142 37 6MB
English Pages 160 [161] Year 2022
The Ontology of Perceptual Experience
The Ontology of Perceptual Experience Sebastián Sanhueza Rodríguez
LEXINGTON BOOKS
Lanham • Boulder • New York • London
Published by Lexington Books An imprint of The Rowman & Littlefield Publishing Group, Inc. 4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706 www.rowman.com 86-90 Paul Street, London EC2A 4NE Copyright © 2022 by The Rowman & Littlefield Publishing Group, Inc. All rights reserved. No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage and retrieval systems, without written permission from the publisher, except by a reviewer who may quote passages in a review. Reprinted by permission from Springer Nature. Sanhueza Rodríguez, S. (2018). Perceptual Experience and Aspect. Acta Analytica, 33(1), 103-120. Republished with permission of Brill. Sanhueza Rodriguez, S. (2016). A Processive View of Perceptual Experience. Grazer Philosophische Studien, 93(1), 130-151. permission conveyed through Copyright Clearance Center, Inc. Reprinted by permission of John Wiley and Sons. Sanhueza Rodriguez, S. (Forthcoming). Wading through the Heraclitean Waters of Experience. Pacific Philosophical Quarterly. British Library Cataloguing in Publication Information Available Library of Congress Cataloging-in-Publication Data Available 978-1-7936-1685-2 (cloth : alk paper) 978-1-7936-1686-9 (ebook) ∞ ™ The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI/NISO Z39.48-1992.
This book is dedicated to my mother, Tatiana, and to the memory of my father, Pablo.
Contents
Acknowledgmentsix 1 The Dynamic Structure of Perceptual Experiences
1
2 Experiential Heracliteanism
19
3 Experiential Non-Heracliteanism
37
4 The Individuation of Experiences
57
5 Experience and Causation
77
6 Applications beyond the Ontology of Perception
95
7 Perceptual Experience and Language
119
Bibliography137 Index145 About the Author
149
vii
Acknowledgments
Since my graduate days, I’ve been extremely fortunate to meet a wide array of people whose generosity helped to improve the line of thought eventually crystallized in this book. Different bits of this work have been presented to audiences in Santiago, San José of Costa Rica, Salvador de Bahía, Edinburgh, and Rijeka. Far more people than I can recall have provided advice that enhanced what would otherwise have been an even rougher monograph, but the following names come to mind: Solveig Aasen, José Tomás Alvarado, Sebastián Briceño, Elena Cagnoli, Henry Clarke, Kathrine Cuccuru, Kênio Estrela, Alex Geddes, José Antonio Giménez, Christoph Hoerl, Ed Lamb, Rory Madden, Chris Meyns, Ed Nettel, Lucy O’Brien, Helen Robertson, Carlo Rossi, Josephine Salverda, Arthur Schipper, Marteen Steenhagen, Helen Steward, Javier Vidal, Lee Walters, Hong Yu Wong, and Zoltan Istvan Zardai. I was also extremely fortunate to start thinking about the ideas at the heart of this project as a doctoral student at University College London. At the time, this institution concentrated a stellar line-up of philosophers of mind: I thank Mike Martin, Christopher Peacocke, and Ian Phillips for supervising my work and encouraging intellectual pursuits that didn’t necessarily fit into more fashionable patterns. Beyond that institutional role, the following philosophers were also unendingly generous with their time and constructive criticism: Thomas Crowther, Mark Kalderon, Matthew Soteriou, and Charles Travis. Again, I’m in personal debt to Katherine and Paul Snowdon for all their human and intellectual support throughout my time in Oxford: their friendship is missed every day—yes, even in spite of Molly the Cat’s passive-aggressive indifference toward my existence. They also remind me that, whereas a life of philosophical creativity may be all well and good, only a life of connection and empathy is really meaningful. Last but not least, this ix
x
Acknowledgments
book is specially dedicated to my parents: I only wish I had finished it sooner so that my father, looking forward to it, had been able to see its completion. For the development of this book, I am indebted to the generous financial and logistic support of an Early Scholar (Iniciación) FONDECYT Fellowship (Project No. 11200074: “The Ontology of Mind and Causation”). The associated funding not only allowed me acquire the necessary equipment and bibliography to conclude this project but also allowed me to count with the much needed logistic and editorial assistance of Camila Riquelme. There wouldn’t be a book without the logistic and motivational support of Lexington Books, either. In particular, I’m extremely grateful to the editorial team led by Jana Hodges-Kluck: an early scholar like me couldn’t have asked for a more patient, courteous, and helpful editor than her. This book—in particular, chapters 2, 3, and 7—include material I have published elsewhere. I am grateful to Brill, Springer, and Wiley, respectively, for permission to reproduce, with very few modifications, the following articles: • Sanhueza Rodriguez, S. (2016). A Processive View of Perceptual Experience. Grazer Philosophische Studien, 93(1), 130-151. https://doi.org/10 .1163/18756735-09301007. • Sanhueza Rodríguez, S. (2018). Perceptual Experience and Aspect. Acta Analytica, 33(1), 103-120. https://doi.org/10.1007/s12136-017-0323-6. • Sanhueza Rodriguez, S. (Forthcoming). Wading through the Heraclitean Waters of Experience. Pacific Philosophical Quarterly. https://doi.org/10 .1111/papq.12340.
Chapter 1
The Dynamic Structure of Perceptual Experiences
This work has two main objectives. At a general level, it strives to parse a number of ways in which we may think about perceptual experiences qua dynamic phenomena. I examine two ontological approaches: on the one hand, Experiential Heracliteanism, a stance characterized by the claim that perceptual phenomena are fundamentally or necessarily dynamic, and, on the other, Experiential Non-Heracliteanism, a rejection of the previous Heraclitean thought in the sense that it acknowledges the dynamic profile or structure of experiences, but refuses to conceive it as necessary. Against this general backdrop, I turn to the more specific task of making a modest case on behalf of a non-Heraclitean view. In particular, I argue for two claims: (a) that a necessarily dynamic understanding of perceptual experience is more controversial than usually depicted in the relevant literature on the subject; and (b), that the non-Heraclitean stance is actually more plausible than how current Heracliteanists depict it. While not quite so ambitious, this work thus fulfills what I deem to be the necessary task of providing a general framework for an ontological debate still in its very early stages. To kick things off, this introductory chapter unpacks the book’s basic themes and overall structure. BEYOND PERCEPTUAL CONTENT AND PHENOMENOLOGY This book examines the question what perceptual experiences are or—to put it in a slightly more semantic tone—what we talk about when we talk about perceptual experiences and does so by focusing on what may be termed their dynamic structure or profile. To unpack this question and the crucial notion of 1
2
Chapter 1
dynamic structure at stake, I turn to Zeno Vendler’s discussion of seeing (cf. Vendler, 1957). This sample of philosophical inquiry is not only significant as a historical precedent for later debates in the ontology of mind, but it also provides an elegant starting point for laying out the more distinctive elements of my own approach to the aforementioned question. To begin with, let us introduce the kind of example that Vendler uses in order to illustrate his remarks on seeing. Imagine a vigilant sailor—call him Jim—who stands on deck and looks out for a particular star on a cloudy night. At first, poor visibility prevents him from spotting the sought-for object: he just can’t see a thing! Complete darkness dominates Jim’s visual field for a while—let us say, from t1 to t9. However, as the sky starts to clear up, Jim once again becomes capable of discriminating his surroundings: at first, he only manages to recognize a few outlines; then, shapes become clearly visible, and finally, after spotting the relevant star, his gaze remains fixed on it for a given period of time—say, from t10 to t20. To shed some philosophical light on a perceptual scenario like this, Vendler depicts a hypothetical exchange along the following lines: “[a] sailor on deck looking ahead remarks, ‘It is pitch dark, I don’t see anything.’ After a while, ‘Now I see a star.’ We ask him, ‘What has happened?’ ‘The cloud’s gone.’ ‘But what else happened?’ ‘Nothing else.’ Of course many things happened in the world and in the sailor. But his seeing is not one of them” (Vendler, 1957: 160). While Vendler does not make the point entirely explicit, it seems fair to assume that this method of cross-questioning works roughly as follows: at any subinterval within the period of time relevant to the perceptual scenario—that is, between t1 and t20—a hypothetical interviewer (an idealized third person or perhaps even Jim himself) could raise questions which prompt Jim to issue reports about what happens or what takes place when she sees or visually perceives his surroundings. The philosophical significance of this method thereby seems to derive from its ability to issue answers that shed some light on the perceptual phenomena at stake. That said, it is also important to add that I am more interested in Vendler’s example than with the particular method he relies on to assess it. The previous example provides a useful even if somewhat crude backdrop against which the subject matter and the main question of the present project may be specified. While it is clear that Vendler focuses on seeing or visual perception, this book examines a different though intimately related phenomenon—namely, perceptual experience. My primary subject matter is thus more general in two respects: on the one hand, it relates to perceptual phenomena in any of its modalities, not visual ones only; and, on the other, it concerns experiencing rather than perceiving. So while Jim’s case is certainly an example of visual perception, the claims derived from it are not visuocentric: bearing in mind that statements about a particular perceptual modality
The Dynamic Structure of Perceptual Experiences
3
