The result of the change in () is complex because it is one thing to be a man and another thing to know music. It is, however, still one thing that results from the change and one thing that enters into it, because being a man and being ignorant of music (or, after the change, being musical) are one “in reality” or “in fact”. We cannot presume, it seems to me, that the acquired form, to musikon, is what we would ordinarily think of as a property – what Aristotelian forms are should be left open at this stage of our inquiry. Whatever they are, however, they explain perception, i.e. the change in the sense: sight passes from being po- tentially red to being actually red by the acquisition of a new form, the red, by the underlying matter, i.e. the eye. Psychological states generally have a hylomorphic structure: they are either “enmattered forms” (logoi enuloi) or “forms in matter” (logoi en hulē) (a), depending on the reading of the ms. (cf. Shields : ). Such affections (passiones, ta pathē) allow for two different descriptions:

διαφερόντως δ’ ἂν ὁρίσαιντο ὁ φυσικὸς [τε] But the student of nature and the dialecti- The natural scientist and the dialectician καὶ ὁ διαλεκτικὸς ἕκαστον αὐτῶν, οἷον ὀργὴ cian would define each of these differently, would define each of these affections differ- τί ἐστιν· ὁ μὲν γὰρ ὄρεξιν ἀντιλυπήσεως ἤ τι e.g. what anger is. For the latter would de- ently, for example, what anger is. The dialec- τοιοῦτον, ὁ δὲ ζέσιν τοῦ περὶ καρδίαν αἵματος fine it as a desire for retaliation or something tician will define it as desire for retaliation, λυπήσεως ἤ τι τοιοῦτον, ὁ δὲ ζέσιν τοῦ περὶ of the sort, the former as the boiling of the or something of this sort, while the natural καρδίαν αἵματος καὶ θερμοῦ. τούτων δὲ ὁ μὲν blood and hot stuff round the heart. Of these, scientist will define it as boiling of the blood τὴν ὕλην ἀποδίδωσιν, ὁ δὲ τὸ εἶδος καὶ τὸν the one gives the matter, the other the form and heat around the heart. Of these, one de- λόγον. ὁ μὲν γὰρ λόγος ὅδε τοῦ πράγματος, and principleL. For this is the principleL of scribes the matter and the other the form and ἀνάγκη δ’ εἶναι τοῦτον ἐν ὕλῃ τοιᾳδί, εἰ ἔσται· the thing, but it must be in a matter of such the account. For this is the account of the (a-b) and such a kind if it is to be. (Aristotle : thing, but it is necessary that it be in matter –) of this sort if it is to exist. (Aristotle : )

Whether or not Aristotle claims at the end of De An. I. that all psychological states involve matter, he certainly claims that some do, e.g. anger. Anger is given a hylomorphic structure: it is the boiling of the blood (material cause) and for retaliation, and hence a type of desire (final cause). Generalising this account to perception, we may say that of some perception of red, sight is the matter and the red the form. The matter is not the eye, the sense-organ, but its capacity, sight. Whatever it is that ‘receives’ the form, is ‘acted upon’ and ‘becomes isomorphic’, it certainly is not the sense-organ (the aisthêtêrion), but it’s capacity or potential, i.e. the faculty of perception (the aisthêtikon) (cf. a- quoted above). In my perception of the red of a tomato, say, my eye realises its potentiality to ‘receive’ the sensible form of the tomato: with respect to this

. Can we speak of (), () and () as three changes, jointly ‘making up’ the change of Socrates’ becoming musical? If so, Aristotle would be saying that substantial change is prior to qualitative change in the following sense: every qualitative change is ‘composed’ or ‘made’ out of simpler existential changes (which shouldn’t be called “substantial”, because they may involve only forms, as does ()). . Shields (: xxxiv) makes this quite clear: “Form reception results in perception only when it occurs in a living being enodwed with the capacity of perception (aisthêsis).”

potentiality, it is affected by the tomato insofar as the latter has a certain sensible form, even if, with respect to its matter, i.e. to what has that potentiality,it is affected by the tomato insofar as the latter has a certain matter. Matter affects the eye, form the eye’s potentiality, sight. In a process of perceiving, sight is the matter and what is perceived is the form. This is compatible, however, with sight being the form of the eye, which is its matter:

ὁ δ’ ὀφθαλμὸς ὕλη ὄψεως, ἧς ἀπολειπούσης The eye is matter for sight, and if this fails it is The eye is the matter of sight; if sight is lost, οὐκέτ’ ὀφθαλμός, πλὴν ὁμωνύμως, καθάπερ no longer an eye, except homonymously, just it is no longer an eye, except homonymously, ὁ λίθινος καὶ ὁ γεγραμμένος. (b-) likeaneyeinstoneorapaintedeye. (Aristotle in the way that a stone eye or painted eye is. : ) (Aristotle : )

Aristotle does not just say there that the stone eye is not an eye though it may appear to be one (in the way in which a false friend is not a friend nor a fake promise a promise), but also that the stone eye is structurally, with respect to its form, ‘almost’ an eye, except that it does not have the kind of matter it needs. We thus have at least two hylomorphic layers: an eye is a pupil that sees (pupil+sight), and a seeing is sight that has a perceptible form (sight+perceptible), in the same way in which Socrates is a body that lives (body+soul) and his life is his soul that knows (soul+knowledge). It is thus fitting that Aristotle begins his “new start” at the beginning of book II with a general statement of hylomorphism:

λέγομεν δὴ γένος ἕν τι τῶν ὄντων τὴν οὐσίαν, Now we speak of one particular kind of exis- We say that among the things that exist one ταύτης δὲ τὸ μέν, ὡς ὕλην, ὃ καθ’ αὑτὸ οὐκ tent thing as substance, and under this head- kind is substance, and that one sort is sub- ἔστι τόδε τι, ἕτερον δὲ μορφὴν καὶ εἶδος, καθ’ ingwesospeakofonething qua matter, which stance as matter, which is not in its own right ἣν ἤδη λέγεται τόδε τι, καὶ τρίτον τὸ ἐκ τού- in itself is not a particular, another qua shape some this; another is shape and form, in ac- των. ἔστι δ’ ἡ μὲν ὕλη δύναμις, τὸ δ’ εἶδος ἐντε- and form, in virtue of which it is then spoken cordance with which it is already called some λέχεια, καὶ τοῦτο διχῶς, τὸ μὲν ὡς ἐπιστήμη, of as a particular, and a third qua the product this; and the third is what comes from these. τὸ δ’ ὡς τὸ θεωρεῖν. (a-) of these two. And matter is potentiality, while Matter is potentiality, while form is actuality; form is actuality – and that in two ways, first and actuality is spoken of in two ways, first as as knowledge is, and second as contemplating knowledge is, and second as contemplating is. is. (Aristotle : ) (Aristotle : )

Aristotle compares what is added to knowledge over contemplation to the difference between sleeping and wak- ing: unexercised knowledge is “prior in generation” to its exercise in contemplation; contemplation is the second, knowledge the first actuality of a rational animal. The latter is the first actuality of the potentiality had by some- one who has a capacity for knowledge, but does not yet know (nor, a fortiori, exercise that knowledge). The soul is not just an actuality, i.e. a form, but a ‘first actuality’ of this type:

διὸ ἡ ψυχή ἐστιν ἐντελέχεια ἡ πρώτη σώματος Hence the soul is the first actuality of a natu- Hence, the soul is the first actuality of a nat- φυσικοῦ δυνάμει ζωὴν ἔχοντος. (a-) ral body which has life potentially. (Aristotle ural body which has life in potentiality. (Aris- : ) totle : )

A body that has life in potentiality is a body which has organs, functionally determined parts the functions of which are to sustain what makes the body natural, i.e its inner principle of change – its soul. It is because soul and body are related as form and matter and as actuality and potentiality, that the question of their distinctness does not arise:

διὸ καὶ οὐ δεῖ ζητεῖν εἰ ἓν ἡ ψυχὴ καὶ τὸ σῶμα, Hence too we should not ask whether the soul For this reason it is also unnecessary to inquire ὥσπερ οὐδὲ τὸν κηρὸν καὶ τὸ σχῆμα, οὐδ’ ὅλως and body are one, any more than whether the whether soul and body are one, just as it is un- τὴν ἑκάστου ὕλην καὶ τὸ οὗ ἡ ὕλη· τὸ γὰρ ἓν wax and the impression are one, or in general necessary to ask this concerning the wax and καὶ τὸ εἶναι ἐπεὶ πλεοναχῶς λέγεται, τὸ κυ- whether the matter of each thing and that of the shape, nor generally concerning the mat- ρίως ἡ ἐντελέχειά ἐστιν. (b-) which it is the matter are one. For, while unity ter of each thing and that of which it is the and being are so spoken of in many ways, that matter. For while one and being are spoken which is most properly so spoken of is the ac- of in several ways, what is properly so spoken tuality. (Aristotle : ) of is the actuality. (Aristotle : ) . We will later (on p. in sct. ..) come back to this important conception of autonomous causation in Aristotle.

If we call some natural thing, with its own inner principle of change, like Socrates “one” or a “being”, it is of this principle, his essence and substance that we speak most properly: it is because he has a soul that he has being (is alive) and unity (is an organic, functionally determined body). When we speak of his body, or of what he potentially is, we speak of the capacities of what he now is in actuality, i.e. we speak of a human body – absent the soul (e.g. when dead) that body would no longer be what it is, and we would call it “body” or “human body” only homonymously. Aristotle is more interested in what perception does than he is in what perception is. The psychē is the “form of the living body” and thus the essence of every plant, animal and human: having a psychē is being alive (De An., a-). Its capacities are nutrition, perception, phantasia and thought. Perception distinguishes animals from plants; they need it, in particular touch, the sense of food, because they have to move for food or, if stationary, will not feed unless triggered by the perception of food (Everson : ): this is explained by the fact that perception brings with it pain and pleasure and these entail wanting, epithumia, which is a “desire for that which is pleasant” (De An., II., b and II., a). Aristotle distinguishes different capacities, according to how ‘life’ is said in many ways – of plants, animals and humans. Nutrition, or rather the “motion in relation to nourishment, decay and growth” (a) is common to everything that lives. Perception is what is primarily responsible for something to be an animal, and among the perceptive faculties mostly touch. “Touch” is understood by Aristotle in a much broader way than we do, as whatever brings us into contact with what is solid in our environment, i.e. whatever is dry, wet, hot or cold. When we perceive the dry and the hot and we desire it, we are hungry; when we perceive the cold and the wet, we are thirsty. We desire them by our desiderative faculty, which is had by anything that perceives, i.e. every animal. The perceptual faculty presupposes the nutritive faculty in many ways: ontologically, in that nothing that perceives is unable to grow; object-wise, because the most basic form of perception (touch) is the perception of nourishment; but also constitutionally, in that the nutritive faculty is present in the perceptive one in potentiality in the way the triangle is present in potentiality in the bisectable square. Though in humans, the nutritive faculty has its actuality in perception, it is separable in plants and it is in virtue of it that they are alive. Already in plants, it makes for some kind of ‘desire’, though Aristotle is quick to point out that it is a kind of teleological function, not an exercise of the desiderative faculty:

φυσικώτατον γὰρ τῶν ἔργων τοῖς ζῶσιν, ὅσα …for it is the most natural function in living For the most natural among the functions be- τέλεια καὶ μὴ πηρώματα ἢ τὴν γένεσιν αὐτο- thins, such as are perfect and not mutilated or longing to living things, at least those which μάτην ἔχει, τὸ ποιῆσαι ἕτερον οἷον αὐτό, ζῷον do not have spontaneous generation, to pro- are complete and neither deformed nor spon- μὲν ζῷον, φυτὸν δὲ φυτόν, ἵνα τοῦ ἀεὶ καὶ τοῦ duce another thing like themselves – an ani- taneously generated, is this: to make another θείου μετέχωσιν ᾗ δύνανται· πάντα γὰρ ἐκεί- mal to produce an animal, a plant a plant – such as itself, an animal an animal and a plant νου ὀρέγεται, καὶ ἐκείνου ἕνεκα πράττει ὅσα in order that they may partake of the everlast- a plant, so that it may,insofar as it is able, par- πράττει κατὰ φύσιν (a-b) ing and the divine in so far as they can; for all take of the everlasting and the divine. For that desire that, and for the sake of that they do is what everything desires, and for the sake of whatever they do in accordance with nature. that everything does whatever it does in ac- (Aristotle : ) cordance with nature. (Aristotle : )

How are we to understand perception as uptake of form? Here is Shields’ account:

A perceiving subject S perceives some sensible object o if and only if: (i) S has the capacity C requisite for receiving o’s sensible form F; (ii) o acts upon C by enforming it; and (iii) C becomes isomorphic with o’s sensible form by becoming itself F.(: xxxiv)

Several questionable and/or anachronistic features should be noted: it is unclear, first, whether there is some one unique thing which is the sensible form of o; “acts”, second, should be understood in line with Aristotle’s general (and complicated) account of causation; third, “isomorphic” should not be understood as “related by some structure-preserving one-to-one mapping”, as it is understood nowadays, but as “iso-morphic”, i.e. co-formed; fourth, “becoming F” is not necessarily a good account of what it is to become enformed by the sensible form F (and it may thus be highly misleading of speaking of the eye, or sight, exemplifying redness; on this, more below).

.. Aristotelian Colours

As he does in the case of change in the Physics, Aristotle starts his investigation of perception in De Anima II., which “seems to be a kind of alteration” (alloiosis), by a distinction between potentiality and actuality – when speaking about potentiality, it is true to say, as some do, that only like affects like; whereas with respect to actuality, unlike affects unlike:

διαιρετέον δὲ καὶ περὶ δυνάμεως καὶ ἐντελε- But we must make distinctions concerning po- One must also draw a distinction concern- χείας· νῦν γὰρ ἁπλῶς ἐλέγομεν περὶ αὐτῶν. tentiality and actuality; for at the moment we ing potentiality and actuality. For we have ἔστι μὲν γὰρ οὕτως ἐπιστῆμόν τι ὡς ἂν εἴ- are speaking of them in an unqualified way. just now been speaking of them without qual- ποιμεν ἄνθρωπον ἐπιστήμονα ὅτι ὁ ἄνθρω- For there are knowers in that we should speak ification. In the first case, something is a πος τῶν ἐπιστημόνων καὶ ἐχόντων ἐπιστή- of a man as a knower because man is one knower in the way in which we might say μην· ἔστι δ’ὡς ἤδη λέγομεν ἐπιστήμονα τὸν of those who are knowers and have knowl- that a human knows because humans belong ἔχοντα τὴν γραμματικήν· ἑκάτερος δὲ τούτων edge; then there are knowers in that we speak to the class of knowers and to those things οὐ τὸν αὐτὸν τρόπον δυνατός ἐστιν, ἀλλ’ ὁ straightaway of the man who has knowledge which have knowledge; but in the second case, μὲν ὅτι τὸ γένος τοιοῦτον καὶ ἡ ὕλη, ὁ δ’ ὅτι of grammar as a knower. (Each of these has we say directly that the one who has gram- βουληθεὶς δυνατὸς θεωρεῖν, ἂν μή τι κωλύσῃ a capacity but not in the same way – the one matical knowledge knows. These are not in τῶν ἔξωθεν· ὁ δ’ ἤδη θεωρῶν, ἐντελεχείᾳ ὢν because his kind, his stuff, is of this sort, the the same way potential knowers; instead, the καὶ κυρίως ἐπιστάμενος τόδε τὸ Α. ἀμφότε- other because he can if he so wishes contem- first one because his genus and matter are ροι μὲν οὖν οἱ πρῶτοι, κατὰ δύναμιν ἐπιστήμο- plate, as long as nothing external prevents of a certain sort, and the other because he νες <ὄντες, ἐνεργείᾳ γίνονται ἐπιστήμονες,> him.) There is thirdly the man who is already has the potential to contemplate whensoever ἀλλ’ ὁ μὲν διὰ μαθήσεως ἀλλοιωθεὶς καὶ πολ- contemplating, the man who is actually and he wishes, so long as nothing external hin- λάκις ἐξ ἐναντίας μεταβαλὼν ἕξεως, ὁ δ’ ἐκ in the proper sense knowing this particular A. ders him. Yet another sort of knower is the τοῦ ἔχειν τὴν ἀριθμητικὴν ἢ τὴν γραμματι- Thus, both the first two, < being > potential one already contemplating, who is in actual- κήν, μὴ ἐνεργεῖν δέ, εἰς τὸ ἐνεργεῖν, ἄλλον knowers, < become actual knowers >, but the ity and strictly knowing this A. In the first τρόπον. (a-b) one by being altered through learning and fre- two cases, then, those knowing in potentiality quent changes from an opposite direction, the come to be knowers in actuality, but the first other by passing in another way from the sate one by being altered through learning, with of having arithmetical or grammatical knowl- frequent changes from a contrary state; and edge without exercising it to its exercise. (Aris- the other, from having arithmetical or gram- totle : ) matical knowledge and not actualizing it to actualizing in another way. (Aristotle : )

Aristotle here contrasts two transitions, both from (some) potentiality to (some) actuality: episteme Humans are by their nature potentially knowers, i.e. possessors of knowledge (episteme); their being knowers is the actualisation of a potentiality which they have by nature, in virtue of man being rational animal. In this sense, all animals are perceivers, in virtue of animal being perceiving living thing. Itisalsoin this sense that anything red is seen as red, for red is appearing red. theoria Someone who has acquired knowledge becomes able to exercise it in contemplation (theoria): when some- one knowingly contemplates, e.g. by speaking or by solving a mathematical problem, she actualises a skill she has previously acquired. In humans, such exercise is subject to the will, but not so in other animals: in such animals, the capacity will be triggered autonomously in the absence of interfering factors, in virtue of the capacity being natural, i.e. of the nature of the animal. It is in this sense that animals see red, just by laying their eyes upon something red, in the absence of interfering factors; and it is in this sense that something red comes to be seen as red, under the right circumstances. Even though the first change is grounded in the nature of the thing changing (the actualisation of the potential is the realisation of the teleological function of the animal in question), it is a real change, “from an opposite direction”, i.e. an alteration of the changing thing. The second change, which happens autonomously in the absence of interfering factors, is not a real change, but the mere making manifest of a pre-existing and prior ability. When my eye thus meets the red tomato, my seeing it and it’s being seen by me are not real changes: I was already there to see it this way and it was there to be seen by me in this way. The objects of perception are either perceived in virtue of themselves (kath’hauta) or accidentally (kata sumbebekos) (De An., II.). Among the first are the proper sensibles (“that which cannot be perceived by any other sense”, colour of sight, sound of hearing, flavour of taste, that to which “the essence of each sense is naturally relative”) and the

common sensibles (movement, rest, number, figure, magnitude). The second are accidents of the thing which one perceives and which is the object of perception as such: in this second sense, not only the white thing, but also the son of Diares may be called “object of my perception” (De An. II., a-), the colour and the accidental sensible (the son of Diares) forming an “accidental unity” (ie. white son of Diares). The sense-organ is not affected by Darius, or any other physical object, but by Darius-qua-white. We have already seen that what actualises its potentiality to see red is not the eye but its potentiality (to see red), i.e. sight. Interestingly, Aristotle draws a corresponding distinction on the object-side of the perception relation: what is seen is not the object itself but its visible appearance, i.e. what Aristotle calls a “proper sensible”:

Λεκτέον δὲ καθ’ ἑκάστην αἴσθησιν περὶ τῶν We must speak first of the objects of percep- In the case of each sense, it is necessary to αἰσθητῶν πρῶτον. λέγεται δὲ τὸ αἰσθητὸν τρι- tion in relation to each senseA. But objects speak first about perceptible objects. Percepti- χῶς, ὧν δύο μὲν καθ’ αὑτά φαμεν αἰσθάνε- of perception are so spoken of in three ways; ble objects are spoken of in three ways: in two σθαι, τὸ δὲ ἓν κατὰ συμβεβηκός. τῶν δὲ δυοῖν of these we say that we perceive two in them- cases we say perceptible objects are perceived τὸ μὲν ἴδιόν ἐστιν ἑκάστης αἰσθήσεως, τὸ δὲ selves, and one incidentally. Of the two, one in their own right, and in one co-incidentally. κοινὸν πασῶν. is special to each senseA, the other common Of the first two, one is exclusive to an individ- λέγω δ’ ἴδιον μὲν ὃ μὴ ἐνδέχεται ἑτέρᾳ αἰσθή- to all. ual sense and the other common to them all. σει αἰσθάνεσθαι, καὶ περὶ ὃ μὴ ἐνδέχεται ἀπα- I call special-object whatever cannot be per- By exclusive I mean what cannot be perceived τηθῆναι, οἷον ὄψις χρώματος καὶ ἀκοὴ ψόφου ceived by another senseA, and about which by another sense and about what one cannot καὶ γεῦσις χυμοῦ, ἡ δ’ ἁφὴ πλείους [μὲν] ἔχει it is impossible to be deceived, e.g. sight as be deceived. For example, sight is of colour, διαφοράς, ἀλλ’ ἑκάστη γε κρίνει περὶ τούτων, colour, hearing sound, an taste flavour, while hearing is of sound, and taste is of flavour, καὶ οὐκ ἀπατᾶται ὅτι χρῶμα οὐδ’ ὅτι ψόφος, touch has many varieties of object. But at any whereas touch has a number of different ob- ἀλλὰ τί τὸ κεχρωσμένον ἢ ποῦ, ἢ τί τὸ ψοφοῦν rate each judges about these, and is not de- jects. In any case, each sense discerns these ἢ ποῦ. τὰ μὲν οὖν τοιαῦτα λέγεται ἴδια ἑκά- ceived as to the fact that there is colour or and is not deceived that there is colour or that στης […] sound, but rather as to what and where the there is sound – as opposed to what or where κατὰ συμβεβηκὸς δὲ λέγεται αἰσθητόν, οἷον εἰ coloured thing is or as to what or where the the coloured or sounding thing is. […] τὸ λευκὸν εἴη Διάρους υἱός· κατὰ συμβεβηκὸς object which sounds is. […] Something is said to be an object of percep- γὰρ τούτου αἰσθάνεται, ὅτι τῷ λευκῷ συμβέ- An object of perception is spoken of as inci- tion co-incidentially if, for example, the white βηκε τοῦτο, οὗ αἰσθάνεται· διὸ καὶ οὐδὲν πά- dental, e.g. if the white thing were the son thing should be the son of Diares. There is σχει ᾗ τοιοῦτον ὑπὸ τοῦ αἰσθητοῦ. of Diares; for you perceive this incidentally, co-incidental perception of him, because he τῶν δὲ καθ’ αὑτὰ αἰσθητῶν τὰ ἴδια κυρίως since this which you perceive is incidental to coincides with the white thing, of which there ἐστὶν αἰσθητά, καὶ πρὸς ἃ ἡ οὐσία πέφυκεν the white thing. Hence too you are not af- is perception. For this reason, one is not af- ἑκάστης αἰσθήσεως. (a-) fected by the object of perception as such. fected by an object of perception insofar as it Of the objects which are perceived in them- is such a thing as the son of Diares. selves it is the special-objects which are ob- Among things perceived in their own right, jects of perception properly, and it is to these exclusive objects are properly perceptible ob- that the essence of each senseAis naturally rel- jects; and it is to these that the essence of each ative. (Aristotle : ) sense is naturally relative. (Aristotle : - )

Aristotle here draws a threefold distinction in terms of two distinctions, between per se (kath’hauto) and incidental (kata sumbebekos) perception, and between its proper (idion) and common (koinon objects: kath’hauto, idion These are the so-called “proper sensibles”: they are perceived in their own right (directly, not by their appearances being perceived) and by just one sense only. While we can both hear, smell, taste and touch the tomato, the colour of the tomato is such that it can only be seen. kath’hauto, koinon Of these so-called “common sensibles”, Aristotle lists here five: motion, rest, number, shape, magnitude. Motion, for example, is “an object for perception for both touch and sight” (a-). kata sumbebekos Things seen ‘incidentally’ or by ‘coincidental perception’ include things such as the son of Diares. The son of Diares is white – coincidentally, that is: accidentally and by way of having the white as a quality. The white or perhaps this white is a proper object of perception, and as it coincides with the son of Diares, ‘through it’ the man is seen as well. The proper sensibles are specific to each sense and are, by that sense, always perceived veridically and kath’hauto; they determine, but are not determined by their sense, the essence of which is “naturally relative to them” (De An., II., a).

. Hamlyn remarks ad loc: “It is noteworthy here that Aristotle turns the conceptual point about the connexion between an sense and its object into one concerned with matters of fact, ie. one about what affects a given sense.” (Hamlyn : ) I think this gives us further reason not to take the connection to be conceptual.

The perceptions of proper sensibles are almost always true, in the sense that the sense proper to the sensible is not deceived, though some belief based on the true perception may still be false (Everson : ):

Aristotle’s claim that we cannot be deceived in our perception of the proper sensibles is not the weak claim that our beliefs about our own experience are incorrigible but the strong claim that, in respect of the proper sensibles, our senses guarantee a truthful report of our perceptual environment. (Everson : )

While perception and the perceptible are correlatives (Cat. ), the relation is not symmetric, but directed from the perceptible objects to the perception, because the former, but not the latter, could exist in the absence of the other one (Cat. a-): the perceptible objects themselves are the agents of perception (Met., Γ. , b-a), and the capacity they actualise when they act upon perceivers is grounded in non-dispositional properties: “the proper sensibles will bring about perception in virtue of being what they are: they are intrinsically such as to produce the relevant changes in the organs” (Everson : ). To posteriority of the senses to their kath’hauta objects, the proper sensibles, does not make the latter in any way ‘internal’ to or ‘subsist in’ the former. It rather constrains their material constitution:

The organ [of sight] needs to be transparent because […] it is the transparent which, because it is colourless, is receptive of colour […]. Coloured substances and transparent substances thus stand as agents and patients kath’hauta: it is the nature of colour to produce change in what is transparent – just as it is in the nature of what is transparent to be affected by colour. (Everson : )

The proper sensibles are thus kath’hauto causes of a change in a perceiving body – a change which is the perception. In all types of perception, the medium is that of a transmission of form. It is a form that is taken on in the alteration of the sense that is the perception. This is put in terms of “things in so far as they are of a certain kind” in passage quoted above (De An., II., a). In the visual perception of a red surface, e.g., the eye is the matter, and redness its form. Redness, however, as the proper sensible of sight, is an external object, and a particular (De An., II., b). It is “capable of setting in motion that which is actually transparent, and this is its nature” (De An., II., a). That only the form, but not the matter of the red object is received, makes it a form of perception; plants, which do not perceive, lack this differential ability because they do not have sense-organs and thus no senses. Of the nature of (some particular) redness, a proper sensible of sight, we are, I think, given the following picture: it is a form of an external object, which exists prior to and independently of our perceiving it. Because it is capable of setting air, which is a medium of sight, into motion, it may be received by a sense, ie. sight, the nature of which it is to be receptive to it, ie. potentially red. When the red object is perceived, it shares its form with the eye – the sense (sight) is affected by the essence of the proper sensible, while the sense-organ is affected in the sense of being of a certain extended magnitude. The sense-organ (eye) and the sense (sight) are “the same, although what it is for them to be such is not the same” (De An., II., a): sight is a potentiality of the eye, its function and essence; it is sight that becomes red, i.e. is affected only be the form (of the red object), while the eye, which is affected in a different way, is in relation not just to the form, but also the matter of the red object, the bearer of e.g. the reflectance properties that causally explain why it is red I see. – So redness is a way for sight to be!

. Everson (: –) argues against Hamlyn (cf. his comments on p. of Hamlyn ()) and Sorabji (: ) that the dependence of the sense on its objects is not definitional, logical or grammatical, but causal. This also defuses the worry that Aristotle’s priority claim is implausible for the case of touch, the objects of the latter to be too varied to define the sense of touch (Hamlyn (: –) and Sorabji (: )). . This interpretation makes best sense of the otherwise quite enigmatic beginning of II., of which I have not been able to find a convincing interpretation. Hamlyn’s use of scare quotes is telling: “Aristotle goes on to ‘explain’ the reception of form in terms of the affection of a sense by things in virtue of their form, i.e. in perception the sense is affected by an object just in so far as it is of the relevant form and not because it is what it is.” (Hamlyn : , ad a)

But do we not see other things than colours and shininess, as e.g. “size and shape, motion or rest, texture, depth, or the location of things” Sorabji (: )? Yes, Aristotle would say,we do see them, but not kath’hauto, but instead by seeing colours (or shininess). The difference between (kath’hauta) and (kata sumbebekos) perception is very important for an account of the percep- tual medium. The medium of sight is that which is transparent; it is set into motion by colours, and then becomes light. Because light is a sort of colour of the medium, we also see the medium, even though it is “not strictly speaking visible in itself” (De An., II., b), we see it because “the colour of each thing is always seen in light”. Like sight, audition is a relation between two things through a medium: “actual sound is always of something in relation to something and in something” (De An., II., b). For the perception of sounds, the medium plays a more active role than for the perception of colours: the differences between densities of quantities of air helps explain the diversity of sounds. Aristotle goes as far as identifying the sound with the movement of the air (De An., II., a) Like sound, smell moves an intervening medium (De An. II., a), and as with sounds, the movements of this medium are identified with the smells smelt (the proper sensibles of the sense of smell). The only sense that apparently does not require an external medium is touch, of which taste is a special kind. The medium of touch (and taste) is flesh, and of taste particularly the flesh of the tongue. The medium is touched as well, though it is not itself active, as are the media in the case of sight, audition and smell. It is not strictly speaking true that colours are the proper sensibles of sight:

Οὗ μὲν οὖν ἐστιν ἡ ὄψις, τοῦτ’ ἐστὶν ὁρατόν, That of which there is sight, then, is visible. That of which there is sight is the visible. The ὁρατὸν δ’ἐστὶ χρῶμά τε καὶ ὃ λόγῳ μὲν ἔστιν What is visible is colour and also something visible is both colour and something which it εἰπεῖν, ἀνώνυμον δὲ τυγχάνει ὄν· δῆλον δὲ which may be described in wordsL, but hap- is possible to describe in words, but which has ἔσται ὃ λέγομεν προελθοῦσι. τὸ γὰρ ὁρατόν pens to have no name; what we mean will be no name. (What we mean will be clear as we ἐστι χρῶμα, τοῦτο δ’ ἐστὶ τὸ ἐπὶ τοῦ καθ’ αὑτὸ clear as we proceed. For the visible is colour, proceed.) ὁρατοῦ· καθ’ αὑτὸ δὲ οὐ τῷ λόγῳ, ἀλλ’ ὅτι and this is that which overlies what is in itself The visible is colour, and that which is on the ἐν ἑαυτῷ ἔχει τὸ αἴτιον τοῦ εἶναι ὁρατόν. πᾶν visible – in itself visible not by definitionL, but surface of what is visible in its own right – in δὲ χρῶμα κινητιἑαυτῷ ἔχει τὸ αἴτιον τοῦ εἶ- because it has in itself the cause of its visibility. its own right not by definition, but because it ναι ὁρατόν. πᾶν δὲ χρῶμα κινητικόν ἐστι τοῦ Every colour is capable of setting in motion contains within itself the cause of its being vis- κατ’ ἐνέργειαν διαφανοῦς, καὶ τοῦτ’ ἐστὶν αὐ- that which is actually transparent, and this is ible. Every colour is capable of setting in mo- τοῦ ἡ φύσις· διόπερ οὐχ ὁρατὸν ἄνευ φωτός, its nature. For this reason it is not visible with- tion that which is actually transparent; and ἀλλὰ πᾶν τὸ ἑκάστου χρῶμα ἐν φωτὶ ὁρᾶται. out light, but the colour of each thing is al- this is its nature. Consequently, nothing is vis- (a-b) ways seen in light. (Aristotle : ) ible without light. Rather, the colour of each thing is always seen in light. (Aristotle : )

MUCH MORE ON THE TRANSPARENT, AND ON LIGHT SUMMARY In all types of perception, the medium is that of a transmission of form. It is a form that is taken on in the alteration of the sense that is the perception:

…the sense (aisthēsis) is that which can receive perceptible forms without their matter, as wax receives the imprint of the ring without the iron or gold, and it takes the imprint which is of gold or bronze, but not qua gold or bronze. Similarly too in each case the sense is affected by that which has colour or flavour or sound, but by these not in so far as they are what each of them is spoken of as being, but in so far as they are things of a certain kind and in accordance with their principle. (De An., II., a)

In the visual perception of a red surface, e.g., the eye is the matter, and redness its form. Redness, however, as the proper sensible of sight, is an external object, and a particular (De An., II., b). It is “capable of setting in motion that which is actually transparent, and this is its nature” (De An., II., a). That only the form, but not the matter of the red object is received, makes it a form of perception; plants, which do not perceive, lack this differential ability because they do not have sense-organs and thus do not have senses. . In principle, such an account would seem to be possible for the visual case too: in this way, Aristotle could explain (at least some) visual illusions in terms of the perceptual medium.

Such a conception of colours as the proper sensibles of sight is realist in at least three different, interlocking, ways: • The form red is a form of mind-independently existing objects and is the actualisation of a power they have independently of being perceived. • The actualisation of the form red does not depend on and is not brought about by perception; red is actu- alised by light, the medium of sight, and the way it is actualised is ‘reciprocal’ to the power of light to make things visible, not the power of perceivers to see; red is essentially visible, but not essentially seen. • In perception, the perceiver is passive; the change that is perception occurs in the perceiver; the perceived object and its colour do not change in perception: nothing mind-independent is altered when and because perception occurs. Aristotelian colours are not, however, objective in Burge’s sense. They are ‘constitutively explained’ by what it is to see them; the account of their nature essentially mentions human perception, and human perception is essentially veridical:

The constitutive explication takes the direction of the constitution relation to be asymmetric. The standards for being veridical that are parts of the natures of perceptual states are constitutively de- pendent on attributes in the environment. The attributes in the environment are not constitutively dependent on those veridicality conditions. (Burge : )

In contrast to this picture, an Aristotelian view will hold that what is seen and the seeing of it are not just mutually interdependent, but that they stand in a relation of mutual grounding: what is seen is essentially visible and of a nature fully actualised only in normal human perception, while normal human perception is what it is in virtue of its power to fully realise what is there to be seen. This essential interdependence of colours and normal human sight is compatible, however, with human perception being objective in the sense in which objectivity is produced by the visual system itself:

Objectivity is the product of separating what occurs on an individual’s sensory surfaces from the significance of those stimulations for specific attributes and particulars in the broader environment. In this way, perception is the product of objectification. (Burge : )

From an Aristotelian perspective, objectification occurs in incidental perception, in the step from the seeing of the white of Darius to the seeing of white Darius, via the seeing of Darius-qua-white. Even though colours are not response-dependent in the usual, counterfactual way, they are still essentially tied to how they look to us, to objective but relational appearance properties:

The reference-fixing responses in us a priori associated with a secondary property are constitutive of the essence of the property; the property is, constitutively, a disposition to actually cause those properties. (García-Carpintero : )

The reference-fixing responses in us are appearance-properties, properties had by mind-independent things in virtue of how we see them. Such perspectival appearance properties will not not features of perspective-independent things, but rather of appearances, appearances of perspective-independent things, to be sure, but not things that are themselves independent of perspectives. It is the appearance of the stick half-immersed in water that looks bent-from-here and the same appearance that may look straight-from-here with respect to some other location or observer. It is the change in appearance that is responsible for the fact that once the stick is taken out of the water and seen by both of us in clear air, it may no longer be said to appear bent in any sense. Because appearances are grounded in the things appearing, together with their environment, they are appearances of things – things with which we are in perceptual contact when we perceive the appearances. There is still a distinction between proper and common sensibles to be drawn, however: . Rather than of colours we should speak here of “chromatic profiles”, so as to preserve Aristotle’s claim that they are the proper sensibles of sight: we also see black and white things, transparent things and things that do not have surfaces, such as shadows, holes, holograms and rainbows. I will ignore this complication in the following.