cannot be immediately extrapolated to other modalities, my remarks strive to be general enough to avoid potentially problematic extrapolations. Again, the distinction between seeing and experiencing—or being aware of—something, is well known by now in the philosophical literature. Perceptual experiences are typically conceived of as necessary but nonsufficient ingredients of perception proper, in the sense that they pick up on the conscious component of perceptual phenomena—a component that does not, however, settle whether a subject does in fact perceive (see, hear, among others) her surroundings. Whereas perceptual experiences are said to be either veridical or hallucinatory in the sense that they may either succeed in presenting us with our surroundings or fail to do so, perception by definition implies successful informational interaction between a subject and his/her actual surroundings. As such, it has also been suggested that a subject may undergo perceptual experiences of a given feature without necessarily seeing it (cf. Wolgast, 1960; Lowe, 1996: 92ff.). Or again, unlike perceptual experiences, seeing or visual perception has been described as factive: if S sees O or that p, it follows that O exists or that p is true (cf. Crane and French, 2017). However, in spite of all these differences, perception and perceptual experience stand in such an intimate relationship as to guarantee that Jim’s case is still perfectly relevant and useful for present purposes. As Michael Tye puts it, it is natural to suppose that “[s]eeing something entails the presence of a visual experience. I cannot see X unless X looks some way to me; and for X to look some way to me, it must cause in me a visual experience” (Tye, 2003: 34–35). Since perceptual experience thus constitutes a key component of perception proper, perceptual cases like that of Jim’s may be tailored to illustrate claims about perceptual experience or awareness too. If Jim uninterruptedly sees a star from t10 to t20, he visually experiences it throughout that period of time. As previously stated, the type of question I am primarily concerned with is what perceptual is or, to put it on a slightly more semantic note, what we talk about when we talk about perceptual experiences. This initial formulation has to be specified further, though—and, as I shall explain presently, Vendler’s work will once again provide a helpful cue. After all, the previous question could point in a number of different directions. For example, some philosophers of mind might take it to tap into the popular debate between representationalist and relationalist theories of perception (e.g., Papineau, 2021). Crudely put, such a debate turns on how the natural world is conveyed to the subject of perceptual experiences: whereas some approaches have reasons to address that question in terms of mental vehicles capable of opening up the world to the subject via contents or otherwise representational devices, others rely on a fundamental or primitive relation of acquaintance between mind and world. To the extent that both approaches take a stance on the intimate link between perceptual experiences and their objects—that is, elements of the
4
Chapter 1
world—they do certainly go some way toward addressing the question what perceptual experiences are. Again, contemporary philosophers of mind have also turned their attention to what is often termed the phenomenal or qualitative character of experience, a feature intended to pick up on the characteristic ways in which members of a given species experience the world vis-à-vis members of other species (cf. Farrell, 1950; Nagel, 1974). Since the characteristic way in which a subject experiences the world may reasonably be deemed a defining aspect of perceptual experiences, a study of their phenomenal character could also go a long way toward addressing what perceptual experiences are. That said, the ontological question I raise here does not strive to launch another contribution in the already robust literature on perceptual content or phenomenal character. Instead, this book is actually concerned with an issue of categorial identification. Let me unpack this claim. In a trend that goes as far back as Aristotle, passing through all the way down to and beyond Kant, metaphysicians have sought to shed some light on our understanding of reality by classifying its elements into general categories. Among the latter categories, one typically finds the notions of substance, event, state, property, relation, among others. Vendler’s remarks belong to an Anglo-American tradition that not only kept that line of philosophical interest alive but also took it further by applying it to the study of our “mental geography” (cf. Ryle, 1937, 1949; Strawson, 1970; Chisholm, 1996; Lowe, 2005; Westerhoff, 2005). So when asking what happens when Jim sees a star, Vendler does not seem particularly concerned with the objects or the phenomenal character of Jim’s seeing: to put it bluntly, he is concerned with what Jim’s seeing itself is—that is, with the fundamental ontological category under which Jim’s seeing falls. To approach this concern, Vendler resorts to a framework that includes the categories of activity, accomplishment, achievement, and state (cf. Vendler, 1957: 144–147). As explained later on in section 3, I shall not resort to the very same categorial framework or conceive its significance for our understanding of perceptual experiences the same way Vendler does. That said, the crucial point at this stage is the minimal thought at the basis of Vendler’s more specific and controversial claims, namely, that perceptual phenomena may be ontologically assessed in a way that does not simply collapse into a discussion about perceptual content or one about phenomenological character. Indeed, this book constitutes an attempt at fleshing out this thought in relation to the study of perceptual experiences. Like Vendler, I examine how perceptual experiences could be philosophically assessed in terms of certain ontological categories under which they could be classified: this is the line of inquiry which the broad question I initially posed—that is, what we talk about when we talk about perceptual experiences—intends to motivate. As noted at the very beginning of this chapter, this question frames two objectives to be pursued throughout this book. On
The Dynamic Structure of Perceptual Experiences
5
a more general note, I set out to outline the basic resources of an ontological discussion that need not collapse into the better-known debates about the content and phenomenal character of perceptual phenomena. Two key stances will thereby emerge: on the one hand, Experiential Heracliteanism (EH), a view according to which the ontological structure of perceptual experiences should be seen as fundamentally or necessarily dynamic, and, on the other, Experiential Non-Heracliteanism (ENH), a stance that acknowledges the dynamic character of experience while rejecting its necessary modality. Against this backdrop, I go on to make a modest case on behalf of ENH. Indeed, my case will be quite modest insofar as I do not seek to settle a debate, but to open one. Over the past two decades, ontologists of mind have been drawn toward some version of EH, as the latter seems to be well suited to capture what such philosophers take to be the characteristic temporal structure of perceptual or otherwise experiential phenomena. Perhaps as a result of this natural inclination, a non-Heraclitean stance has received little if any critical attention—even in spite of counting with some relevant historical precedent in Vendler’s work itself. As such, I strive to put some pressure on the current status quo by suggesting that EH may not be as obvious as it currently seems and presenting ENH in a more favorable light. In short, while the first objective is aimed at parsing different ways of categorizing perceptual phenomena, the more specific one is intended to show that there is more room for maneuver in this landscape than suggested by the current literature. Section 3 will describe the relevant ontological categories at play here in a little more detail. To conclude this section, I just want to add that the present work will by and large avoid standard debates about the content and the phenomenal character of perceptual experiences. One reason for doing so is rather pedestrian: I strive to fill a void precisely left by the overwhelming amount of attention that those discussions received over the past few decades. I suspect that a point along these lines also drives a new wave of literature on the ontology of mind (e.g., Steward, 1997, 2011; O’Shaughnessy, 2000; Crowther, 2009a, 2009b; Soteriou, 2007, 2011, 2013, 2017, 2018; Bartlett, 2018). For example, Matt Soteriou has noticed the aforementioned sort of omission when contemporary philosophers of perception focus only on certain phenomenological features of temporal experience: [ . . . ] the focus of enquiry tends to be directed more or less exclusively on the nature of these phenomenal, what-it-is-like properties, and it is rarely made explicit how we are to think of the temporal profile of the experiences to which such properties are attributed. For example, it’s not often made explicit whether the bearers of these phenomenal properties are mental events, or mental states, or mental processes. Moreover, there isn’t much discussion of the relevance that these ontological considerations might have in accounting for our knowledge
6
Chapter 1
of what it is like to have conscious experiences. Implicit is the assumption that such matters have little to contribute to a philosophical account of sensory consciousness. (Soteriou, 2013: 2)
Another reason for avoiding standard debates about perceptual content and phenomenology relates to the emergence of certain epistemological cracks in current debates in the philosophy of perception. Over the past three decades, for example, representationalist and relationalist views of perception have clashed time and time again on the question which proposal best accommodates our introspective evidence about the intentional or the phenomenological character of perceptual experience (cf. Tye, 1995; Martin, 2002). The sociological problem is that it is not even clear what the relevant evidence is supposed to be! A key assumption at the heart of those discussions is that introspection constitutes a relatively uncontroversial source of knowledge about the content and phenomenology of experiences. But is that assumption well-grounded? Philosophical fashions are cyclical. Just as G.E. Moore’s first-person method of introspective inquiry met significant pressure earlier in the twentieth century, aspects of the introspective transparency of experience that used to be taken as obvious or sacrosanct only recently, are now facing serious criticism (cf. Spener, 2012; Zimmerman, 2012; Frischhut, 2013).1 True: the present remarks do not scratch the surface of such a delicate issue like the legitimacy of introspection as a source of facts about perceptual content and phenomenology. But they do to some extent contextualize my methodological reticence to approach certain popular themes throughout this book: to the extent that a key source of facts about perceptual content and phenomenology is far from uncontroversial, it seems reasonable to look for alternative ways of shedding some philosophical light on perceptual experiences. The ontological approach I introduce here intends to constitute one of those alternative paths. At the same time, my methodological bias is not meant to imply that an ontological assessment of perceptual phenomena is ultimately independent of discussions concerning perceptual content and phenomenology. On the contrary, I take it for granted that they potentially complement each other. As previously suggested, a systematic discussion of the relevant experiential features will no doubt shed light on the nature of experiences. But, conversely, how could we expect to assess certain features of a given item going in blind about the type or category under which that item falls? At least a partial understanding of how it should be categorized also seems to determine our general understanding of its distinctive features. Specifying whether x is a color, a sound, or an object, bears on the relations in which x may stand vis-àvis other features or items. For example, classifying x as a color rules out that it could relate to a substance the same way in which two substances interact
The Dynamic Structure of Perceptual Experiences
7
with each other. Likewise, how could we pretend to understand features like perceptual content and phenomenology without even a basic bearing on the categorial profile of their bearer—that is, perceptual experiences themselves? To revisit the intentional and phenomenological features of experience in a new and potentially fruitful light, it thus seems incumbent on philosophers of mind to engage in a categorial assessment of the relevant mental phenomena. In a somewhat more general context, this suggestion is echoed by Geoffrey Lee. After surveying different debates concerning the temporal structure of conscious phenomena, he concludes that we can’t hope to get the right view of the constitution of the stream of consciousness without thinking carefully about the possible metaphysical forms it could have, and about the relationship it stands in to its neural realization in the brain. (Lee, 2014: 19)
Claims like these seem indicative of a growing ontological sensitivity among philosophers of mind: to carve perceptual phenomena at their joints, it seems necessary to rely on something else beyond considerations from introspection. In short, the main bulk of this book will avoid the no doubt necessary question how an ontology of perception bears on our understanding of perceptual content and phenomenology. This is not to say, however, that I do not take such implications to exist: toward the end of this project (cf. chapter 6), I shall provide a few cues as to how the reader could explore them. To reinforce the foregoing motivational remarks, I want to say a little more about the claim that perceptual experiences could be assessed in a way that transcends standard discussions of either its intentional or its phenomenological features. In particular, I shall illustrate this claim by turning to a well-known controversy about perceptual content—namely, Gareth Evans’ critique of doxastic accounts of perception on account of their apparent “over-intellectualization” of perceptual content—and explaining how it could be revisited in an ontological key, as per the previous suggestion. After laying out these “recitations tending to produce belief” (Turing, 1950: 455), I will finally refer to the significance of a categorial approach for the ontological study of perceptual experience at stake. THE ONTOLOGICAL SIGNIFICANCE OF DOXASTIC VIEWS OF PERCEPTIONS I previously floated the idea that, by focusing on questions of perceptual content and phenomenology alone, philosophers of perception have neglected the assessment of the very experiential vehicles those features are features
8
Chapter 1
of. To illustrate this claim, I will turn to a fairly well-known debate between classic doxastic views of perception and Gareth Evans’s objection from over-intellectualization. This debate is relevant for present purposes because the opposing views it involves could find themselves at cross purposes precisely on account of the varying subject matters I am concerned with: whereas Evans’ remarks may be read as supporting a nonconceptualist view of perceptual content, at least some versions of the doxastic account seem to characterize the ontological profile of perceptual phenomena in a way that transcends their individuation in terms of content and phenomenology. Since the present section is by no means devoted to argue for or against a doxastic view, my exegetical characterization of the main opponents will be extremely schematic.2 Flourishing around the late 1960s and the early 1970s, doxastic accounts of perception described perceptual phenomena in terms of the acquisition of or dispositions to acquire beliefs or belief-like states about the perceivable environment. D.M. Armstrong, for instance, held that “perception is nothing but the acquiring of true or false beliefs concerning the current state of the organism’s body and environment. [ . . . ] Veridical perception is the acquiring of true beliefs, sensory illusion the acquiring of false beliefs”3 (Armstrong, 1968). Right off the bat, this proposal met significant criticism, as it apparently committed to a controversial take on perceptual content. To the extent that doxastic theorists compared the way in which experiences represent the perceivable environment to that in which beliefs do so, a natural objection—crystallized in Gareth Evans’ objection from over-intellectualization— highlights the thought that beliefs present us with the world in a far more cognitively sophisticated way than that in which experiences do so. In particular, Evans feels that explaining perception in terms of belief “gets things the wrong way round. It is as well to reserve ‘belief’ for the notion of a far more sophisticated cognitive state: one that is connected with (and, in my opinion, defined in terms of) the notion of judgement, and so, also, connected with the notion of reasons. The operations of the informational system are more primitive” (Evans, 1982: 124; also cf. McDowell, 1994: 60). The objection thus seems to run as follows: notions like belief and judgment are closely related to the possibility of ascribing concepts to the relevant cognitive subjects, but conditions of concept-possession intuitively seem to be more stringent than those of experience-ownership—after all, we are ready to identify perceiving organisms (e.g., newborn babies, dogs and bats) that fail to qualify as subjects of propositional attitudes in the traditional sense of the expression; hence, it seems intuitively misplaced to characterize experiences in terms of concept-dependent states like beliefs. Although it is by no means obvious in what sense the conditions of concept- or belief-possession are more cognitively demanding than those of experience-possession, it is
The Dynamic Structure of Perceptual Experiences
9
natural to read the objection as targeting the way in which the doxastic view models the contents of perceptual experience.4 Thus understood, the worry seems to be that characterizing how our senses deliver information to us as concept-dependent fails to reflect the cognitive sophistication of perceptual experiences vis-à-vis beliefs. After all, the realm of items which a subject could experience and the way in which she could experience them, could surely transcend the conceptual repertoire she is endowed with. As previously mentioned, this section is not primarily concerned with the exegesis of the historical debate in which Armstrong or Evans participated. As such, I do not explore here how specific doxastic theorists go about addressing the objection from over-intellectualization.5 My main concern is the question whether the study of perceptual experiences could be understood in a way that does not necessarily rely on the notions of perceptual content or phenomenology: the previous debate intends to provide one way of illustrating that question. Hence, on the assumption that Evans’ remarks on the nonconceptual character of experience thereby provide as good an example as any of a debate that gravitates toward the notion of perceptual content, my next move is to show how his apparent target may actually move on a different level of discussion. In a nutshell, the suggestion is that a doxastic view of perception may be construed as a philosophical account that specifies key features of experiential attitudes: by comparing perception and belief, what a doxastic view does for us is to bring features like the nonfactive character of experience to the fore. The latter feature is what crucially allows doxastic views of perception to resist subjectivist approaches like the sense-datum theory of perception. Let us unpack this thought. In broad lines, sense-datum views characterized perceptual phenomena in terms of primitive relations of awareness between a subject and the objects of perception, where the relevant objects were conceived as mind-dependent, private items. One important motivation for introducing such items—also known as sense-data or sensa—into our psychological ontology concerned a traditional philosophical comparison of veridical experiences and anomalous ones (e.g., illusions and hallucinations): whereas ordinary experiences suggest that we are aware of the mind-independent, medium-sized world; illusory and hallucinatory phenomena seem to indicate that the apparent objects of experience need not exist in our surroundings. By stressing the introspective or otherwise subjective similarities between both types of experiences, the sense-datum theorist applies the same explanatory model to both experiential categories: whether normal or anomalous, experiences are conceived as psychological relations of awareness between subjects and mind-dependent, private items. That said, doxastic views emerge as an attempt to account for perceptual phenomena without committing to the existence of private, mind-dependent
10
Chapter 1
objects of perception. To undermine the line of reasoning leading up to the introduction of sense-data, they challenge a key but implicit premise behind the previous understanding of illusion and hallucination. The assumption at stake may be briefly stated as follows: if something seems F to a subject, then something with that quality exists. This claim—nowadays known as the Phenomenal Principle (cf. Robinson, 1994: 32)—relates to the feature of factivity. As traditionally understood, if a given psychological attitude toward a given content p is factive, then p is true whenever a given subject instantiates that attitude. Knowledge states provide a standard example: if you know that 2 + 2 = 4, then it is true that 2 + 2 = 4. According to doxastic views of perception, however, we need not commit to the Phenomenal Principle: just like beliefs, perceptual experiences might be thought to involve nonfactive attitudes—that is, attitudes where things seem a certain way even though they fail to be so. Beliefs are intuitively nonfactive in the sense that, although we always take the beliefs we endorse to be true, it does not automatically follow that their contents convey true propositions: one may believe that p even when p is not the case. By comparing perception and belief, the doxastic theorist strives to show that perceptual experiences may likewise constitute counterexamples to the Phenomenal Principle. According to the doxastic theorist, the sense-datum theory mistakenly assumes that, if something F appears to a subject, then there must be an object which instantiates F. Like beliefs, experiences might be the kind of mental phenomena capable of (re) presenting things that need not exist. In this specific regard, perception is like belief. If the Phenomenal Principle thus breaks down, then it becomes unclear that the classic comparison of veridical and hallucinatory experiences necessarily lead up to the introduction of mind-dependent objects like sense data into our world-view. Indeed, describing perceptual phenomena without invoking such controversial items seems to be a key ontological motivation behind the emergence of doxastic views. To achieve this goal, the doxastic theory compares perception and belief, but it does not do so in order to hint at a conceptualist view of perceptual content, but to frame perceptual experiences as nonfactive states. As previously mentioned, this section aimed to revisit the classic debate between doxastic views of perception and the objection from intellectualization so as to illustrate the possibility of undertaking an ontological assessment of perceptual experiences that does not gravitate toward the features of perceptual content or phenomenal character. For present purposes, the doxastic view is helpful insofar as it deploys a response against sense-datum theories grounded on the nonfactivity of perceptual experiences, where the latter feature need not specify or individuate perceptual contents, but the very experiential phenomena to which contents are ascribed. As previously mentioned, this neglected experiential layer seems to be coming back into focus