“What puts vision apart from hearing, smell, and taste is that we do not conceive of the visible world as offering us objects of visual awareness and attention distinct from (but coincident with) the concrete objects that we also see.” (Martin : ), citing O’Shaughnessy (: –)

The chromatic profile of a coloured thing (its visual appearance, its look) is more closely tied to it than its taste or its visual shape: the being-the-appearance-of relation here is not causal, but constitutive. Looks are nothing but looks of things, i.e. things looking a certain way. In other terms: for a to have a visual look that appears red is both, for the look to be red and for the thing itself to appear red. Two possible sources of error present in the case of taste are thereby eliminated: it is not possible that only its look, but not the thing looking a certain way appears red, and the look itself cannot just appear red but fail to be red.

.. Representation by Exemplification

If perception is alteration, then in what does this alteration consist? It is on that question (which, for reasons laid out above, is quite tangential to Aristotle’s main interest in De Anima) that an influential debate of interpreters has focussed: • According to the ‘literalist’ interpretation of Aristotle (Slakey (), Sorabji (, ), Everson ()), “when a sense organ is activated and perception occurs, the organ is altered so that it literally becomes like its (proper) object: it takes on the property of the sensible which affects it” (Everson : ). The material alteration of the body by perception is either identical with (Slakey : ) or constitutes (Sorabji : ) the perceptual activity. • According to the ‘spiritualist’ interpretation (Burnyeat : ), the change is a “becoming aware of some sensible quality in the environment”. Shields (: ) says that the sense organ will symbolize sensible qualities in one way or another. But what does this mean? How does the eye do its symbolising? The circularity of Burnyeat’s ‘spiritualist’ account becomes apparent when he says (italics mine): “…receiving the warmth of a warm thing without its matter means becoming warm without really becoming warm; it means registering, noticing, or perceiving the warmth without really becoming warm” (: ) Hamlin puts much stress on the fact that the actualisation which is the alteration of the sense-organ that occurs in perception is of a hexis, not of a dunamis. CHECK Contrary to Perler (: ), this is not primarily the question whether perception is an at least partly physiological or entirely immaterial transmission of form. Neither do I think that it is helpful to contrast the literal with the ‘rep- resentational’ becoming red of the sense organ (Shields : ): the choice is between two types of representation: by exemplification, or by some other, less direct, means. I also think that it is beside the point to point out that exemplification is not sufficient for representation (Shields : ): of course it is not, only exemplifications of red in some kind linked to others represent something as being red. It is also beside the point to support the non-sufficiency claim by attentional phenomena: “That [literal alteration is sufficient for perception] seems an odd and unsustainable claim, since, for example, my flesh can become warm without my perceiving warmth; indeed, the view thus interpreted seems to entail that I perceive every sensible quality in whose presence I find myself, with no allowance for the phenomena of selective attention, inattention, distraction, and the like.” (Shields : ) Attention is a passive faculty: it presupposes some content already independently available. The question rather is: in virtue of what does the perceiving animal, or its eye(s), represent the apple as red? Aristotle’s answer is: because of the nature of redness, and the eye’s nature of being receptive to it. It is not immediately clear that this very un-orthodox theory of perception is afflicted by the main problem of its contemporary descendant: the argument from illusion, which crucially depends on ‘finding a home for’ what we merely seem to see in illusionary cases.

This theory combines two very attractive features: a naturalistic account of the experience we undergo when and because we perceive, and an explanation how perception and its objects may be mutually dependent and ‘correlative’, without making the objects ‘subjective’ or dependent on their being perceived. Burnyeat () thinks that the Aristotelian theory of perception needs not to be taken seriously because it presup- poses something like perspectival facts:

One might say that the physical material of animal bodies in Aristotle’s world is already pregnant with consciousness, needing only to be awakened to red or warmth. (: )

. Mode and Content

.. Relations and Representations

.. Hedging Russell’s Retreat

When things appear to us a certain way, we may report this by saying that they have, or produce in us, such- and-such appearances, whether or not they really are as they appear. It is often thought that this manoeuvre (the ‘Russellian retreat’) increases justified confidence: Although we (often) do not know how things are, or are ‘in themselves’, we may at least be confident that they appear to be a certain way. The epistemological gains, however, come at a metaphysical price: what are these appearances, and what in their nature explains that they are more accessible than what they are appearances of? In my contribution to the conference, I would like to explore an alternative picture according to which belief how things appear is in some sense weaker, not stronger, than belief how things are. Let us start with an assertion, by someone at a certain type, pointing to a sheet of paper in front of her, of

() This is red.

This assertion expresses a belief with the content that the sheet in question is red and is true iff the sheet is red. Suppose now that the same person, in the same circumstances and with respect to the same sheet, says

() This looks red.

I will suppose, for the moment, that the person could have made an assertion relevantly equivalent to with any of the following: “This looks red to me.”, “This seems / appears red [to me].”, “This has a red look [for me].”, “Of this object, one has [I have] a red impression. I will also assume that is true iff the sheet looks red, i.e. iff the sheet has a property, looking red (or perhaps: looking red to someone in such-and-such circumstances), we may call an “appearance property”. But what is expressed by ? It is commonly assumed that , like , expresses a belief, differing from only with respect to its content: while ascribes to the sheet the property being red, ascribes to it the different property looking red. On most views of such properties, application conditions for their predicates are easier to fulfill than the application conditions for the predicates they embed: to be justified in , I have to be justified in thinking that the sheet is really red, while justification fo is easier to come by. I want to epitomise this view by the label “Russell’s retreat”. Russell’s retreat is the view that in asserting not just but as well, I take a certain type of epistemic risk (perhaps because expresses some defeasible evidence for ): when challenged with respect to , it may make sense to ‘fall back to’, or ‘retreat’ to the weaker assertion . In contrast to this picture suggested by Russell’s retreat, I want to suggest that is expressive not of belief, but of another state of mind – of conjecturing, taking it to be possible, or even of surmising. As my starting-point and motivation, consider what Sellars says about “looks”-statements in Empiricism and the Philosophy of Mind:

…to say that a certain experience is a seeing that something is the case, is to do more than describe the experience. It is to characterize it as, so to speak, making an assertion of claim, and – which is the point I wish to stress – to endorse that claim. […] …the statement “X looks green to Jones” differs from “Jones sees that x is green” in that whereas the latter both ascribes a propositional claim to Jones’ experience and endorses it, the former ascribes the claim but does not endorse it. (Sellars : , -)

If does not endorse the claim that the sheet is red, what does it do? Sellars says that it “report[s] the fact that my experience is, so to speak, intrinsically, as an experience, indistinguishable from a veridical one of seeing [that the sheet is red]” (: , -). This may well be. But what is endorsed by ? Certainly not what is reported: asserting I am not endorsing any claim about what would be indistinguishable to what, under such-and-such circumstances and for such-and-such subjects. What I am endorsing in asserting , plainly, is that this looks red to me. Does this bring us back to the appearance- properties with respect to which we are supposed to enjoy some epistemological privilege? Not necessarily, I want to argue. There are three question I would like to take up in turn: . Does the content of (or rather: of the mental state expressed by ) involve an appearance-property? . Do the correctness conditions of involve an appearance property? . Do the truth conditions of involve an appearance property? Construing “seems” as a (type of) copula, it seems to me, gives us the material for a negative answer to the first question. Motivation for a negative answer to the second question may be found, I want to suggest more tentatively, may be found in Kant’s distinction between judgments of perception and judgments of experience. Even more speculatively, grounds for a negative answer to the third question may be found (or at least looked for) in a certain view about the mind of the Pyrrhonian skeptic or, alternatively, in a view about what God has to know to know everything. Broad (: ) distinguishes between what he calls the “Multiple Relation Theory of Appearing” and the “Sen- sum Theory”. According to the former, which he finds suggested (though not worked out) in the writings of Dawes Hicks and G.E. Moore, “appearing to be so and so is a unique kind of relation between an object, a mind, and a characteristic”. “Seems” is treated as a “mode of copuation”, “allow[ing] Moore to acknowledge the possibility than an object of awareness might be united with [red] only relative to my current experience” (Martin : ). According to this ‘theory’, the way in which a region of space has a colour (is “pervaded” by it) is a three-place re- lation, “involving the pervading colour, the pervaded region, and another region which we might call the “region of projection”.” (: ).

.. Perception as intake of form

Putting history to work. While I think it is plausible on its own and combines, I hope to show, in an internally coherent way the main insights from the tradition, I give two further arguments in favour of the theoretical fruit- fulness of an account of perception as intake of form. The first concerns the explanatory potential of such a theory with respect to some of the stock problems of the contemporary philosophy of mind. To substantiate this claim, I would like to sketch its solutions to three classical problems: (i) the argument from illusion; (ii) the problem of how perceptual content is individuated; (iii) the metaphysical status of appearances more generally. The second reason is that the form-intake account of perception generalises naturally to all forms of representation. While such an argument will necessarily remain cursory, I still hope that attention to other types of meaning will help to clarify the metaphysics of the representation relation, esp. with respect to the case perhaps furthest apart from direct sensory stimulation, linguistic representation. Back to the problem of conflicting appearances. Let us consider one version of the argument from illusion: A stick in front of me looks bent. A stick in front of me looks straight. There is at most one stick in front of me. No stick can be both bent and straight.

∴ In at least one case, my perceptual experience is non-veridical. In the non-veridical perceptual experience, I am presented with a mind-dependent intentional object. Both perceptual experiences have the same type of intentional object. ∴ The intentional object of both experiences is mind-dependent. Sticks are not mind-dependent. ∴ I neither case do I see the stick. Hence nothing that may look to have contrary properties is ever seen – a reductio. The most promising way out, I think, is to block the deduction of (), while preserving a sense in which both () and () may be true. Adverbialists reinterpret () and () as i I am appeared to bent-ly. i I am appeared to straight-ly. Even though we do not then have analogues of () and (), () still seems to follow: If am appeared to bent-ly and straightly, then in at least one case, my perceptual experience is non-veridical. The sense data theorist reinterprets () and () as ii It appears to me: something is bent*. ii It appears to me: something is straight*. Bentness* and straightness* are not properties of sticks, but of sense-data: if they can both be had by the same sense-datum, then we do not have an analogue of (); if they cannot, then we do not have an analogue of (). The victory is Pyrrhonic, however, because even if they can perhaps provide a reading of “seeing” on which it is true that I see the stick, I will not on their view see its bentness or straightness. Perspectival realists also deny () by reinterpreting () and (). Their version is: iii The stick appears bent-to-me. iii The stick appears straight-to-me. This, I think is the right response to the argument. It combines the adverbial account of the (allegedly) contrary properties with the introduction of appearance properties. These appearance properties are perspectival – what in the world could have them? The answer is straightforward: appearances. Kant is right in thinking that the perception of appearance properties is the perception of appearances, where such appearances are grounded in the things they are appearances of. We should not, however, in the way of sense-data theorists, taken these appearances to be the direct objects of perception – rather, it is their grounds that are seen through them. This explanation of appearance properties thus takes us to iv The stick-as-seen-by-me appears-bent. iv The stick-as-seen-by-me appears-straight. We thus have two perceptions (and two perceived appearances), but still just one stick, grounding them both. () does not follow if the two appearances really have their appearance properties. Such a reply to the argument from illusion, I surmise, is not only independently plausible, but also generalises to the analoguous arguments by McTaggart against temporal A-determinations and and by Frank Jackson in favour of perceptual qualia. Back to the content-determination problem. Virtually any state can be used as a representation: you can decide to use a red flag to represent danger and you can take your aching muscles as a sign that you should not have walked that far. But these representational contents seem to derive from contingent dispositions to interpret the relevant states in certain ways. By constrast, our propositional attitudes and perceptions seem to have their content originary, not in virtue of being interpreted. If there is intrinsic content (content-properties exemplified intrinsically), a natural question to ask is what they are. How do mental representations acquire their accuracy conditions, in virtue of which do they represent the world as being some way rather than another? One superfi- cially tempting proposal is that the representational content of an internal representation is determined by simple counterfactual co-variation relations, perhaps backed by causal laws and past causal dependence (Stampe ; Dretske ). For reasons laid out in the ‘overview’ section above, however, this will not do: co-variation relations, causal or not, cannot be what determines the representational content of mental states. Not only does causal

covariation not allow for the possibility of error, neither does it account for the determinacy of representational content. Causal co-variation accounts of mental representation will take ‘horse’ to represent not the natural kind horse, but the disjunctive kind horse-or-horsey-images-on-the-retina-or thoughts-of-cowboys-or-cows-on-a-dark- night-or-donkeys-in-the-distance-or-…(Fodor ). Toaccount for determinate and fallible accuracy conditions of mental states, we are thus forced to examine the representation relation directly, and to characterise it intrinsically: as with the accuracy conditions invoked by vision science (according to Burge), perceptual states are individuated in part by their function of representing particular distal features of the environment (cf. eg. Burge : -). Burge, however, leaves open the question in what exactly this function consists – the Aristotelian account sketched above may help here, letting the determinate objets of perception in the distal environment themselves, as well as the perceptual milieu, do some of the work of bestowing their perceptual representations with determinate content. Representational properties as intrinsic, but relational. In previous work, I have discussed the famous definitions of intrinsicness proposed by David Lewis (a: , : , : ) and Lewis & Langton ()), criticising them for not doing justice to the intimate connection between parthood and intrinsicness (F is an intrinsic property of a if it is a part of it; a is a part of b if there is a region of intrinsic match between a and b). In all Lewis- definitions, “how a thing is by itself” is translated into “how a thing would be if it were lonely”. This transition, however, is far from being mandatory. Another possible way to spell out the “by itself” clause, as the examples given e.g. by Sider (: ) (being green or being feet from some red thing – an intrinsic property (only) of green objects) and by Sider (: ) (so-called ‘maximal’ properties F like being a house such that large parts of an F are not themselves F, but F still counts intuitively as intrinsic) show,is to count those features of a thing as intrinsic that are determined by what goes on inside its borders, i.e. by how its parts are and in what relations they stand (cf. Humberstone : ). It is in this sense of internality, not of compatibility-with-loneliness, that representational properties may plausible taken to be intrinsic. That they are intrinsic, however, does not rule out that they are relational, i.e. individuated with respect to relations (Hochberg : , cf. also Dunn ; a). Some relational properties, such as having a as a part, creating the world (of God), the value of Diana’s dress (Rabinowicz & Rønnow-Rasmussen ), being of a crime (of some punishment) and, most importantly for us, being of a song (of a singing) are such that they are had (or lacked) by things in virtue of how these things are in themselves. The most important class of intrinsic, but relational properties are representational properties like meaning that Fa, representing a to be F or thinking of a as F: they are intrinsically exemplified by some thing x iff x exemplifies the property independently of how matters stand with respect to other things than x – no further properties have to be exemplified for other things for my thought, e.g., to represent a to be F. Allowing for intrinsic, but relational representation also clarifies the common distinction between derived and originary ‘intentionality’ (as it is mistakenly called). Many things may be said to have content, but most of them do so indirectly: they have content in virtue, for example, of having been produced in a certain way or with certain intentions, or of standing in some relation to other things that have content. The most important such relation is that of some things expressing other things. It is in virtue of expressing my beliefs that my utterances have content, and – subject to certain constraints – the beliefs expressed determine what content they have. Most contentful things thus have their content extrinsically: they mean what they do in virtue of other things having a certain content. At some point, however, the bucket must stop: if there are any representational properties at all, some things must have them intrinsically (Searle : -, -). Because they are representational, however, they are relational even when exemplified intrinsically: they represent something other than themselves, creating a relation between their bearer and the things they make their bearer be about. If my thought, for example, represents a to be F, it stands in the relation of aboutness to a and in the predication relation to the universal F. It is in virtue of these relations that my thought can stand in for a’s being F, and be – in some sense further to be specified – a substitute of this external fact.

Chapter

Intentionality and Representation

. The Emergence of Content

.. Appearances and phantasiai

It is only in De Anima III. that Aristotle confronts the problem of error, criticising the ‘ancients’ who said that reasoning (noein) and understanding (phronein are like perception “of like by like”:

καίτοι ἔδει ἅμα καὶ περὶ τοῦ ἠπατῆσθαι αὐ- Yet they should at the same time have said Even so, they ought at the same time to have τοὺς λέγειν, οἰκειότερον γὰρ τοῖς ζῴοις, καὶ something about error, for this is more char- said something about error, since this is the πλείω χρόνον ἐν τούτῳ διατελεῖ ἡ ψυχή· διὸ acteristic of animals and the soul spends more more typical state in animals and the soul ἀνάγκη ἤτοι, ὥσπερ ἔνιοι λέγουσι, πάντα τὰ time in this state; hence on their view either spends more time in this condition. For this φαινόμενα εἶναι ἀληθῆ, ἢ τὴν τοῦ ἀνομοίου all appearances must be true, as some say, or reason it is necessary either, as some say, that θίξιν ἀπάτην εἶναι, τοῦτο γὰρ ἐναντίον τῷ τὸ error must be a contact with the unlike, for all appearances are true, or that error is con- ὅμοιον τῷ ὁμοίῳ γνωρίζειν· δοκεῖ δὲ καὶ ἡ this is the opposite of recognizing like by like. tact with what is unlike, since this is the oppo- ἀπάτη καὶ ἡ ἐπιστήμη τῶν ἐναντίων ἡ αὐτὴ But error and knowledge seem to be the same site of coming to know like by like. It seems, εἶναι (a–b) in respect of the opposites. (Aristotle : ) however, that both error and knowledge of opposites are the same. (Aristotle : -)

That imagination may be of things that are not present and can thus give rise to error – be false in a certain way – is because it involves combination, in the same way incidental perception does:

τὸ γὰρ ψεῦδος ἐν συνθέσει ἀεί· καὶ γὰρ ἂν τὸ For falsity always depends upon a combina- For falsity is always in combining. For even if λευκὸν μὴ λευκὸν <φῇ, τὸ λευκὸν καὶ> τὸ tion; for even if someone says that white is one says the white is not white, one has com- μὴ λευκὸν συνέθηκεν· (b) non-white he combines < white and > non- binedwhiteandnotwhite. (Aristotle : ) white. (Aristotle : )

The result of such combining are ‘images’, and these are ‘produced’ by perception in the following way:

τὸ μὲν οὖν αἰσθάνεσθαι ὅμοιον τῷ φάναι μό- φήσῃ ἢ ἀποφήσῃ, φεύγει ἢ διώκει· διὸ οὐδέ- Perceiving, then, is like mere assertion and νον καὶ νοεῖν· ὅταν δὲ ἡδὺ ἢ λυπηρόν, οἷον κα- ποτε νοεῖ ἄνευ φαντάσματος ἡ ψυχή. (a) thought; when something is pleasant or ταφᾶσα ἢ ἀποφᾶσα διώκει ἢ φεύγει· καὶ ἔστι painful, { the soul } pursues or avoids it, as it τὸ ἥδεσθαι καὶ λυπεῖσθαι τὸ ἐνεργεῖν τῇ αἰ- were asserting or denying it; and to feel plea- σθητικῇ μεσότητι πρὸς τὸ ἀγαθὸν ἢ κακόν, ᾗ sure or pain is to be active with the perceptive τοιαῦτα. mean towards the good or bad as such. καὶ ἡ φυγὴ δὲ καὶ ἡ ὄρεξις ταὐτό, ἡ κατ’ ἐνέρ- Avoidance and desire, as actual, are the same γειαν, καὶ οὐχ ἕτερον τὸ ὀρεκτικὸν καὶ τὸ φευ- thing, and that which can desireK and that κτικόν, οὔτ’ ἀλλήλων οὔτε τοῦ αἰσθητικοῦ· which can avoidK are not different either ἀλλὰ τὸ εἶναι ἄλλο. from each other or from that which can τῇ δὲ διανοητικῇ ψυχῇ τὰ φαντάσματα οἷον perceiveK; but what it is for them to be such αἰσθήματα ὑπάρχει, ὅταν δὲ ἀγαθὸν ἢ κακὸν is different.

To the thinking soul images serve as sense- Perception is similar, then, to bare assertion desire and the capacity for avoidance do not perceptions (aisthēmata). And when it asserts and to thinking. But whenever there is some- differ either from one another or from the per- or denies good or bad, it avoids or pursues it. thing pleasant or painful, it by, so to speak, af- ceptual faculty, though they do differ in be- Hence the soul never thinks without an image. firming or denying, pursues or avoids. And it ing. (Aristotle : ) is the case that being pleased or being pained Images belong to the rational soul in the man- are the actualization of the mean of the per- ner of perceptions, and whenever it avoids or ceptual faculty in relation to what is good or denies that something is good or bad, it pur- bad insofar as they are such. sues or avoids. Consequently, the soul never And avoidance and desire are the same, in re- thinks without an image. (Aristotle : ) spect of their actuality; and the capacity for

Aristotle here is not just saying that “images (phantasmata) function in the rational soul just as perceptions (aisthêmata) function in the perceptual soul” (Shields : ). Rather, he says that images involve combination already at the perceptual level, insofar as they represent something as pleasant or painful, to be pursued or to be avoided. Much later, the Stoics were to cash this out as “apprehension”, as reported by Diogenes Laertius:

Perception (aisthēsis) is said by the Stoics to be () the pneuma extending to the senses (aisthēseis), () the apprehension through them, () the makeup of the sense organs, in which some people are defective. (D.L. .)

The crucial new ingredient is the distinction between appearance (phantasia) and assent (sunkatathesis):

Of appearances, some are rational and some nonrational: rational are those of rational animals, non- rational, of animals that are nonrational. The rational ones are thoughts, the nonrational have no name. (D.L. .)

Annas characterises the difference as one of conceptualisation:

Perceiving is thinking, not the reception of raw data. Perceiving and thinking are not separate fac- ulties; for humans there is no way of taking in information about the world that does not involve thinking.(: )

Such a distinction is also clearly present in Origen:

Things with soul move themselves when an appearance (phantasia) occurs and calls forth an impulse (hormē). Now in some animals when appearances occur and call forth impulse it is their nature which deals with appearances (phusis phantastikē) in an orderly way and moves the impulse, as when in the spider there is an appearance of web spinning and impulse follows to spin a web; it is its nature which deals with appearances that calls it forth to this in an orderly way,and nothing else in the animal other than this nature dealing with appearances has been convinced. […] But a rational animal (logikon zōion) has reason as well as nature dealing with appearances, reason which judges the appearances and nullifies some while accepting others, so as to lead the animal in accordance with them. (De princ. ..-, quoted after Annas (: –))

Such phantasiai are now understood as being the results of perception, what is received by perceivers in virtue of their perceiving, as an “imprint” (tupōsis) in the soul. Different Stoics understand the nature of that imprint in different ways, but all agree that there is one:

Cleanthes understood the imprint in terms of recess and projection, just like the imprint of seals on wax, but Chrysippus thought such a thing absurd; for firstly, he says, when the mind is presented with (phantasioumenēs) a triangular and a square thing on one occasion, then the same body will at one and the same time have [round it] triangular and square shapes simultaneously – or even circular – which is absurd. […] So he himself supposed that “imprint” was used by Zeno in place of “alteration.” so

that the definition is like this: “An appearance is an alteration of the soul”; so we no longer get the absurdity of the same body’s receiving far too many alterations at one and the same time when many appearances come together in us; for just as the air receives countlessly many different blows together when many people are talking at once, and immediately sustains many alterations, so the hēgemonikon [the dominant part of the perceptual soul] will undergo something analogous to this when it receives various presentations (poikilōs phantasioumenon). (Sextus Empiricus, Math. .-, cited after Annas (: ))

These presentations are now understood as re-presentations, i.e. as sayables, lekta:

A lekton subsists in a way corresponding to a rational appearance, and a rational appearance is one where one can establish in language the object of the appearance. (Sextus Empiricus, Math. .., cited after Annas (: ))

Do we here already have a belief-theory of perception? Annas thinks so:

Thus in a perception there is an appearance which contains or realizes content, and an assent in the person’s soul to a statement which articulates this content. Perception is a taking in of, and recognition of, information; to have a perception is to assent to the truth of a statement which is true or false, and thus to have a corresponding belief. (: )

.. Seeables and Sayables

STILL TO COME

.. Living by Appearances

Pyrrhonian sceptics faced a similar question. Diogenes Laertius IX - objects to skeptics professing to live without beliefs that they have to “reject what is apparent”. Sextus replies:

…we do not overturn anything which leads us, without our willing it, to assent in accordance with a passive appearance – and these things are precisely what is apparent. When we investigate whether existing things are such as they appear, we grant that they appear, and what we investigate is not what is apparent but what is said about what is apparent – and this is different from investigating what is apparent itself. (PH I x, : )

. Against Reification

.. Speaking of the Looks of Things

.. Acts and their Objects

.. Modes and their Contents

. Veridicality Conditions

.. ..

.. ..

.. ..

Chapter

The Good and the Pleasant

. Emotion and Valence

.. Feeling the Danger

Suppose I am standing in a Roman amphitheatre, about to be eaten by a lion. Looking at the hungry, angry, ferocious lion before me, I note certain characteristics, of me, the lion and the situation, that induce me to believe that the lion is dangerous (to me, in this situation) and also to see it as dangerous. In feeling fear of the lion, the world, and some aspect of it in particular, is presented to me as dangerous. By feeling the fear, not just as a consequence of it, I am motivated to flee – this marks a contrast to the judgement that the lion is dangerous. By feeling the fear, not just as a consequence of it, I am in contact with danger and in a position to acquire the concept of danger – this marks a contrast to the perception of the lion as dangerous.

Aspect-perception involves a curious doubling, or re-duplication, of content. There is, one the one hand, the (supposedly) real danger in front of me facing me in the shape of a ferocious lion; this is the danger my fear is directed towards and intentionally represents. On the other hand, danger is also an aspect of how the world is given to me: (the feeling of) danger is an aspect of my situation, a feature of how I am when faced with the lion. Though not much more than a metaphor, representation-as brings out this double aspect: I represent the dangerous lion as dangerous, and thus correctly. In Aquinas’ theory, perception, the central and paradigmatic case of representation, is the combination of two processes, which together constitute the uptake of form without matter. There is, (i), the the getting of the (inner) form: by the application of a representational faculty to something out there in the world, something inner is created (the form of the thing outside in its ‘intentional existence’, ie. a structural property of the brain); by being in such a state, we locate ourselves with respect to things outside of us. There is also, (ii), the the seeing of the (outer) form: our brain being in such a state causes us to see the form, ie. to reach out in more or less successful ways to the things we take to be of this form; as a result of this activity, the form out there gets to be seen. It gets to be seen as such-and-such, and in certain ways, in virtue of the way the seeing is done. In different types of representation, the importance of these components varies. In sensation, the first of these components is

much more important than the second; the act-aspect is mere awareness, which, as Brentano correctly recognised, is often already contained in the representational state. With imagination, emotion, judging, thinking, and mere supposing – states which do not only have representational properties, but exhibit intentionality as well – the second, mind-to-world aspect becomes more important. Even with such ‘higher’ forms of cognition, however, the first component never completely disappears: all such mental acts are properly understood as reactions. Like perception and emotion, belief is a reaction: the representational state is an entertaining of a thought, and the mental act is a type of Stoic assent (in the case of belief; dissent in the case of disbelief). To believe, Augustine says, is ‘cum assensione cogitare’ (de praed. sanct. ). In the same way that thinking, by the ‘affirmative’ mode, gives beliefs, it may, by the ‘interrogative’ mode, give questions, and by the ‘mode of conjecture’, suppositions. What is belief a reaction to? Brentano may help us here, drawing our attention to the possibility that inner representation is containment (i.e., works roughly in the way in which a quotation name represents what it quotes by surrounding it by quotation marks). Quoting De Anima III. , p. , b, , he remarks that a doubling of intentional objects is only needed if we individuate them by the representational states, but not if we individuate them (correctly) by the mental acts directed at it:

“…the presentation [Vorstellung] of the sound is connected with the presentation of the presentation of the sound in such a peculiarly intimate way that the former, by obtaining [indem sie besteht], contributes inwardly to the being of the other [innerlich zum Sein der anderen beiträgt].” (bk. , ch. , §, : )

The resulting picture is the following: There is thinking going on, and some of it is partitioned into a belief of a certain content by an attitude of assent the subject directs at its own mental state. Such a belief, however, is not a representational vehicle in the traditional sense, that allows the subject to represent something. Rather, the belief, as a mental particular (and thus distinguished from the state of believing), is itself produced by (and thus posterior to) the representing of form. The attitude of assent directed at the immanent object is at the same time, and by the same act, a representation of itself as a psychological state, i.e. a belief. It thus projects its own form onto an amorphous mental process of thinking. There is a peculiar interweaving [“eigentümliche Verwebung”, : ] of the object of inner presentation (the belief) and the presentation itself (the thinking), and both belong to the same act, the believing.

.. Emotions and their Correctness Conditions

Success- or, more generally, correctness conditions explain why emotions such as fear represent the world without being propositional attitudes. Emotions, at least in standard cases, stand ‘between’ perceptions and judgements with respect how ‘explicitly’ representational they are and thus afford a model to understand how something may have a mind-to-world direction of fit without being a description of the world that is either true or false. Contrary to some, we thus distinguish emotions both from axiological judgements (“This is dangerous”) and from axiological perceptions (seeing the dog as dangerous). Suppose I am standing in a Roman amphitheatre, about to be eaten by a lion. Looking at the hungry, angry, ferocious lion before me, I note certain characteristics, of me, the lion and the situation, that induce me to believe that the lion is dangerous (to me, in this situation) and also to see it as dangerous. In feeling fear of the lion, the world, and some aspect of it in particular, is presented to me as dangerous. By feeling the fear, not just as a consequence of it, I am motivated to flee – this marks a contrast to the judgement that the lion is dangerous. By feeling the fear, not just as a consequence of it, I am in contact with danger and in a position to acquire the concept of danger – this marks a contrast to the perception of the lion as dangerous. Taking, with Aquinas, danger to be the formal object of fear, explains this double character of emotions: that they side with perceptions in being themselves motivational, but with judgements in making demands on the world that go beyond how things appear to us to be. That the formal object of fear, danger, is something I evaluate negatively

. I am setting aside here other, non-standard, types of emotional episodes, such as they occur in cases of cinema emotions, vicarious emotions (such as those, perhaps, involved in empathy), emotional contagion etc.

in addition explains why fear is a negative emotion, an emotion I would rather not have. That I would rather not have it is a consequence of my metaphysically and logically prior desire to live in a world that is safe (for me); given, however, that I live in a dangerous world, fear is often the appropriate emotion to have. Emotions share with perceptual experiences that they may persist even if they are known to be incorrect: even if I know that my fear of the lion is based on a hallucination, I still feel it, in the same way I still ‘see’ the lines in the Muller-Lyer illusion as of different length even when I have measured them and found out not to be. They differ from perceptions, however, in that they may be incorrect in two different ways, by being based on an illusionary or hallucinatory perception (the lion is in a cage or there is no lion at all), but also by leading us astray about the axiological properties of their direct object, representing a harmless cat or a robot as dangerous. In this latter case, I not only, as in an illusionary or hallucinatory perceptual experience, have an inaccurate representation of the lion, but I also have the wrong reaction to it – whereas, at least in cases of persistent and not self-induced illusions, there is nothing wrong with what I do as a perceiver when I ‘see’ the stick half-immersed in water as bent (how else should I see it?). In some cases, moreover, emotions can be perfectly appropriate even in the absence of any direct object, as with Angst or the fear of a forthcoming exam (which is itself present, not just an anticipation of some future episode of fear). In such cases at least, whether or not an emotion is appropriate not just as a reaction to the belief but as a reaction to (axiological aspects of) the world depends on the presence or absence of its formal object: it is in this sense that the formal object determines the correctness conditions of the emotion. Like perceptions, emotions provide prima facie and pro tanto reasons for axiological beliefs, whether or not they are themselves justified by those beliefs being true.The axiological beliefs are intentionally about the formal object: they represent it as being present even when it is not, e.g. when I inappropriate fear the harmless house-spider in front of me. Insofar as the belief inherits this directedness from the emotion that bases it, the emotion itself exhibits such intentionality, or ‘presence-in-absence’ of the formal object. It is this intentional presence of the formal object that explains why it is irrational to continue to have ‘recalcitrant’ emotions, i.e. emotions known to be inappropriate. It is irrational because the combination of my knowledge and ofmyemotionalstatecommitmetotakingtheworldinawayitcannotbe. Wehaveherea sui generis incompatibility of different attitudes, which is not an incompatibility of their contents; the axiological properties that justify my emotional reaction and to which, as their intentional objects, my emotions give me access to are not represented by them, nor by their content. Even though not represented, such properties are made available to beliefs.That (J) and (K) depend in the same way on (I) can thus explain why (J) justifies (K): (I) The lion is dangerous. (J) I fear the lion. (K) I believe that the lion is dangerous. Because (J), for its correctness, depends on (I), basing (K) on (J) allows for the transmission of warrant: the justifica- tion I have for (J) may be credited to (K), and may account for my knowledge that the lion is dangerous. Emotions, in this way, are intentional (directed towards some formal object, even when the formal object is absent and the emotion inappropriate), about an axiological feature of the world, without thereby ascribing or predicating of some worldly item that it exemplifies the feature in question.

. I understand this last claim contrastively: given that I react emotionally to the situation at all, fear is one of the right such reactions. We will return to “appropriate” below. . I leave open the additional claim that emotions are individuated by their formal objects rather than, e.g., by how they feel or by how it is to have them.