The Dynamic Structure of Perceptual Experiences
11
in a new wave of literature in the philosophy of mind, but probably passed unnoticed among philosophers of mind who, like Evans, primarily strived to specify perceptual experiences in terms of their contents. In a somewhat roundabout way, the foregoing remarks have strived to shed a little more light on the scope of the present study of experience qua an ontology study. My aim has not been to espouse a doxastic view of perception, but to highlight a line of ontological reflection that could enrich mainstream debates that have appeared otherwise indifferent toward the ontology of perception up to now. The next section gets back on track by introducing the basic categorial resources the present book relies on. A CATEGORIAL FRAMEWORK FOR PERCEPTUAL PHENOMENA The ontological approaches outlined throughout the following chapters heavily rely on what Matt Soteriou has termed occurrent states, a hybrid category in the sense that it involves states which constitutively depend on processes of a yet-to-be-specified kind. To introduce these categorial blocks, let us go back to Vendler’s original proposal. Vendler’s treatment of perceptual phenomena fits into a venerable ontological tradition striving to map specific domains of items within a wider picture of reality. The widespread characterization of perceptual experiences as events reflects this tradition: not only does it codify the intuitive multiplicity of experiences across time and subjects but it also frames experiences in temporal rather than spatial terms. Vendler voices the latter conception by bringing up a number of categories intended to pick up on different temporal schemata or structures. And although his motivations for doing so are not entirely clear, his approach does fit with a general philosophical tendency to assess our psychological lives in temporal terms (e.g., Heidegger, 2010; Kant, 2007: A33/B49; Plato, 2010). For present purposes, I find it helpful to rehearse Vendler’s proposal against the backdrop of what Helen Steward terms a temporal strategy to the study of our mental ontology: There is room for dispute about whether or not, and in what sense, mental phenomena are physical, whether they are spatially located, and whether they have subjects, and if so, what those subjects might be. [ . . . ] But there is no controversy about the temporality of mental phenomena—about the fact that they take place in, or persist through, time. (Steward, 1997: 75–76)
Even if the ascription of temporal features to psychological phenomena turns out to be controversial, it seems a safer starting point than relying on spatial
12
Chapter 1
categories. Against this background, the categories of processes and states specially stand out, as they pick up on different ways of persisting through— or “filling”—time (cf. Vendler, 1957: 143–149; Steward, 1997: 73). This point is far more delicate than it may initially seem, for clean and heavy-duty criteria for distinguishing both categories in temporal terms are hard to come by. For the time being, let us stick to the traditional way of phrasing the relevant distinction: whereas processes occur, unfold, or take time, states only obtain or continue to obtain; whereas there are only parts of a process at each moment before its completion, states exist wholly present at each moment of time throughout which they obtain.6 Vendler himself latches onto the relevant categories mainly via examples: in particular, he compares the processes of running and writing, on the one hand, and, on the other, the state of knowing (cf. Vendler, 1957: 144–145). Indeed, running and writing are uncontroversial processes; they are temporally protracted in the sense that they go on, unfold, or take time. This point may also be put by saying that such processes have temporal phases or parts, parts which come into existence successively: when a subject runs or writes, her running or writing is not given as a whole at each of the instants throughout which she runs or writes; such processes come into existence successively until they terminate when the subject stops running or writing. States, by contrast, do not share the same mode of existence. Vendler grants that they are temporally protracted: “one can know or believe something, love or dominate somebody, for a short or long period” (Vendler, 1957: 146). Hence, processes and states do not seem to vary on account of temporal duration. Whatever the relevant distinction exactly come down to, it is typically stated in the following terms: whereas processes go on, states obtain at a time, and exists over time by being wholly present at each moment they obtain. In a significant departure from modern work on the ontology of perception, this book will frame the previous categories more generally, namely, as dynamic rather than temporal schemata. For, while Steward’s temporal strategy no doubt captures an attractive approach to the study of perceptual experiences, I believe that it should refer to a conception of experiential phenomena as dynamic—that is, as a form of change. This claim is partially grounded on an intuitive understanding of an experience as a specific way in which the environment changes a subject’s psychological economy. In addition to this, it is also hinted at by the common if loose characterization of perceptual experiences as mental events, as the latter category precisely reflects the idea of change at a high level of ontological generality.7 Finally, the following chapters will hopefully show that the ontological discussion at stake here is best understood in terms of a tension between competing ways of specifying the dynamic character of perceptual experience. As explained later on in chapter 2, Brian O’Shaughnessy illustrates a Heraclitean view of
The Dynamic Structure of Perceptual Experiences
13
experience by means of a thought experiment concerning mental freeze; that is, a scenario where a conscious subject’s mental life abruptly stops in its tracks. And while he strives to draw claims about the temporal structure of experience from that scenario, it is also clear that the latter thought experiment concerns the fact that experience cannot exist in a world devoid of change. Against the previous categorial backdrop, Vendler’s remarks about seeing also suggest different ways of thinking about perceptual experience. Recall that, when interrogating our vigilant sailor, Vendler claimed that many things happened in the world and in the sailor—seeing not being one of them. For sure, he is not thereby committing to an eliminativist take on perceptual phenomena. Instead, the thought seems to be that a description of a temporally protracted perceptual phenomenon like Jim’s seeing of a bright star from t1 to t10, need not rely on the notion of process; in principle, it could make do with mental states broadly conceived as the instantiations of properties or relations. Indeed, Vendler does recognize that very complex physical and neurobiological processes underpin a description of perceptual phenomena. To describe what happens when Jim perceives a star, one certainly has to identify the material substances and properties thus involved—Jim himself and the relevant star, for example. We also have to provide a physical story about the luminous particles that ultimately bridge the gap between both bodies. Furthermore, a neurobiological description should specify how the environmentally received information is processed from Jim’s retinae all the way down to his primary visual cortex. What Vendler contests when claiming that nothing happens in the sailor’s mind while perceiving the star, is that seeing’s reliance on physical and psychological yet unconscious processes does not imply that seeing itself is a process. According to Vendler, temporally extended seeing—that is, being perceptually aware of worldly items or states of affairs—should be understood as the obtaining of a mental state (cf. Vendler, 1957: 155–158). Drawing on Vendler’s remarks, EH could initially be characterized as the view that perceptual experience should be conceived as a process, a categorization intended to highlight the dynamic structure of experience. Likewise, ENH would capture the view that perceptual experiences could be states: while dynamic, experiences may be described in terms of nondynamic ingredients (e.g., properties, relations, and objects). Matters are more complicated than this, however. In particular, Vendler crucially oversimplifies the relationship between processes and states by taking them to be mutually exclusive. For all their differences, writers like Vendler and O’Shaughnessy coincide on this assumption, from which it follows that EH simply amounts to a processive view of experience, whereas ENH comes down to a stative or state-based ontology. But, as I explain in chapter 3, Matt Soteriou’s recent writings on the subject persuasively exhibit a more nuanced understanding of
14
Chapter 1
the relevant categories. In a nutshell, the crucial thought is that EH as well as ENH may draw on occurrent states, that is, states which constitutively depend on processes of potentially different kinds. This category suggests that processes and states could be intimately related, since they need not pick up on different phenomena as much as on different stages of the same dynamic phenomena. Accordingly, occurrent states will help to highlight the fact that the tension between EH and ENH does not turn on whether these views conceive perceptual experiences as dynamic or not, but on what kind of dynamic phenomena they take experiences to be. Vendler’s original remarks also suggest a challenge that inspires a key motivation behind this book. As previously mentioned, Vendler develops a story where temporally extended seeing is conceived as a condition instantiated by a subject for a certain period of time, or, in other words, as a state obtaining—as opposed to unfolding—in a determinate subject. Since he takes the notions of process and state to be mutually exclusive, Vendler naturally criticizes a processive view of seeing. I do not share Vendler’s specific reasons for rejecting a processive view of perceptual phenomena, but I do believe that one of his concerns is sound: even if a processive view of seeing is internally coherent or otherwise appealing, it raises the question whether processes of a special mental kind are needed in order to describe a temporally extended perceptual scenario. A key motivation behind Vendler’s analysis of seeing is the belief that nothing need happen in a subject’s mind when she is perceptually aware of her surroundings, in the sense that no special mental processes have to be invoked to describe the mental life of someone like Jim during t10–t20. Many things happen in the world and in Jim’s head: complex physical processes bridge the space and time between a bright star and Jim; equally complex neurobiological or subpersonal processes come into play when Jim processes the relevant visual information. But in addition to all of this, it is unclear that the very perceptual phenomenon of Jim’s temporally extended seeing should be understood as a processive item apart from those already invoked. To borrow D.M. Armstrong’s notion of state of affair (cf. Armstrong, 1997; Thau, 2002), for example, one could expand on Vendler’s stative story saying that Jim’s seeing simply consists in the instantiation of an informational relation between Jim and the relevant star. No special perceptual processes are needed in this story. On grounds of ontological—or at the very least categorial—simplicity, Vendler thus poses a challenge against the need for introducing special perceptual processes within a story of seeing. Taking the previous line of reasoning one step farther, I think it suggests a structurally similar challenge against EH. While I do not take processes and states to be mutually exclusive, I will stress throughout the following chapters that Heraclitean writers like O’Shaughnessy and Soteriou invoke a very distinctive kind of processes in order to categorize perceptual experiences. Like