.. Emotions and Values

. Desire

.. The Aboutness of Desire

.. The Guise of the Good

.. The Value of Emotions

. Correctness Conditions

.. Correctness Conditions and Intentionality two objections to the theory of correctness conditions: correctness conditions are not truth-evaluable and one central type of correctness condition, for judgement and belief, is superfluous. Correctness conditions for psychological states that exhibit intentionality (from Mulligan (: )):

x desires to Fx x ought to F “Tun-sollen” x wishes that p It ought to be the case that p “Sein-sollen” x values y y is valuable x admires y y is admirable x fears that y y is dangerous x ‘values’ that p That p is valuable, is a “Wertverhalt” x regrets that p It is regrettable that p x is ashamed that p It is shameful that p x prefers y to z y is better than z x judges (believes) that p The state of affairs that p obtains The proposition that p is true x conjectures that p It is probable that p x has an interrogative attitude toward p It is questionable whether p x doubts whether p It is doubtful whether p x is certain that p It is certain that p

Talking about correctness conditions includes an explanatory claim:

If xcorrectly judges that p, then x correctly judges that p because the state of affairs that p obtains. If xcorrectly judges that p, then x correctly judges that p because the proposition that p is true. If x correctly conjectures that pä, then x correctly conjectures that p because it is probable that p

While mental states and acts or their contents represent (conceptually) their satisfaction conditions, at least some of them do not represent their correctness conditions. To undergo an emotion is indeed to stand in an intentional relation to value, but the relation is not belief nor does it involve any representation (thought) of value.

.. Intrinsicness and Relationality

As commonly introduced, intrinsic properties of a . are / account for / ground ‘how a is by itself’, are exemplified by a ‘in virtue of the way it is in itself ’; . make for genuine similarity, are ‘non-disjunctive’, have ‘non-gerrymandered’ extensions; . are shared by a and its duplicates / replicas / perfect copies.

It is is the first feature that excludes having a brother, for that property of me involves my brother; it is the second that excludes grue, for being grue is being green and examined before t or being blue and examine at or after t; it is the third that excludes not being accompanied by a unicorn, for that property could be lacked by a perfect replica of the entire universe, existing in a larger world that also contains some unicorn. In some metaphysical systems, intrinsic properties are supposed to play certain theoretical roles. They . are qualitative natures of combinatorial units; . make for real, as opposed to Cambridge change; . do not entail, nor are entailed by the existence of any other things wholly distinct from their bearers. Combinatorial units – “substances” in one sense of this term – are the elements that are recombined when describ- ing alternative possibilities. Socrates becoming taller than Simmias is a real change for Socrates when Socrates grows, but a Cambridge change for him if Simmias becomes smaller. A (neo-)Humean ban on necessary connec- tions between distinct existences is restricted to their intrinsic properties: that my being such that Socrates is white entails that he is white is not a reason to deny me that property, but rather a reason to think it is not non-relational. It has turned out surprisingly difficult to turn this intuitive notion into a precise definition. According to the most discussed proposal, by Lewis & Langton (), a property is intrinsic if it does not distinguish between things that have the same pure, non-disjunctive and non-co-disjunctive properties that are independent of loneliness and of accompaniment. At least prima facie, the relational/non-relational contrast is different. Non-relational properties of a . do not ‘essentially mention’ other things than a; . do not ‘stem from’ metaphysically / conceptually / explanatorily prior relations a has; . are ‘genuinely monadic’; Saying of me that I am Michael’s brother ‘essentially mentions’ Michael – which is why it is an ascription of an ‘impure’ property. It is also a derivative property, presupposing (and mentioning) the prior relation of brotherhood that obtains between me and him, and it is for this reason not ‘genuinely monadic’, but a de-relativisation of a relation, i.e. a relational property. Relational properties are also supposed to play certain theoretical roles. Non-relational properties . are wholly qualitative: their nature is exhausted by how the things that have them are; . are non-haecceitistic: may be shared by distinct indiscernibles; . are pure, i.e. entirely general, i.e. do not involve particulars. If Michael has an indiscernible twin, everything qualitative ascribable to me by predicating “…is Michael’s brother” could still be true of me, and I could still lack the property (being rather the brother of his twin); conversely, my qualitatively identical twin could lack it if Michael does not coexist with him; hence, a general description of the world, not involving names or referential devices, will not fix whether or not I am Michael’s brother (rather than the brother of his indiscernible twin). Attempts at defining relationality of properties have mainly focussed on purity. A property P is called “impure” iff there is a relation R and a y, such that whenever anything, x, has the property, it also stands in relation R to y. Metaphysically, relational properties have been characterised as properties that are individuated with reference to relations (Hochberg : ): to say that, generally and as a matter of logical truth, if a = b, then λx(aRx) = λx(bRx), we need to quantify over relations.

. A property is pure iff its exemplification does not imply the existence of anything else than the thing exemplifying it. Something is accompanied iff it does not coexist with a contingent wholly distinct thing and it is lonely iff it coexists only with its proper parts (if it has any). A property is independent of loneliness (accompaniment) iff it is both possible that is is had and that it is lacked by a lonely (accompanied) thing. A property is disjunctive iff it can be expressed by a disjunctive predicate but is not natural and much less natural than either of its disjuncts. The pure, non-disjunctive and non-co-disjunctive properties independent of loneliness and accompaniment are called “basic intrinsic” by Lewis and Langton. Their definition says that a property is intrinsic iff it supervenes on basic intrinsic properties, or, equivalently, iff it never differs between duplicates (where two things are duplicates iff they have the same basic intrinsic properties). . The reason why loving-Superman and loving-Clark-Kent is one and the same property (and Lois Lane, as a matter of logic, exemplifies one iff she exemplifies the other), is that Superman is Clark Kent; therefore, the properties are not atomic, but derelativisations of the prior relation of loving.

The two distinctions crosscut. Here are some examples of the relational intrinsic: • having a as a part: this is intrinsic because it only turns on how its bearer is by itself, but relational, because it mentions a (and not it’s duplicate!) as its part. • the value of Diana’s dress: this is its intrinsic value because it is not determined in terms of what you can buy for it, why you want to have it or any other external determinants, but is the value it has in virtue of what it is in itself, i.e. in virtue of being Diana’s dress; it is still relational, however, because a dress worn by her indiscernible twin would be (much?) less valuable. • being of a crime of some punishment: this is intrinsic if the punishment is reserved for this crime, and intimately depending on it, as e.g. a specifically destined act of reparation is; it is still relational, however, for the punishment relates the punished to the crime they committed: qualitatively the same punishment for another crime would relate the punished to something else. Here are some examples of the non-relational extrinsic: • not being accompanied by a unicorn: this is an extrinsic property of everything there is because everything there is could be just as it is and a unicorn exist in addition; it is not a relational property, however, because no relation can relate you to something that does not exist. • being all there is: this is extrinsic by the same token, but also non-relational, unless you posit ontologically dubious and (arguably) paradoxical existing ‘totality states of affairs’. • being surprising of an event: this is extrinsic, because an event has it only if embedded in a certain context, within which it is surprising; it is non-relational, however, because you can wholly and completely appreciate this characteristic without remembering or otherwise knowing what has gone before. According to what Chisholm (: ) calls “Brentano’s Thesis” – that intentionality is the mark of the mental – intentional properties are extrinsic, but non-relational. They are extrinsic, because they are signs, but non- relational, because they are characterised by “intentional inexistence”: psychological states may exist even in the absence of what they are about. The representationality of some properties has to be sharply distinguished from their intentionality. A property of something is intentional iff it is taken to be about something else than itself. It is so taken to be if we attribute to it conditions under which it may be said to be correct. Correctness conditions specify the intentional content, but – being conditions – do not themselves require this content to be satisfied. Because they are outward-directed, and cannot be accounted for without reference to their intentional objects, intentional states are extrinsic: they are what they are in virtue of participating in a complex process, which not only involves their objects and their bearer, but also a process of interpretation or understanding. Intentionality is the flip-side of representation: whereas representational properties are intrinsic, but relational, intentional properties are extrinsic, but non-relational. Taking up an attitude towards the cognitive base turns the latter’s relatum into the former’s ‘intentional object’. Such a ‘conversion’ of the relatum of a representational state into the intentional object on which an intentional state depends for its existence without being related to it, is what happens in Kantian ‘synthesis’: when I see a thing as red and white, redness and whiteness hang together by being aspects of the one thing my perception relates me to; when I, however, only imagine a red and white thing, the link can not come from the object alone – it must be ‘constructed’ by my faculty of imagination, and it is so constructed by my imagining one thing as both red and white. This intentional object will therefore be extrinsic, depending for its existence on my act of taking my representational state in a certain way. Virtually any state can be used as a representation: you can decide to use a red flag to represent danger and you can take your aching muscles as a sign that you should not have walked that far. Such representational contents, however, derive from contingent dispositions to interpret the relevant states in certain ways. Since you could equally well take red flags or aching muscles to represent something different (that the communists are marching

. It is only as an analysis of intentionality, not of representationality, I think, that Aristotle’s theory of thoughts being likenesses of objects has any plausibility . . This has been particularly stressed by Charles W.Morris: “The properties of being a sign, a designatum, an interpreter, or an interpretant are relational properties which things take on by participating in the functional process of semiosis.” (: )

or that your work-out has been successful, for example), neither the red flag nor the aching muscles have their representational content intrinsically. Independently of what properties they intrinsically have, we use them as signs for other things, we bestow on them the representational powers they have. Not all powers of representation, however, are derivative in this sense. Our propositional attitudes and our perceptions, for example, do not seem to derive their representational contents from other states by the use we make of them: they have their content originally, not in virtue of being interpreted by some other mental state. Many things may thus be said to have content, but most of them do so indirectly: they have content in virtue, for example, of having been produced in a certain way or with certain intentions, or of standing in some relation to other things that have content. The most important such relation is that of some things expressing other things. It is in virtue of expressing my beliefs that my utterances have content, and – subject to certain constraints – the beliefs expressed determine what content they have. That some representational properties are exemplified intrinsically by some things follows from the following ar- gument: (i) Some things have representational properties. (ii) If something exemplifies a representational property extrinsically, it does so in virtue of there being something else that bestows it with this representational property. (iii) In order for something to bestow something else with a representational property, the first thing needs to exemplify this representational property itself. (iv) The transmission of representational powers can neither go on forever, nor go in circle: it must be started by something. (v) A thing that has a representational property that is not bestowed upon it by something else exemplifies it intrinsically. Take the representational property that rabbits are present, exemplified by my utterance of “Lo, a rabbit”. It is exemplified extrinsically: my utterance could be just as it is by itself and express another belief, or no belief at all. By (ii), this representational property is bestowed upon my utterance by something else – something which, by (iii), has it itself. By (iv), we conclude that the regress must stop, and by (v) we know that it must stop with something that intrinsically means that rabbits are present. Representational properties like meaning that Fa, representing a to be F or thinking of a as F are intrinsically exemplified by some thing x iff x exemplifies the property independently of how matters stand with respect to other things than x – no further properties have to be exemplified for other things for my thought, e.g., to represent a to be F. Even when they are exemplified intrinsically, however, representational properties are still relational: they connect their bearers to the things they are about. If my thought, for example, represents a to be F, it stands in the relation of aboutness to a and in the predication relation to the universal F. It is in virtue of these relations that my thought can stand in for a’s being F, and be in some sense a substitute of this external fact.

.. The Category of the Intrinsic-Relational

. Different accounts of this relation of standing-in have been proposed, from Aristotle’s ‘being-a-token-of’ – “It is not possible to converse by bringing in the objects themselves, but instead of the objects we use words as tokens”, Sophistici Elenchi , a- – to the scholastic modes of objective existence.

Chapter

From Intelligible Species to Ideas

. The medieval problem: intentionality

The medievals both broadened and focussed the problem, extending it to thoughts, linguistic signs and even (the contents of) the knowledge of God, but also concentrated largely on the change in the perceiver rather than its efficient cause. Perler (: –) speaks of a confluence of the following traditions, which all came to center on “intentio”: • The interpretation of the De Anima claim that perception of a is the taking in of the form of a without its matter. Albert the Great claims that the form (ie. a colour, a sound) is taken in not in its ‘material being’ (esse materiale), but in its ‘spiritual being’ (esse spirituale), or rather just its intentio, which, with Aquinas, became the claim that the form has ‘intentional being’ in the soul. • The interpretation of the De Interpretatione claim that spoken and written words are signs for ‘impressions in the soul’ (passiones animae), which are themselves representations of external things, identified with the ‘words in the heart’ (verba in corde) that Augustine had claimed to be prior to spoken or written words, and which were – via the arabic translations of al-Farabi – translated as ‘intentiones’. • The interpretation of the ‘perspectivalist’ tradition of optics, which explains visual perception by the trans- portation, normally through air, of so-called ‘species in medio’, which Roger Bacon claimed to be ‘inten- tiones’ of the thing itself which have ‘weaker being’. • The interpretation of Augustine’s claim that God foreknows in virtue of having eternal ideas which (only) have an ‘intelligible’ or ‘intentional’ mode of being. The species theory, most clearly in Aquinas, sides with the ‘spiritualist’ interpretation of Aristotle:

Est autem duplex immutatio: una naturalis, et alia spiritualis. Naturalis quidem, secundum quod forma immutantis recipitur in immutato secundum esse naturale, sicut calor in calefacto. Spiritualis autem, secundum quod forma immutantis recipitur in immutato secundum esse spirituale; ut forma coloris in pupilla, qua not fit per hoc colorata. Ad operationem autem sensus requiritur immutatio spiritualis, per quam intentio formae sensibilis fiat in organo sensus. Alioquin, si sola immutatio nat- uralis sufficeret ad sentiendum, omnia corpora naturalia sentirent dum alterantur. (ST, I, q. , art. , corp.) Es gibt aber eine zweifache Veränderung: eine natürliche und eine geistige. Eine natürliche erfolgt, wenn die Form des Verändernden mit einem natürlichen Sein im Veränderten aufgenommen wird, so wie die Hitze im Erhitzten aufgenommen wird. Eine geistige aber erfolgt, wenn die Form des Verändernden mit einem geistigen Sein im Veränderten aufgenommen wird, so wie die Form der Farbe in der Pupille aufgenommen wird, die ja dadurch nicht gefärbt wird. Für die Tätigkeit des Wahrnehmungssinns ist aber eine geistige Veränderung erforderlich, durch die eine Intention der wahrnehmbaren Form im Wahrnehmungsorgan entsteht. Andernfalls, wenn eine natürliche Verän-

derung allein zum Wahrnehmen genügte, würden alle natürlichen Körper wahrnehmen, solange sie verändert werden. (translation by Perler (: ))

Aquinas’ central claim is that cognition is assimilation: “…omnis cognitio est per assimilationem scientis ad scitum” (De veritate, q. , art. ). By ‘assimilating’ itself to its object, the intellect does not leave the mental realm, however. It replicates mind-independent reality, but does not literally become it. Its reality is the replicated one, intentionality (being directed towards, referring to) becomes self-reference, cognition becomes self-knowledge:

“…intelligibile in actu est intellectus in actu” (ST I, q. , art. , corp) “…intellectum in actu est intellectus in actu…” (ST I, q. ,art. ,ad )

For there to be cognition, three conditions are individually necessary and jointly sufficient: doubling of forms whatever cognises has two forms: its own and the form of that which it cognises; intentional being the form that is ‘taken up’ in cognition has a special way of being: esse intentionale or esse spirituale, as opposed to esse naturale; perception by the organ the form must be taken up by something that is naturally perceiving; What I find striking here is that the direction of explanation is completely reversed: rather than starting with a special type of mind-independent objects (proper sensibles) and explain perception as what is naturally relative to them, we start with an independently demarcated class of beings – minds – and ask how they may be perceptive of other things, i.e. what it is for them to have intentional states. Among these things – assuming they have the right organs and are capable of receiving ‘extra’ forms –, the question of intentionality becomes the question what it takes for some form to have spiritual/intentional being. It is with respect to the contrast between ‘natural’ and spiritual/intentional being that different senses are distin- guished: the less ‘natural’ change some sense undergoes in cognising, the ‘higher’ it is. How did this change occur? The ‘spiritual’ or ‘intentional’ mode of being is then characterised as relational:

Unter dem natürlichen Sein ist die Existenzweise zu verstehen, die ein Engel an sich hat, unabhängig von irgendeiner kognitiven Relation zu etwas anderem. Im Gegensatz dazu ist unter dem geistigen Sein jene Existenzweise zu verstehen, die ein Engel hat, insofern er von einem anderen Engel mit Hilfe einer Species erkannt wird, also die Existenzweise in einer kognitiven Relation zu etwas anderem. Kurz gesagt: Der Engel hat an sich immaterielles und natürliches Sein; in einer kognitiven Relation hat er im- materielles und geistiges Sein. (Perler : )

Relationality is then ascribed to what is cognised, as that which is taken up is relative to what is taking it up. Here we have a complete reversal of the Aristotelian picture: it is not sight that is relative to colours, but colours that are said to be relative to sight! King () distinguishes four theses about representationalism discussed in the Middle Ages: . Conformality: The mental representation and the represented have the same form. . Likeness: The mental representation resembles, or is a likeness, of the represented item. . Causal covariance: The mental representation is caused by the represented item. . Sign: The mental representation signifies the represented item.

. ST I, q. , art. I, corp.: “…cognoscentia a non cognoscentibus in hoc distinguuntur, quia non cognoscentia nihil habent nisi formam suam tantum; sed cognoscens natum est habere formam etiam rei alterius…” . Sentencia libri De anima II, : “…et ideo forma recipitur in paciente sine materia in quantum paciens assimilatur agenti secundum formam et non secundum materiam; et per hunc modum sensus recipit formam sine materia, quia alterius modi esse habet forma in sensu et in re sensibili: nam in re sensibili habet esse naturale, in sensu autem habet esse intentionale siue spirituale.” . Sentencia libri De anima II, : “…odorare est sic pati aliquid ab odore quod senciat odorem, aer autem non sic patitur ut senciat, quia non habet potenciam sensitivam, set sic patitur ut sit sensibilis, in quantum scilicet est medium in sensu.”

To explain the non-symmetry of representation, conformality theory needs additional resources: the most promis- ing such account – Aquinas’: that it’s the mode of the form’s presence that makes the difference – says that the form is present in the intellect ‘spiritually’ or ‘intellectually’, ie. without being the form of it – as it is present e.g. in the intervening air. Aquinas also bites the bullet to say that the air really ‘is perceptive of’ the colour. Duns Scotus’ version of the conformality account holds that the form that is the thinking (an accident of the intellect) is not the same as the form that determines what the thinking is about (by conformality), but is directed at or includes the latter. The act/content distinction he thus introduces is between the ‘subjectively’ present (accident of) thinking and the ‘objectively’ present external form. But what is these esse objectivum (or esse representativum, esse deminutum) forms of external things have when they are in our minds? Likeness can either be taken literal or pictorial: in the first sense, it amounts to sameness of quality, and makes the soul literally coloured; in the second sense, it presupposes an account of representation by pictures: what aspects of the pictures are taken to represent aspects of the thing pictured? Introducing (primitive?) correspondence rules will not help:

When a transformation-rule is applied to some item, the result is, ideally, something with features that systematically correspond to properties of the original item. What is it that such transformation-rule preserves? Well, the natural answer is: form. (Nowadays, people say ‘structure’ but that’s an acceptable translation of forma.)” (King : )

Ockham realised, however, that even isomorphism + conformity will not help with the singularity of representation: “Ockham’s point is that images, conscious or not, are by their nature applicable to many – that the correspondence- rules aren’t guaranteed to have unique inverses (i.e. the rules don’t in general yield one-to-one mappings).” (King : ) The causal covariance account needs to be supplemented by the ‘sign’ account to provide determinate enough contents. Their combination yields a “mediaeval version of functionalism, the idea that determinate content is fully specified by inputs (covariance) and outputs (linguistic role)”, where this is taken to be their place in the general categories of a mental language. Once the mental language and its network is in place, there is no need for intelligible species any more: mental representation can just be the mental events themselves.

. Cogitationes: Material for Doubt

.. The Cartesian Project

Descartes, as it is well known, wanted to rebuild the sciences on a secure foundation he tried to uncover in meta- physics. In his Meditationes de Prima Philosophiae he gives us an account of what he takes to be the “premier principe de la philosophie” (AT VI 22−23) and what I will henceforth call the Cogito. The privilege of the Cogito resides in its resistance to what Descartes calls “scepticism”. Descartes’ sceptic is not only someone who doubts that we have any knowledge, i.e. any beliefs meeting a certain given standard of justification, but what one might call a Pyrrhonian sceptic, someone who doubts the existence of a standard of knowledge, not only of beliefs meeting such a standard. Descartes thought it worth while to (try to) prove the sceptic wrong not because he doubted that we have any knowledge. Instead he aimed to show that and how we can defend our knowledge claims against someone who questions their legitimacy. He wanted to prove our entitlement to them by showing that they are not only true, but justifiedly taken to be so. Not satisfied with the mere fact that the sceptic is wrong – because there is something we know – he wanted to show why the sceptic is wrong. To achieve this, he had to show more than just that we know something; he had to show that we know that we know something, i.e. that we know that the sceptic is wrong. To do so, he had to show not only that there

. All references to Descartes are to the standard edition by Adam & Tannery (). . The distinction is Sextus’s: “Pyrrhonian skepticism is a more radical position [than academic skepticism]. It holds that the academics are dogmatic even to affirm the impossibility of certain knowledge and denies that propositions differ in their probability. A judgement of probability can be made only by someone who possesses a standard of knowledge and truth. But the existence of such a standard is just what the pyrrhonians question. Since probability cannot guide our choices, they propose to follow custom.” (Curley : )

is something we know, but that there is a knowledge claim we can defend even against the most hard-headed of sceptics. This is the task of the Cogito: to prove the untenability of universal doubt by refuting the evil demon hypothesis and thereby to establish a standard of knowledge. To induce, in us and in the Cartesian thinker, the most general and far-reaching doubt one may entertain (AT VII 13) and thus to uncover the most certain of our beliefs, Descartes mounts a sceptical argument in three steps in Meditation One, leading to the ‘bracketing’ (on grounds of dubitability) of larger and larger classes of truths, for reasons Descartes takes to be valid and rational (AT VII 26−30). In a first step, the Cartesian thinker of the Meditations notes the existence of sensory illusions:

Nempe quidquid hactenus ut maxime verum admisi, vel a sensibus, Tout ce que j’ay receu iusqu’à present pour le plus vray & assuré, je vel per sensus accepi; hos autem interdum fallere deprehendi, ac pru- l’ay appris des sens, ou par les sens: or j’ay quelquefois éprouvé que dentiae est nunquam illis plane confidere qui nos vel semel deceperunt. ces sens estaient trompeurs, & il est de la prudence de ne se fier jamais (AT VII 15−18) entierement à ceux qui nous ont une fois trompez. (AT IX/I )

The conclusion drawn in the last clause, never to rely on sense perception (“a sensibus”) nor on testimony (“per sensus”), may seem a little too quick. It makes sense, however, if we place it in its epistemological context: if my senses are not always reliable, then this method of forming beliefs is not fool-proof; barring further information, I must consider it as at any instant potentially unreliable. I am not justified in assuming the reliability of my senses unless I have at my disposal a way of knowing that they are reliable at the times when they are. I cannot, however, detect such optimal sensory conditions without relying on my senses: bracketing their evidence, I cannot establish their reliability. This is why I am forced to conclude A from A: A It is possible that my senses deceive me. A It is always possible that my senses deceive me. The Cartesian thinker has to go further than even this, however. He cannot justifiably assert of any instance of sense perception that his senses did not deceive him on that occasion. To do this, he would have to know that conditions were optimal – which he cannot, if he has not prior sensory assurance that they were. Because the proposition that they conditions are optimal is itself doubtable, he must treat it as if it were false (AT VII 4−10)– he must assume the worst case scenario, that his senses always deceive him: A It is possible that my senses always deceive me. Many have doubted that the Cartesian thinker could rationally entertain something like A – if I need my senses to establish that they deceived me on another occasion, they argue, the hypothesis of permanent sense deception does not make sense. Though it is clear that Descartes is not committed to A and that neither A nor A can plausibly said to follow from A, such worries, I think, misunderstand the nature of Descartes’ project. To get a clearer picture of why the Cartesian thinker, in virtue of noticing that he cannot resist, in principled grounds, the generalisation of A to A, is warranted in supposing A, we need to say more about what Descartes means by “doubting”. Though Descartes requires reasons for doubting, he requires those to be neither true nor assertable by the Cartesian thinker. He explicitly says to Bourdin that reasons for doubting may themselves be doubtable:

Eae enim sunt satis validae rationes ad cogendum nos ut dubitemus, There may be reasons which are strong enough to compel us to doubt, quae ipsae dubiae sunt, nec proinde retinendae, ut jam supra notatum even though these reasons are themselves doubtful, and hence are not est. Atque validae quidem sunt, quandiu nullas alias habemus, quae to be retained later on, as I have just pointed out. The reasons are dubitationem tollendo certitudinem inducant. (AT VII 27-3) strong so long as we have no others which produce certainty by re- Verissimum enim est, nihil admittendum esse ut verum, quod non pos- moving the doubt. (Descartes : ) simus probare esse verum, cùm de eo statuendo vel affirmando quaes- The maxim ‘We should not admit anything as true unless we can prove tio est; sed, cùm tantùm de effodiendo vel abdicando, sufficit quòd it is true’ is perfectly correct when it is a question of establishing or af- suspicemur. (AT VII 18−22) firming some proposition; but when it is merely a matter of renounc- ing a belief (or digging out a trench), then mere suspicion is all that is required. (Descartes : )

. The transition is even hastier in the Discours: “Ainsi, à cause que nos sens nous trompent quelquefois, je voulus supposer qu’il n’y avait aucune chose qui fût telle qu’ils nous la font imaginer.” (AT VI 30-3)

They are reasons for doubting in the sense that they present the Cartesian thinker with scenarios he is in no position to exclude. The Cartesian thinker cannot exclude the possibility of universal sense deception merely on the basis of the observation that the assertability of “my senses deceive me” requires some veridical sense perceptions. But neither can he deny their veridicality – when they are reliable – justify my knowledge claims based on them. So the sceptic needs to take another step. This second step is known as the “dream argument”:

Quasi scilicet non recorder a similibus etiam cogitationibus me aliàs Mais, en y pensant soigneusement, je me ressouviens d’avoir esté sou- in somnis fuisse delusum; quae dum cogito attentius, tam plane video vent trompé, lors que je dormais, par de semblables illusions. Et nunquam certis indiciis vigiliam a somno posse distingui, ut obstu- m’arrestant sur cette pensée, je voy si manifestement qu’il n’y a point pescam, & fere hic ipse stupor mihi opinionem somni confirmet. (AT d’indices concluans, ny de marques, assez certaines par où j’en puisse VII 17−22) distinguer nettement la veille d’avec le sommeil, que j’en suis tout etonné; & mon estonnement est tel, qu’il est presque capable de me persuader que je dors. (AT IX/I )

Contrary to the first, this second step rules out the possibility that there are conditions under which knowledge claims about material objects are justified. Trusting my senses to establish their reliability is a perfectly respectable procedure; but I am never entitled to trust my senses to answer the question whether I might be sleeping now. I can only know that p by present sensory evidence if certain conditions C hold – but I can find out whether C, for the conditions C′ which have to obtain for that may differ from C. Not so with dreaming: I cannot find out that I am dreaming – for if I am dreaming, I cannot find out anything whatsoever. As I can never find out that I am dreaming (at that time), I can never know that I am awake (at that time), not being able to rule out a possibility I would not detect if it were actual. Whereas the argument from sensory illusion undermined any knowledge claims based on sensory evidence, the dream argument carries the doubt further by making any knowledge about material things unclaimable: it thereby undermines any contingent knowledge claim whatsoever, including most examples of what has been called the contingent a priori. There is, for any such proposition, whether a priori or not, a circumstance in which it would be false and I may dream that this circumstance obtains. As with the first step, the Cartesian thinker infers B from B, and then finds no grounds to rule out B: B It is possible that I am now dreaming. B It is always possible that I am dreaming (then). B It is possible that I am always dreaming. That I am always dreaming, then, describes an epistemic possibility for the Cartesian thinker of the Meditations – not in the sense of describing a world in which he, for all he knows, might be (for he may also know that he is, from time to time, awake), but in the sense of expressing a belief he can never rule out on the basis of present evidence alone. The dream argument, then, leaves almost nothing untouched, sparing only necessary truths, which are true in all circumstances, and a fortiori true in all circumstances one may dream to obtain. Getting rid of these is the aim of the next step. The third and final step in Descartes’ argument is the malin génie hypothesis:

Verum tamen infixa quaedam est meae menti vetus opinio, Deum haec omnia non aliter quàm nunc mihi videantur existere? Imô etiam, esse qui potest omnia, & a quo talis, qualis existo, sum creatus. Unde quemadmodum judico interdum alios errare circa ea quae se perfec- autem scio illum non fecisse ut nulla plane sit terra, nullum coelum, tissime scire arbitrantur, ita ego ut fallar quoties duo & tria simul addo, nulla res extensa, nulla figura, nulla magnitudo, nullus locus, & tamen

. What then about the coherentist criterion Descartes provides us with in the Sixth Medititation? It only works given the assumption that there is a time t when I am not sleeping – and this we know only after having established the legitimacy of at least one knowledge claim: ”Potest verò Atheus colligere se vigilare ex memoriâ anteactae vitae; sed non potest scire hoc signum sufficere ut certus sit se non errare, nisi sciat se a Deo non fallente esse creatum.” (AT VII 11−14) . It is not required that my conception is a coherent one: most of us dream very weird things. It might seem difficult to imagine dreaming that I am not here now or that I am not Philipp Blum but in fact it is not. I may dream that I am dead or a multiply located universal, that I look into my passport and find another name written in it, that I look into the mirror and see nothing etc. That reasons to doubt are not required to be coherent overall scenarios, as long as they achieve their intended effect which is helping us to get rid of our beliefs, marks another respect in which pyrrhonian differs from academic scepticism.

vel numero quadrati latera, vel siquid aliud facilius fingi potest?” (AT Toutesfois il y a longtemps que j’ay dans mon esprit une certaine opin- VII 1−11) ion, qu’ilyaunDieuquipeuttout, &parquij’ayestécreéé&produit tel que je suis. Or qui me peut avoir assuré que ce Dieu n’ait point fait qu’il n’y ait aucune terre, aucun Ciel, aucun corps estendu, au- cune figure, aucune grandeur, aucun lieu, & que neantmoins j’aye les sentimens de toutes ces choses, & que tout cela ne me semble point ex- ister autrement que je le voy? Et mesme, comme je juge quelquefois que les autres se méprennent, mesme dans les choses qu’ils pensent scavoir avec le plus de certitude, il se peut faire qu’il ait voulu que je me trompe toutes les fois que je fais l’addition de deux & de trois, ou que je nombre les costez d’un carré, ou que je juge de quelque chose encore plus facile, si l’on se peut imaginer rien de plus facile que cela. (AT IX/I )

The Cartesian thinker here entertains the possibility that he might be fooled even in things he considers most certain – due to the manipulation of an omnipotent evil demon who perverts his epistemic instincts and makes him spontaneously assent to propositions which are in fact false. The evil demon hypothesis is thus more than the supposition that I might err in the most certain of my beliefs – it is the far scarier supposition that I might err in them precisely because I hold them to be the most certain. It is for this reason that the Cartesian thinker is led to infer C from C, and is in no position (yet) to exclude C: C It is possible that I am deceived in what I consider most certain. C In all my beliefs am I possibly deceived. C It is possible that I am deceived in all my beliefs. Because C states an epistemic possibility, all our knowledge claims are undermined, including our presumed knowledge of simple mathematical (AT VIII/I 8−20), logical and other necessary truths, throwing the Cartesian thinker into a state of despair. The doubt induced by consideration of the evil demon hypothesis is thus maximal. More generally, how can the Cartesian thinker doubt the certainty (or the knowledge) that p without thereby doubting p? Such doubt, if possible at all, would certainly not qualify as ‘methodical’. We will see below that, by proceeding a facilioribus ad difficiliora, the analytic method brings with it a distinction between two senses of “primary”: it starts from what is prima facie primary, i.e. most familiar to us, carrying us to what is primary in itself, the true ground on which a given body of (alleged) knowledge rests. The world, in itself, i.e. independently of our beliefs about it, has a certain epistemological structure; some truths are by themselves more evident, certain and epistemologically basic than others. The epistemic activities of humans, on the other hand, exhibit another, and possibly divergent, pattern. Custom, training and talent bring it about that some truths are more accessible, more easily graspable and more familiar to (some of) us. It is the task of the right method to make these two orderings match, i.e. to make what is most familiar in itself most familiar to us. This is the rationale of the method of doubt. The evil demon hypothesis is the suspicion that the epistemic capacity underwriting the possibility of such a tran- sition is fatally flawed: that we may in principle be incapable of matching our epistemic instincts, what we find plausible or evident, to what really is plausible or evident (in itself). The sceptical scenario, then, is that what I find most plausible might be false precisely because I find it plausible.