The Dynamic Structure of Perceptual Experiences
15
Vendler, I strive to put some pressure on the need for introducing such items into our general ontology. To wrap things up, I have resorted to Vendler’s framework, first, to ground the ontological question what we talk about when we talk about perceptual experiences, and, second, to highlight a potential line of criticism against EH to be pursued later on. Now that the basic elements of the relevant dialectic have been laid out, the remainder of this book will be devoted to exploring EH and ENH in more detail. A GLIMPSE OF THINGS TO COME To close this chapter, I shall briefly summarize the overall structure of the present work. This book consists of seven chapters, which may be divided into four parts depending on the general goals they contribute to. First, I pursue the primary objective of this work, namely, to outline a framework for assessing the ontological structure of perceptual experiences. Chapter 2 introduces and critically assesses what I have termed here EH: that is, the view—or perhaps the family of views—according to which perceptual experience is fundamentally dynamic in the sense that it cannot be described—let alone explained—in nondynamic terms. The line of criticism I press against this view crucially concerns the fact that EH thereby relies on a potentially problematic type of experiential processes in order to capture the dynamic structure of perceptual experience. This thought is then reinforced in chapter 3, where I introduce an alternative ontological approach, namely, ENH. As suggested through a comparison with Matt Soteriou’s nuanced version of EH, ENH is actually structurally similar to its Heraclitean counterpart. The main difference between both ontologies precisely turns on the fact that, without loss of descriptive power, the Non-Heracliteanist might articulate the dynamic structure of perceptual experience without relying on the aforementioned processes. As such, the previous line of criticism is complemented as follows: the special notion of processes EH resorts to, is not only potentially problematic; it may also turn out to be unnecessary in order to accommodate the dynamic character of experience. Second, I probe the potential impact of the previously introduced framework on more traditional ontological themes, for example, individuation and causation. Thus, chapter 4 examines what bearing that framework has on the question how temporally extended experiences should be individuated. Over the past few years, this issue has motivated a debate between two competing positions: on the one hand, the view that a subject undergoes a single, protracted experience for as long as it goes on uninterruptedly; and, on the other, accounts that divide temporally extended experiences into
16
Chapter 1
temporal parts in a more or less arbitrary way. Neither option seems entirely attractive: while the former is understandably counterintuitive, the latter seems ontologically callous to the extent that it simply gives up on providing principled criteria of individuation. By taking the distinctive ontological profile of occurrent states seriously, however, this chapter suggests a potential challenge to a key assumption underpinning that discussion: when conceived as states, perceptual experiences are not the type of entities that are counted at all. In other words, the suggestion is that the cardinality question (“are temporally extended experiences one or many?”) should not be solved, but avoided entirely, like an instantiated hue of red or belief held by one person throughout time, perceptual experiences should not be counted as one or as many because they are not the kind of item that should be counted at all. Then, moving on beyond the theme of individuation, chapter 5 turns to experience and causation. Like other types of mental phenomena, perceptual experiences are often thought of as mental events which stand in relations of efficient causation vis-à-vis other (mental as well as physical) events. While customary, this depiction comes at a cost, though; as the philosophical literature on the causal closure of the physical has made abundantly clear, mental phenomena run the risk of becoming ontologically redundant in a world the causal structure of which is primarily conceived along efficient lines. This chapter will in turn outline how approaches like EH and ENH could provide an alternative path to the previous portrayal. After all, occurrent states do not fit into our natural worldview via relations of efficient causation to other natural events, but in virtue of their constitutive dependence on phenomena that may or may not stand in relations of efficient causation vis-à-vis other events. Accordingly, an understanding of perceptual experiences as occurrent states suggests an approach to mental causation that owes more to the Aristotelian notion of formal cause than to the standard notion of efficient causation at work in contemporary philosophy of mind. Third, I seek to shed further light on the significance of the present framework by outlining its potential impact on a number of philosophical issues that lie beyond the more traditional ontological or metaphysical themes explored in the previous chapters. A bit more specifically, chapter 6 suggests how the debate between views like EH and ENH might relate to the mind-body problem; our understanding of a key epistemological feature of perceptual experiences, namely, their assertive character; and the currently popular discussion on the temporal structure of experience, which really covers temporal features of what is (re)presented in experience rather than temporal features of experiential vehicles themselves. For the first two themes, my remarks will run along the following general lines:
The Dynamic Structure of Perceptual Experiences
17
i. certain experiential features—for example, the relation between physical and psychological phenomena or the assertive character of experience— raise recalcitrant problems in the philosophical literature. ii. traditional debates concerning those issues tend to assume that a satisfactory philosophical account should resort to a story concerning the (re)presentational and/or the phenomenological character of perceptual experience. iii. however, an ontological approach to experience along the lines here outlined provides a potentially fruitful resource for revisiting the previous debates, as it suggests the existence of potentially explanatory elements that transcend their purely (re)presentational and phenomenological properties of experience. As for the third theme, I shall suggest that the temporal phenomenology of experience may actually be silent on its ontological structure. Finally, the book closes addressing a methodological question taking into account a methodological thought that perhaps appeals philosophers of the Oxbridge tradition, chapter 7 examines whether linguistic constructions throw light on the ontological structure of perceptual experience. I suggest that they do not.
NOTES 1. I highly recommend Anil Gomes’ historical discussion of the place of introspective evidence in the philosophy of perception (cf. Gomes, 2017). 2. For a more thorough assessment of this debate, cf. Sanhueza Rodríguez, 2020. 3. For other classic and more recent statements of the doxastic view, cf. Pitcher, 1971; Roxbee Cox, 1971; Sibley, 1971; Craig, 1976; Heil, 1984; Byrne, 2009; Glüer, 2009, 2018. 4. For sure, this is not the only way of reading the relevant objection. The debate between conceptualists and nonconceptualists which Evans is helping to shape at the time need not be conceived as turning on the question how the contents of experience should be conceived (cf. Crowther, 2006; Heck Jr, 2000; Speaks, 2005). For present purposes, I only want to suggest that it is not far-fetched to read Evans’ line of criticism as one that targets a particular understanding of perceptual content. 5. That said, a case on behalf of the doxastic view could run as follows: Armstrong and Pitcher did not take perceptual beliefs to stand for cognitively sophisticated states: on the contrary, they made efforts to deny that that is the case. As such, their accounts are not really affected by Evans’ intended criticism. Their views and the latter’s critique talk past each other insofar as they trade in terms of different notions of concept and belief. What Armstrong and Pitcher have in mind is a relatively
18
Chapter 1
rudimentary notion of concept as a discriminatory capacity that human adults, babies, and even certain lower-level creatures (e.g., cats and dogs) possess: on that basis, they build their perceptual beliefs as cognitively rudimentary attitudes in virtue of which subjects are sensitive to certain types of environmental information. Evans, meanwhile, grounds his understanding of beliefs on the notion of concepts which we know and love—that is, those abstract items which set fully developed adults apart from babies and snails. The belief-attitudes Armstrong and Pitcher refer to need not relate to the notions of judgment and of reasons; those Evans refers to, have to. In vernacular philosophical language, we should perhaps say that Armstrong and Pitcher are not talking about beliefs, but about a structurally similar albeit still different kind of dispositional state. This is the reason why the objection from over-intellectualization does not target the relevant doxastic accounts of perception. 6. As previously noted, Vendler in fact uses four categories: activities, accomplishments, achievements, and states. I merge the first two categories into that of process here, as accomplishments only pick up on processes in which intentional agents engage. Achievements refer to instantaneous or semi-instantaneous events, so they are not specially relevant for present purposes. 7. D.M. Armstrong suitably captures this characterization of events by describing them as “the coming to be or passing away of a state, or the initiating or terminating of a process” (Armstrong, 1968: 130–131).
Chapter 2
Experiential Heracliteanism
Recent work on perception is slowly steering away from debates exclusively concerned with the (intentional, phenomenological) properties and cognitive function of perceptual experience, so as to address the fundamental ontological question what perceptual experiences are (cf. O’Shaughnessy, 2000; Steward, 2011; Soteriou, 2007, 2011, 2013, 2018; Crowther, 2009a, 2009b, 2010; Lee, 2014; Di Paolo, Buhrmann, and Barandarian, 2017). On the plausible assumption that perceptual awareness exists in time and thereby involves changes in our psychological lives, the temporal and dynamic structure of experience crucially bears on how we should ontologically categorize it. While experience may intuitively involve temporally extended changes, there is plenty of room for debating whether its dynamic character rests on dynamic or nondynamic components. Heraclitus is held responsible for claiming that certain aspects of reality flow like a river (cf. fragments 50 and 51 in Kahn, 1979: 53), and the general thought behind the metaphor seems to be that change is a fundamental feature of (at least some aspects of) reality. Accordingly, a Heraclitean doctrine of perceptual experience may be understood as one according to which perceptual awareness is fundamentally dynamic: change is a fundamental or primitive element of at least a certain aspect of experience.1 EH will stand here for any view according to which perceptual experiences depend on at least some necessarily dynamic elements. Experiential Non-Heracliteanism (ENH) rejects that non-Heraclitean views that model the dynamic character of perceptual experience in terms of nondynamic elements (e.g., states, properties, and relations). This chapter will outline and critically assess two versions of EH: on the one hand, a rather extreme version according to which perceptual experiences are exhaustively described as necessarily dynamic processes, and, on the other, 19
20
Chapter 2
a more moderate version that models perceptual phenomena as states that in turn constitutively depend on necessarily dynamic processes. I shall argue here that, by encoding the notion of change into the fundamental structure of experiences, EH seems to obscure our understanding of perception as a natural dynamic phenomenon. The present assessment will be divided into four parts. First, I introduce my main target, namely, O’Shaughnessy’s and Soteriou’s Heraclitean characterizations of perceptual experience. Second, I critically assess two of O’Shaughnessy’s key considerations on behalf of EH: on the one hand, the claim that pure conceptual analysis reveals that perceptual experiences must be understood as processes rather than states—a line of reasoning I show to be inconclusive, and, on the other, a thought experiment concerning the total freeze of a mental life—a fiction that turns out to be silent on whether experiences are Heraclitean or non-Heraclitean. Third, I turn to describe two general problems faced by Heraclitean proposals: on the one hand, the necessarily dynamic processes they posit seem to obscure the very dynamic feature that they seek to account for, and, on the other, EH also obscures the ontological status of experiences by blurring their conditions of individuation. I take these arguments to be fully effective against the more extreme version of EH. My critical considerations are not in turn completely effective against the more moderate version of DH, but have enough firepower to motivate the exploration of a more conservative ontology of experience if available. Such a view is precisely what I intend to introduce and defend in chapter 3. EXPERIENCES AS NECESSARILY DYNAMIC PROCESSES For O’Shaughnessy, perception and action are varieties of experience (cf. O’Shaughnessy, 2000: 38), as such, he does not think that all experiences are perceptual. The present study is only concerned with the perceptual aspect of that wider account. A good starting point is the fact that, according to O’Shaughnessy, a crucial feature of perceptual experiences qua experiential is their essentially or necessarily dynamic character. Thus, he claims that “[c] haracteristically the contents of experience are in flux, and necessarily experience itself is in flux, being essentially occurrent in nature” (O’Shaughnessy, 2000: 43); or again, that “[i]t is not the mere existence of flux [ . . . ] in the case of experience that is distinctive: it is the necessity of flux” (O’Shaughnessy, 2000: 44). This characterization highlights two things: first, the dynamic or changing nature of perceptual experiences; and, second, the necessity of that character.
Experiential Heracliteanism
21
The relevant dynamic component translates into a fully Heraclitean conception of perceptual experiences (cf. O’Shaughnessy, 2000: 42). According to O’Shaughnessy, perceptual experiences are temporally extended events or happenings: such events are temporally structured in the sense that they have a beginning, a middle, and an end. More importantly, these happenings are of a processive kind, that is, they go on for a certain period of time. Thus, if Jim sees a bright star from t1 to t10, a perceptual event constituted by a mental process extending from t1 to t10 takes place in his mind. O’Shaughnessy also takes the notion of perceptual process to be intimately related to that of perceptual event; after all, he claims that perceptual awareness may be conceived in terms of processes and events. Both ontological categories are not equivalent, but they are closely related (cf. O’Shaughnessy, 2000: 44). In particular, events are complete only when processes conclude. Since parts of a potential event come into existence—that is, are realized—as the process which constitutes that event unfolds, one might actually say that an ongoing process is an incomplete event. A complete event is, in turn, the result of a process that has already stopped. Indeed, it seems natural to claim that a certain event, the Battle of Trafalgar, began at 11:45 on October 21, 1805, and ended five hours later; at each moment, the fighting through which the battle was successfully realized was going on. This suggests that, although intimately related, events and processes constitute different ontological categories. A fairly popular take on the present point is that events are related to processes via the notion of constitution: just as count-quantifiable, spatial entities (e.g., a statue, a tree) are constituted or realized by mass-quantifiable stuff (e.g., wood, bronze), certain events should be conceived as count-quantifiable, temporally extended items which are made of or constituted by massquantifiable processes (cf. Armstrong, 1968: 131; Steward, 1997: 94–97; Crowther, 2011). While temporally extended events may be understood as temporal particulars—that is, they exist in time, are temporally structured and count-quantifiable—processes may be conceived as the matter or stuff out of which such particulars are constituted. This suggestion is attractive because it captures intuitive contrasts between the notions of process and event: processes, not events, go on; process-talk allows for adjective or adverbial qualifications which event-talk does not—for example, the humming of my computer may be persistent or continuous (cf. Steward, 1997: 95); unlike processes, events do not stop but only come to an end (cf. Steward, 1997: 95). Events are count-quantifiable—we can speak, for example, of one or two songs, of one or two battles—whereas processes are only massquantifiable—there is not one or two hummings, one or two runnings, but only more or less humming, more or less running (cf. Steward, 1997: 96–97; Crowther, 2011). In line with these remarks, O’Shaughnessy himself caps the
22
Chapter 2
previously quoted passage by saying that perceptual processes are “the very stuff or phenomenal matter of events.” That said, the relevant processive view also contains a modal qualification which is crucial for understanding what kind of perceptual processes, and hence events, are at stake. As stated at the beginning of this section, O’Shaughnessy thinks that perceptual experiences are necessarily or essentially processive. Why? Because he conceives them as processive or occurrent “to the core” (O’Shaughnessy, 2000: 49). To highlight this point, he draws a line between experiences, on the one hand, and, on the other, nonexperiences or “the sector that encompasses the relatively stable unexperienced mental foundation (e.g., cognitive, evaluative, etc.) upon which experience occurs” (O’Shaughnessy, 2000: 42–43). The relevant contrast is not one between a dynamic and a static sector of the mind. After all, changes can take place among nonexperiential states: for example, our beliefs or memories could fade, reemerge, or otherwise change over time. The crucial point is not merely that perceptual experiences involve change, for nonexperiential states do so as well. The thought is that experiences are processive down to their ultimate parts; no matter how you go about analyzing perceptual awareness, you always end up with processes. By thus expanding on the processive view, O’Shaughnessy comes to share a claim suggested by Vendler, namely, that perceptual processes and perceptual states are mutually incompatible. Let me expand on this point. Throughout the defense of his processive stance, O’Shaughnessy raises the question whether perceptual experiences could allow for a dual ontological analysis, that is, whether they could be analyzed in stative as well as processive terms (cf. O’Shaughnessy, 2000: 46). This possibility is suggested by the fact that physical as well as psychological, but nonexperiential processes may be so analyzed (cf. O’Shaughnessy, 2000: 44–47). For example, O’Shaughnessy concedes that certain movements across space may be conceived as processes constituted by objects standing in certain states: “constituting a process like moving out of states like being at a position in space at a particular time, is not in competition with constituting such a process out of parts the same in kind as itself” (O’Shaughnessy, 2000: 45). The point is not that either analysis is acceptable, but that it is necessary to invoke both process-parts and state-parts in order to capture the kind of events or changes that certain movements across space are. Again, some psychological albeit nonexperiential phenomena (e.g., certain instances of forgetting, coming to understand, or deciding) also seem to be processes with states at their core. O’Shaughnessy by no means thinks that physical and psychological but nonexperiential change will always be processive: according to him, processes in general are changes always exhibiting some form of continuity among their constituting parts (cf. O’Shaughnessy, 2000: 47), but instances of changes
Experiential Heracliteanism
23