. That the possibility envisaged is one of a perversion of what one might call our‘epistemic instincts’, our spontaneous and almost inevitable belief in certain very simple and (seemingly) evident propositions, is made even clearer in the Discours and the Principes: “Et parce qu’il y a des hommes qui se méprennent en raisonnant, même touchant les plus simples matières de géométrie, et y font des paralogismes, jugeant que j’étais sujet a faillir autant qu’aucun autre, je rejetai comme fausses toutes les raisons que j’avais prises auparavant pour démonstrations.”(AT VI 3−9); “Dubitabimus etiam de reliquis, quae antea pro maximè certis habuimus; etiam de Mathematicis demonstrationibus, etiam de iis principiis, quae hactenus putavimus esse per se nota: tum quia vidimus aliquando nonnullos errasse in talibus, & quaedam pro certissimis ac per se notis admisisse, quae nobis falsa videbantur; tum maximè, quia audivimus esse Deum, qui potest omnia, & à quo sumus creati.” (AT VIII/I 8−15) . Descartes calls it “summa de omnibus dubitatio” (AT VII 13). Cf. also his comments on the relevant passage to Burman: “Reddit hîc [VII 22] auctor hominem tam dubium, et in tantas dubitationes conjicit ac potest…” (AT V ) The inclusion of mathematics is explicitly stated at AT VIII/I 8−20. Curiously, both Kennington (: ) and Soffer (: ) have denied that mathematical statements are doubted in Meditation One, partly on dubious methodological and systematic grounds which will be discussed below. . As with the other two sceptical hypotheses, it does not have to be coherent to achieve its intended effect: it does not have to depict a possibility which might in fact obtain. It is enough if it describes a situation which the Cartesian thinker is in no position to rule out. Descartes

The Cartesian sceptic (the philosophical opponent of the Meditations) is not someone who believes or says that everything we believe is false. Instead, he believes that all our knowledge claims (our beliefs of the form “I know that p”) are unwarranted, i.e. that there are possible situations (A, B, C) we are not justified to exclude where they are false – not so much because what we claim to know would be false in that situation, but because we would, in the imagined circumstances, not know it. The possibility we are called upon to exclude by the sceptic, then, is not one in which what we believe to know is false, but the possibility that our knowledge claims are true only by chance, that the link between belief and truth underwriting them, while obtaining in the actual world, is not within our epistemic reach and cannot be conclusively established to hold. The doubt, then, is that if we know something, then we know it only by chance. The sceptic thereby challenges our entitlement to all our knowledge claims – not by producing a scenario in which they are false, but one that makes them unjustified and thus unclaimable in the methodological context of a refoundation of our whole system of beliefs. Descartes calls beliefs that may be rationally entertained even in such an epistemic state of total doubt “clear and distinct”. The question then becomes whether there are any clear and distinct beliefs, whereas the Cogito is the claim that the belief I express by “I think; therefore I am” falls into this category and indeed is its paradigm exemplar. By showing us how to overcome the most general doubt, Descartes provides us with a foundation and justification of the analytic method – it is in this sense that he validates reason. Bourdin, in the Seventh Objections remarks that nothing can be indubitable in the Cartesian system, since everything had been rationally doubted by the Cartesian thinker in Meditation One. In reply, Descartes says that Bourdin’s mistake is to think that dubitability is an intrinsic property of knowables:

“Notandumque ipsum ubique considerare dubitationem & certi- “It should be noted that throughout he [Bourdin] treats doubt and cer- tudinem, non ut relationes cognitionis nostrae ad objecta, sed ut pro- tainty not as relations of our thought to objects, but as properties of prietates objectorum quae perpetuo ipsis inhaereant, adeo ut ea, quae the objects which inhere in them all the time. This means that if we semel dubia esse cognovimus, non possint unquam reddi certa.” (AT have once realized that something is doubtful, it can never be rendered VII 17−22) certain.” (Descartes : –)

Descartes here alludes to the relational account of certainty he gave in the Second Replies according to which whether a truth is certain or not depends on our relations to it, not only on the truth itself:

“Vel enim, ut vulgo omnes, per possibile intelligitis illud omne quod non “If by ‘possible’ you mean what everyone commonly means, namely repugnat humano conceptui; quo sensu manifestem est Dei naturam, ‘whatever does not conflict with our human concepts’, then it is man- prout ipsam descripsi, esse possibilem, quia nihil in ipsâ supposui nisi ifest that the nature of God, as I have described it, is possible in this quod clare & distincte perciperemus debere ad illam pertinere, adeo sense, since I supposed it to contain only what, according to our clear ut conceptui repugnare non possit. Vel certe fingitis aliquam aliam and distinct perceptions, must belong to it; and hence it cannot con- possibilitatem ex parte ipsius objecti, quae, nisi cum praecedente con- flict with our concepts. Alternatively,you may well be imagining some veniat, nunquam ab humano intellectu cognosci potest, ideoque non other kind of possibility which relates to the object itself; but unless plus habet virium ad negandum Dei naturam sive existentiam, quàm this matches the first sort of possibility it can never be known by the ad reliqua omnia, quae ab hominibus cognoscuntur, evertenda.” (AT human intellect, and so it does not so much support a denial of God’s VII 19-3) nature and existence as serve to undermine every other item of human knowledge.” (Descartes : )

The distinction between what ‘non repugnat humano conceptui’ (“ne répugne point à la pensée humaine” (AT IX/I )) and what is possible ‘ex parte ipsius objecti’ is made in reply to the charge that the ontological proof is premissed on the coherence of the idea of God (AT VII 12−14). Descartes’ point, as the context makes clear, is that the ontological proof only presupposes that the idea we have of God is de facto coherent, not that it is known to be coherent. Only the latter, but not the former, presupposes a proof of God’s benevolence. The coherence of the idea we have of God is put on a par with necessary truths and the Cogito itself, things which have been doubted in Meditation One:

himself takes a malin génie to be impossible: both because he shows its incoherence in the Second Meditation and because it is incompatible with the true nature of God. Cf. his remarks to Burman: “Loquitur hîc [AT VII 25] auctor contradictoria, quia cum summâ potentiâ malignitas consistere non potest.” (AT V ) It has to be noted, however, that showing that the malin génie hypothesis is impossible, is not Descartes’ main aim – even a hypothesis entertainable only per impossibile undermines our knowledge claims, if it keeps popping up. We have to be able to show that it is impossible, thereby vaccinating ourselves against it. . To my ears, “repugnat” and “répugne” are stronger than “conflict”.

“Par enim jure, quo negatur Dei naturam esse possibilem, quamvis “For as far as our concepts are concerned there is no impossibility in nulla impossibilitas ex parte conceptûs [“de la part du concept ou de the nature of God; on the contrary, all the attributes which we include la pensée”] reperiatur, sed contrà omnia, quae in isto naturae divinae in the concept of the divine are so interconnected that it seems to us to conceptu complectimur, ita inter se connexa sint, ut implicare nobis be self-contradictory that any one of them should not belong to God. videatur aliquid ex iis ad Deum non pertinere, poterit etiam negari Hence, if we deny that the nature of God is possible, we may just as possibile esse ut tres anguli trianguli sint aequales duobus rectis, vel ut well deny that the angles of a triangle are equal to two right angles, or ille, qui actu [“actuellement”] cogitat, existat; & longe meliori jure ne- that he who is actually thinking exists; and if we do this it will be even gabitur ulla ex iis quae sensibus usurpamus vera esse, atque ita omnis more appropriate to deny that anything we acquire by means of the humana cognitio, sed absque ullâ ratione, tolletur.” (AT VII 3−13) senses is true. The upshot will be that all human knowledge will be destroyed, though for no good reason.” (Descartes : )

The Principles are said to satisfy the desideratum that their axioms can be understood by themselves:

…ces Principes doivent avoir deux conditions: […] l’autre, que ce soit d’eux que depende la connois- sance des autres choses, en sorte qu’ils puissent estre connus sans elles [”‘les autres choses”’], mais non pas reciproquement elles sans eux; & qu’apres cela il faut tacher de déduire tellement de ces principes la connoissance des choses qui en dependent, qu’il n’y ait rien, en tout la suite des deductions qu’on en fait, qui ne soit tres-manifeste. (AT IX/II -)

In the characterisation of the dichotomy of methods in the Second Replies quoted above, Descartes/Clerselier para- phrases the “necessity” of the conclusions of the analytic method by their being clear “de soy-mesme”. We cannot but assent to what we clearly and distinctly perceive – such perceptions are epistemically irresistible to us. At the same time, however, we can doubt what we could, but do not, so perceive; and, what is more, we can suppose that the beliefs in question do not merit their status, that their irresistibility is not backed up by their truth but rather the effect of an evil demon fooling us by perverting our epistemic instincts. Not only some of his contemporaries, but also many of Descartes’ commentators believe that such universal a doubt undermines the stability of the whole Cartesian project. There is, however, nothing in general problematic with rule-circular justification, nor is there any general problem in establish, by using a richer range of ressources, what can be done with less. We may also quite unproblematically entertain metaphysical hypothesis which we could not entertain if they were actual. We may talk, e.g., about possible worlds where no language-using creatures exist.

.. The Cogito

A major exegetical problem with the Cogito is that there are many versions. The version in the Meditations, which will mostly concern us, is the following:

Sed mihi persuasi nihil plane esse in mundo, nullum coelum, nullam Sed est deceptor nescio quis, summe potens, summe callidus, qui terram, nullas mentes, nulla corpora; nonne igitur etiam me non esse? de industriâ me semper fallit. Haud dubie igitur ego etiam sum, si me Imo certe ego eram, si quid mihi persuasi. fallit & fallat quantum potest, nunquam tamen efficiet, ut nihil sim quamdiu me aliquid esse cogitabo. Adeo ut, omnibus satis superque pensitatis denique statuendum sit hoc pronuntiatum, Ego sum, ego existo,

. For a vivid description of the kind of epistemic irresistibility in question, cf. Kemmerling (). . Cf. Kennington (: –): “Finally, the finitude of the power of the Evil Genius is demanded by the argument if the Evil Genius were omnipotent, the law of noncontradiction would be suspended, and all further reasoning would have to cease.” Similar qualms have been mounted by R. Walker (: ): “…if truth is a matter of correspondence [that’s how he rightly interprets Descartes] no argument could be a decisive refutation of the malin génie hypothesis […] since any argument has to rely upon assumed premisses and principles of inference.” Carriero (: –): “Moreover, viewing Descartes as seeking to defend us from the epistemic onslaught of an omnipotent being leaves him engaged in a project that is futile on its face. It would be a quite weak omnipotent being indeed who lacked the power to convince us of the self-evidence of whatever premise he chose or of the validity of whatever reasoning he wanted. Therefore, if the anti-skeptic should ever produce an argument beyond our criticism, the skeptic would still be free to plead that our inability to criticize the anti-skeptic’s argument is simply the result of the deceiver’s nefarious activity.” and Olson (: ): “Descartes did hold the curious view that God freely created the eternal truths and could have made contradictories true. So it seems that, if God is a deceiver, then the law of non-contradiction is dubitable. Thus, if God is a deceiver, no argument could be formulated. Indeed, all reasoning would have to cease.” . It is, e.g., standard mathematical practice to reason about a formal system drawing on resources not available in the system under scrutiny. We may give intuitionistically unacceptable proofs of the completeness of some system of intuitionistic logic, prove various theorems about proof systems having no or only very weak induction axioms by “ordinary mathematical induction” (cf. e.g. Smullyan : for an instructive example).

quoties a me profertur, vel mente concipitur, necessario esse verum. Mais je me suis persuadé qu’il n’y a rien du tout dans le monde, qu’il (AT VII 2−12) n’y avait aucun ciel, aucune terre, aucuns esprits, ni aucuns corps, ne me suis-je donc pas aussi persuadé que je n’étais point? Non certes, j’étais sans doute si je me suis persuadé, ou seulement si j’ai pensé quelque chose. Mais il y a un je ne sçay quel trompeur tres-puissant & tres-rusé, qui employe toute son industrie à me tromper tousiours. Il n’y a donc point de doute que je suis, s’il me trompe; et qu’il me trompe tant qu’il voudra il ne saurait jamais faire que je ne sois rien, tant que je penserai être quelque chose. De sorte qu’après y avoir bien pensé, et avoir soigneusement examiné toutes choses, enfin il faut conclure, et tenir pour constant que cette proposition: Je suis, j’existe, est nécessaire- ment vraie, toutes les fois que je la prononce, ou que je la conçois en mon esprit. (AT IX/I )

This is the Cartesian thinker speaking, recalling his universal doubt of the last Meditation and realising that he did not doubt his own existence and that it is indeed impossible do so – not only that he cannot doubt without existing, but that he cannot even try (and therefore take it to be possible) to do so, i.e. that the thought “I do not exist” is – in a particular way – necessarily false. The cited passage makes it clear that the indoubtability of “I exist” is given a peculiar status. Even though they are not explicitly mentioned by their canonical description, the necessary truths were subject to the general doubt evoked in the first paragraph. This peculiar status is not exhausted by other a priori or performatively “self- verifying” claims, nor by such on which the Cartesian thinker has first person authority. What, then, is it that distinguishes “I am” from other candidate sentences as “I am here now” (Röd : – ), “I am in pain” or “I am making a statement” (Frankfurt : )? The first may be false, e.g., in cases of deferred reference or context shift – which may even occur in dreams. The second is not indoubtable, but at most indoubtable when true. The third, even if it is indoubtable, cannot be the only indoubtable statement. Even “This is a statement”, however, is less indoubtable than “I am”, for statements are ontologically dependent entities, not able to exist without someone who makes them. Nobody thus can rationally think that this is a statement without thinking that something different from this statement exists. The peculiar status of “sum”, then, resides in its ontological inconsequentiality, i.e. the fact that it may rationally taken to be the only true existence claim. Recognising the epistemic irresistability of his belief that he exists (his “sum”-intuition, as I will call it), gives the Cartesian thinker a belief of the truth of which he cannot help being convinced. This was not was Descartes was after, however: the “fundamentum inconcussum” he tries to uncover is a belief one can claim to know even when faced with the most witty sceptic producing the most general sceptical scenario imaginable. This paradigmatic knowledge claim is not “sum” alone – doubtable, if only per impossibile, and indeed doubted by the first person of the Meditations at the end of Meditation One – but “I think; therefore I am”.

. As the section label in the Principes might suggest: “VII. Non posse a nobis dubitari, quin existamus dum dubitamus; atque hoc esse primum, quod ordine philosophando cognoscimus.” (AT VIII 31−32) / “. Que nous ne sçaurions douter sans estre, & que cela est la premiere connoissance certaine qu’on peut acquerir.” (AT IX/II ) Descartes makes it clear in his replies to Hobbes and Gassendi that existential generalisation is not what he is interested in at this step of the argument (AT VII 25-8 and 6−18, cf. also AT II 26-21). . The existence of the corporeal world, the subject matter of geometry, is a mathematical fact. . This raises the interesting question of the alleged primacy of singular over general beliefs, a thesis Descartes relied on when arguing that he did not (syllogistically) infer “I think; therefore I am” from “whatever thinks, is” but was convinced of the truth of the former by a simple mental insight (AT VII 12-2, IX/I 25-5, V ). To evaluate this claim of epistemological primacy of singular beliefs, let us consider the following example of Harrison: “If a man believes that he believes that the earth is flat, it is possible for him to believe that this is the only thing he believes. […] But if he believes at least one thing, it is not possible for him to believe that this is the only thing he believes.” (Harrison : ) Someone who believes that the earth is flat and that this is his only belief has at least one false belief. The person trying to believe that the very belief he is trying to entertain is his only one faces a more difficult problem, however: he cannot ascribe the belief in question to himself. For any such self-ascription would ascribe two beliefs, that he has only one belief and that this is the one. . Given the epistemic priority of singular over general beliefs mentioned in fn. , this is why “I am” is epistemically irresistable in a way “there is at least one thing” is not: “I am” retains its plausibility even when taken to be my only belief, while “there is at least one thing” requires a specialisation. . I take this to be roughly equivalent to “Ego sum, ego existo, quoties a me profertur, vel mente concipitur, necessario esse verum.” “Cogito ergo sum” appears in the Discours version of the Cogito: “Mais, aussitost aprés, je pris garde que, pendant que je voulois ainsi penser que tout estoit faux, il falloit necessairement que moy, qui le pensois, fusse quelque chose. Et remarquant que cete verité: je pense, donc je suis, estoit si

In what does this differ from the “sum”-intuition? In that it gives us a reason to think that the intuition we cannot help having is right. This does not mean, however, that “therefore” has the force of a conditional, a turnstile or indicates some kind of conclusion. “Cogito” provides adequate grounds for “sum”: its truth is necessary and sufficient for recognising “sum”’s indoubtability. With “cogito ergo sum” we not only express an intuition, but make a knowledge claim – and a peculiar one in that we cannot make it falsely. Even this, however, will not suffice: proving the sceptic wrong by showing how to make a true knowledge claim even in the scenario the sceptic considers possible is not enough. Descartes wants to show his reader how to prove the sceptic wrong, provide him with a tool to prove his own existence (not the one of the Cartesian thinker) even when under the impression of the most general doubt imaginable. This, in my opinion, crucial step from “sum”, the belief we cannot but take to be true, through “cogito ergo sum”, which is the paradigmatic knowledge claim, to the claim that “cogito ergo sum” is paradigmatic in this sense is taken only at the beginning of the Third Meditation:

() Quid verô? Cùm circa res Arithmeticas vel Geometricas aliquid “Mais lorsque je considérais quelque chose de fort simple et de fort valde simplex & facile considerabam, ut quôd duo & tria simuljuncta facile touchant l’arithmétique et la géométrie, par exemple que deux sint quinque, vel similia, nunquid saltem illa satis perspicue intuebar, et trois joints ensemble produisent le nombre de cinq, et autres choses ut vera esse affirmarem? semblables, ne les concevais-je pas au moins assez clairement pour assurer qu’elles étaient vraies?

() Equidem non aliam ob causam de iis dubitandum esse postea judi- Certes si j’ai jugé depuis qu’on pouvait douter de ces choses, ce n’a cavi, quàm quia veniebat in mentem forte aliquem Deum talem mihi point été pour autre raison, que parce qu’il me venait en l’esprit, que naturam indere potuisse, ut etiam circa illa deciperer, quae manifestis- peut-être quelque Dieu avait pu me donner une telle nature, que je me sima viderentur. trompasse même touchant les choses qui me semblent les plus mani- festes.

() Sed quoties haec praeconcepta de summâ Dei potentiâ opinio mihi Mais toutes les fois que cette opinion ci-devant conçue de la souveraine occurrit, non possum non fateri, si quidem velit, facile illi esse efficere puissance d’un Dieu se présente à ma pensée je suis contraint d’avouer ut errem, etiam in iis quae me puto mentis oculisquàm evidentissime qu’il lui est facile, s’il le veut, de faire en sorte que je m’abuse, même intueri. dans les choses que je crois connaître avec une évidence très grande.

() Quoties verô ad ipsas res, quas valde clare percipere arbitror, me Et au contraire toutes les fois que je me tourne vers les choses que je converto, tam plane ab illis persuadeor, ut sponte erumpam in has pense concevoir fort clairement, je suis tellement persuadé par elles, voces: que de moimême je me laisse emporter à ces paroles:

() fallat me quisquis potest, nunquam tamen efficiet ut nihil sim, Me trompe qui pourra, si est-ce qu’il ne saurait jamais faire que je ne quandiu me aliquid esse cogitabo; vel ut aliquando verum sit me nun- sois rien tandis que je penserai être quelque chose; ou que quelque quam fuisse, cùm jam verum sit me esse; jour il soit vrai que je n’aie jamais été, étant vrai maintenant que je suis,

() vel forte etiam ut duo & tria simul juncta plura vel pauciora sint ou bien que deux et trois joints ensemble fassent plus ni moins que quàm quinque, vel similia, inquibus scilicet repugnantiam agnosco cinq, ou choses semblables, que je vois clairement ne pouvoir être manifestam. d’autre façon que je les conçois.

() Et certe cùm nullam occasionem habeam existimandi aliquem Et certes, puisque je n’ai aucune raison de croire qu’il y ait quelque Deum esse deceptorem, nec quidem adhuc satis sciam utrùm sit Dieu qui soit trompeur, et même que je n’aie pas encore considéré aliquis Deus, valde tenuis &, ut ita loquar, Metaphysica dubitandi ra- celles qui prouvent qu’il y a un Dieu, la raison de douter qui dépend tio est, quae tantùm ex eâ opinione dependet. seulement de cette opinion, est bien légère, et pour ainsi dire méta- physique.

() Ut autem etiam illa tollatur, quamprimum occurret occasio, exam- Mais afin de la pouvoir tout à fait ôter, je dois examiner s’il y a un inare debeo an sit Deus, &, si sit, an possit esse deceptor; hac enim re Dieu, sitôt que l’occasion s’en présentera; et si je trouve qu’il y en ait ignoratâ, non videor de ullâ aliâ plane certus esse unquam posse.” un, je dois aussi examiner s’il peut être trompeur: car sans la con- naissance de ces deux vérités, je ne vois pas que je puisse jamais être certain d’aucune chose.” ferme & si assurée, que toutes les plus extravagantes suppositions des Sceptiques n’estoient pas capables de l’esbransler, je jugay que ie pouvois la recevoir, sans scrupule, pour le premier principe de la Philosophie, que je cherchois.” (AT VI 15−24) . This seems to me to explain why Descartes calls both “sum” and “cogito ergo sum” pieces of intuitive knowledge (for the latter: AT VII 7−11, IX/I 20−25), while at the same time demonstrating his own existence from the sole premiss of his thinking. . Again, this does not make it a premiss of some sort of formal argument. Suppose it is not doubted that there is a sole tired person, of unknown sex, in the house. Given this, “The sole person in the house is male” provides an adequate ground for “he is tired”.

This is the passage I claim to be the Cogito. While having his “sum”-intuition, the first person of the Meditations (A) proves his existence. Recalling the impossibility of doubting his existence and looking back to his “sum”-intuition, the Cartesian thinker (B) proves “cogito ergo sum”. For us, the foreknowledgeable and charitable readers (C), this amounts to a proof of the existence of a thinking substance. Here is a reconstruction of the argument: A B C clear and distinct beliefs are in- B takes “cogito ergo sum” to be in- () doubtable (by me) doubtable (in itself) no other beliefs are claimed by my “cogito ergo sum” is not () to be indoubtable (in itself) with doubtable (in itself), unless B more right than these is fooled the evil demon hypothesis seems I can, however, only entertain it per B can at least entertain it () coherent impossibile I cannot help having the “sum”- but even in that I could be fooled () intuition by an evil demon I clearly and distinctly perceive A can indeed not be fooled while () that I cannot be fooled while I am having this intuition having this intuition I cannot even doubt the indoubta- A is clearly and distinctly perceiv- “sum” is in fact indoubtable (in it- () bility of “sum” ing his unability to doubt “sum” self), so B is right A uncovered a hidden contradic- the evil demon hypothesis is not () tion in the evil demon hypothesis clear and distinct B is not fooled, “cogito ergo sum” () is clear and distinct (in itself) It is in () that A has the relevant insight that the hypothesis that he does not exist is “repugnant to reason”, i.e. contains a hidden contradiction. B, however, who already had the “sum”-intuition before, knows from the beginning that the only way to doubt one’s own existence, if there is one, is per impossibile. He detects the hidden contradiction in the evil demon hypothesis, noticing that even such a being could not make A falsely believe that he does not exist. B observes that any project of doubting one’s own existence fails, thereby grasping the (necessary) truth of “I think, therefore I am”. C observes that B has maximal evidence for this, for it resists his trying to assume the existence of an evil demon. If it would be false, then, this could only be due to the workings of precisely such an evil demon – a hypothesis effectively ruled out by A. So, C concludes, that it was because of his knowledge of “I think; therefore I am” that B was able to rule out the evil demon hypothesis. That’s why “I think; therefore I am” is not only knowable and true but also provides the grounds to effectively rule out scepticism. Meeting the sceptical challenge was not an end in itself for Descartes, but a means to prove his central metaphysical doctrines, i.e. substance dualism, the existence of God and the immortality of the soul, from their “true reasons”. How then does our reconstructed Cogito help him in that? In two ways: by grounding the synthetic and deductive order arising from the analytic method, from the most evident (in itself) back to its consequences, which are familiar to us, and by establishing a certain dependence of the Cartesian thinker’s existence from the fact that he is thinking. There is an important connection between clarity and distinctness and dubitability: even though truths that are in themselves clear and distinct need not be clear and distinct for us, and thus can be doubted, they can only be doubted per impossibile, even though this does not need to be always transparent to us. It is, however, transparent to us if we correctly apply the analytic method. Herein lies an important presupposition: that the analytic method, as it is employed in the first Meditation allows us to keep track of epistemological, and perhaps even metaphysical dependencies. This is what Descartes explicitly claims in the seventh set of Replies, commenting on “Certissimum est hujus sic praecise sumpti notitiam non pendere ab iis quae existere nondum novi; non igitur ab iis ullis, quae imaginatione effingo.” (AT VII 1−2):

. While A’s insight, the “sum”-intuition, is categorical, B becomes aware of the truth of a conditional: assuming its antecedent in a thought experiment (staging A as his counterpart), he notes that the conclusion is inevitable. It is only C who is in a position to note that B’s thought experiment is doomed to fail, thereby establishing that B can truly claim to know “I think; therefore I am”.

…scripsi enim, fieri non posse ut ea, quam jam habeo, notitia rei, …I wrote that the knowledge I already have of something I know to quam novi existere, pendeat a notitiâ ejus, quod existere nondum novi. exist cannot possibly depend on the knowledge of things of whose ex- […] Nam sane perspicuum est, rei, quae cognoscitur ut existens, noti- istence I am as yet unaware. […] It is transparently clear that the tiam illam, quae jam habetur, non pendere a notitiâ ejus, quod non- already acquired knowledge of a thing which is recognized as exist- dum cognoscitur ut existens; quia hoc ipso quòd aliquid percipiatur ing does not depend on the knowledge of that which we have not yet ut pertinens ad rem existentem, necessario etiam percipitur existere. recognized as existing; for the very fact that something is perceived to (AT VII 24−26;30-5) belong to an existing thing necessarily implies tha tit is perceived to exist. (Descartes : )

The Cogito, in which the analytic method ‘culminates’, then aligns the ordo cognoscendi with the ordo essendi, allow- ing Descartes to proceed in a way such that he does not have to presuppose anything which was not previously demonstrated. The transition, within this order, from thinking (entertaining the evil demon hypothesis) to ex- istence (realising that the evil demon hypothesis is not entertainable without pragmatic inconsistency), becomes indicative of real dependence. The dependence of the Cartesian thinker’s existence on the first-person’s thinking is, first, dialectical. Only by granting the premiss that he is thinking can we prove her existence. Second, it is epistemological: we, and the Cartesian thinker, need no other belief about him than the one that he is thinking to prove that he exists – in this way, we prove his existence as a thinking thing. Third, it is one of Descartes’ later doctrines that any such ‘distinction of reason’, where we clearly and distinctly perceive that two things can exist without each other, is underwritten by a ‘real difference’ – the ontological independence of two substances. We are not, of course, at this stage of the argument in a position to know that some such doctrine is true; but it is enough that it is true. If it is true (and this in turns depends on the validity of the ontological proof of the existence of God), then we can truly claim in retrospect to have demonstrated the existence of res cogitans. And this is quite much, too much perhaps, but in any case another issue. The situation is importantly different with the truth-rule, however. The truth rule is, but the real distinction is not established by the Cogito. If the first Meditation instantiates (a correct way of applying) the analytic method, then we are justified in reserving our highest epistemic modality (“clear and distinct”) for just those truths that withstand even the strongest of all possible doubts. We have seen reasons both to believe in the antecedent and to take Descartes to be convinced of it, so the question of the universality of maximal doubt becomes the question whether there are any clear and distinct beliefs at all. The Cogito is the claim that the belief I express by “I think; therefore I am” falls into this category and indeed is its paradigm exemplar. By showing us how to overcome the most general doubt, Descartes provides us with a foundation and justification of the analytic method – it is in this sense that he validates reason. He also achieves something else: not only is everything that is clearly and distinctly perceived doubted at the end of Meditation One (this is built into the evil demon hypothesis), but it is also the case that whatever can only be doubted by this most general doubt is (by itself) clear and distinct.

.. Ideas as contents

The official definitions:

Ideae nomine intelligo cujuslibet cogitationis formam illam, per cujus Par le nom d’idée, j’entens cette forme de chacune de nos pensées, par immediatam perceptionem ipsius ejusdem cogitationis conscious sum; la perception immediate de laquelle nous avons connoissance de ces adeo ut nihil possim verbis exprimere, intelligendo id quod dico, quin mesmes pensées. En telle sorte que je ne puis rien exprimer par des ex hoc ipso certum sit, in me esse ideam ejus quod verbis illis signifi- paroles, lorsque j’entens ce que je dis, que de cela mesme il ne soit cer- cantur. (AT VII -) tain que j’ay en moy l’idée de la chose qui est signifiée par mes paroles. (AT IX/I )

Our Idea of God:

“Nec certè quisquam talem ideam Dei nobis inesse negare potest, nisi qui nullam planè Dei notitiam in humanis mentibus esse arbitretur.” (AT VIII/I -) “…si on prend le mot d’idée en la façon que j’ay dit tres-expressement que je la prenois […], on ne sçaurait nier d’avoir quelque idée de Dieu, si ce n’est qu’on die qu’on n’entend pas ce que signifient ces mots: la chose la plus parfaite que nous puissions concevoir; car c’est ce que tous les hommes apellent Dieu.” (AT IX/I -)

. Ideas

.. Lockean ideas

Locke introduces ‘idea’ as “whatsoever is the object of the understanding when a man thinks …or whatever it is which the mind can be employed about in thinking” (I.i.). There are ideas of ‘sensation’ and of ‘reflection’, those the mind acquires “when it turns its view inward upon itself, and observes its own actions about those ideas it has” (II.vi.). Abstract ideas are created by omission of detail (something Berkeley found impossible because for him an idea must ‘resemble’ its object by possessing the property it represents). Simple ideas, by contrast, are never created:

That the Mind, in respect of its simple Ideas, is wholly passive, and receives them all from the Existence and Operations of Things, such as Sensation or Reflection offers them, without being able to make any one Idea, Experience shows us. (II.xxii,)

Mistakes arise only in the combination of simple to complex ideas: simple ideas are “natural and regular produc- tions of Things without us, really operating upon us; and so carry with them all the conformity which is intended; or which our state requires” (IV.iv.). Even though Locke introduces ideas as the objects of thinking and sensation, some commentators have taken them to be processes. Kemmerling thinks they are processes insofar as these have representational content:

Das kognitiv Entscheidende an jedem Vorgang des Perzipierens ist sein repräsentationaler Gehalt. Indem über “die Idee” gesprochen wird (die ontologisch gesehen nichts anderes ist als der Vorgang ihres Perzipiertwerdens), wird der Gehalt des Vorgangs thematisiert. Wird also z.B. über die Idee von Rund gesprochen, dann wird damit der Gehalt des Vorgangs einer Rund-Perzeption thematisiert. (Kemmerling : )

But how can an idea be its own content? To reconcile Locke’s two desiderata – to have ideas both as contents of acts of thinking as as their immediate objects –, Lenz (: ) speaks of two perspectives, distinguishing “ideas as material” from “ideas as parts of episodes of thinking”. Simple ideas are signs of their qualities because they causally covary with them:

For the Objects of our Senses, do, many of them, obtrude their particular Ideas upon our minds, whether we will or no […] As the Bodies that surround us, do diversly affect our Organs, the mind is forced to receive the Impressions; and cannot avoid the Perception of those Ideas that are annexed to them. (II.i.)

Simple ideas thus have both their representational power and their veridicality just in virtue of their causal origin: their truth consists in “such Appearances, as are produced in us, and must be suitable to those Powers, [God] has placed in external Objects, or else they could not be produced in us” (II.xxxii.).

.. Berkeleyan radicalisation

Berkeley takes the idea that ideas are the objects of sensation as his starting point, and denies that they have the causes we think they have; as Locke said of the idea of substance, things themselves are just bundles of ideas:

It is evident to anyone who takes a survey of the objects of human knowledge, that they are either ideas actually imprinted on the senses; or else such as are perceived by attending to the passions and oper- ations of the mind; or lastly, ideas formed by help of memory and imagination – either compounding, dividing, or barely representing those originally perceived in the aforesaid ways. – By sight I have the ideas of light and colours, with their several degrees and variations By touch I perceive hard and

soft, heat and cold, motion and resistance, and of all these more or less either as to quantity or degree. Smelling furnishes me with odours; the palate with tastes; and hearing conveys sounds to the mind in all their variety of tone and composition. – And as several of these are observed to accompany each other, they come to be marked by one name, and so to be reputed as one thing.(§ )

He then presents his idealism as a consequence of the rejection of abstract ideas:

For as to what is said of the absolute existence of unthinking things without any relation to their being perceived, that is to me perfectly unintelligible. Their esse is percipi, nor is it possible they should have any existence out of the minds or thinking things which perceive them. […] For can there be a nicer strain of abstraction than to distinguish the existence of sensible objects from their being perceived, so as to conceive them existing unperceived? Light and colours, heat and cold, extension and figures – in a word the things we see and feel – what are they but so many sensations, notions, ideas, or impressions on the sense? and is it possible to separate, even in thought, any of these from perception? For my part, I might as easily divide a thing from itself. (§§ ,)

The reply to invoke representation as the bridge between the realm of ideas and that of things he rejects citing the so-called ‘likeness principle’:

But say you, though the ideas themselves do not exist without the mind, yet there may be things like them whereof they are copies or resemblances, which things exist without the mind, in an unthinking substance. I answer, an idea can be like nothing but an idea; a colour or figure can be like nothing but another colour or figure. If we look but ever so little into our thoughts, we shall find it impossible for us to conceive a likeness except only between our ideas. (§ )

He then explicitly generalises the argument against the reality and the mind-independence of secondary to all qualities. Berkeley’s so-called ‘Master Argument’ is supposed to show that the very notion of an unperceived object is contra- dictory:

But say you, surely there is nothing easier than to imagine trees, for instance, in a park, or books existing in a closet, and no body by to perceive them. I answer, you may so, there is no difficulty in it: but what is all this, I beseech you, more than framing in your mind certain ideas which you call books and trees, and at the same time omitting to frame the idea of any one that may perceive them? But do not you your self perceive or think of them all the while? This therefore is nothing to the purpose: it only shows you have the power of imagining or forming ideas in your mind; but it doth not shew that you can conceive it possible, the objects of your thought may exist without the mind: to make out this, it is necessary that you conceive them existing unconceived or unthought of, which is a manifest repugnancy. When we do our utmost to conceive the existence of external bodies, we are all the while only contemplating our own ideas. (§ )

The problem with this argument is that it already presupposes that there is no stable way to draw a difference between the content of what is imagined and what follows from the fact that something is being imagined. It cannot be saved by supposing that ‘imagine’ is factive and that we cannot perceive that no-one perceives x without perceiving x (this is the suggestion of Barkhausen & Haag (: )); at best, it produces a stand-off. The Dialogues (III, -) suggest another route to Berkeleyan idealism: as the mind combines ideas from different senses in Molyneux’s situation to construct the idea of some one thing, so do we call “things” (hypothetical) sources of ideas. As our interest lies in the connection of ideas and this is all we can ever have epistemic access to, we have no, and cannot ever have any, epistemic warrant to go ‘beyond’ the bundle of ideas. In his work on optics, Berkeley came to the conclusion that we do not see distance, but feel it, in virtue of feeling the movements of our own eyes. Talk of an external body that is both felt and seen is nothing but a construction of our mind (Saporiti

: ). In the Principles, this bundling is generalised: the things we see have nothing but objective being, they are these bundles of ideas themselves.

.. Humean reconciliation?

Hume divides introspectible mental items (“perceptions”) into two classes:

Those perceptions which enter with most force and violence we may name impressions, and under this name I comprehend all our sensations, passions, and emotions, as they make their first appearance in the should. By ideas I shall mean the faint images of these in thinking and reasoning. (I.i.)