like movement across space, forgetting, and deciding, may be either continuous or discontinuous. Physical and psychological but nonexperiential changes involve a processive and a stative analysis only when they are temporally continuous. But what about experiential processes? Could they also be analyzed in terms of processes involving temporally continuous state-parts? According to O’Shaughnessy, they could not: he argues that psychological states cannot be constitutive components of perceptual experiences (cf. O’Shaughnessy, 2000: 44, 47), and it should be relatively clear that this negative claim is related to the irreducibly processive character of experiential flux: after all, if experiences are irreducibly processive—that is, if they are processive or occurrent to the core—they could not be analyzed into stative components; mental states cannot underpin experiences because these psychological phenomena are necessarily processive. At a crucial point in the previously quoted passage, O’Shaughnessy relates the temporal structure of aural experiences to the temporal structure of their corresponding objects. Indeed, states may lie at the heart of physical and nonexperiential processes, but that is so only because such changes are not processive through and through. According to O’Shaughnessy, however, there are no experiential states. To motivate his view on perceptual experiences, O’Shaughnessy relies to a greater or lesser extent on at least two motivations: on the one hand, he thinks that a head-on analysis of the concept of experience vindicates experiential processes over experiential states; and, on the other, he invokes a thoughtexperiment—viz. a case of “total mental freeze”—the purpose of which is to highlight the necessary dynamic character of perceptual phenomena and, accordingly, the obvious appeal of a processive view. O’Shaughnessy takes the correctness of his processive view to count against a stative position. I shall go back to these motivating thoughts in the next section, but let us first turn to Soteriou’s more moderate version of EH. Soteriou summarizes his experiential ontology as follows: When one perceives a static scene, the phenomenally conscious state that obtains is an “occurrent” state whose obtaining over an interval of time is constitutively dependent on the occurrence of an event/process that takes that interval of time to occur. The event/process in question is not an event/process one is perceptually aware of—for in the case imagined one is aware of a static scene. The event/process in question is, rather, a phenomenally conscious event/ process that occurs during the time over which the phenomenally conscious state obtains. The “occurrent” phenomenally conscious state obtains in virtue of, and for the duration of, the occurrence of the phenomenally conscious event/ process. (Soteriou, 2013: 141)
24
Chapter 2
Perceptual experiences are mental states constitutively dependent on processes of a phenomenally conscious kind. To bring out the full import of this claim, it is important to note that Soteriou does not conceive the categories of process and state as mutually exclusive. Processes are thought of as continuous, temporally extended changes, the latter being in turn conceived as successive transitions among states.2 Soteriou describes occurrent states as states the existence of which is underpinned by that of events or processes (cf. Soteriou, 2011: 496; 2013: 104; Steward, 2018). In particular, he characterizes them in terms of three features: First, their obtaining constitutively depends on the occurrence of such events or processes of a certain kind. A liquid’s temperature, for example, may be understood as a state in which that substance finds itself when and because it undergoes certain processes at a molecular level. Since the relation of constitution need not be reductive, the relevant states are not necessarily explained away in terms of their constituting processes. So occurrent states depend on certain processes, but need not collapse into the latter. Second, such states are occurrent precisely because they depend on processes or otherwise episodic items. Although it is unclear to me whether the relevant states literally inherit the occurrent character of their constituents,3 the aforementioned constitutive bond at least seems to guarantee that the relevant states are occurrent in an important even if derivative sense. In other words, such states at the very least presuppose certain sorts of occurrences in their corresponding subjects. Third, occurrent states and their underlying processes could be heterogeneous in the sense that they need not belong to the same kind. A liquid’s heat does not depend on “hot” processes, but on molecular ones. My back pain does not rely on processes of the same subjective or phenomenological kind, but on processes of a muscular and nervous kind. Again, certain forms of perceptual monitoring (e.g., watching) may involve processive and nonprocessive events of different kinds (e.g., spotting, looking for, staring at) (cf. Vendler, 1957: 151, 158–159; Crowther, 2009b). Soteriou’s proposal models perceptual experiences as states on a number of grounds. For reasons that go beyond the scope of this work, Soteriou endorses a relational view of perception in virtue of which it becomes natural to conceive perceptual awareness as instantiations of phenomenally conscious properties in a given subject (cf. Soteriou, 2013: ch. 1, 2 & 4; Soteriou, 2016). Again, he takes a stative ontology of experience to accommodate the phenomenological fact that perceptual awareness apparently ranges over temporally extended objects (cf. Soteriou, 2011: 489–490, 493; 2013: 94). Finally, he has even defended a stative view on the basis of grammatical considerations concerning the relationship between perceptual verbs and the continuous tense (cf. Soteriou, 2007: 551). That said, Soteriou also thinks that, to the extent that they obtain at a time or over a period of time, states
Experiential Heracliteanism
25
fail to capture an important dimension of perceptual experience, namely, its occurrent or dynamic character. As such, he also espouses a processive approach. The general thought behind this stance is in fact quite persuasive: whether directed at a typical changing landscape or at a rare static scene, our perceptual lives are fundamentally dynamic, at least in the sense that they could not exist in a world where our mental lives came to a total stand-still (cf. O’Shaughnessy, 2000: 42–43). Even when presenting us with static scenes, perceptual experiences would make us aware of the passage of time.4 In other words, to accommodate the sensory phenomenology of temporal passage, Soteriou models perceptual experiences as phenomenally conscious processes. Here lies the crux of Soteriou’s ontological proposal: the relevant processes are not of any common kind; they are precisely what commit him to a Heraclitean ontology of experience. As previously mentioned, physical and psychological but unconscious processes may be understood in terms of the continuous transition among states. In order to accommodate the dynamic character of perceptual experiences, however, Soteriou introduces processes that allow for no such a stative description: according to him, perceptual experiences involve phenomenally conscious processes that are in turn constituted by further processes. Soteriou’s experiential processes are like no other standard form of (physical or psychological) process because, unlike the latter, they cannot be understood in terms of the continuous transition among states over time: their internal ontological structure resembles Heraclitus’ river—that is, they are dynamic through and through. Although Soteriou’s allegiance to EH may be less obvious than that of O’Shaughnessy’s, his defense of perceptual relationalism to a good extent rests on the thought that representationalism has to conceive perceptual experiences as state-dependent processes, a conception that he in turn finds unappealing (cf. Soteriou, 2013: 37ff.). The phenomenally conscious processes thus invoked are no ordinary processive items: in virtue of their special modal character, they constitute a sui generis set of entities even in a worldview that includes those (physical or psychological but nonconscious) processes we are familiar with. This section has thus outlined two versions of EH, that is, the stance according to which perceptual experiences are (partially or wholly) constituted by processes of a necessarily dynamic kind. As previously mentioned, O’Shaughnessy’s more extreme view resorts to two motivations: on the one hand, he thinks that a head-on analysis of the concept of experience vindicates experiential processes over experiential states; and, on the other, he invokes a thought-experiment—viz. a case of ‘total mental freeze’—the purpose of which is to highlight the necessary dynamic character of perceptual phenomena and, accordingly, the obvious appeal of a processive view. In the
26
Chapter 2
next section, I critically assess both motivations. Then, I shall outline two problems for both versions of EH. CONCEPTUAL ANALYSIS AND THOUGHT EXPERIMENTS To begin with, O’Shaughnessy argues that perceptual experiences could not be conceptually analyzed or anatomized in terms of temporally continuous transitions from state to state, but only in terms of processes. The modal qualification of this statement is important: O’Shaughnessy’s claim is not just that perceptual experiences may be analyzed along processive lines, but that they have to insofar as they are occurrent to the core. Since my target is only the modally stronger claim, the present section strives to show that O’Shaughnessy has not said enough to rule out an alternative analysis of perceptual experiences. Since O’Shaughnessy focuses on undermining the stative view, his argumentative strategy is to a good extent negative. He writes: [ . . . ] one is inclined to believe that (say) hearing a sound consists in the obtaining of a relation, that of awareness, between a mind and a sound. Accordingly, one might suppose that there exists an experience which is the realization in time of a state, viz. the relation of awareness between a mind and a sound. This is to strictly model “He hears the sound” upon “He touches the wall.” But “He touches the wall” is ambiguous between an event consisting in the establishing of a relation, and the relation itself. By contrast, “He hears the sound” exhibits no such ambiguity: it describes an event, and never designates a relation. A fortiori the event of hearing a sound does not consist in the realization at or over a time of a relation of hearing the sound. This event occurs at an instant if the sound is instantaneous, and over an interval if the sound is temporally extended; then in the latter case it will need to be renewed instant by succeeding instant, as happens when listening is going on. (O’Shaughnessy, 2000: 49)
In the present context, O’Shaughnessy conceives “relations of awareness” as states in which subjects may stand relative to their surroundings; this is, I suspect, why he denies that perceptual experiences pick up on such relations. Accordingly, the question at stake is whether he conclusively shows that the notion of perceptual experience should necessarily be analyzed in terms of mental processes, as opposed to relations of perceptual awareness. My answer will be negative, and, as a result of that, I conclude that the present motivation is unsatisfactory.