Ideas and impressions are linked by Hume’s copy thesis: “all our simple ideas are caused by previous impressions which they resemble in every respect except the degree of vivacity which defines the difference between them” (Bennett : ). He says that “all our simple ideas proceed, either mediately or immediately, from their cor- respondent impressions” is the “first principle [he] establish[es] in the science of human nature” (I.i.-). It is tempting to understand this ‘procession’ as a representation relation: ideas represent their impressions (e.g. Wild a: ). The “general proposition” he “establishes” in I.i. – “That all our simple ideas in their first appear- ance are deriv’d from simple impressions, which are correspondent to them, and which they exactly represent.” (I.i..) The impressions themselves do not represent anything. Or, at least, if they did we would not know:

A like reasoning will account for the idea of external existence. We may observe, that ’tis universally allow’d by philosophers, and is besides pretty obvious of itself, that nothing is ever really present with the mind but its perceptions or impressions and ideas, and that external objects become known to us only by those perceptions they occasion. To hate, to love, to think, to feel, to see; all this is nothing but to perceive. Now since nothing is ever present to the mind but perceptions, and since all ideas are deriv’d from something antecedently present to the mind; it follows, that ’tis impossible for us so much as to conceive or form an idea of any thing specifically different. from ideas and impressions. Let us fix our attention out of ourselves as much as possible: Let us chase our imagination to the heavens, or to the utmost limits of the universe; we never really advance a step beyond ourselves, nor can conceive any kind of existence, but those perceptions, which have appear’d in that narrow compass. This is the universe of the imagination, nor have we any idea but what is there produc’d. (I.i..)

In Berkeleyan fashion, this seems to confuse historical with representational properties of ideas. We may call it a “phenomenological attitude” (Wild a: ) and claim that “in the phenomenological attitude, impressions do not have external objects” (Wild b: ) – but this leaves open the question whether they do have external objects. Accepting Berkeley’s criticism of Locke’s abstract ideas, Hume explains that particular ideas can be used in a gen- eral way (I.i..). Such a use of ideas also explains the difference between the ideas of judgement and of imagination:

An idea assented to feels different from a fictitious idea, that the fancy alone presents to us: And this different feeling I endeavour to explain by calling it a superior force, or vivacity,or solidity,or firmness, or steadiness. This variety of terms, which may seem so unphilosophical, is intended only to express that act of the mind, which renders realities more present to us than fictions, causes them to weigh more in the thought, and gives them a superior influence on the passions and imagination. Provided we agree about the thing, ’tis needless to dispute about the terms. The imagination has the command over all its ideas, and can join, and mix, and vary them in all the ways possible. It may conceive objects with all the circumstances of place and time. It may set them, in a, manner, before our eyes in their true colours, just as they might have existed. But as it is impossible, that that faculty can ever, of itself, reach belief, ’tis evident, that belief consists not in the nature and order of our ideas, but in the

manner of their conception, and in their feeling to the mind. T confess, that ’tis impossible to explain perfectly this feeling or manner of conception. We may make use of words, that express something near it. But its true and proper name is belief, which is a term that every one sufficiently understands in common life. And in philosophy we can go no farther, than assert, that it is something felt by the mind, which distinguishes the ideas of the judgment from the fictions of the imagination. It gives them more force and influence; makes them appear of greater importance; infixes them in the mind; and renders them the governing principles of all our actions.

Whether or not an idea represents something external to it is just a matter of how it feels to entertain it.

Chapter

Colours and Values

. Conflict

.. Different types of conflict

General motivation: is some dualism of objects and how they are needed to account for impressions of multiplicity? Straightforward application to perception: object constancy. However, we have constancy phenomena for prop- erties too: colours, velocity. Compare (i) I believe that p and I believe that ¬p. (ii) I intend to ϕ and I intend to ¬ϕ. (iii) I desire to ϕ and I desire to ¬ϕ. (iv) I am afraid that p and I am glad that p. (v) I am afraid that p and I believe that it is not dangerous to me that p. (vi) I intend to ϕ and I believe that it is impossible to ϕ. (vii) I desire to ϕ and I believe that it is not good for me to ϕ. (viii) I intend to ϕ, and I can ϕ,andIdonotdo ϕ.

.. Resisting Counterevaluatives

Counterevaluatives are sentences that claim evaluative matters to be different from how they actually are. Coun- terevaluatives have recently received a lot of attention in discussions of the so-called “puzzle of imaginative resis- tance” (Gendler ). Emotions are said to involve formal objects, in the way fear involves the fearful (Kenny : ). It is with respect to their formal objects that an emotion’s appropriateness may be assessed (cf. Goldie : ) David Hume writes in On the Standards of Taste:

“Where speculative errorsmaybefound in thepolite writingsofanyage orcountry,theydetract but little from the value of those compositions. There needs to be but a certain turn of thought or imagination to make us enter into all the opinions which then prevailed and relish the sentiments or conclusions derived from them. But a very violent effort is requisite to change our judgment of manners, and excite sentiments of approbation or blame, love or hatred, different from those to which the mind from long custom has been familiarized […] I cannot, nor is it proper that I should, enter into such [vicious] sentiments.”

Though the exact interpretation of this passage is controversial, Hume distinguishes between the reactions we have towards assertions that contradict our descriptive and evaluative beliefs respectively. At least part of the puzzle this raises can also be illustrated by the following:

Try to imagine the following fictional mini-story to be true: “In killing her baby, Giselda did the right thing; after all, it was a girl.”

When faced with such an imperative, we face what we may call “imaginative resistance”. If this is accepted as a fact, then the following puzzle arises:

“…the puzzle of explaining our comparative difficulty in imagining fictional worlds that we take to be morally deviant.” (Gendler : )

For the sake of exposition let us indulge in the fiction that we imagine worlds, possible or impossible, and let us call the world we are to imagine the Giselda-world. In the following, I am going to assume that we have indeed comparative difficulty in imagining morally deviant fictional worlds and will discuss several explanations philosophers have given of this fact, before offering my own. Let us first note, however, how vague our starting point really is. Gendler frames it in terms of an asymmetry in make-belief:

“I cannot bring myself to believe that murder is right – but I cannot bring myself to believe that the earth is flat either. When it comes to make-belief, however, we seem more inclined to find ourselves stumped in the one case than in the other.” (Gendler : )

In a footnote, she then notes:

“Walton points out (personal correspondence) that my use of ‘make-believing’ seems ambiguous be- tween two readings. If I make-believe that p, I may be: (a) accepting that p as been successfully made fictional (that is, accepting that the author has succeeded in presenting a story in the context of which a certain proposition is true) or (b) pretending that p (that is, entertaining or attending to or considering the content of p, in the distinctive way required by imagination).” (Gendler : , fn. )

Weatherson () has called (a) the alethic and (b) the imaginative puzzle of imaginative resistance. In the following, I will concentrate on (b), though I take our answer to the (b)-problem to have implications on (a) as well – at least if works of fiction are invitations to imagine (Walton ). I do not claim that we never succeed in accepting morally deviant propositions to be fictionally true – context and rhetoric may help us to do so. I think, however, that there is a comparative – not necessarily an absolute – difficulty in imagining morally deviant worlds and that this explains why context and rhetoric are required for this. A range of solutions In recent philosophical discussions, imaginative resistance has been explained by: (i) our refusal to imagine (Gendler); (ii) our difficulty of imagining (Hume, Moran, Currie); or (iii) our inability to imagine certain propositions (Walton, Weatherson, Yablo). The argument of Gendler () consisted of two parts: she first argued that our difficulty in imagining the Giselda- world is not due to its being logically or conceptually impossible, because we succeed in imagining (at least some) logically or conceptually impossible worlds and because we face imaginative resistance also with respects to worlds that are clearly logically and conceptually possible. Though I am not very much convinced of her first reason, the

. (Cf. Walton : ). I follow Gendler (: ) in taking this to be a – somewhow contrived – example of an alleged fictional truth evincing imaginative resistance, ignoring its possible literary contexts.

second – which I take to be sound – suffices to discard conceptual impossibility as the only source of imaginative resistance. In a second part, Gendler sketched a positive account, turning on the alleged fact that the reader of fictions “feels being asked to export a way of looking at the actual world which she does not wish to add to her conceptual repertoire” (Gendler : ). Unfortunately, Gendler’s positive proposal is rather vague – unwillingness to add a certain perspective to one’s ‘conceptual repertoire’ can explain unwillingness to imagine the Giselda-world only if it is properly distinguished both from believing that murder is right and from supposing that it is. The assumption that the author of the Giselda- story presumably wanted us to convince that murder is right is gratuituous – we are not required to make it in order to experience imaginative resistance. The order of explanation is rather in reverse: it is because we experience imaginative resistance that we are unwilling to accept the author’s invitation to change our moral beliefs if really such an invitation is made. But in the same way, we are unwilling to change our non-moral belief if some author would want to convince us that really, pigs can fly. The impossibility hypothesis, as Gendler (: ) calls it – i.e. the view that we cannot imagine the Giselda-world because it is conceptually impossible – is just one way of defending an inability-diagnosis of imaginative resistance. A more general diagnosis is that imaginative the Giselda-worlds goes against our belief in certain dependence ties:

“Our reluctance to allow moral principles we disagree with to be fictional is just an instance of a more general point concerning dependence relations of a certain kind.” (Walton : )

The dependence relations here are those holding between the subvening base of non-moral properties and the moral properties supervening on them in the sense that no two things could differ in moral properties without also differing in non-moral ones. Weatherson (: ??) has argued that we experience imaginative resistance in all and only the cases where the following supervenience principle is violated (he calls it “virtue”):

“If p is the kind of claim that, if true, must be true in virtue of lower-level facts, and if the story is about those lower-level facts, then it must be true in the story that there is some true proposition r which is about those lower-level facts such that p is true in virtue of r.” (Weatherson : )

Weatherson () claims that radically divergent epistemic evaluations and attributions of mental states and con- tent also violate this principle and are therefore difficult to imagine. The problem, however, is that our belief in supervenience does not account for our “inability to understand fully what it would be like for [the moral facts] to be different” (Walton : ), or, as Gendler (: ) phrases it, of why “we cannot make sense of what it would be for something to be both an instance of murder and an instance of something that is morally right, and for it to be morally right because it is an instance of murder”. For our belief that some supervenience relation holds is itself descriptive and therefore readily available for suspension in imagination. But it is not easier to imagine the truth of

“In a world where the total distribution of physical qualities over space-time is importantly different from how it is in the actual world, Giselda did the right thing in killing her baby; after all, it was a girl.”

But supervenience-violation is neither necessary nor sufficient for imaginative resistance. It is not necessary for we experience imaginative resistance also in cases where no supervenience relations are violated, e.g. where the lower-level facts are imagined to be radically different from what they are. It is not sufficient because many violations of supervenience-relations are readily imagined. We imagine, for ex- ample, that one morning, Gregor Samsa, waking up from anxious dreams, discovered that in bed he had been changed into a monstrous verminous bug – even if we think that Gregor Samsa is essentially a person. This is even clearer in the case of entities of other categories: their spatio-temporal location is plausibly essential to events, but we readily imagine my talk having taken place minutes earlier than in fact it actually does.

. I am indebted here and in the following to Tyler Dogget’s talk on imaginative resistance at MIT in spring .

Stephen (Yablo : ) argues that we have difficulty in accepting a story where maple-leafs are oval. He generalises this to grokking or response-enabled concepts. A response-enabled concept like “oval” is picked out, but not analysable in terms of our perceptual responses: while we pick out oval things as those that look roughly egg-shaped, it is not necessary that oval things look egg-shaped – even in a world where perceivers have radically different perceptual mechanism than we do, there still would be oval things. The question whether some things in a possible world are oval is not to be decided by how they look to their world-mates, but by how they look to us. Currie (: ) draws an interesting contrast between belief-like and desire-like imagining. Belief-like imagining is supposing – what we do if we evaluate indicative conditionals by the Ramsey test. In belief-like imagining, the inferential patterns of belief contents are preserved, even though imagined propositions are not required to be consistent with each other and with our beliefs. Desire-like imaginings, according to Currie (: –) occur when “we imagine ourselves in [a] situation and then, in imagination, we decide to do something”. The desire of the theatre-goer that Othello not kill Desdemona is of this kind – and it is not the desire that, in this fiction, Desdemona not be murdered. The latter is a real (but odd) desire, the first a desire-in-imagination (Currie : ). It is in conjunction with desire-like imaginings that belief-like imaginings have emotional consequences. Imaginative resistance, according to Currie, arises when we choose to preserve the normal connection between belief and desire (and between belief-like and desire-like imagining) at the cost of desire-like imagining:

“…if it is difficult [for the reader] to have the desire-like imagining that female infants be killed, she can have the belief-like imagining that female infanticide is right only at the expense of the harmony between belief-like and desire-like imagining which is the natural stance of the intelligent and sensitive reader.” (Currie : )

As Currie (: –) acknowledges, this diagnosis does not cover all cases: it does not explain our difficulty in imagining the lame joke to be funny or imagine some pain we would experience if tortured by gentle contact with a fluffy chair.

“it is harder, much harder, to get people to desire in imagination against the trend of their own real desires than it is to get people to believe in imagination against the trend of what they really believe” (Currie : )

Gendler characterises the cases of imaginative resistance Hume had in mind as “cases involving valenced nor- mative evaluations: we are asked to assess something as mannerly or unmannerly, praiseworthy or blameworthy, loveable or hateable, where each of the pairs identifies two points along a normative spectrum where one end is desirable and the other is not”. We need, in other words, a logic of “make-believedly” that applies to emotive and not just conative actions. An alternative to the voluntarist explanation: . Evaluative judgments are tied up with feelings. . Feelings either cannot be simulated or they can be simulated only in rather special ways (as some kind of make-believe feelings, as perhaps they are when we go to the movies). Compare this with the imaginative resistance of a vegetarian to “if you haven’t had a good steak for days, you just feel terribly hungry”. Compare: () Suppose that murder is right. () Imagine that murder is right. () Accept as fictionally true that murder is right.

. It should apply to them ‘directly’, not just in a round-about way via a detour through fictional truths that include “such propositions as that one is now experiencing fear and pity for the tragic hero, or that one feels the satisfactions of vengeance when rough justice is meted out to the deserving” (Moran : ). Imagining that the proposition “I am feeling fear of the lion” is true (or imagining it as true), however, is clearly different from make-believedly fear the lion: feeling fear-at-the-movies is not an activity involving propositions.

() Pretend that murder is right. () Make-believedly approve of murder. (), even by Gendler’s lights, is easy. () is what she takes to be difficult, understanding it as (at least entailing) (). She takes the difficulty of () to explain the difficulty of (). I think the difficulty (if any) of () is explained by the difficulty of (). How wide is the phenomenon of imaginative resistance? A lot of cases have been mentioned in the literature: (i) We have difficulty in imagining, of a really lame joke, that it is funny (Walton : –). (ii) We have difficulty in imagining, of sour milk, that it smells good (Gendler : attributes it to Carl Ginet). (iii) We have difficulty in imagining, of a monster truck rally, that is is sublimely beautiful (Yablo : ). The following are easy: (E) Suppose that this tomato looks green. (E) Suppose that this joke is funny. (E) Suppose that sour milk smells good. (E) Suppose that you are a zombie. (E) Suppose that murder is right. But the following are difficult: (D) Make-believedly see this tomato as green. (D) Make-believedly laugh about this joke. (D) Make-believedly like this smell. (D) Make-believedly be all dark inside. (D) Make-believedly approve of this murder. Distinguish . fear, understanding, assertion, having as formal objects the dangerous / meaning (?) / truth. . quasi-fear (in movie theaters), quasi-understanding (e.g. empathy), quasi-assertion (on stage); same psycho- logical feel, but different connection with action and reasoning. . make-believe / as-if fear, understanding, assertion (supposition): no or different psychological feel You can make-believedly understand “kdflkdslf” if reading a story in which you do, but you do not quasi-understand it. You can make-believe fear and suppose fearing it, but you cannot quasi-fear something you do not at the same time take to be dangerous.

.. The Negation Problem

Non-cognitivists’ central claim: Moral sentences don’t describe. Expressivists’ central claim (Schroeder : ): Moral sentences express not beliefs, but non-descriptive, desire- like states of mind: “…‘murder is wrong’ stands to disapproval of murder in the same way as ‘grass is green’ stands to the belief that grass is green.” (Schroeder : ) (= the ‘Basic Expressivist Maneuver’) Advantages of expressivism over other noncognitivist views (emotivism, prescriptivism): . explains motivational force of moral thoughts; . applies not only to moral language, but also to moral thought; . avoids the two main objections to speaker subjectivism (the ‘modal problem’ – murder would be wrong even if I didn’t disagree with it – and the ‘disagreement problem’ – if you say that murder is wrong, I disagree with you if I say that what you say is false); . respects the prima facie similarity of moral and descriptive language. Problems of expressivism:

. Schroeder says “noncognitive”, but Horgan/Timmons have non-descriptive, but cognitive beliefs.

. what’s the expression relation? (what do the relations between descriptive sentences and beliefs and between moral sentences and noncognitive, desire-like states of mind have in common?) . why are “murder is wrong” and “murder is not wrong” inconsistent? . why are “murder is wrong” and “murder is not wrong” logically inconsistent? . what is the meaning of moral predicates (necessary to give a semantics for the quantifiers)? . what mental states are expressed by logically complex moral sentences? Problem: () is probably not accepted by Gibbard (cf. Gibbard (), reply to van Roojen); it is certainly not acceptedbyBrandom. Atleastonanaturalreading, () already commits us to the univocity of negation in descriptive and normative language. Thesis of the book: these questions may be answered, but only in such a way that we need to give an expressivist analysis of belief in terms of a further noncognitive attitude and a new, non-standard semantics of descriptive language, which cannot account for the expressive power of natural language. To solve the negation problem, expressivists have to assume that “intention is subject to a basic, noncognitive kind of inconsistency,the kind of inconsistency on which expressivists ought to be hope to be able to model their account of the inconsisteny between the mental state expressed by ‘murder is wrong’ and that expressed by ‘murder is not wrong’ ” (Schroeder : ). The negation problem: To explain why the attitude of ϕing murdering ascribed to Jon in (n′) is inconsistent with the one ascribed to him in (w′): w′ Jon disapproves of murdering. n′ Jon does not disapprove of murdering. n′ Jon ϕs murdering. n′ Jon disapproves of not murdering. The need to provide for a ϕ for (n′) is motivated by the suggested analogy with: w Jon thinks that murdering is wrong. n Jon does not think that murdering is wrong. n Jon thinks that murdering is not wrong. n Jon thinks that not murdering is wrong. Problem is that we need a sentential, not just a predicational operator, allowing for a three-fold distinction between ¬□p, □¬p and ¬□¬p. Hence, we need dual operators. But such dual operators are readily available: In these and similar cases, there are four types of inconsistency: . logical: the opposition between (w) and (n) and between (w′) and (n′), due to the semantics of “¬”. . objectual: the opposition between (w) and (n) and between (w′) and (n′), due to the fact that the same attitude takes ‘contrary’ objects; why these objects are contrary has further, case-specific reasons (for actions: that they cannot be done both by the same person at the same time; for plans: that they are not realisable at the same time) . “type A”: analoguous to the cases of belief and intention (‘personal inconsistency’) . “type B”: neither of the other three

. With most other normative terms (‘is rational’, ‘is the thing to do’, ‘ought’), the attitude is rather one of commendation / advising against. For example: v′ Jon recommends giving. m′ Jon does not recommend giving. m′ Jon advises against giving. m′ Jon recommends not giving. compared to v Jon thinks that it is rational to give. m Jon does not think that it is rational to give. m Jon thinks that it is not rational to give. m Jon thinks that it is rational not to give.

. “Beliefs can be inconsistent in a way that goes over and above the inconsistency of their contents.” (Schroeder : )

Whether or not an opposition is of type B, depends on what we count as analoguous to the opposition between “Bp” and “B¬p”. The distinction between (n) and (n) is the one between: • disapproval / tolerance • recommending / advising against • agnosticism / atheism • non-acceptance / rejection • obligation / permission • hope / regret • anticipation / memory Strategy : deny that (n′) needs an expressivist analysis. Possibilities: • deny that (n′) is in the relevant sense an evaluative sentence; idea: ethics prescribes, deals with what is obligatory and what is forbidden; it only deals indirectly with what is permissible, by staying silent about it; • the predicate that needs analysis in (w′) is ‘thinking-to-be-wrong’: this is an evaluative predicate, because it motivates, expresses a desire-like state etc. The ‘contrary’ predicate, ‘thinking-to-be-not-wrong’, however, is not motivating; • compare the case with imperatives, desires and promises: v′ Close the door! // I desire to p. // I promise to ϕ. m′ I don’t order you to close the door. // I don’t desire to p. // I don’t promise to ϕ. m′ ?? // ?? // ?? m′ Don’t close the door. // I desire not to p. // I promise not to ϕ. • point out that this is a problem for everyone; on the multiple-relation view of judgement, for example, you can stay in the judgement relation to a and F (w), or stay in the judgement relation to a and ̸ F (n), or fail to stay in the judgement relation (n), but nothing like (n) is possible: there is no content to be negated. Strategy : provide an expressivist analysis of (n′). (n′) is analysed as “I disapprove of disapproval of murder.” Arguments: • This is not an atomic attitude, it is complex, and has a structure. • It is an accident that of polar opposites, we tend to take the ‘positive’ one to be basic. Instead of privileging assertion, and analysing rejection as the assertion of the negation à la Frege, we could take rejection to be basic Why expressivists need not be too much troubed by type B ‘inconsistency’: • Whether or not an opposition is of type B, depends on what we count as analoguous to the opposition between “Bp” and “B¬p”. • We may deny that (n′) is in the relevant sense an evaluative sentence. Ethics prescribes, deals with what is obligatory and what is forbidden; it only deals indirectly with what is permissible, by staying silent about it. • The predicate that needs analysis in (w′) is ‘thinking-to-be-wrong’: this is an evaluative predicate, because it motivates, expresses a desire-like state etc. The ‘contrary’ predicate, ‘thinking-to-be-not-wrong’, however, is not motivating. • Compare the case with imperatives, desires and promises: v′′ Close the door! // I desire to p. // I promise to ϕ. m′′ I don’t order you to close the door. // I don’t desire to p. // I don’t promise to ϕ. m′′ ?? // ?? // ?? m′′ Don’t close the door. // I desire not to p. // I promise not to ϕ.

. Primary and Secondary Qualities

.. The Posteriority of Sensible Qualities

Throughout the history of philosophy,we encounter a metaphysical distinction between more and less fundamental properties of perceptible things, which is supposed to explain the reliability of our judgements attributing them. There are different ways of drawing this distinction, which, for systematic and interpretational reasons, have to be carefully kept apart: Error theory Only the A properties are real; our perception of them explains why we falsely think we perceive B properties. Dispositionalism Our perception of A properties is to be explained in terms of things being disposed to appear to us in certain ways, B properties being the categorical basis of such dispositions, and Role A properties being these dispositions of things perceived to appear in certain ways; Realizer A properties being properties of the type whatever configuration of B properties grounds the disposition to appear in this way; Ground A properties being the actual configurations / sets / fusions of B properties grounding the dispo- sition; Realism B properties are real, and had by the things perceived, but Intra-object Causation In things perceived, A properties cause B properties, Intra-object Constitution The A properties of things perceived constitute their B properties, The importance of this contrast for modern philosophy, in particular the thinking of Descartes and of Locke, can hardly be over-estimated.Despite its huge importance, the characterisation itself has proved surprisingly elusive (cf. Nolan a: ). What is it that lies at the bottom of the distinction between properties such as extension, size, shape, motion and position on the one hand and colour, sound, taste, odour, heat, cold etc. on the other? It is often said that primary qualities are primary in the order of explanation (Smith ). The notion of ‘explanation’ in play here is of a distinctively ontological kind. Bennett (: ) lists the following five theses that Locke affirms of secondary qualities and denies of primary ones: . They are dispositions to cause a characteristic kind of sensory state in percipients. “Secondary qualities …are nothing but the powers those substances have to produce several ideas in us by our senses” (Essay II.xxiii.). “[We speak] as if light and heat were really something in the fire more than a power to excite these ideas in us; and therefore are called qualities in or of the fire. But these [are] nothing in truth but powers to elicit such ideas in us” (xxxi.). . They are not in outer objects: “Yellowness is not actually in gold” (xxiii.). . They are not intrinsic to the objects that have them, but rather are relations between those objects and something else: The yellowness, solubility, etc. of gold […] are nothing else but so many relations to other substances, and are not really in the gold considered barely in itself” (II,xxiii.). . They are in minds rather than in outer objects: “Light, heat, whiteness or coldness are no more really in them than sickness or pain is in manna. Take away the sensation of them; let not the eyes see light, or colours, nor the ears hear sounds; let the palate not taste, nor the nose smell; and all colours, tastes, odours, and sounds, as they are such particular ideas, vanish and cease” (viii.). . The ideas of them do not resemble anything in the physical world, as do ideas of primary qualities: “The ideas of primary qualities of bodies are resemblances of them …; but the ideas produced in us by these secondary qualities have no resemblance of them at all. There is nothing like our ideas existing in the bodies themselves” (viii.). Berkeley uses the internality of secondary qualities to argue for the mind-dependence of all properties of objects:

I shall farther add, that after the same manner, as modern philosophers prove certain sensible qualities to have no existence in matter, or without the mind, the same thing may be likewise proved of all other

sensible qualities whatsoever. Thus, for instance, it is said that heat and cold are affections only of the mind, and not at all patterns of real beings, existing in the corporeal substances which excite them, for that the same body which appears cold to one hand, seems warm to another. Now why may we not as well argue that figure and extension are not patterns or resemblances of qualities existing in matter, because to the same eye at different stations, or eyes of a different texture at the same station, they appear various, and cannot therefore be the images of any thing settled and determinate without the mind? Again, it is proved that sweetness is not really in the sapid thing, because the thing remaining unaltered the sweetness is changed into bitter, as in case of a fever or otherwise vitiated palate. Is it not as reasonable to say, that motion is not without the mind, since if the succession of ideas in the mind become swifter, the motion, it is acknowledged, shall appear slower without any alteration in any external object. (§ )

Hume thinks that the notion and rejection of secondary qualities characterises “modern philosophy”:

The fundamental principle of that philosophy is the opinion concerning colours, sounds, tastes, smells, heat and cold; which it asserts to be nothing but impressions in the mind, deriv’d from the operation of external objects, and without any resemblance to the qualities of the objects. […] This principle being once admitted, all the other doctrines of that philosophy seem to follow by an easy consequence. For upon the removal of sounds, colours, heat, cold, and other sensible qualities, from the rank of continu’d independent existences, we are reduc’d merely to what are call’d primary qualities, as the only real ones, of which we have an adequate notion. (I.iv.)

.. The Molyneux Problem and Common Sensibles

.. Distinguishing “red” from “funny”

The notion of response-dependence was introduced by Mark Johnston () as a generalisation of the notion of secondary qualities (cf. Menzies (); Casati & Tappolet (); Yates () for overviews of the debate). Johnston (: ) characterises it as follows (the requirement that the “basic equation”, i.e. the biconditional, is necessarily is not present in Johnston’s original criterion, but clearly needed and commonly added in the literature, cf. also Pettit () and Wright () for similar characterisations):

A predicate F is response-dependent iff there are ‘substantial’ specifications of S, C and R such that it is necessary and a priori that, for all x,“x is F” is true iff x has the disposition to produce in all of S a mental response R under conditions C

While much work has been done on the modal and epistemological status of such ‘basic equations’ (Smith & Stoljar ), often in connection with the so-called ‘missing explanation argument’ (cf. Johnston (: –), () and (: –), cf. the discussion in Blackburn (), López de Sa (), McFarland (), Menzies & Pettit (), Miller (), Miller (), Miller () and Johnston (: –) for a new version) and their rigidifications [cf.][]haukioja:, they have rarely been discussed in the context of metaphysical explanation (cf. Ásta [Ásta Kristjana Sveinsdóttir] () for an interesting first step). The “missing explanation argument” Johnston (: –) deploys against any “response-dependent account of value or reason” (: ) is that it cannot account for “the felt independence of value from value-directed responses” (: ) that underlies Socrates’ objection to Euthyphro’s proposed definition of piety as what the gods love, which consists in pointing out that while

(Eu1) The gods love x because x is pious. is true,

(Eu2) x is pious because the gods love x.

is false. In Johnston’s opinion, the friend of response-dependence may accept instances of this argument, but not its most general conclusion:

…now the Socratic objector will take the decisive step, insisting that whatever is a reason is not a reason because we would stably take it to be so as we approach ideal conditions. Rather, in the most fortunate case, we would take it to be a reason under such conditions just because it is a reason. And now our kind of response-dependent theorist must dig in. The hyper-objectifying error has at last been manifested. The objector in effect wants hyper-external reasons, reasons which could in principle outrun any tendency of ours to accept them as reasons, even under conditions of increasing information and critical reflection. (Johnston : )

Even if this reply is accepted, there are (at least) two problems with an account of secondary qualities as response- dependent: first, it is a classification of predicates, and second, there are ‘basic equations’ for all (Pettit , ) or at least all basic concepts (Pettit : ). The first problem is particularly important because, arguably, the whole point of Locke is that our ordinary concepts for secondary qualities are response-dependent while the properties are not. The idea of the second criticism, in short, is that many concepts expressing what we intuitively take to be primary, and hence non-response-dependent qualities F have (or had) their reference fixed by some descriptive material F∗ such that all instances of “x is F iff x is F∗” is known a priori by speakers competent with “F”. It amounts to refuse the invitation of Johnston (: , n. ) to exclude reference-fixing descriptions by fiat. Fodor puts the situation colourfully:

“…the concept chair expresses that property that things have in virtue of striking minds like ours as appropriately similar to paradigmatic chairs. (Fodor : )

Another worry is more damaging: with what right are we assuming that there are independently ‘substantive’ specifications of the response R? Is it not at least possible that all that red things have in common is that they appear red, to some under some circumstances? There would thus appear to be a vicious circularity: the property that is analysed dispositionally figures in the content of the response allegedly constitutive of it. We should accept that there is a circularity, but deny that it is vicious:

[The dispositionalist] may point out that a circular account of a property can still be true, and indeed informative, despite its circularity. For instance, to define courage as a disposition to act courageously is to give a circular definition, a definition that cannot convey the concept of courage to anyone who does not already have it. Even so, courage is a disposition to act courageously, and this definition may reveal something important about the property – namely that it is a behavioural disposition. (Boghossian & Velleman : )

Boghossian and Velleman claim that the circularity is vicious because the context in which “red” appears on the right-hand side of an instance of the dispositionalist biconditional is intensional:

Not only does [the instance of the dispositionalist biconditional for “red”] fail to tell us which colour red is, then; it also precludes visual experience from telling us which colour an object has. The former failure may be harmless, but the latter is not. […] the dispositionalist about colour not only invokes the content of colour experience in explicating that content; he places that content in a relation to itself that it is impossible for it to occupy. (Boghossian & Velleman : –) . I can think of no other way to ‘remove’ the circularity than introducing so-called “primed” or “appearance” properties. Even if they can independently made sense of, however, their introduction only helps if they themselves are, rather arbitrarily, exempted from the analysis, making it doubtful, as Boghossian & Velleman (: ) stress, that they are ever exemplified. If they are not so exempted, the account is either circular (if they are analysed in terms of colour-properties) or leads to an infinite regress (if they are analysed in terms of further, higher-order appearance properties). . They qualify the context as “intentional”, though they presumably mean “intensional”.

. The Question of Objectivity

.. Faultless Disagreement

The (alleged) phenomenon of so-called ‘faultless disagreement’ has recently been used to argue for relativism: only relativising truth or content to ‘contexts of assessments’ can explain, it has been said, why there is no real dispute about matters of taste, even though taste ascriptions purport to describe reality. I argue that another, more plausible, diagnosis is available: the ‘reality’ in question is partly mind-dependent, and our reference to it is aspectival. This explains, or so I argue, the data and allows for a plausible liberal account of how we should deal with such disagreements. Faultless disagreement, it is commonly held, might arise as follows: utterance u1 by person A: “This orange is sweet.” utterance u2 by person B: “This orange is bitter.” Let us also suppose that the disagreement is faultless at least to the extent that (i) A and B are equally well justified in their assertions; (ii) both report their opinions sincerely and with the intention of contradicting the other and (iii) they would stick to their assertions if they had tasted the bit of the orange the other is in fact chewing. Let us also suppose that the circumstances are normal, in that neither A or B is in a state usually taken to disqualify them (or to significantly lower their credibility) in matters of taste (e.g. neither has just brushed their teeth), and that they their faculties of taste are physiologically alike. It is commonly, though not universally, agreed that such faultless disagreements in matters of taste actually occurs in real life and I will presuppose so in the following. The best reason, in my view, to believe in the existence of faultless disagreement is so to say political: given their irresolvability in principle, we want to characterise such disagreements as faultless, so as not to be obliged to take sides and to attribute blame or praise in an asymmetric way. Matching in-principle-irresolvability with rational faultlessness thus not only allows, but provides an argument for tolerance: that the disagreement is faultless is a reason that it should ultimately be avoided. As things stand, however, there are faultless disagreements, or at least I will assume so. The philosophers’ problem, as often, is to explain how this is possible. The main task of a philosophical account of faultless disagreement is to show how both A and B may disagree (at least in the minimal sense in which it would be wrong for A to stand by u1 and ‘say the same thing’ as B) and both be ‘right’ (in the sense required for their disagreement to be faultless). This can only be done by distinguishing the object of their disagreement from what they are right about. It is in terms of such a conceptual distinction that we have to steer a middle course between the Scylla of objectivism (allowing for disagreement, but not for faultlessness) and the Charybdis of subjectivism (allowing for faultlessness, but not for disagreement). The distinction needed should account, in the first place, for the blamelessness of both A and B: such an account should explain why, given their situation and their communicative ends, it makes sense for them to make their

. A dissenting voice is Democritus who believes that sweet things are sweet because they entirely consist of round and large atoms, and bitter things are bitter because they have only spherical, smooth, scalene and small atoms. It is not clear to me how Democritus could even explain the appearance of faultlessness of disagreements like the one between A and B. . Of the “de gustibus non est disputandum”, Stigler & Becker (: ) say that “the venerable admonition not to quarrel over tastes is commonly interpreted as advice to terminate a dispute when it has been resolved into a difference of taste”. It should be noted, however, that they oppose this interpretation. . I will remain neutral, however, about the question how far my – and the relativists’ – discussion about taste-disagreements generalises. It does carry over, it seems to me, to knowledge attributions (where I would epistemic profiles have play the rôle of so-called ‘standards of knowledge attribution’), but probably not to future contingents. . This way of setting up the problem is perhaps not by all lights the standard one. It is often said that the task is to account for alleged linguistic “data”. While I do not think it matters for my purposes here, I nevertheless think that the real problem already arises in the armchair. First, the “data” have been challenged and the jury on the empirical relevance of all this is still out. Second, making sense of the conceptual possibility of faultless disagreement is prior to explaining its actuality. Third, and most importantly, we should keep in mind that what philoso- phers take to be “accounts” (of psychological reality, presumably) linguists take to be “models” (to be judged by their predictive force and theoretical usefulness). It may very well be (and, from the armchair, even looks quite probable) that different and theoretically non-equivalent such models will be empirically equivalent, which would then again land us in the armchair (which we did not want to leave anyway).

respective utterances, and it should do so symmetrically. The broader significance of the point is to distinguish between blameworthiness for norm-violation and excusability – an agent may be excusable and still blameworthy, or blameless but not excusable (this is not Williamson’s terminology, who correlates blamelessness with justifica- tion).

The characteristic relativist move is to draw this distinction in terms of the evaluations of u1 and u2 for truth. This, relativists typically claim, not only makes faultless disagreement possible, but is also the best explanation of its reality. The central relativist claim is thus: relativism the evaluations of u1 and u2 for truth are not invariant

Correlative to different sources of this variance (or ‘relativity’) in the evaluations of u1 and u2 for truth, we may distinguish different types of relativisms: content relativism what is evaluated when we evaluate u1 and u2 is not invariant ontological relativism what u1 and u2 are evaluated with respect to is not invariant truth relativism what u1 and u2 are evaluated for is not invariant Content relativists find the variance relativists look for in what is said in the two utterances, distinguishing the two different and non-contradictorily true things said in u1 and u2 from the more restrictive content A and B disagree about. Truth relativists deploy a notion of relative truth that applies to both u1 and u2, while still being contrary in the sense of allowing for disagreement. In contrast, ontological relativists locate the relativity or variance in what u1 and u2 are about or in what makes them true. As with the two other versions of relativism, the main task for the ontological relativist is to say what it is A and B disagree about. In contrast to them, however, she has the resources to draw the relevant distinction at the ontological level. Is is here, I will argue, that her distinctive advantage lies. Ontological relativism has, as far as I know, not been squarely addressed in recent discussions of semantically motivated forms of relativism. The reason for this, I suspect, is that the debate has usually been framed as a discussion of the resources available to the contextualist to accommodate the apparent data. A common complaint against contextualism is then that it collapses into subjectivism, in that it has to claim that the utterances of A and B are covertly or implicitly about themselves:

If [contextualism] were right, there would be an analogy between disputes of inclination and the ‘dispute’ between one who says ‘I am tired’ and her companion who relies, ‘Well, I am not’ (when what is at issue is one more museum visit). (Wright : ) According to [simple indexicalism], when I say ‘Blair ought to go to war’, I assert that my moral code requires Blair to go to war. I am therefore talking about my moral code. However, this seems wrong: I talk about Blair and what he ought to do, and not about my moral code and what it requires Blair to do. (Kölbel : ) (cf. also Kölbel (: )) The contextualist takes the subjectivity of a discourse to consist in the fact that it is covertly about the speaker […]. Thus in saying that apples are “delicious”, the speaker says, in effect, that apples taste good to her (or to those in her group). (MacFarlane : ) On this [contextualist] view, “It’s wrong to cheat” involves ellipsis, or a place holder indicating a set of standards, a code, whatever. What [its] use says depends on what has been elided or what is being assigned to the place holder. (Richard : ) …any indexical proposal distorts the content of the utterances, for it claims that Anna and Barbara

. Blamelessness here has to be restricted to ‘primary norms’; even when an agent complies with these, she may still be blameworthy in a wider sense for violating derivative norms. Timothy Williamson explores such derivative norms in “Justifications, Excuses, and Sceptical Scenarios” (ms., available on this webpage, presented as “Legality and Law-Abidingness” in Geneva on September , ). . There are, of course, others, and perhaps even more prominent arguments for relativism, e.g. the claim that it best explains the so-called disagreement and retraction data. Insofar these arguments are not versions of the argument from faultless disagreement (some of them are, the (lack of) disagreement being diffferent stages of the same person), I do not have the space to discuss them here (but cf. also fn. ).

. Such a distinction can be drawn in very different ways. Some theorists advocate broadening the notion of what is said to include much more than literal and speaker meaning,

. So-called “indexical relativism” (cf. Kölbel ) is a form of contextualism and does not satisfy (relativism) above.

assert (and believe) propositions that concern their own standards, even though it seems that they are merely comparing Depp and Pitt without their assertions and thoughts having any reflective content. (Kölbel : ) The second option is to adopt a form of [contextualism] according to which the candidate disputes are really faultless, but they are only apparent disagreements. In this case the view is that the judgments conveyed by A and B are, or are equivalent to, those expressed by sentences involving reference to A and B themselves. (Iacona : )

If content itself is relative in this way, there is no common content for the speakers to disagree about: Cappelen & Hawthorne () call this the ‘problem of shared content’. While this problem is not unique to contextualism (as we have seen, all three brands of relativism face it too), I will grant for the purposes of this paper that contextualists are ill-equipped to solve it: A and B are just not talking about themselves. What I take issue with, however, is that this argument rules out ontological relativism too. While ontological relativists hold that u1 and u2 are to be evaluated for truth with respect to different bits of reality, they are not committed to the claim that the utterances are (explicitly or implicitly) about these different bits. Such a distinction between aboutness and truthmaking is familiar in other areas of metaphysics: while physicalists claim that ascriptions of mental states are made true by physical states, they do not have to (and better do not) claim that such ascriptions are about physical states. I propose to explain the faultlessness of the disagreement between A and B in terms of their utterances being about different aspects of the taste-profile of the orange. Even though the aspects they perceive are different, they are aspects of the same orange, which is the truthmaker of their claims if they have a truthmaker, which explains that there is disagreement in the first place. Facts about aspects of objects which are such that they may have conflicting aspects, i.e. aspects that cannot be manifested ‘together’ (in a sense to be explained much more below), are plausible called “non-objective”. I thus explain faultless disagreement by the non-objectivity of its subject matter. I want to sketch how ontological relativists may allow for faultless disagreement by claiming that the conflict between A and B is faultless because u1 and u2 are about perspectival facts: A and B are both talking about the taste of the orange, but because taste is a secondary quality of oranges, the aspects of its taste profile that they are talking about are different: in and of itself, the orange does not have a taste, it’s taste is a relational property, implicating tasters. Even though they are talking about different things, however, A and B do disagree, because the different perspectival facts are still grounded in the same item of reality – the orange. Disentangling in this way aboutness and truthmaking allows, I will further claim, for a politically satisfactory accommodation of faultless disagreements. Before moving on to the positive account, we have to get clearer on the desiderata for a theory of faultless disagreement about taste and on the reasons why the extant accounts do not quite meet them.

.. Arguing about taste

Whoever has organised conference meals for philosophers knows that normally they do not care much about tastes. Here as in other areas, however, philosophers are barely exemplars of the enlightened rational ideal man: it is certainly possible to sensibly argue about taste and tastes, many rational people do and much can turn on the outcome of such discussions. As people having brought up in rigidly food-conscious or pathologically smell-averse communities can attest, there are rational limits to taste fanaticism: though disagreements about taste may be sensible, even productive, there should not be allowed to become struggles for life and death. But where is the line

. To answer this charge, contextualists can draw on the rich literature on the semantics/pragmatics interface and invoke the recently much studied mechanisms of ‘free enrichment’, whereby contextual information is made available for the hearer without thereby becoming part of the semantic content of the utterance. . This includes, in my view, contextualists who externalise the way in which their standards are determined indexically by ideal judges of taste or the standards of taste one should have. Such proposals either claim that A and B talk about these things in u1 and u2, or they claim that these indexically determined elements enter into the presuppositions of what they are saying. The first claim is false, the second very questionable, as the disagreement seems to persist even when these presuppositions are explicitly cancelled: “I know I am not an ideal judge of this, but the orange is sweet – no, it is bitter”; “I know I do not have the standard of taste I should have, but the orange is sweet – no it is bitter”. . This is an exact reversal of the direction of explanation championed in the broadly Dummettian tradition, by e.g. Crispin Wright and Max Kölbel (, : ). It may be that the predominance of the Dummettian take on the realism/antirealism debate has occluded ontological relativism from the view of its participants.

between sensible and insensible disagreements to be drawn? Reflection on this question may help us delineate the desiderata of a theory of faultless disagreement about matters of taste. A first desideratum is that an account of taste disputes should leave open, but not entail the existence of expertise in the matter disputed. Distinguish blamelessness from faultlessness: I know that I have many false beliefs about electricity, including the belief that electricity literally flows, recognise this as metaphorical and a literal reading of it as magical thinking. I also know how to find out. But it may still be permissible for me not to correct my errors, even though it would be good if I were to. I am at fault, but not to be blamed. The case with applications of predicates of personal taste (‘delicious’) to wines, say, is exactly analogous. It would be good for me to know more, to have judgments that are better justified, deeper in some sense, but as a plain user of wine, and plain user of electricity, I cannot be blamed for my opinions, though they may still be false and I therefore at fault. The claim to expertise that adds ‘oomph’ to a dispute about taste cannot be part of what is – explicitly or implicitly – under dispute, however: these matters, who is an expert, how things taste according to a given standard or how things should taste to us, may very well be objective, or are at least not subjective in the same way as tastes are supposed to be. Another approach that has recently been advocated – the so-called “superiority account” defended in Zakkou – holds that u1 and u2 do not only convey (or at least: do not only express the propositions) that the orange is sweet and that it is bitter respectively, but also something like “A’s standard of sweetness is best” and “B’s standard of bitterness (or, perhaps, of sweetness?) is best”. The ‘directly’ expressed propositions are then counted as both true, but non-contrary, and at least one of the contrary pragmatically conveyed propositions will be false. This divide-and-conquer strategy risks inheriting the weaknesses of both proposals it combines: (i) to account for the non-contrariness of the propositions directly expressed, they have to have contextual elements, which makes them at least partly about A and B; (ii) to account for the faultlessness of the disagreement with respect to the pragmatically conveyed propositions, more has to be said. Another problem, however, is more serious: Even if we somehow manage to individuate standards of sweetness finely enough to map the patterns of disagreement (and lack thereof) we observe in discussions of taste, it is not clear what the content of the propositions allegedly pragmatically conveyed is supposed to be. What does it mean to say that some standard of taste (or, more particularly, of sweetness) is best? The very possibility of faultless disagreement seems to undermine any attempt of spelling this out. A second desideratum is to do justice to the intimate connection our judgments of taste have with our sensibility: judgments of taste are perceptual. Desideratum (i) straightforwardly rules out the so-called “presuppositional account”, defended inter alia by Dan López de Sa (: –, , and This is as it should be, as it is simply implausible that utterances of trigger the presupposition that A’s and B’s tastes are the same. Even if they would, moreover, they would presumably be cancelled in the case of transparent and explicit disagreement. Such a theory, I argue, promises to meet the following four desiderata for an account of disagreement in the realm of taste: . allows for disagreement not conditional on the presupposition that the tastes are the same; . allows for culinary learning and teaching; . allows for a distinction between deeper and more shallow disagreement and places disagreement about culinary tastes on this scale between disagreements about the morally right and the comic.

. This problem has received surprisingly little attention, though Kölbel (: ) discusses it. . Zakkou suggests that the pragmatically conveyed propositions are merely presupposed, not believed. Even if we grant that conflict of presuppositions accounts for disagreement, this inherits the problems of the presuppositional account: in particular, it does not account for disagreement in situations where the presupposition is explicitly cancelled.

Desideratum (i) straightforwardly rules out the so-called “presuppositional account”, Desideratum (ii) rules out subjectivism: if A’s and B’s respective utterances are interpreted as reports only of private impressions, if they are Desidaratum (iii) puts pressure on truth and content relativism.

.. Aspectival reference

There is a grain of truth in what relativists say about the content of ‘it’s tasty’ - it’s not about the speaker (nor about judges, nor does it involve a presupposition of commonality) - but perspectival facts can explain all this The reason is: taste, colour and sound properties of things are extrinsic, and different taste / sound / colour profiles belong to the same thing; these are perspectival entities, lesser entities, and they being the truthmakers of the relativistic claims explains in what sense their perception is part of how the world reveals itself to us and how the incompatibility of different perspectives is not at bottom one of content, but of mode. My broader aim is to characterise a genuinely metaphysical form of relativistic content, where each party states a perspectival fact, and is thus epistemically faultless. Because the different perspectival facts are still grounded in the same item of reality, there nevertheless is something for the contestants to disagree about. We should avoid the postulation of mysterious standards and codes, and take the truthmakers of relative judgments to be secondary qualities – real, but non-fundamental properties of the external world. Faultless disagreement is not so much about truth itself (otherwise it could not be faultless), but about the truthmakers of our ascriptions of secondary qualities. When you and I disagree about whether liquorice is delicious, we represent different perspectival facts – in this sense, our disagreement is faultless. It is, however, still a disagreement about some real properties of liquorice, those responsible for you finding it tasteful and me finding it disgusting, and it is in these independently and objectively existing properties that our respective perspectival objects are grounded. We should take the truthmakers of relative judgments to be secondary qualities – real, but non-fundamental proper- ties of the external world. Faultless disagreement is not so much about truth itself (otherwise it could not be faultless), but about the truthmakers of our ascriptions of secondary qualities. When A and B disagree about whether the orange is sweet, they represent different perspectival facts – in this sense, their disagreement is faultless. It is, however, still a disagreement about some real properties of liquorice, those responsible for you finding it tasteful and me finding it disgusting, and it is in these independently and objectively existing properties that our respective perspectival objects are grounded. Compare utterance () with utterance ():

() p

() The right thing to think about these matters, according to me, is: p

Plausibly, many utterances of commit me to . In cases where I could not correctly (ie. truly) utter but still utter , the sincerity condition for assertions is violated. But contrast and with the following utterances of my interlocutor:

(’) ¬p

(’) The right thing to think about these matters, according to me, is: ¬p

While ’ contradicts , ’ does not contradict .

I said above that relativism (in my sense of the term) is characterised by the claim that to make sense of two utter- ances being faultlessly in disagreement, they have to be differently evaluated for truth. In all the extant literature, this has been taken to mean that they are, in a sense to be specified further, both true. In this last section, I want to explore another alternative: both faultlessly disagreeing utterances are in fact false, and this has consequences for how the disagreeing parties should conduct their argument.

Chapter

Belief

. Content Attribution

.. Correctness and Truth

By being tied to correctness and truth respectively, intentionality and representation differ quite markably in what may be called their ‘metaphysical status’.

.. Indeterminacy of Content

We have, I hope, motivated the theoretical need to proceed (i), via intentionality, from accuracy conditions of pictorial or sense-datum representations to correctness conditions of attitudinal representations and (ii), via rep- resentation, from such correctness conditions of property ascription or ‘feature-placing’ to the truth-conditions of full-blown predications. In this section and the following, we have to pay attention to the degree of indeterminacy of ‘content’ that is thereby introduced. Distinguishing correctness from truth allows distinguishing two sources in which indeterminacy in (the attribution of) content to representational states may arise. There may be different, equally good (i.e. bad) candidates to account for their intentionality, different things, features or facts they may be said to be about. Many mental states, while they are directed and directed towards things (features, fact) of a certain sort, may not meaningfully be said to be directed towards anything in particular. In addition, and independently of this indeterminacy, there may be no meaningful way of attributing to them one particular, rather than another quite different, set of truth- conditions. While they do impose a condition on the world for their truth, the condition does not correspond to any determinate way or set of ways the world might be.

.. The Determination of Aboutness

Even though intentionality (directedness, aboutness), as has been argued, is a genuine feature of many of our attitudinal reactions to the world, it is in many cases very much indeterminate what a mental state is directed towards. FOR EMOTIONS. PROBLEM OF SUPERVENIENCE For emotions, the problem of indeterminacy arises in the following way. When I fear the lion, my fear is directed towards and a reaction to (present or absent) danger, but is appropriate only if the lion really is dangerous. Lions, however, are not dangerous tout court – they are dangerous in virtue of exhibiting other properties, such as being hungry and aggressive, having sharp teeth, not being separated from their putative prey by e.g. a cage etc.

One problem with non-existent things is that there is a huge (and moreover: an indeterminate) number of them. We all have heard of Pegasus, the winged horse that was, according to Greek mythology, captured by Bellerophon. How many winged horses do not exist? What about the Easter Bunny? DIRECTED, BUT NOT TOWARDS ANYTHING IN PARTICULAR I OWE YOU A HORSE

. Content

.. The Attribution of Representational Content

The indeterminacy introduced in the second step is of a different kind, both with respect to its nature and to its theoretical explanation. The problem of the indeterminacy of representational content arises from the combination of two factors. On the one hand, the world makes more distinctions than we do; this follows from the fact that at least in a large number of cases, we do not ‘make’ distinctions at all, but rather latch onto distinctions that are already there, privileging some by marking them representationally, plastering over many others. On the other hand, what distinctions are relevant varies with time, across people, even with contexts: even with respect to our own representational repertoire, we almost always underexploit its capacity to make distinctions. When we attribute representational content to some state and cannot specify it in terms of representational content possessed by some other state, the conditions we impose on the world as those under which the state is true are, by the second factor, coarser-grained as other conditions we could also impose. By the first factor, these other, more fine-grained conditions do generally pick out different ways in which the world could make the state true. With what justification, therefore, do we take the broader and not the more narrow conditions to specify the content in question? Within a theory of mental representation this problem arises with respect to the question how exactly mental representations acquire their truth conditions. One superficially tempting proposal is that the representational content of an internal representation is determined by simple counterfactual co-variation relations, perhaps backed by causal laws and past causal dependence (Stampe ; Dretske ). Causally-backed co-variation relations are ubiquitous: tree rings causally co-vary with the age of the tree, smoke causally co-varies with fire, and utterances of ‘Fire!’ causally co-vary with the presence of a speaker at the time and location of the utterance. It is therefore widely agreed that simple causal co-variation relations cannot be what determines the representational content of mental states. First, causal co-variation cannot adequately account for the possibility of error. A belief or perception can misrepresent the world – but a tree ring cannot misrepresent the age of the tree. In general, if we identify that which is represented by a given state type with whatever causally co-varies with it, then the existence of the mental state will guarantee the existence of the thing represented. A second related problem is that such ‘natural’ representation cannot explain the determinacy of representational content. Your ‘horse’ concept is accurate of all and only horses, but its occurrence in your thinking causally co-varies with many other things, including the proximal retinal stimuli that lead you to identify horses, as well as many non-horses in the distal environment. So the worry is that causal co-variation accounts of mental representation will take ‘horse’ to represent not the natural kind horse, but the disjunctive kind horse-or-horsey-images-on-the-retina-or thoughts-of-cowboys-or-cows-on- a-dark-night-or-donkeys-in-the-distance-or-… (Fodor ). If we want to explain the determinate content and fallible truth conditions we attribute to mental states, we need a less permissive representation relation than mere causal co-variation. To further illustrate this obvious point, consider a much-discussed example: magnotactic bacteria. Certain anaer- obic bacteria possess internal structures – magnetosomes – that are sensitive to magnetic fields. In the bacterium’s home environment, these structures normally align with the earth’s magnetic field, causing the bacterium to move downwards towards oxygen-poor water, which is essential to its survival. So there is a reliable co-variation – backed by causal laws and selected for by evolutionary processes – between the alignment of the magnetosomes with the

direction of a magnetic field, the North Pole, the ocean depths, oxygen-free water, survival-friendly conditions, and so on. Should we say that the alignment of the magnetosomes in these bacteria is a representation with the content, say, magnetic north is in that direction, or oxygen-poor water is over there, or it’s safe down there? The reason not to attribute them such contents is not in the first place their lack of the supposedly necessary conceptual capacities. The main reason is that attributing them such representational properties is not doing any real explanatory work. As many theorists have pointed out, we can give a full explanation of the movements of the bacterium in one di- rection rather than another by providing a mechanical account of (i) why magnetosomes are sensitive to magnetic fields, (ii) how the presence of a magnetic field controls their alignment within the organism, (iii) how alignment of the magnetosomes affects the orientation of the organism as a whole, and (iv) how the organism’s flagella propel it in the direction in which it’s oriented. There is no need to mention representational states with intrinsic truth con- ditions in this explanation – the mechanical account is a complete proximal explanation the organism’s reaction to the stimulus. Not all cases of measurement thus involve representation. Which ones do? With respect to the accuracy conditions invoked by vision science, Burge has argued that they are plausibly construed as being intrinsic to the perceptual states themselves: perceptual states are individuated in part by their function of representing particular distal features of the environment. Moreover, Burge argues that individuating perceptual states in this way is crucial to the explanations of perceptual psychology – vision science, in Burge’s view, is organized around explaining how representations of the world are constructed from impoverished sensory input:

The fundamental mode of explanation in the perceptual psychology of vision is to explain ways in which veridical representations of the environment are formed from and distinguished from registra- tion, or encoding, of proximal stimulation. Veridicality, fulfillment of representational function, is the central explanandum of visual psychology. Illusions are explained as lapses from normal representa- tional operation, or as the product of special environmental conditions. (Burge : -)

Assuming that perceptual states do have intrinsic accuracy conditions, what exactly do they represent? A frog snaps at a passing fly. Assuming that its visual system is representing some feature of the world, what exactly is the content of that representation? Does the frog’s visual system represent something as specific as the content that there is a house fly (musca domestica) at location x? Or does it represent that there is a packet of frog food at location x? Or does it just represent that there is a small, dark, moving object at location x? Pace Burge, the explanatory function of content attribution itself does not decide among these alternatives. In many cases, the possible psychological explanations themselves compete. Consider alternative explanations of a five-year-old’s choice of ice cream over broccoli: . The child chooses ice cream over broccoli because in the past she perceived ice cream as sweet and she likes sweet things. . The child chooses ice cream over broccoli because in the past she perceived ice cream as having concentrated nutrients and she likes things with concentrated nutrients. . The child chooses ice cream over broccoli because in the past she perceived ice cream as containing large proportion of mono- or disaccharides and she likes things containing large proportion of mono- or disac- charides. While these explanations differ in what they attribute to the child, and what they predict about its future behaviour, there may simply fail to be, not only in practice, but even in principle, any way of singling out the correct one. What is represented intrinsically by whatever ground the child has for her choice may be radically, metaphysically and not just epistemically indeterminate.

.. Thinking that x is F

Aspect-perception involves a curious doubling, or re-duplication, of content. There is, one the one hand, the (supposedly) real danger in front of me facing me in the shape of a ferocious lion; this is the danger my fear is directed towards and intentionally represents. On the other hand, danger is also an aspect of how the world is given to me: (the feeling of) danger is an aspect of my situation, a feature of how I am when faced with the

lion. Though not much more than a metaphor, representation-as brings out this double aspect: I represent the dangerous lion as dangerous, and thus correctly. In Aquinas’ theory, perception, the central and paradigmatic case of representation, is the combination of two processes, which together constitute the uptake of form without matter. There is, (i), the the getting of the (inner) form: by the application of a representational faculty to something out there in the world, something inner is created (the form of the thing outside in its ‘intentional existence’, ie. a structural property of the brain); by being in such a state, we locate ourselves with respect to things outside of us. There is also, (ii), the the seeing of the (outer) form: our brain being in such a state causes us to see the form, ie. to reach out in more or less successful ways to the things we take to be of this form; as a result of this activity, the form out there gets to be seen. It gets to be seen as such-and-such, and in certain ways, in virtue of the way the seeing is done. In different types of representation, the importance of these components varies. In sensation, the first of these components is much more important than the second; the act-aspect is mere awareness, which, as Brentano correctly recognised, is often already contained in the representational state. With imagination, emotion, judging, thinking, and mere supposing – states which do not only have representational properties, but exhibit intentionality as well – the second, mind-to-world aspect becomes more important. Even with such ‘higher’ forms of cognition, however, the first component never completely disappears: all such mental acts are properly understood as reactions. Like perception and emotion, belief is a reaction: the representational state is an entertaining of a thought, and the mental act is a type of Stoic assent (in the case of belief; dissent in the case of disbelief). To believe, Augustine says, is ‘cum assensione cogitare’ (de praed. sanct. ). In the same way that thinking, by the ‘affirmative’ mode, gives beliefs, it may, by the ‘interrogative’ mode, give questions, and by the ‘mode of conjecture’, suppositions. What is belief a reaction to? Brentano may help us here, drawing our attention to the possibility that inner representation is containment (i.e., works roughly in the way in which a quotation name represents what it quotes by surrounding it by quotation marks). Quoting De Anima III. , p. , b, , he remarks that a doubling of intentional objects is only needed if we individuate them by the representational states, but not if we individuate them (correctly) by the mental acts directed at it:

“…the presentation [Vorstellung] of the sound is connected with the presentation of the presentation of the sound in such a peculiarly intimate way that the former, by obtaining [indem sie besteht], contributes inwardly to the being of the other [innerlich zum Sein der anderen beiträgt].” (bk. , ch. , §, : )

The resulting picture is the following: There is thinking going on, and some of it is partitioned into a belief of a certain content by an attitude of assent the subject directs at its own mental state. Such a belief, however, is not a representational vehicle in the traditional sense, that allows the subject to represent something. Rather, the belief, as a mental particular (and thus distinguished from the state of believing), is itself produced by (and thus posterior to) the representing of form. The attitude of assent directed at the immanent object is at the same time, and by the same act, a representation of itself as a psychological state, i.e. a belief. It thus projects its own form onto an amorphous mental process of thinking. There is a peculiar interweaving [“eigentümliche Verwebung”, : ] of the object of inner presentation (the belief) and the presentation itself (the thinking), and both belong to the same act, the believing.

.. Representation-as

We have noted above the close kinship of my fear of the lion, whereby I represent it as dangerous, and of my perception of the lion as dangerous. Though it should not be identified with my fear (for it may occur without it, as e.g. when I am courageous), such aspect-perception often accompanies and derives from the emotional state. This basic structure of representation-as is perhaps most clearly discernible in the case of emotion: when I am afraid of the lion, I locate myself with respect to it (putting it, so to say, on my mental map of the world), but I also reach out to an axiological aspect of the world, the lion’s being dangerous. It is in virtue of this formal (or, in Brentanian terms, ‘secondary’) object of my fear that I represent the dog as dangerous (Mulligan a, b, ; Teroni ). This representation, however, is not just grounded in the ‘mode’ of my entertaining a certain content,

or the ‘attitude’ I have towards the direct object of my state, the lion. It is, rather, a reaction to an independently given feeling of value, which reveals to me the dangerousness of the lion. Only by construing the emotion as a reaction, can we explain how it contingently has the valence it has, i.e. why, for non-perverted people, fear is not a pleasant emotion to have, and why being afraid feels a certain way. That the attitude I take towards the cognitive base comes with a feeling explains, I think, the phenomenology of the mental. This ‘inner side’ of our mental states, which enables and sometimes constitutes our awareness of them, is a matter of how they feel. To account for how they feel may be one way of cashing out what it means to say that ‘there is something it is like’ to be in them, though the latter notion is so slippery that it is hard to be sure. That my taking up an attitude towards a representational state that relates it to a formal object is quite unmysterious and ubiquitous phenomenon; while it requires, as its ground, that the attitude has a determinate nature, it does not require either phenomenal content nor autonomous phenomenal properties. CONSTRUCTION OF THE PERCEPTUAL OBJECT (ie the thing I see as being so-and-so) – IS LATER, IN INTENTIONAL OBJECTS

. As Teroni and Deonna would have it, cf. their ; ; ).

Bibliography

Abela, Paul, . Kant’s Empirical Realism. Oxford: Oxford University Press.

Adam, Charles & Paul Tannery (editors) , . Œuvres de Descartes. edition. Paris: Librairie philosophique Jean Vrin. Tomes I – XI.

Adickes, Erich, . Kant und das Ding an sich. Berlin: Pan Verlag Rolf Heisse.

Alanen, Lilli K., . Descartes’s Concept of Mind. Cambridge, Massachusetts: Harvard University Press.

Allais, Lucy, . Kant’s One World: Interpreting ‘Transcendental Idealism’ . British Journal for the History of Philosophy (): –.

Allais, Lucy, . Intrinsic Natures: A Critique of Langton on Kant [on Langton ()]. Philosophy and Phenomenological Research (): –.

Allais, Lucy, . Kant’s Idealism and the Secondary Quality Analogy. Journal of the History of Philosophy (): –.

Allison, Henry E., . Kant’s Transcendental Idealism: An Interpretation and Defense. edition. New Haven, Connecticut: Yale University Press. Second edition: Allison ().

Allison, Henry E., . Kant’s Transcendental Idealism: An Interpretation and Defense. edition. New Haven, Connecticut: Yale University Press. Revised and enlarged edition of Allison ().

Almér, Alexander & Dag Westerståhl, . Review of Cappelen & Hawthorne (). Linguistics and Philosophy (): –.

Almog, Joseph, . The What and the How. The Journal of Philosophy (): –.

Almog, Joseph, . The Structure-in-Things: Existence, Essence and Logic. Proceedings of the Aristotelian Society (): –.

Almog, Joseph, John R. Perry & Howard K. Wettstein (editors) , . Themes from Kaplan. Oxford: Oxford University Press.

Annas, Julia, . Hellenistic Philosophy of Mind. Berkeley, California: University of California Press.

Aquila, Richard E., . Representational Mind: A Study of Kant’s Theory of Knowledge. Studies in Phenomenology and Existential Philosophy, Bloomington, Indiana: Indiana University Press.

Aristotle, . De Anima. Books II and III (with certain passages from Book I). Clarendon Aristotle Series, ed. J.L. Ackrill, Oxford: Clarendon Press. Translated, with introduction and notes, by W.D. Hamlyn.

Aristotle, . Eudemian Ethics Books I, II, and VIII. edition. Clarendon Aristotle Series, ed. J.L. Ackrill and Lindsay Judson, Oxford: Clarendon Press. Edited and translated with a commentary by Michael Woods; first edition: Woods ().

Aristotle, . The Complete Works of Aristotle – The Revised Oxford Translation. Princeton, New Jersey: Princeton University Press.

Aristotle, . De Anima. edition. Clarendon Aristotle Series, ed. Lindsay Judson, Oxford: Clarendon Press. Translated, with an introduction and commentary, by Christopher Shields.

Armstrong, David M., . A Materialist Theory of the Mind. London: Routledge & Kegan Paul.

Armstrong, David M., . A World of States of Affairs. Cambridge: Cambridge University Press.

Armstrong, David M., . Truth and Truthmakers. Cambridge Studies in Philosophy, Cambridge: Cambridge University Press.

Ásta [Ásta Kristjana Sveinsdóttir], . Siding with Euthyphro: Response-Dependence and Conferred Properties. European Journal of Philosophy (): –.

Atkin, Albert, . There’s no place like “here” and no time like “now”. American Philosophical Quarterly (): –.

Audi, Paul, a. A Clarification and Defense of the Notion of Grounding. In Correia & Schnieder (), pp. –.

Audi, Paul, b. Grounding: Toward a Theory of the In-Virtue-Of Relation. The Journal of Philosophy (): –.

Audi, Paul, . Explanation and Explication. In Daly, Christopher John (editor) The Palgrave Handbook of Philosophical Methods, pp. –. Basingstoke, Hampshire: Palgrave Macmillan.

Bach, Kent, . Perspectives on Possibilities: Contextualism, Relativism, or What? In Egan & Weatherson (), pp. –.

Baghramian, Maria, . Relativism. London: Routledge.

Baghramian, Maria (editor) , . The Many Faces of Relativism. London: Routledge.

Barkhausen, Max & Johannes Haag, . Abstrakte Ideen und die Argumentation für den Idealismus. In Perler & Haag (a), pp. –.

Barnes, Jonathan, Malcolm Schofield & Richard Sorabji (editors) , . Articles on Aristotle: Volume (Psychology and Aesthetics). London: Duck- worth.

Beebee, Helen & Julian Dodd (editors) , . Truthmakers. The Contemporary Debate. Mind Association Occasional Series, Oxford: Oxford University Press.

Bennett, Jonathan, . Learning from Six Philosophers: Descartes, Spinoza, Leibniz, Locke, Berkeley, Hume; vol. . Oxford: Clarendon Press.

Berkeley, George, . A treatise concerning the principles of human knowledge. Dublin: Rhames.

Berkeley, George, . A treatise concerning the principles of human knowledge. Oxford: Oxford University Press. Edited by Jonathan Dancy.

Bermúdez, José-Luis, . The Paradox of Self-Consciousness. Cambridge, Massachusetts: The MIT Press.

Betti, Arianna, . Explanation in Metaphysics and Bolzano’s Theory of Ground and Consequence. Logique et Analyse (): –.

Bird, Graham H., . Kant’s Theory of Knowledge: an outline of one central argument in the Critique of Pure Reason. London: Humanities Press.

Bird, Graham H., . Review of Langton (). The Philosophical Quarterly (): –.

Blackburn, Simon, . Circles, Finks, Smells and Biconditionals. In Tomberlin (), pp. –.

Block, Ned, . Advertisement for a Semantics for Psychology. In French et al. (), pp. –.

Blum, Philipp, . A World of Truthmakers. In Monnoyer, Jean-Maurice (editor) Metaphysics and Truthmakers, pp. –. Number in Philosophische Analyse / Philosophical Analysis, Heusenstamm b. Frankfurt: Ontos Verlag. Published under the name of “Keller”.

Blum, Philipp, . Review of Armstrong (). Mind (): –. Published under the name of “Keller”.

Blum, Philipp, a. Connectives, prenectives and dishonoured cheques of metaphysical explanation. In Reboul (). URL http://www. philosophie.ch/kevin/festschrift/. Reprinted in Reboul (: –).

Blum, Philipp, b. Racisme et bêtise. In Tappolet, Christine, Fabrice Teroni & Anita Konzelmann Ziv (editors) Les Ombres de l’Âme. Penser les émotions négatives, pp. –. Genève: Éditions Markus Haller. Published under the name of “Keller”.

Boghossian, Paul Artin, . Fear of Knowledge: Against Relativism and Constructivism. Oxford: Oxford University Press.

Boghossian, Paul Artin (editor) , . Content & Justification. Philosophical Papers. Oxford: Oxford University Press.

Boghossian, Paul Artin, . Our Grasp of the Concept of Truth: Reflections on Künne [on Künne ()]. Dialectica (): –.

Boghossian, Paul Artin, . The transparency of mental content revisited [on Stalnaker ()]. Philosophical Studies (): –.

Boghossian, Paul Artin & David J. Velleman, . Colour as a Secondary Quality. Mind : –. Reprinted in Byrne & Hilbert () and in Boghossian (: –).

BonJour, Laurence, . Epistemology. Classic Problems and Contemporary Responses. edition. Lanham, Maryland: Rowman & Littlefield Publishers, Inc. First edition: ?.

Bottani, Andrea, Pierdaniele Giaretta & Massimiliano Carrara (editors) , . Individuals, Essence and Identity. Themes in Analytic Metaphysics. Number in Topoi Library, Dordrecht: Springer Verlag.

Brentano, Franz, . Psychologie vom Empirischen Standpunkte, volume I. Leipzig: Dunker und Humblot. Second edition: Brentano ().

Brentano, Franz, . Psychologie vom Empirischen Standpunkt, volume I. edition. Leipzig: Felix Meiner Verlag. Edited by Oskar Kraus.

Brewer, Bill, . Perception and Reason. Cambridge: Cambridge University Press.

Brewer, Bill, . Perception and its Objects. Oxford: Oxford University Press.

Broad, Charlie Dunbar, . Scientific Thought. International Library of Psychology, Philosophy and Scientific Method, London: Kegan Paul, Trench, Trübner & Co.

Broad, Charlie Dunbar, . The Mind and Its Place In Nature. International Library of Psychology, Philosophy and Scientific Method, London: Kegan Paul, Trench, Trübner & Co.

Brogaard, Berit, a. Introduction. Synthèse (): –.

Brogaard, Berit, b. Perceptual Content and Monadic Truth: On Cappelen & Hawthorne (). Philosophical Books (): –.

Brook, Andrew, . Kant and the Mind. Cambridge: Cambridge University Press.

Brooks, David Havard Macleod, . Secondary Qualities and Representation. Analysis (): –.

Brunschwig, Jacques & Martha Craven Nussbaum (editors) , . Passions & Perceptions. Studies in Hellenistic Philosophy of Mind. Proceedings of the Fifth Symposium Hellenisticum, Champagnole, . Cambridge: Cambridge University Press.

Buchdahl, Gerd, . Metaphysics and the Philosophy of Science. The Classical Origins: Descartes to Kant. Oxford: Basil Blackwell Publishers.

Buchdahl, Gerd, . Transcendental Reduction: A Concept for the Interpretation of Kant’s Critical Method. In Funke, Gerhard & Joachim Kopper (editors) Akten des . Internationalen Kant-Kongresses, volume I, pp. –. Berlin: Walter De Gruyter.

Buchdahl, Gerd, . Reduction – realization: A key to the structure of Kant’s thought. Philosophical Topics (): –. Reprinted in Mohanty & Shahan (: –).

Buchdahl, Gerd, . Metaphysical and internal realism: The relations between ontology and methodology in Kant’s philosophy of science. In Marcus, Ruth Barcan, Georg J.W. Dorn & Paul Weingartner (editors) Logic, Methodology, and Philosophy of Science VII: Proceedings of the Seventh International Congress of Logic, Methodology, and Philosophy of Science, Salzburg, , pp. –. Number in Studies in Logic and the Foundations of Mathematics, Amsterdam: North-Holland Publishing Co.

Buchdahl, Gerd, . Kant and the Dynamics of Reason. Oxford: Blackwell-Wiley. van der Burg, Floris, . Davidson and Spinoza. Mind, Matter and Morality. Aldershot, Hampshire: Ashgate Publishing Limited.

Burge, Tyler, . Individualism and Self-Knowledge. The Journal of Philosophy (): –. Reprinted in Burge (: –).

Burge, Tyler, . Origins of Objectivity. Oxford: Oxford University Press.

Burge, Tyler, . Cognition Through Understanding. Number in Philosophical Essays, Oxford: Clarendon Press.

Burnyeat, Myles F., . Is an Aristotelian Philosophy of Mind Still Credible? A Draft. In Nussbaum & Rorty (), pp. –.

Byrne, Alex & David R. Hilbert (editors) , . Readings on Colors. Volume : The Philosophy of Color. Cambridge, Massachusetts: The MIT Press.

Byrne, Alex & Heather Logue (editors) , . Disjunctivism: Contemporary Readings. Cambridge, Massachusetts: The MIT Press.

Cameron, Ross P., a. There Are No Things that are Musical Works. The British Journal of Aesthetics (): –.

Cameron, Ross P., b. Truthmakers and Ontological Commitment: or, how to deal with complex objects and mathematical ontology without getting into trouble. Philosophical Studies (): –.

Cameron, Ross P., c. Turtles all the Way Down: Regress, Priority and Fundamentality. The Philosophical Quarterly (): –.

Campbell, Keith, . Abstract Particulars. Oxford: Basil Blackwell Publishers.

Cappelen, Herman, a. Content Relativism and Semantic Blindness. In García-Carpintero & Kölbel (), pp. –.

Cappelen, Herman, b. The Creative Interpreter: Content Relativism and Assertion. In Hawthorne (), pp. –.

Cappelen, Herman & John Hawthorne, . Relativism and Monadic Truth. Oxford: Oxford University Press.

Cappelen, Herman & John Hawthorne, a. Précis of Cappelen & Hawthorne (). Philosophical Studies (): –.

Cappelen, Herman & John Hawthorne, b. Reply to Lasersohn (), MacFarlane (b) and Richard () [on Cappelen & Hawthorne ()]. Philosophical Studies (): –.

Cappelen, Herman & John Hawthorne, c. Reply to Glanzberg, Soames and Weatherson [critics of Cappelen & Hawthorne ()]. Analysis (): –.

Cappelen, Herman & John Hawthorne, d. Summary of Cappelen & Hawthorne (). Analysis (): –.

Carriero, John P., . The First Meditation. Pacific Philosophical Quarterly : –. Cited after reprint in Chappell (: –).

Casati, Roberto, . La scoperta dell’ombra. Milano: Arnoldo Mondadori Editore S.p.A.

Casati, Roberto, . The shadow club. New York: Alfred A. Knopf. Translation of Casati () by Abigail Asher.

Casati, Roberto, . Minor entities: surfaces, holes and shadows. In Le Poidevin, Robin, Peter M. Simons, Andrew McGonigal & Ross P. Cameron (editors) The Routledge Companion to Metaphysics, pp. –. Routledge Philosophy Companions, London: Routledge.

Casati, Roberto & Jérôme Dokic, . La philosophie du son. Nîmes: Éditions Jacqueline Chambon.

Casati, Roberto & Jérôme Dokic, . Sounds. In Zalta, Edward N. (editor) The Stanford Encyclopedia of Philosophy. Stanford, California: The Metaphysics Research Lab, Center for the Study of Language and Information. Version of July , .

Casati, Roberto & Christine Tappolet (editors) , . Response-Dependence. Number in European Review of Philosophy, Stanford, California: CSLI Publications.

Casati, Roberto & Achille C. Varzi, . Holes and Other Superficialities. Cambridge, Massachusetts: The MIT Press.

Casati, Roberto & Achille C. Varzi, . Spatial Entities. In Stock, Oliviero (editor) Spatial and Temporal Reasoning, pp. –. Dordrecht: Kluwer Academic Publishers.

Casati, Roberto & Achille C. Varzi, . Parts and Places: The Structures of Spatial Representation. Cambridge, Massachusetts: The MIT Press.

Casati, Roberto & Achille C. Varzi, . Counting the Holes. Australasian Journal of Philosophy (): –. Reprinted in Jackson & Priest (: –).

Castañeda, Hector-Neri, . The Structure of Morality. ?: Charles C. Thomas.

Castañeda, Hector-Neri, . Perception, belief, and the structure of physical objects and consciousness. Synthèse : –.

Chalmers, David J., David Manley & Ryan Wasserman (editors) , . Metametaphysics. New Essays on the Foundations of Ontology. Oxford: Oxford University Press.

Chappell, Vere C. (editor) , . Descartes’s Meditations, Critical Essays. Lanham, Maryland: Rowman & Littlefield Publishers, Inc.

Chisholm, Roderick M., . Intentionality and the Theory of Signs. Philosophical Studies (): –.

Chisholm, Roderick M., . Beyond Being and Nonbeing. Philosophical Studies (): –.

Chisholm, Roderick M., . Coming into Being and Passing Away: Can the Metaphysician Help? In Spicker, Stuart F. & H. Tristram Englehardt, Jr. (editors) Philosophical Medical Ethics: Its Nature and Significance, pp. –. Dordrecht: Kluwer Academic Publishers. Reprinted in Chisholm (a: –).

Chisholm, Roderick M., . Brentano’s Conception of Substance and Accident. Grazer Philosophische Studien : –. Reprinted in Chisholm (b: –).

Chisholm, Roderick M., a. Beginnings and Endings. In Chisholm (b), pp. –. Revised version of Chisholm ().

Chisholm, Roderick M., b. Brentano and Meinong Studies. Number – in Studien zur ’Osterreichischeń Philosophie, Amsterdam: Rodopi.

Chisholm, Roderick M., c. Properties and States of Affairs Intentionally Considered. In Leinfellner, Werner, Eric Russert Kraemer & Jeffrey C. Schank (editors) Proceedings of the th International Wittgenstein Symposium: Language and Ontology, pp. –. Number in Schriftenreihe der Österreichischen Ludwig Wittgenstein Gesellschaft, Wien: Hölder-Pichler-Tempsky. Reprinted in Chisholm (a).

Chisholm, Roderick M., . What is the Problem of Objective Reference? Dialectica (–): –.

Chisholm, Roderick M., a. Brentano and Intrinsic Value. Minneapolis, Minnesota: University of Minnesota Press. Republished as Chisholm (a).

Chisholm, Roderick M., b. Correct Emotion. In Chisholm (a), pp. –. Republished as Chisholm (a).

Chisholm, Roderick M., a. Brentano and Intrinsic Value. Modern European Philosophy, Cambridge: Cambridge University Press. Republi- cation of Chisholm (a).

Chisholm, Roderick M., b. Scattered Objects. In Thomson, Judith Jarvis (editor) On Being and Saying: Essays For Richard Cartwright, pp. –. Cambridge, Massachusetts: The MIT Press. Reprinted in Chisholm (a: –).

Chisholm, Roderick M., a. On Metaphysics. Minneapolis, Minnesota: University of Minnesota Press.

Chisholm, Roderick M., b. Why Singular Propositions? In Almog et al. (), pp. –.

Chisholm, Roderick M., . Events Without Times. An Essay On Ontology. Noûs (): –.

Chisholm, Roderick M., . Ontologically Dependent Entities. Philosophy and Phenomenological Research (): –.

Chisholm, Roderick M., . A Realistic Theory of Categories: An Essay on Ontology. Cambridge: Cambridge University Press.

Clarke, Desmond M., . Descartes’ Theory of Mind. Oxford: Clarendon Press. van Cleve, James, . Problems from Kant. Oxford: Oxford University Press.

Collins, Arthur W., . Possible Experience: Understanding Kant’s Critique of Pure Reason. Berkeley, California: University of California Press.

Copp, David, . Morality, Normativity, and Society. New York: Oxford University Press.

Corkum, Phil, . Aristotle on Nonsubstantial Individuals. Ancient Philosophy (): –.

Correia, Fabrice, . Existential Dependence and Cognate Notions. Thèse de doctorat, Département de philosophie, Université de Genève, Genève. Book publication: Correia ().

Correia, Fabrice, . Husserl on Foundation. Dialectica (): –.

Correia, Fabrice, . Existential Dependence and Cognate Notions. Analytica: Investigations in Logic, Ontology, and the Philosophy of Language, München: Philosophia Verlag. Book publication of Correia ().

Correia, Fabrice, . Generic Essence, Objectual Essence, and Modality. Noûs (): –.

Correia, Fabrice, . Grounding and Truth-Functions. Logique et Analyse (): –.

Correia, Fabrice & Benjamin Sebastian Schnieder (editors) , . Metaphysical Grounding. Understanding the Structure of Reality. Cambridge: Cam- bridge University Press.

Crane, Tim, . The Problem of Perception. In Zalta, Edward N. (editor) The Stanford Encyclopedia of Philosophy. Stanford, California: The Metaphysics Research Lab, Center for the Study of Language and Information. Version of March , .

Cronin, Timothy J., . Objective Being in Descartes and in Suarez. Roma: Gregorian University Press. Reissued as ?.

Curley, Edwin M., . Descartes Against the Skeptics. Oxford: Basil Blackwell Publishers.

Currie, Gregory, . Desire in Imagination. In Gendler & Hawthorne (), pp. –.

Dalbosco, Claudio Almir, . Ding and sich und Erscheinung. Perspektiven des transzendentalen Idealismus bei Kant. Würzburg: Königshausen & Neumann.

Daly, Christopher John, . Skepticism about Grounding. In Correia & Schnieder (), pp. –.

David, Marian A., . Armstrong on Truthmaking. In Beebee & Dodd (), pp. –.

Davidson, Donald & Gilbert H. Harman (editors) , . Semantics of Natural Language. Number in Synthese Library, Dordrecht: D. Reidel Publishing Co.

Della Rocca, Michael, . Representation and the Mind-Body Problem in Spinoza. Oxford: Oxford University Press.

Deonna, Julien Amos & Fabrice Teroni, . Qu’est-ce qu’une émotion? Chemins Philosophiques, Paris: Librairie philosophique Jean Vrin.

Deonna, Julien Amos & Fabrice Teroni, . The Emotions. A Philosophical Introduction. London: Routledge. Substantially enlarged edition and translation of Deonna & Teroni ().

Deonna, Julien Amos & Fabrice Teroni, . Emotions as Attitudes. Dialectica (): –. Special issue “Beyond Perceptualism”, edited by Sabine A. Döring and Anika Lutz. deRosset, Louis, . Getting Priority Straight. Philosophical Studies (): –. deRosset, Louis, . What is the Grounding Problem? Philosophical Studies (): –.

Descartes, René, . The Philosophical Writings of Descartes, vol. II. Cambridge: Cambridge University Press. Translated and edited by John Cottingham, Robert Stoothoff and Dugald Murdoch.

Devitt, Michael, . Coming to Our Senses: A Naturalistic Program for Semantic Localism. Cambridge: Cambridge University Press.

Di Maria, Theodore, . Kant’s conception of the self: Applying the dual-aspect reading of the phenomena/noumena distinction to the self. PhD dissertation, Marquette University, Milwaukee, Wisconsin.

Dietz, Richard, . Epistemic Modals and Correct Disagreement. In García-Carpintero & Kölbel (), pp. –.

Dokic, Jérôme, . Le dualisme forme/contenu et la théorie de la perception. Cahier de la Maison des Sciences de l’Homme ().

Dokic, Jérôme, . Qu’est-ce que la perception ? Chemins Philosophiques, Paris: Librairie philosophique Jean Vrin.

Dowell, Janice L., . A Flexible Contextualist Account of Epistemic Modals. Philosophers’ Imprint ().

Dretske, Fred I., . Knowledge and the Flow of Information. Cambridge, Massachusetts: The MIT Press. Reissued in the David Hume Series of Philosophy and Cognitive Science Reissues series of CSLI Press, Stanford.

Dretske, Fred I., . Explaining Behaviour: Reasons in a World of Causes. Cambridge, Massachusetts: The MIT Press.

Dryer, Douglas Poole, . Kant’s Solution for Verification in Metaphysics. London: Allen & Unwin.

Dunn, Michael J., . Relevant Predication I: the Formal Theory. The Journal of Philosophical Logic (): –.

Dunn, Michael J., a. Relevant Predication II: Intrinsic Properties and Internal Relations. Philosophical Studies (): –.

Dunn, Michael J., b. Relevant Predication III: Essential Properties. In Dunn, Michael J. & Anil Gupta (editors) Truth or Consequences: Essays in Honor of Nuel Belnap, pp. –. Dordrecht: Kluwer Academic Publishers.

Easton, Patricia (editor) , . Logic and the Working of the Mind. The Logic of Ideas and Faculty Psychology in Early Modern Philosophy. Atascadero, California: Ridgeview Publishing Co.

Ebbersmeyer, Sabrina (editor) , . Emotional Minds. The Passions and the Limits of Pure Inquiry in Early Modern Philosophy. Berlin: Walter De Gruyter.

Egan, Andy, a. Appearance Properties? Noûs (): –.

Egan, Andy, b. Secondary Qualities and Self-Location. Philosophy and Phenomenological Research (): –.

Egan, Andy, . Epistemic Modals, Relativism, and Assertion. Philosophical Studies (): –.

Egan, Andy, . Disputing about Taste. In Feldman & Warfield (), pp. –.

Egan, Andy, Andy Hawthorne & Brian Weatherson, . Epistemic Modals in Context. In Preyer & Peter (), pp. –.

Egan, Andy & Brian Weatherson (editors) , . Epistemic Modality. Oxford: Oxford University Press.

Einheuser, Iris, . Three Forms of Truth Relativism. In García-Carpintero & Kölbel (), pp. –.

Empiricus, Sextus, . Outlines of Scepticism. edition. Cambridge: Cambridge University Press. Edited by Julia Annas and Jonathan Barnes.

Engel, Pascal & Richard Rorty, . A quoi bon la vérité? Paris: Grasset.

Erismann, Christophe & Alexandrine Schniewind (editors) , . Compléments de substance. Études sur les propriétés accidentelles offertes à Alain de Libera. Problèmes & Controverses, Paris: Librairie philosophique Jean Vrin.

Evans, Gareth, . Things Without the Mind: A commentary on chapter two of Strawsons Individuals. In van Straaten, Zak (editor) Philosophical Subjects: Essays Presented to P.F. Strawson, pp. –. Oxford: Clarendon Press. Reprinted in Evans (: –).

Evans, Gareth, . Understanding Demonstratives. In Parret, Herman & Jacques Bouveresse (editors) Meaning and Understanding, pp. –. Berlin: Walter de Gruyter. Reprinted in Evans (: –) and in Ludlow (: –).

Evans, Gareth, . Collected Papers. Oxford: Clarendon Press.

Everson, Stephen, . Aristotle on Perception. Oxford: Clarendon Press.

Falkenstein, Lorne, . Kant’s Argument for the Non-Spatiotemporality of Things in Themselves. Kant-Studien (): –.

Feldman, Richard H. & Ted A. Warfield (editors) , . Disagreement. Oxford: Oxford University Press.

Fine, Kit, . Dependent Objects. Unpublished manuscript.

Fine, Kit, . Essence and Modality. In Tomberlin, James E. (editor) Philosophical Perspectives : Logic and Language, pp. –. Oxford: Basil Blackwell Publishers. The Second Philosophical Perspectives Lecture.

Fine, Kit, a. The Logic of Essence. The Journal of Philosophical Logic (): –.

Fine, Kit, b. Senses of Essence. In Sinnott-Armstrong, Walter, Diana Raffman & Nicholas Asher (editors) Modality, Morality, and Belief: Essays in Honor of Ruth Barcan Marcus, pp. –. Cambridge: Cambridge University Press.

Fine, Kit, . Semantics for the Logic of Essence. The Journal of Philosophical Logic (): –.

Fine, Kit, . The Question of Realism. Philosophers’ Imprint (). Reprinted in Bottani et al. (: –).

Fine, Kit, a. Modality and Tense. Philosophical Papers. Oxford: Oxford University Press.

Fine, Kit, b. Necessity and Non-Existence. In Fine (a), pp. –.

Fine, Kit, c. Reference, Essence and Identity. In Fine (a), pp. –. Written up in the spring of as a talk for the conference ‘Themes from Kaplan’ .

Fine, Kit, . Response to Horwich (). Dialectica (): –.

Fine, Kit, a. The Question of Ontology. In Chalmers et al. (), pp. –.

Fine, Kit, b. The Role of Variables. In Almog, Joseph & Paolo Leonardi (editors) The Philosophy of David Kaplan, pp. –. Oxford: Oxford University Press.

Fine, Kit, . Some Puzzles of Ground. Notre Dame Journal of Formal Logic (): –.

Fine, Kit, . Guide to Ground. In Correia & Schnieder (), pp. –. von Fintel, Kai & Anthony S. Gillies, . An Opiniated Guide to Epistemic Modality. In Gendler, Tamar Szabó & John Hawthorne (editors) Oxford Studies in Epistemology, volume II, pp. –. Oxford: Oxford University Press. von Fintel, Kai & Anthony S. Gillies, . CIA leaks. The Philosophical Review (): –. von Fintel, Kai & Anthony S. Gillies, . ‘Might’ Made Right. In Egan & Weatherson (), pp. –.

Fodor, Jerry A., . Semantics, Wisconsin Style. Synthèse (): –. Reprinted in Silvers () and in Fodor (b: –).

Fodor, Jerry A., . Psychosemantics: The Problem of Meaning in the Philosophy of Mind. Cambridge, Massachusetts: The MIT Press.

Fodor, Jerry A., a. The Elm and the Expert: Mentalese and Its Semantics. Cambridge, Massachusetts: The MIT Press.

Fodor, Jerry A., b. A Theory of Content and Other Essays. Cambridge, Massachusetts: The MIT Press.

Fodor, Jerry A., . Concepts: Where Cognitive Science Went Wrong. Oxford: Clarendon Press.

Francescotti, Robert M. (editor) , . A Companion to Intrinsic Properties. Berlin: Walter de Gruyter.

Frankfurt, Harry G., . Demons, Dreamers and Madmen. Indianapolis, Indiana: Bobbs-Merrill Company Inc. New edition: Frankfurt ().

Frankfurt, Harry G., . Demons, Dreamers and Madmen. The Defense of Reason in Descartes’s Meditations. Princeton, New Jersey: Princeton University Press. Reissue of Frankfurt (), with a foreword by Rebecca Goldstein.

French, Peter A., Theodore E. Uehling, Jr. & Howard K. Wettstein (editors) , . Midwest Studies in Philosophy : Contemporary Perspectives in the Philosophy of Language. Minneapolis, Minnesota: University of Minnesota Press. Second, enlarged edition: French et al. ().

French, Peter A., Theodore E. Uehling, Jr. & Howard K. Wettstein (editors) , . Midwest Studies in Philosophy : Contemporary Perspectives in the Philosophy of Language. edition. Minneapolis, Minnesota: University of Minnesota Press. First edition: French et al. ().

French, Peter A., Theodore E. Uehling, Jr. & Howard K. Wettstein (editors) , . Midwest Studies in Philosophy : Studies in the Philosophy of Mind. Minneapolis, Minnesota: University of Minnesota Press.

French, Steven, . The Structure of the World. Metaphysics and Representation. Oxford: Oxford University Press.

Friebe, Cord, . Über einen Einwand gegen die Zwei-Aspekte-Interpretation von Kants Unterscheidung zwischen Erscheinung und Ding an sich. Zeitschrift für philosophische Forschung (): –.

García-Carpintero, Manuel, . A Non-Modal Conception of Secondary Properties. Philosophical Papers (): –.

García-Carpintero, Manuel & Max Kölbel (editors) , . Relative Truth. Oxford: Oxford University Press.

Geach, Peter Thomas, . Mental Acts. Their Content and Their Objects. London: Routledge & Kegan Paul.

Gendler, Tamar Szabó, . The Puzzle of Imaginative Resistance. The Journal of Philosophy (): –. Reprinted in Gendler (: –).

Gendler, Tamar Szabó, . Intuition, Imagination, and Philosophical Methodology. Oxford: Oxford University Press.

Gendler, Tamar Szabó & John Hawthorne (editors) , . Conceivability and Possibility. Oxford: Oxford University Press.

Gibbard, Allan F., . Thinking How to Live. Cambridge, Massachusetts: Harvard University Press.

Glanzberg, Michael, . Context, content, and relativism. Philosophical Studies (): –.

Glanzberg, Michael, . Semantics and Truth Relative to a World. Synthèse (): –.

Glanzberg, Michael, . More on Operators and Tense [comments on Cappelen & Hawthorne ()]. Analysis (): –.

Glauser, Richard, . Berkeley et les philosophes du XVIIe siècle. Paris: Mardaga.

Goldberg, Nathaniel, . Response-Dependence, Noumenalism, and Ontological Mistery. European Journal of Philosophy (): –.

Goldie, Peter, . The Emotions: A Philosophical Exploration. Oxford: Oxford University Press.

Goldman, Alvin I., . Epistemic Relativism and Reasonable Disagreement. In Feldman & Warfield (), pp. –. Reprinted in Goldman (: –).

Goldman, Alvin I., . Reliabilism and Contemporary Epistemology. Essays. New York: Oxford University Press.

Gorman, Michael, . The Essential and the Accidental. Ratio N.S. (): –.

Gowans, Christopher W., . Moral Relativism. In Zalta, Edward N. (editor) The Stanford Encyclopedia of Philosophy. Stanford, California: The Metaphysics Research Lab, Center for the Study of Language and Information. Version of February , ; substantive revision Tue Dec , .

Griffin, Nicholas, . Russell’s Multiple Relation Theory of Judgment. Philosophical Studies (): –.

Groff, Ruth, . Ontology Revisited. Metaphysics in Social and Political Philosophy. London: Routledge.

Grüne, Stefanie, . Blinde Anschauung: die Rolle von Begriffen in Kants Theorie sinnlicher Synthesis. Frankfurt a.M.: Vittorio Klostermann.

Guyer, Paul, . Kant and the Claims of Knowledge. Cambridge: Cambridge University Press.

Hales, Steven D. (editor) , . A Companion to Relativism. Blackwell Companions to Philosophy, Oxford: Blackwell-Wiley.

Hamilton, Andy, . The Sound of Music. In Nudds & O’Callaghan (), pp. –.

Hamlyn, David Walter, . Aristotle: De Anima. Books II and III (with certain passages from Book I). edition. Clarendon Aristotle Series, ed. J.L. Ackrill, Oxford: Clarendon Press. Translated with introduction and notes.

Hamlyn, David Walter, . Aristotle: De Anima. Books II and III (with certain passages from Book I). edition. Clarendon Aristotle Series, ed. J.L. Ackrill, Oxford: Clarendon Press. Translated with introduction and notes, with a report on recent work and a revised bibliography by Christopher Shields; st edition: Hamlyn ().

Hanna, Robert, . Kant and the Foundations of Analytic Philosophy. Oxford: Oxford University Press.

Hare, Caspar, . Self-Bias, Time-Bias, and the Metaphysics of Self and Time. The Journal of Philosophy (): –.

Hare, Caspar, . On Myself, and Other, less Important Subjets. Princeton, New Jersey: Princeton University Press.

Harman, Gilbert H., . Thought. Princeton, New Jersey: Princeton University Press.

Harman, Gilbert H., . Explaining objective color in terms of subjective reactions. In Villanueva, Enrique (editor) Philosophical Issues : Perception, pp. –. Atascadero, California: Ridgeview Publishing Co.

Harrison, Jonathan, . The Incorrigibility of the Cogito. Mind (): –. Reprinted in Moyal (: –).

Haugeland, John, . Having Thought: Essays in the Metaphysics of Mind. Cambridge, Massachusetts: Harvard University Press.

Haukioja, Jussi, . The Modal Status of Basic Equations. Philosophical Studies (): –.

Hausman, Alan & David B. Hausman, . Descartes’s legacy: minds and meaning in early modern philosophy. Toronto: University of Toronto Press.

Hawthorne, John, . Metaphysical Essays. Oxford: Clarendon Press.

Hawthorne, John (editor) , . Philosophical Perspectives : Philosophy of Language. Oxford: John Wiley and Sons, Inc.

Heck, Jr., Richard Gustave, . MacFarlane on Relative Truth. In Sosa, Ernest (editor) Philosophical Issues : Philosophy of Language, pp. –. Oxford: Basil Blackwell Publishers.

Hestevold, H. Scott, . Boundaries, Surfaces and Intinuous Wholes. The Southern Journal of Philosophy : –.

Hochberg, Herbert, . A Refutation of Moderate Nominalism. Australasian Journal of Philosophy (): –. Elaborated into Hochberg ().

Hochberg, Herbert, . A Refutation of Moderate Nominalism. In Russell, Moore and Wittgenstein. The Revival of Realism, pp. –. Number in Philosophische Analyse / Philosophical Analysis, Heusenstamm b. Frankfurt: Ontos Verlag.

Hogan, Desmond, a. How to Know Unknowable Things in Themselves. Noûs (): –.

Hogan, Desmond, b. Noumenal Affection. The Philosophical Review (): –.

Hogan, Desmond, c. Three Kinds of Rationalism and the Non-Spatiality of Things in Themselves. Journal of the History of Philosophy (): –.

Holt, Edwin Bissell, . The Concept of Consciousness. London: George Allen & Unwin.

Horwich, Paul, . The Quest for REALITY. Dialectica (): –. Reprinted, in substantially revised form, in Horwich (: –).

Horwich, Paul, . Truth-Meaning-Reality. Oxford: Oxford University Press.

Humberstone, I. Lloyd, . Intrinsic / extrinsic. Synthèse : –.

Hume, David, . A Treatise of Human Nature. London: Penguin Books. Repr. .

Hume, David, a. A Treatise of Human Nature, volume . Oxford: Oxford University Press. Edited by David Fate and Mary J. Norton.

Hume, David, b. A Treatise of Human Nature, volume . Oxford: Oxford University Press. Edited by David Fate and Mary J. Norton.

Hüttemann, Andreas, . Descartes’ Kritik an den realen Qualit’aten:́ das Beispiel der Schwere. Archiv für Geschichte der Philosophie (): –.

Iacona, Andrea, . Faultless or Disagreement. In García-Carpintero & Kölbel (), pp. –.

Jackson, Frank, . Perception: a Representative Theory. Cambridge: Cambridge University Press.

Jackson, Frank & Graham Priest (editors) , . Lewisian Themes: The Philosophy of David Lewis. Oxford: Clarendon Press.

Jaworski, William, . Structure and the Metaphysics of Mind. How Hylomorphism Solves the Mind-Body Problem. Oxford: Oxford University Press.

Johnston, Mark, . Dispositional Theories of Value. Proceedings of the Aristotelian Society, Supplementary Volume : –.

Johnston, Mark, . Explanation, Response-Dependence and Judgement-Dependence. In Menzies (), pp. –.

Johnston, Mark, . Verificationism as Philosophical Narcissism. In Tomberlin (), pp. –.

Johnston, Mark, . Are Manifest Qualities Response-Dependent? The Monist (): –.

Jolley, Nicholas, . Causality and Mind. Essays on Early Modern Philosophy. Oxford: Oxford University Press.

Kalderon, Mark Eli, . Form without Matter. Empedocles and Aristotle on Color Perception. Oxford: Oxford University Press.

Kant, Immanuel, . Kritik der reinen Vernunft. edition. Riga: Johann Friedrich Hartknoch.

Kant, Immanuel, . Kritik der reinen Vernunft. edition. Riga: Johann Friedrich Hartknoch.

Kaplan, David, . On the Logic of Demonstratives. The Journal of Philosophical Logic (): –. Also published in French et al. (: –).

Kaplan, David, . Demonstratives. In Almog et al. (), pp. –. Widely circulated from on.

Kaplan, David, a. Thoughts on Demonstratives. In Yourgrau, Palle (editor) Demonstratives, pp. –. Oxford: Oxford University Press.

Kaplan, David, b. Words. Proceedings of the Aristotelian Society, Supplementary Volume : –.

Kemmerling, Andreas, . Die Denkbarkeit des ganz eigenen Ichs. Archiv für Geschichte der Philosophie (): –. Als “Die Ich-Idee” wiederabgedruckt in Kemmerling (: –).

Kemmerling, Andreas, . Ideen des Ichs – Studien zu Descartes’ Philosophie. Frankfurt a.M.: Suhrkamp Verlag.

Kemmerling, Andreas, . Locke über die Wahrnehmung sekund’areŕ Qualit’aten.́ In Perler & Wild (), pp. –.

Kennington, Richard, . The Finitude of Descartes’ Evil Genius. Journal of the History of Ideas : –.

Kenny, Anthony John Patrick, . Action, Emotion and the Will. London: Routledge & Kegan Paul. nd edition: Kenny ().

Kenny, Anthony John Patrick, . Action, Emotion and the Will. edition. London: Routledge & Kegan Paul. First edition: Kenny ().

Kim, Jaegwon, . Essays in the Metaphysics of Mind. Oxford: Oxford University Press.

King, Peter O., . Rethinking Representation in the Middle Ages: A Vade-Mecum to Medieval Theories of Mental Representation. In Lagerlund (b), pp. –.

Knuuttila, Simo, . Emotions in Ancient and Medieval Philosophy. Oxford: Clarendon Press.

Knuuttila, Simo & Pekka Kärkkäinen (editors) , . Theories of Perception in Medieval and Early Modern Philosophy. New York: Springer.

Knuuttila, Simo & Ilkka Niiniluoto (editors) , . Methods of Philosophy and the History of Philosophy. Proceedings of the Entretiens of Institute International de Philosophie, Helsinki, August -, . Number in Acta Philosophica Fennica, Helsinki: Societas Philosophica Fennica, Akateeminen Kirjakauppa.

Kölbel, Max, . A Criterion of Objectivity. Theoria (): –.

Kölbel, Max, . Truth Without Objectivity. Oxford: Basil Blackwell Publishers.

Kölbel, Max, . Indexical Relativism vs Genuine Relativism. International Journal of Philosophical Studies (): –. Reprinted in Baghramian (: –).

Kölbel, Max, . How to Spell Out Genuine Relativism and How to Defend Indexical Relativism. International Journal of Philosophical Studies (): –. Reprinted in Baghramian (: –).

Kölbel, Max, . Introduction: Motivations for Relativism. In García-Carpintero & Kölbel (), pp. –.

Kölbel, Max, . Global Relativism and Self-Refutation. In Hales (), pp. –.

Kripke, Saul A., . Naming and Necessity. Oxford: Basil Blackwell Publishers.

Kukla, André & Joel Walmsley,. Mind. A Historical and Philosophical Introduction to the Major Theories. Indianapolis, Indiana: Hackett Publishing Co.

Kulvicki, John V.,. Perceptual Content, Information, and the Primary/Secondary Quality Distinction. Philosophical Studies (): –.

Künne, Wolfgang, . Conceptions of Truth. Oxford: Clarendon Press.

Künne, Wolfgang, . Replies to Paul Boghossian () and Kevin Mulligan (). Dialectica (): –.

Künne, Wolfgang, . Vérité sans vérités? Réponse à Kevin Mulligan. Philosophiques (): –.

Lagerlund, Henrik (editor) , a. Forming the Mind. Essays on the Internal Senses and the Mind/Body Problem from Avicenna to the Medical Enlightenment. Number in Studies in the History of Philosophy of Mind, Dordrecht: Springer Verlag.

Lagerlund, Henrik (editor) , b. Representation and Objects of Thought in Medieval Philosophy. Ashgate Studies in Medieval Philosophy, Aldershot, Hampshire: Ashgate Publishing Limited.

Lagerlund, Henrik & Mikko Yrjönsuuri (editors) , . Emotions and Choice from Boethius to Descartes. Number in Studies in the History of Philosophy of Mind, New York: Springer Science+Business Media.

Langton, Rae, . Kantian Humility: Our Ignorance of Things in Themselves. Oxford: Clarendon Press.

Lasersohn, Peter, . Context Dependence, Disagreement, and Predicates of Personal Taste. Linguistics and Philosophy (): –.

Lasersohn, Peter, . Quantification and Perspective in Relativist Semantics. In Hawthorne (), pp. –.

Lasersohn, Peter, . Context, relevant parts and (lack of) disagreement over taste [on Cappelen & Hawthorne ()]. Philosophical Studies (): –.

Lehrer, Keith, . Analysis and Metaphysics. Essays in Honor of R.M. Chisholm. Number in Philosophical Studies Series, Dordrecht: D. Reidel Publishing Co. Second edition: Lehrer ().

Lehrer, Keith, . Analysis and Metaphysics. edition. Dordrecht: D. Reidel Publishing Co. First edition: Lehrer ().

Lenz, Martin, . Lockes psychologische Logik der Ideen. In Perler & Haag (a), pp. –.

Lewis, David, . General Semantics. Synthèse (–): –. Republished in Davidson & Harman (: –), in Partee (: –) and, with a postscript (Lewis c), in Lewis (b: –).

Lewis, David, . Attitudes de Dicto and de Se. The Philosophical Review : –. Reprinted, with a postscript (Lewis d), in Lewis (b: –).

Lewis, David, . Mad Pain and Martian Pain. In Block, Ned (editor) Readings in the Philosophy of Psychology. Volume I, pp. –. Language and Thought, Cambridge, Massachusetts: The MIT Press. Reprinted, with a postscript (Lewis e), in Lewis (b: –).

Lewis, David, a. New Work for a Theory of Universals. Australasian Journal of Philosophy (): –. Reprinted in Lewis (: –).

Lewis, David, b. Philosophical Papers, volume . Oxford: Oxford University Press.

Lewis, David, c. Postscript to Lewis (). In Lewis (b), pp. –.

Lewis, David, d. Postscript to Lewis (). In Lewis (b), pp. –.

Lewis, David, e. Postscript to Lewis (). In Lewis (b), pp. –.

Lewis, David, . On the Plurality of Worlds. Oxford: Basil Blackwell Publishers.

Lewis, David, . Papers in Metaphysics and Epistemology. Cambridge: Cambridge University Press.

Lewis, David, . Redefining “Intrinsic”. Philosophy and Phenomenological Research (): –.

Lewis, David & Rae Langton, . Defining “intrinsic”. Philosophy and Phenomenological Research (): –. Reprinted in Lewis (: –) and in Francescotti (: –).

Lewis, David & Stephanie R. Lewis, . Holes. Australasian Journal of Philosophy : –. Reprinted in Lewis (b: –).

Liggins, David, . Truth-makers and Dependence. In Correia & Schnieder (), pp. –.

Locke, John, . An Essay concerning Human Understanding. London: by Eliz. Holt, for Thomas Basset, at the George in Fleet Street. Modern edition: Locke ().

Locke, John, . An Essay concerning Human Understanding. Oxford: Clarendon Press. Edited with a foreward by Peter H. Nidditch.

Long, Anthony A., . Greek Models of Mind and Self. Cambridge, Massachusetts: Harvard University Press.

López de Sa, Dan, . Non-Objective Truths. Theoria : –.

López de Sa, Dan, . Response-Dependencies: Colors and Values. Tesis doctoral, Universidad de Barcelona, Departamento de Lógica, Historia y Filosofía de la Ciencia, Barcelona.

López de Sa, Dan, . The Many Relativisms and the Question of Disagreement. International Journal of Philosophical Studies (): –. Reprinted in Baghramian (: –).

López de Sa, Dan, . Presuppositions of Commonality: An Indexical Relativist Account of Disagreement. In García-Carpintero & Kölbel (), pp. –.

López de Sa, Dan, . The Makings of Truth: Realism, Response-Dependence, and Relativism. In Pedersen, Nikolaj Jang Lee Linding & Cory D. Wright (editors) New Waves in Truth, pp. –. New Waves in Philosophy, Basingstoke, Hampshire: Palgrave Macmillan.

Lowe, Edward Jonathan, . Subjects of Experience. Cambridge: Cambridge University Press.

Lowe, Edward Jonathan, . Personal Agency. The Metaphysics of Mind and Action. Oxford: Oxford University Press.

Ludlow, Peter J. (editor) , . Readings in the Philosophy of Language. Cambridge, Massachusetts: The MIT Press.

MacFarlane, John, . Future Contingents and Relative Truth. The Philosophical Quarterly (): –.

MacFarlane, John, a. The Assessment Sensitivity of Knowledge Attributions. In Gendler, Tamar Szabó & John Hawthorne (editors) Oxford Studies in Epistemology, volume I, pp. –. Oxford: Oxford University Press.

MacFarlane, John, b. Making Sense of Relative Truth. Proceedings of the Aristotelian Society (): –.

MacFarlane, John, . Relativism and Disagreement. Philosophical Studies (): –.

MacFarlane, John, . Truth in the Garden of Forking Paths. In García-Carpintero & Kölbel (), pp. –.

MacFarlane, John, . Nonindexical Contextualism. Synthèse (): –.

MacFarlane, John, a. Epistemic Modals are Assessment-Sensitive. In Egan & Weatherson (), pp. –.

MacFarlane, John, b. Simplicity made difficult [on Cappelen & Hawthorne ()]. Philosophical Studies (): –.

MacFarlane, John, . Assessment Sensitivity: Relative Truth and Its Applications. Unpublished manuscript. Version of November , .

Mackie, John Leslie, . Self-Refutation – A Formal Analysis. The Philosophical Quarterly (): –. Reprinted in Mackie (: –).

Mackie, John Leslie, . Logic and Knowledge: Selected Papers Volume I. Oxford: Clarendon Press.

Marshall, Colin R., a. Kant’s Appearances and Things in Themselves as Qua-Objects. The Philosophical Quarterly (): –.

Marshall, Eugene, b. The Spiritual Automaton. Spinoza’s Science of the Mind. Oxford: Oxford University Press.

Martin, M.G.F.,. The Reality of Appearances. In Sainsbury, Richard Mark (editor) Thought and Ontology, pp. –. Milano: Franco Angeli. Reprinted in Byrne & Logue (: –).

Martin, M.G.F., a. An Eye Directed Outward. In Wright, Crispin, Barry C. Smith & Cynthia Macdonald (editors) Knowing Our Own Minds, pp. –. Mind Association Occasional Series, Oxford: Oxford University Press.

Martin, M.G.F., b. Setting Things before the Mind. In O’Hear, Anthony (editor) Current Issues in Philosophy of Mind, pp. –. Number in Royal Institute of Philosophy Supplement, Cambridge: Cambridge University Press.

Martin, M.G.F., . Beyond dispute: sense-data, intentionality and the mind-body problem. In Crane, Tim & Sarah Patterson (editors) History of the Mind-Body Problem, pp. –. Studies in the History of Philosophy, London: Routledge.

Martin, M.G.F.,. Epistemic Openness and Perceptual Defeasibility [on Brewer ()]. Philosophy and Phenomenological Research (): –.

Martin, M.G.F.,. Sensible Appearances. In Baldwin, Thomas (editor) The Cambridge History of Philosophy –, pp. –. Cambridge: Cambridge University Press.

Martin, M.G.F., . What’s in a Look? In Nanay (), pp. –.

Martin, M.G.F., . Moore’s Dilemma. In Coates, Paul & Sam Coleman (editors) Phenomenal Qualities. Sense, Perception, and Consciousness, pp. –. Oxford: Oxford University Press.

Matthews, Robert J., . The Measure of Mind. Propositional Attitudes and Their Attribution. Oxford: Oxford University Press.

McDaniel, Kris, . Kantian ignorance. Unpublished manuscript.

McFarland, Duncan, . Mark Johnston’s Substitution Principle: A New Counterexample? Philosophy and Phenomenological Research (): –.

McFarland, Duncan & Alexander Miller, . Jackson on colour as a primary quality. Analysis : –.

Melia, Joseph, . Truthmaking without Truthmakers. In Beebee & Dodd (), pp. –.

Menzies, Peter (editor) , . Response-Dependent Concepts. Number in Working Papers in Philosophy, Canberra: RSSS Australasian National University.

Menzies, Peter & Philip Pettit, . Found: the Missing Explanation. Analysis (): –.

Miller, Alexander, . Objectivity Disfigured: Mark Johnston’s Missing-Explanation Argument. Philosophy and Phenomenological Research : –.

Miller, Alexander, . Tacit Knowledge. In Hale, Bob & Crispin Wright (editors) A Companion to the Philosophy of Language, pp. –. Blackwell Companions to Philosophy, Oxford: Basil Blackwell Publishers.

Miller, Alexander, . The Missing-Explanation Argument revisited. Analysis (): –.

Mohanty, Jitendra Nath & Robert W. Shahan (editors) , . Essays on Kant’s Critique of Pure Reason. Norman, Oklahoma: University of Oklahoma Press.

Moore, Adrian W., . Points of View. Oxford: Oxford University Press.

Moran, Richard, . The Expression of Feeling in Imagination. The Philosophical Review (): –. Reprinted in Moran (: –).

Moran, Richard, . The Philosophical Imagination. Selected Essays. Oxford: Oxford University Press.

Morganti, Matteo, . Ontological Priority, Fundamentality, and Monism. Dialectica (): –.

Morris, Charles W., . Foundations of the Theory of Signs. Number I, in International Encyclopedia of Unified Science, Chicago, Illinois: University of Chicago Press. Reprinted in Neurath et al. (: –).

Moyal, Georges J.D. (editor) , . René Descartes. Critical Assessments, Critical Assessments of Leading Philosophers, volume II. London: Routledge.

Mulligan, Kevin, . Perception. In Smith & Woodruff Smith (), pp. –.

Mulligan, Kevin, . Perception, Particulars and Predicates. In Fisette, Denis (editor) Consciousness and Intentionality: Models and Modalities of Attribution, pp. –. Number in The University of Western Ontario Series in Philosophy of Science, Dordrecht: Kluwer Academic Publishers.

Mulligan, Kevin, . Brentano on the Mind. In Jacquette, Dale (editor) The Cambridge Companion to Brentano, pp. –. Cambridge Companions to Philosophy, Cambridge: Cambridge University Press.

Mulligan, Kevin, a. Ascent, Propositions, and Other Formal Objects. Grazer Philosophische Studien : –. “Propositions”, ed. by Massimiliano Carrara and Elisabetta Sacchi.

Mulligan, Kevin, b. Facts, Formal Objects and Ontology. In Bottani, Andrea & Richard Davies (editors) Modes of Existence. Papers in Ontology and Philosophical Logic, pp. –. Number in Philosophische Forschung / Philosophical Research, Heusenstamm b. Frankfurt: Ontos Verlag. Papers from the conference “On what (perhaps) there is”, held May at the University of Bergamo.

Mulligan, Kevin, . Intentionality, Knowledge and Formal Objects. In Rønnow-Rasmussen, Toni, Björn Petersson, Jonas Josefsson & Dan Egonsson (editors) Hommage à Wlodek. Philosophical Papers Dedicated to Wlodek Rabinowicz. Lund: Philosophy Department. URL http: //www.fil.lu.se/hommageawlodek.

Mulligan, Kevin, . Husserls Herz. In Frank, Manfred & Niels Weidtmann (editors) Husserl und die Philosophie des Geistes, pp. –. Frankfurt a.M.: Suhrkamp Verlag.

Nanay, Bence (editor) , . Perceiving the World. Oxford: Oxford University Press.

Needham, Paul, . Descartes: An Introduction to the Philosophy of Mind. Stockholm: Akademilitteratur.

Neurath, Otto, Rudolf Carnap & Charles W. Morris (editors) , . International Encyclopedia of Unified Science, volume I, –. Chicago, Illinois: University of Chicago Press.

Nieland, J.W. & Erik Weber, . Metaphysical Explanatory Asymmetries. Logique et Analyse (): –.

Nightingale, Andrea & David Sedley (editors) , . Ancient Models of Mind. Studies in Human and Divine Rationality. Cambridge: Cambridge University Press.

Nolan, Lawrence, a. Introduction. In Nolan (b), pp. –.

Nolan, Lawrence (editor) , b. Primary & Secondary Qualities. The Historical and Ongoing Debate. Oxford: Oxford University Press.

Nudds, Matthew, . Sounds and Space. In Nudds & O’Callaghan (), pp. –.

Nudds, Matthew, . What Sounds Are. In Zimmerman (), pp. –.

Nudds, Matthew & Casey O’Callaghan (editors) , . Sounds and Perception. New Philosophical Essays. Oxford: Oxford University Press.

Nussbaum, Martha Craven & Amélie Oksenberg Rorty (editors) , . Essays on Aristotle’s ‘De Anima’ . Oxford: Clarendon Press.

Oaklander, L. Nathan, . C.D. Broad’s Ontology of Mind. Number in Philosophische Analyse / Philosophical Analysis, Heusenstamm b. Frankfurt: Ontos Verlag.

O’Callaghan, Casey, . Constructing a Theory of Sounds. In Zimmerman (), pp. –.

O’Daly, Gerard, . Augustine’s Philosophy of Mind. Berkeley, California: University of California Press.

Oliver, Alex, . The Metaphysics of Properties. Mind : –.

Olson, Mark A., . Descartes’ First Meditation: Mathematics and the Laws of Logic. Journal of the History of Philosophy (): –.

O’Shaughnessy, Brian, . The Location of Sound. Mind : –.

O’Shaughnessy, Brian, . Consciousness and the World. Oxford: Oxford University Press.

Palmquist, Stephen R., . Kant’s System of Perspectives. An architectonic interpretation of the Critical philosophy. Lanham, Maryland: University Press of America.

Parsons, Josh, . Theories of Location. In Zimmerman, Dean W. (editor) Oxford Studies in Metaphysics, volume III, pp. –. Oxford: Clarendon Press.

Partee, Barbara Hall (editor) , . Montague Grammar. New York: Academic Press.

Pasnau, Robert, . What is Sound? The Philosophical Quarterly (): –.

Paton, Herbert James, . Kant’s Metaphysic of Experience, volume . London: George Allen & Unwin.

Peacocke, Christopher, . Sense and Content. Experience, Thought, and their Relations. Oxford: Clarendon Press.

Peacocke, Christopher, . Thoughts: An Essay on Content. Oxford: Basil Blackwell Publishers.

Peacocke, Christopher, . A Study of Concepts. Cambridge, Massachusetts: The MIT Press.

Perler, Dominik, . Repräsentation bei Descartes. Number in Philosophische Abhandlungen, Frankfurt a.M.: Vittorio Klostermann.

Perler, Dominik (editor) , . Ancient and Medieval Theories of Intentionality. Number in Studien und Texte zur Geistesgeschichte des Mittelalters, Leiden: E.J. Brill.

Perler, Dominik, . Theorien der Intentionalität im Mittelalter. Frankfurt a.M.: Vittorio Klostermann. Zweite Auflage: Perler ().

Perler, Dominik, . Theorien der Intentionalität im Mittelalter. Frankfurt a.M.: Vittorio Klostermann. Erste Auflage: Perler ().

Perler, Dominik (editor) , . Transformations of the Soul. Aristotelian Psychology -. Leiden: E.J. Brill.

Perler, Dominik (editor) , . The Faculties. A History. Oxford Philosophical Concepts, Oxford: Oxford University Press.

Perler, Dominik & Johannes Haag (editors) , a. Ideen. Repräsentationalismus in der frühen Neuzeit. . Kommentare. Berlin: Walter de Gruyter.

Perler, Dominik & Johannes Haag (editors) , b. Ideen. Repräsentationalismus in der frühen Neuzeit.. Texte. Berlin: Walter de Gruyter.

Perler, Dominik & Markus Wild (editors) , . Sehen und Begreifen. Wahrnehmungstheorien in der frühen Neuzeit. Berlin: Walter de Gruyter.

Perry, John R., . The Problem of the Essential Indexical. Noûs (): –. Reprinted with a postscript in Perry (: –), and in Perry (: –).

Perry, John R., . Thought without Representation. Proceedings of the Aristotelian Society, Supplementary Volume : –. Reprinted, with a postscript, in Perry (: –), and in Perry (: –).

Perry, John R., . The Problem of the Essential Indexical and Other Essays. New York: Oxford University Press. Enlarged edition: Perry ().

Perry, John R., . The Problem of the Essential Indexical and Other Essays. Stanford, California: CSLI Publications. Enlarged edition of Perry ().

Pettit, Philip, . Virtus normativa: Rational Choice Perspectives. Ethics : –. Reprinted in Pettit (: –).

Pettit, Philip, . Realism and Response-Dependence. Mind (): –. Reprinted in Pettit (: –).

Pettit, Philip, . Noumenalism and Response-Dependence. The Monist (): –. Reprinted in Pettit (: –).

Pettit, Philip, . Rules, Reasons, and Norms: Selected Essays. Oxford: Clarendon Press.

Prauss, Gerold, . Kant und das Problem der Dinge an sich. Bonn: Bouvier Verlag Herbert Grundmann. Second edition: Prauss ().

Prauss, Gerold, . Kant und das Problem der Dinge an sich. Bonn: Bouvier Verlag Herbert Grundmann. Zweite, verbesserte Auflage.

Preyer, Gerhard & Georg Peter, . Contextualism in Philosophy. Oxford: Oxford University Press.

Quarfood, Marcel, . The Thing in Itself. Methodological Perspective or Metaphysical Entity? In Transcendental Idealism and the Organism. Essays on Kant, pp. –. Number in Stockholm Studies in Philosophy, Stockholm: Almqvist & Wiksell.

Rabinowicz, Włodzimierz & Toni Rønnow-Rasmussen, . A Distinction in Value: Intrinsic and for Its Own Sake. In Spinning Ideas: Electronic Essays Dedicated to Peter Gärdenfors. Lund: Lunds Universitet. Republished as Rabinowicz & Rønnow-Rasmussen ().

Rabinowicz, Włodzimierz & Toni Rønnow-Rasmussen, . A Distinction in Value: Intrinsic and for Its Own Sake. Proceedings of the Aristotelian Society : –. Republication of Rabinowicz & Rønnow-Rasmussen (); reprinted in Rønnow-Rasmussen & Zimmerman (: – ).

Raven, Michael J., . In Defence of Ground. Australasian Journal of Philosophy (): –.

Reboul, Anne (editor) , . Philosophical Papers Dedicated to Kevin Mulligan. Genève: Département de philosophie, Université de Genève. URL http://www.philosophie.ch/kevin/festschrift/. Book publication: Reboul ().

Reboul, Anne (editor) , . Mind, Values, and Metaphysics. Philosophical Essays in Honor of Kevin Mulligan, volume . Cham: Springer Verlag. Book publication of Reboul ().

Récanati, François, . Perspectival Thought. A Plea for (Moderate) Relativism. Oxford: Oxford University Press.

Récanati, François, . Moderate Relativism. In García-Carpintero & Kölbel (), pp. –.

Récanati, François, . Truth-Conditional Pragmatics. Oxford: Oxford University Press.

Reed, Edward S., . From Soul to Mind. The Emergence of Psychology from Erasmus, Darwin to William James. New Haven, Connecticut: Yale University Press.

Reid, Thomas, . The Works of Thomas Reid, D.D., Now Fully Collected, with Selections from His Unpublished Letters. Preface, Notes and Supplementary Dissertations, by Sir W.Hamilton, Bart., [. . .]. Prefixed, Stewart’s Account of the Life and Writings of Reid; with Notes by the Editor. Edinburgh and London: MacLachlan and Stewart; Longman, Green, Longman, Roberts, and Green.

Richard, Mark, . When Truth Gives Out. Oxford: Oxford University Press.

Richard, Mark, . Relativistic content and disagreement [on Cappelen & Hawthorne ()]. Philosophical Studies (): –.

Röd, Wolfgang, . Descartes – Die Genese des Cartesianischen Rationalismus. edition. München: C.H. Beck. überarbeitete und erweiterte Auflage.

Rollinger, Robin D., . Austrian Phenomenology: Brentano, Husserl, Meinong, and Others on Mind and Object. Number in Phenomenology and Mind, Heusenstamm b. Frankfurt: Ontos Verlag.

Rønnow-Rasmussen, Toni & Michael J. Zimmerman (editors) , . Recent Work on Intrinsic Value. Number in Library of Ethics and Applied Philosophy, Dordrecht: Springer Verlag.

Rosefeldt, Tobias, . Review of van Cleve (), Collins () and of Langton (). European Journal of Philosophy (): –.

Rosefeldt, Tobias, . Dinge an sich und sekundäre Qualitäten. In Stolzenberg, Jürgen (editor) Kant in der Gegenwart, pp. –. Berlin: Walter de Gruyter.

Rosen, Gideon, . Metaphysical Dependence: Grounding and Reduction. In Hale, Bob & Aviv Hoffmann (editors) Modality: Metaphysics, Logic, and Epistemology, pp. –. Oxford: Oxford University Press.

Russell, Bertrand Arthur William, . The Analysis of Mind. London: George Allen & Unwin.

Saarinen, Esa, . Weakness of Will in Medieval Thought. Number in Studien und Texte zur Geistesgeschichte des Mittelalters, Leiden: E.J. Brill.

Saporiti, Katia, . Weshalb die Welt so ist, wie wir sie sehen. Berkeleys These der Unfehlbarkeit unserer Wahrnehmung. In Perler & Wild (), pp. –.

Schaffer, Jonathan, . On What Grounds What. In Chalmers et al. (), pp. –.

Schaffer, Jonathan, . Grounding, transitivity, and contrastivity. In Correia & Schnieder (), pp. –.

Schlutz, Alexander M., . Mind’s World. Imagination and Subjectivity from Descartes to Romanticism. Seattle, Washington: University of Washington Press.

Schnieder, Benjamin Sebastian, . The Ability to Render Something False. Proceedings of the Aristotelian Society (): –.

Schnieder, Benjamin Sebastian, a. A Certain Kind of Trinity: Dependence, Substance, Explanation. Philosophical Studies (): –.

Schnieder, Benjamin Sebastian, b. Truth-Making Without Truth-Makers. Synthèse (): –.

Schnieder, Benjamin Sebastian, . Truth-Functionality. The Review of Symbolic Logic (): –.

Schnieder, Benjamin Sebastian, . A Puzzle about ‘Because’. Logique et Analyse (): –.

Schnieder, Benjamin Sebastian, . A Logic for ‘Because’. The Review of Symbolic Logic (): –.

Schroeder, Mark, . Being For: Evaluating the Semantic Program of Expressivism. Oxford: Clarendon Press.

Schulting, Dennis, . Kant’s Idealism: The Current Debate. In Schulting, Dennis & Jacco Verburght (editors) Kant’s Idealism: New Interpretations of a Controversial Doctrine, pp. –. Number in The New Synthese Historical Library, Dordrecht: Springer Verlag.

Schultz, Johann, . Prüfung der Kantischen Critik der reinen Vernunft, volume . Königsberg: Bey G.L. Hartung, K. Preuss, Hofbuchdrucker und Buchhändler.

Schultz, Johann, . Prüfung der Kantischen Critik der reinen Vernunft, volume . Königsberg: Bey Friedrich Nicolovius.

Scott, Dana S., . Advice on Modal Logic. In Lambert, Karel (editor) Philosophical Problems in Logic: Some Recent Developments, pp. –. Number in Synthese Library, Dordrecht: D. Reidel Publishing Co.

Scruton, Roger, . Hearing Sounds. In Zimmerman (), pp. –.

Searle, John R., . Intentionality: an essay in the philosophy of mind. Cambridge: Cambridge University Press.

Searle, John R., . The Rediscovery of the Mind. Cambridge, Massachusetts: The MIT Press.

Sellars, Wilfrid, . The Myth of the Given: Three Lectures on Empiricism and the Philosophy of Mind. In Feigl, Herbert & Michael Scriven (editors) Minnesota Studies in the Philosophy of Science, Volume I: Foundations of Science & the Concepts of Psychology and Psychoanalysis, pp. –. Minneapolis, Minnesota: University of Minnesota Press. Reprinted, under the tile “Empiricism and the Philosophy of Mind”, in Sellars (: –) and in book form as Sellars ().

Sellars, Wilfrid, . Science, Perception and Reality. London: Routledge & Kegan Paul. Reprinted as Sellars ().

Sellars, Wilfrid, . Science, Perception and Reality. Atascadero, California: Ridgeview Publishing Co. Reprint of Sellars ().

Sellars, Wilfrid, . Empiricism and the Philosophy of Mind. Cambridge, Massachusetts: Harvard University Press. Republication of Sellars (), with an introduction by Richard Rorty and a study guide by Robert Brandom.

Senderowicz, Yaron M., . Controversies and the Metaphysics of Mind. Amsterdam: John Benjamins Publishing Co.

Shields, Christopher, . Aristotle. In Zalta, Edward N. (editor) The Stanford Encyclopedia of Philosophy. Stanford, California: The Metaphysics Research Lab, Center for the Study of Language and Information. Revision, July , , of the version of September , .

Shields, Christopher, . Aristotle: De Anima. edition. Clarendon Aristotle Series, ed. Lindsay Judson, Oxford: Clarendon Press. Translated with an introduction and commentary.

Sider, Theodore, . Intrinsic Properties. Philosophical Studies : –.

Sider, Theodore, . Maximality and Intrinsic Properties. Philosophy and Phenomenological Research (): –.

Sider, Theodore, . Writing the Book of the World. Oxford: Clarendon Press.

Sider, Theodore & John Hawthorne, . Locations. Philosophical Topics (): –. Reprinted in Hawthorne (: –).

Silvers, Stuart (editor) , . Rerepresentation: Readings in the Philosophy of Language. Number in Philosophical Studies Series, Dordrecht: D. Reidel Publishing Co.

Simons, Peter M., . Location. Dialectica (): –.

Simons, Peter M., . The Logic of Location. Synthèse (): –.

Slakey, Thomas J., . Aristotle on Sense Perception. The Philosophical Review (): –.

Smith, A.D., . Of Primary and Secondary Qualities. The Philosophical Review (): –.

Smith, Barry, . On Drawing Lines on a Map. In Frank, Andrew U. & Werner Kuhn (editors) Spatial Information Theory: A Theoretical Basis for GIS, International Conference COSIT ’, Semmering, Austria, September -, , pp. –. Number in Lecture Notes in Computer Science, Berlin: Springer Verlag.

Smith, Barry, . Boundaries: An Essay in Mereotopology. In Hahn, Lewis Edwin (editor) The Philosophy of Roderick M. Chisholm, pp. –. Number in The Library of Living Philosophers, LaSalle, Illinois: Open Court Publishing Co.

Smith, Barry, . Fiat Objects. Topoi (): –.

Smith, Barry & Achille C. Varzi, . Fiat and Bona Fide Boundaries. Philosophy and Phenomenological Research (): –.

Smith, Barry & David Woodruff Smith (editors) , . The Cambridge Companion to Husserl. Cambridge Companions to Philosophy, Cambridge: Cambridge University Press.

Smith, Michael A. & David Stoljar, . Global Response-Dependence and Rigidified Descriptions. The Monist (): –.

Smullyan, Raymond M., . Gödel’s Incompleteness Theorems. Number in Oxford Logic Guides, Oxford: Clarendon Press.

Soames, Scott, . True at [comments on Cappelen & Hawthorne ()]. Analysis (): –.

Soffer, Walter, . From Science to Subjectivity: An Interpretation of Descartes’ Meditations. Westport, Colorado: Greenwood Press.

Sorabji, Richard, . Aristotle on Demarcating the Five Senses. The Philosophical Review (): –. Reprinted in Barnes et al. (: –).

Sorabji, Richard, . Body and Soul in Aristotle. Philosophy (): –. Reprinted in Barnes et al. (: –).

Sorabji, Richard, . Intentionality and Physiological Processes. Aristotle’s Theory of Sense-Perception. In Nussbaum & Rorty (), pp. –.

Sorabji, Richard, . Emotion and Peace of Mind. From Stoic Agitation to Christian Temptation (The Gifford Lectures). Oxford: Oxford University Press.

Stalnaker, Robert C., . Our Knowledge of the Internal World. Oxford: Clarendon Press.

Stampe, Dennis W., . Toward a Causal Theory of Linguistic Representation. In French et al. (), pp. –. Reprinted in French et al. (: –).

Stang, Nicholas F., . The Non-Identity of Appearances and Things in Themselves. Noûs (): –.

Stang, Nicholas F., . Who’s Afraid of Double Affection? Philosophers’ Imprint ().

Steward, Helen, . The Ontology of Mind: Events, Processes, and States. Oxford: Clarendon Press.

Steward, Helen, . Actions as Processes. In Hawthorne, John (editor) Philosophical Perspectives : Philosophy of Mind, pp. –. Oxford: John Wiley and Sons, Inc.

Steward, Helen, . Processes, Continuants, and Individuals. Mind (): –.

Stigler, George J. & Gary S. Becker, . De Gustibus Non Est Disputandum. The American Economic Review (): –.

Stojanović, Isidora, . Talking about Taste: Disagreement, Implicit Arguments, and Relative Truth. Linguistics and Philosophy (): –.

Stout, George F., . The Object of Thought and Real Being. Proceedings of the Aristotelian Society : –. Reprinted in Stout (: –).

Stout, George F., . Studies in Philosophy and Psychology. London: MacMillan Publishing Co.

Stout, Rowland, . Was Sally’s Reason for Running from the Bear that She Thought it was Chasing Her? In Sandis, Constantine (editor) New Essays on the Explanation of Action, pp. –. Basingstoke, Hampshire: Palgrave Macmillan.

Stroll, Avrum, . Two Conceptions of Surfaces. In French, Peter A., Theodore E. Uehling, Jr. & Howard K. Wettstein (editors) Midwest Studies in Philosophy : Studies in Metaphysics, pp. –. Minneapolis, Minnesota: University of Minnesota Press.

Stroll, Avrum, . Seeing Surfaces. In French et al. (), pp. –.

Stroll, Avrum, . Moore and Wittgenstein on Certainty. New York: Oxford University Press.

Teroni, Fabrice, . Emotions and Formal Objects. Dialectica (): –.

Titelbaum, Michael G., . The Relevance of Self-Locating Beliefs. The Philosophical Review (): –.

Tomberlin, James E. (editor) , . Philosophical Perspectives : Language and Logic. Oxford: Basil Blackwell Publishers.

Travis, Charles, . Unshadowed Thought: Representations in Thought and Language. Cambridge, Massachusetts: Harvard University Press.

Travis, Charles, . The Face of Perception. In Ben Menahem, Yemima (editor) Hilary Putnam, pp. –. Contemporary Philosophy in Focus, Cambridge: Cambridge University Press.

Travis, Charles, . Occasion-Sensitivity. Selected Essays. Oxford: Oxford University Press.

Travis, Charles, . What are Words For? Unpublished manuscript.

Tye, Michael, . The Metaphysics of Mind. Cambridge: Cambridge University Press.

Walker, Ralph C.S., . Kant. The Arguments of the Philosophers, London: Routledge.

Walker, Ralph C.S., . The Coherence Theory of Truth. London: Routledge.

Walton, Kendall L., . Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Massachusetts: Harvard University Press.

Walton, Kendall L., . Morals in Fiction and Fictional Morality. Proceedings of the Aristotelian Society, Supplementary Volume : –.

Watson, Richard A., . Representational ideas. From Plato to Patricia Churchland. Number in Synthese Library, Dordrecht: Kluwer Academic Publishers.

Waxman, Wayne, . Kant’s Anatomy of the Intelligent Mind. Oxford: Oxford University Press.

Weatherson, Brian, . Morality, Fiction and Possibility. Philosophers’ Imprint ().

Weatherson, Brian, . No Royal Road to Relativism [comments on Cappelen & Hawthorne ()]. Analysis (): –.

Weber, Erik, Jan Willem Wieland & Tim De Mey, . Introduction. Logique et Analyse (): –.

Wedin, Michael V.,. Mind and Imagination in Aristotle. New Haven, Connecticut: Yale University Press.

Wild, Markus, . Hume über das Wahrnehmungsobjekt. In Perler & Wild (), pp. –.

Wild, Markus, a. David Hume: Einleitung und Textpassagen. In Perler & Haag (b), pp. –.

Wild, Markus, b. David Hume: Stellenkommentar. In Perler & Haag (a), pp. –.

Willaschek, Marcus, . Die Mehrdeutigkeit der kantischen Unterscheidung zwischen Dingen an sich und Erscheinungen. Zur Debatte um Zwei-Aspekte- und Zwei-Welten-Interpretationen des transzendentalen Idealismus. In Gerhardt, Volker, Rolf-Peter Horstmann & Ralph Schumacher (editors) Kant und die Berliner Aufklärung, Akten des IX. Internationalen Kant-Kongresses, volume , pp. –. Berlin: Walter De Gruyter.

Williams, J. Robert G., . Requirements on Reality. In Correia & Schnieder (), pp. –.

Wong, David B., . Moral Relativity. Berkeley, California: University of California Press.

Wong, David B., . Pluralistic Relativism. In French, Peter A., Theodore E. Uehling, Jr. & Howard K. Wettstein (editors) Midwest Studies in Philosophy : Moral Concepts, pp. –. Notre Dame, Indiana: University of Notre Dame Press.

Wong, David B., . Natural Moralities. A Defense of Pluralistic Relativism. Oxford: Oxford University Press.

Woods, Michael J., . Aristotle: Eudemian Ethics Books I, II, and VIII. edition. Clarendon Aristotle Series, ed. J.L. Ackrill, Oxford: Clarendon Press. Edited and translated with a commentary by Michael Woods; second edition: Woods ().

Woods, Michael J., . Aristotle: Eudemian Ethics Books I, II, and VIII. edition. Clarendon Aristotle Series, ed. J.L. Ackrill and Lindsay Judson, Oxford: Clarendon Press. Edited and translated with a commentary by Michael Woods; first edition: Woods ().

Wright, Crispin, . Moral Values, Projection and Secondary Qualities. Proceedings of the Aristotelian Society, Supplementary Volume : –. Reprinted in Wright (: –).

Wright, Crispin, . Truth and Objectivity. Cambridge, Massachusetts: Harvard University Press.

Wright, Crispin, . On Being in a Quandary. Relativism, Vagueness, Logical Revisionism. Mind (): –. Reprinted in Wright (: –).

Wright, Crispin, . Saving the Difference: Essays on Themes from ‘Truth and Objectivity’ . Cambridge, Massachusetts: Harvard University Press.

Wright, Crispin, . Intuitionism, Realism, Relativism and Rhubarb. In Greenough, Patrick & Michael Patrick Lynch (editors) Truth and Realism, pp. –. Oxford: Clarendon Press. Proceedings of the St.Andrews Conference on Realism and Truth.

Wright, Crispin, . New Age Relativism and Epistemic Possibility: The Question of Evidence. In Sosa, Ernest & Enrique Villanueva (editors) Philosophical Issues : The Metaphysics of Epistemology, pp. –. Oxford: Basil Blackwell Publishers.

Wright, Crispin, . Relativism about Truth Itself: Haphazard Thoughts about the Very Idea. In García-Carpintero & Kölbel (), pp. –.

Yablo, Stephen, . Coulda, Woulda, Shoulda. In Gendler & Hawthorne (), pp. –.

Yates, David, . Recent Work on Response-Dependence. Philosophical Books (): –.

Zalta, Edward N., . Essence and Modality. Mind (): –.

Zimmerman, Dean W. (editor) , . Oxford Studies in Metaphysics, volume V. New York: Oxford University Press.