Experiential Heracliteanism
27
Turning to the previous question, consider a construction of the following form: (i) S perceives O (as F), where “perceives” could be replaced by “sees,” “hears,” “smells,” etc. If there are any linguistic constructions we use to pick up on perceptual experiences, (i) seems a good candidate. O’Shaughnessy specifically tries to drive an asymmetry between statements concerning hearing and touch. According to him, the reference of a perceptual statement like (ii) He touches the wall, is ambiguous between a duration-less event—the touching of a wall at an instant—and a state—that is, the state in which a subject stands vis-à-vis the object (i.e., the wall) he is in direct contact with. Although I think that (ii) could be read as a statement of tactile perception (Armstrong, 1962; Martin, 1995; Bermúdez, 1998; de Vignemont, 2011), the most charitable reading in the present context seems to be that it concerns a physical relation which may obtain even in the absence of tactile perception, as when a plant or a rock touches a wall. The suggestion which O’Shaughnessy tries to counteract is that one could apply a similar analysis to statements concerning perceptual phenomena. To undercut this move, he contends that a statement of auditory perception, such as (i*) He hears the sound, is not ambiguous; on the contrary, it unequivocally points to a single reading where, at least as far as hearing temporally extended sounds is concerned, the subject’s hearing should be understood in processive terms. Although O’Shaughnessy does not generalize this claim to statements of other sensory modalities, one would expect him to do so for the sake of a Heraclitean view. That said, I do not think that the previous line of reasoning conclusively shows that perceptual experiences have to be analyzed along processive lines: for, on the one hand, an alternative (specifically, a stative) analysis of perceptual statements also seems initially plausible; and, on the other, it is unclear that constructions like (i*) actually demand the introduction of a processive component. Turning to the first point, reflection on our pretheoretical uses of perceptual verbs actually suggests that (i*) and (ii) resemble each other more than
28
Chapter 2
O’Shaughnessy allows. When Jim claims “I see it [i.e., the star],” he may naturally be taken to refer either to the duration-less event of spotting a star or to the temporally extended occurrence of being aware of the star between t1 and t10. Over and above a description of our ordinary linguistic practices, Vendler’s work on seeing also suggests an alternative way of categorizing perceptual experiences. As I previously explained, the basic thought was that visual awareness could be understood as a state obtained in a determinate subject, or, to use O’Shaughnessy’s terminology, one could conceive visual experiences as relations of awareness instantiated by a subject and her surroundings. This conception is, I think, appealing insofar as it draws from a familiar understanding of experiences as being open to the world. Indeed, experiences are naturally taken to constitute a fundamental means for being in contact with the external world, for retrieving information for cognitive processing downstream perception. Thus, conceived, perceptual experiences do not seem to be all that conceptually different from tactile phenomena of the sort expressed in (ii): that is, it does not seem counterintuitive to analyze statements of the form (i) along the stative lines which, as O’Shaughnessy concedes, the analysis of (ii) follows. These brief remarks are of course not intended to vindicate a stative analysis of (i*): all I thereby aim to show is that O’Shaughnessy’s strong modal claim is not obviously true. Recall that he argues for the claim that perceptual experiences necessarily demand a processive account; according to him, they have to be analyzed along processive lines. Rather, my point is just that, to the extent that he does not say enough to undermine what is an otherwise intuitive analysis of experience, O’Shaughnessy has to say much more in order to show that statements of perceptual experiences cannot be analyzed in any but processive terms. In addition to the previous point, it is not even obvious that O’Shaughnessy’s analysis of (i*) is correct: more specifically, his inference to the claim that statements about auditory perception refer to experiential processes seems problematic. At a crucial point in the previously quoted passage, O’Shaughnessy claims that “the event of hearing a sound [ . . . ] occurs at an instant if the sound is instantaneous, and over an interval if the sound is temporally extended.” I am ready to concede a number of things here. To begin with, I grant that a subject may perceive processes, for example, when she perceives a temporally extended sound. Again, a subject’s visual experiences typically are temporally extended. Finally, I could concede that there is a natural even if broad sense in which the length of an experience in objective time depends on the length of the perceived item in objective time. But does any of this entail that, in addition to a perceived process (say, a temporally extended sound) and the neurobiological or subpersonal processes taking place in a subject when she perceives a process over a period of time, there is a process of an entirely different kind—presumably, a phenomenally
Experiential Heracliteanism
29
conscious one—taking place in the subject’s mind? I do not think so. Everything O’Shaughnessy has established thus far is that a subject may perceive processes or otherwise temporally extended items, and that, to perceive such objects, she has to be aware of them for as long as they take. But these facts also seem to be compatible with a stative view of experience, for perceptual states no less than perceptual processes can be temporally extended. If hearing a temporally extended sound may be understood in terms of a relation of awareness between a subject and a worldly item, as a way of being in touch with the world, it is thus unclear that hearing something has to be extended in time by unfolding or by taking time. As the reader may recall, these remarks are nothing more than an expansion of the Vendlerian line of criticism against perceptual “happenings” introduced toward the end of chapter 1. O’Shaughnessy, meanwhile, asks us to believe something far stronger: the suggestion is that perceptual experiences have the exact same temporal structure as the processes a subject would perceive. In fact, if we bear in mind that experiences are supposed to be processive to the core, he is endowing perceptual experiences with a temporal structure that no physical or subpersonal process quite shares. Since a subject may be perceptually aware of processes—say, a temporally extended sound or a jogger’s running across the park-—O’Shaughnessy assumes that the corresponding episode of temporally extended perceptual awareness must be processive. To reach the desired conclusion, he perhaps presupposes a principle to the effect that the temporal features of represented items have to be ascribed to the corresponding vehicles of representation. This seems unlikely, though. Furthermore, Daniel Dennett has persuasively argued that a principle along such lines is far from uncontroversial (cf. Dennett, 1991). O’Shaughnessy’s move is controversial to the extent that all he says about the analysis of temporally extended episodes of perceptual awareness could also be accommodated by a nonHeraclitean approach to experience like the one I shall outline and defend in the next chapter. As such, O’Shaughnessy’s conceptual analysis of perceptual experiences does not conclusively show that perceptual experiences have to be analyzed as mental processes. A second motivation behind O’Shaughnessy’s processive view concerns the alleged difference between “the characters and conditions of identity” of experiences and nonexperiences (O’Shaughnessy, 2000: 44). The general thought is, I think, that perceptual experiences fail to exist in circumstances where processes would fail to do so, as such it would be extremely reasonable to conceive perceptual experiences as mental processes. To spell this point out, O’Shaughnessy invokes the following thought experiment: [ . . . ] the domain of experience is essentially a domain of occurrences, of processes and events. In this regard we should contrast the domain of experience
30
Chapter 2
with the other great half of the mind: the non-experiential half. That is, the sector that encompasses the relatively stable unexperienced mental foundation (e.g., cognitive, evaluative, etc.) upon which experience occurs. While many of the non-experiential contents of this domain could continue in existence when all mental phenomena had frozen in their tracks, say (fancifully) in a being in suspended animation at 0° Absolute, those in the experiential domain could not. (O’Shaughnessy, 2000: 42–43)
According to this fiction, the mental life of a subject is frozen in a way which is intended to highlight different circumstances or conditions under which experiences and nonexperiences may exist: unlike experiences (e.g., perceptual experiences), nonexperiences (e.g., beliefs) could exist in a frozen mental life, that is, a life where no mental changes take place. In the light of their distinctive dynamic character, experiences are naturally classified as processes. As stressed by O’Shaughnessy, the relevant psychological contrast is one between a sector of the mind irreducible to nonprocessive components and a sector of the mind ultimately reducible to nondynamic elements (i.e., state-parts). An initial worry about this fiction is whether it is actually obvious that one could ascribe nonexperiential states (say, beliefs) to a subject in total mental freeze—after all, this fiction comes close to a case of brain death, where it is not implausible to deny cognitive states to the relevant subject. Furthermore, the uses of “can”-constructions are complex enough to raise the question whether the sense in which a subject in total mental freeze could have cognitive states is the same sense as that in which a sleeper or otherwise unconscious person could do so. I admit, though, that this line of attack is extremely delicate, so I present it only as a tentative suggestion. The line of criticism I shall pursue here is that, since O’Shaughnessy’s thought experiment could be accommodated within a non-Heraclitean understanding of experience, it is unclear why the idea of mental freeze forces a processive view upon us. That is, I shall grant that a subject in mental freeze could have a nonexperiential life while lacking an experiential one. More importantly, I agree with O’Shaughnessy that a frozen mental life is one where perceptual experiences cannot exist. Although he grounds this point on the allegedly processive character of experience, one could independently motivate it by recognizing that, if perceptual experiences constitute a key informational channel between mind and world, a frozen mental life—as it were, one in informational lockdown vis-à-vis its environs—is one where a subject does not engage in live transactions with her surroundings; that is, one where perceptual experiences cannot take place. Now, my point is that denying the existence of perceptual experiences in cases of mental freeze does not entail that perceptual experiences are Heraclitean processes.5
Experiential Heracliteanism
31
The previous line of objection may be supported in two related ways. First, one could hold that O’Shaughnessy’s thought experiment does not show that experiences are necessarily processive, but only that experiences and nonexperiences constitute different kinds of states—that is, kinds of states which have different identity conditions. One would certainly have to motivate this line of reasoning, but it seems a promising option. I motivate this line of thought in the next chapter. Second, one could hold that O’Shaughnessy’s fiction does not show that experiences are mental processes, but only that they somehow depend on processes of different kinds. In a slogan, the relevant case only shows no changes whatsoever, no experiences. But it fails to show, first, that there is a one-on-one mapping between experiences and processes, and, second, that the relevant processes are of a specific mental kind. To secure a processive conception, O’Shaughnessy needs to secure these two points: since his thought experiment fails to do so, he fails to secure a processive view. To illustrate the previous remarks, let us assume that perceptual experiences are mental states, and then determine whether one could still make sense of O’Shaughnessy’s mental-freeze case. The subjects of these experiences have bodies which, in turn, implement sensory systems from which perceptual experiences will ensue, as such one could reasonably suppose that a complex number of physiological processes take place in the relevant perceivers. On the basis of these stipulations, the suggestion is that the case of total mental freeze could be accommodated by a stative understanding of perceptual experiences. The reason why this is so is that, although perceptual experiences themselves would not be Heraclitean processes, they could be states that in turn depend on processes of a different—specifically, a nonHeraclitean—kind. The sort of dependence at stake would not be causal, or at least not merely causal: crucially, I am inclined to think that such states would constitutively depend on processes; that is, the latter processes would be constitutive elements of perceptual states. As previously mentioned, the idea of a process-dependent state prominently features in Soteriou’s work and will also characterize the version of ENH I endorse. Now, to the extent that the condition of total freeze could obliterate not only subjective processes but also subjective process-dependent states, a subject could thereby fail to have perceptual experiences not only when the latter are conceived as processes but also when conceived as process-dependent states. Although O’Shaughnessy is rather obscure on this point, I take it that a world where a perceiver’s mental life is frozen has to be one where the psychophysical basis of her experiences is (totally or partially) in a standstill too; this is, I presume, what would account for the subject’s being experientially cut off from her surroundings. In that world, freezing the physical basis of perceptual phenomena would also undermine the mental states which it
32
Chapter 2
grounds. The general notion of process would still have a role to play in the relevant scenario, but it does not follow that experiences have to be happenings of a special processive kind. So I grant that O’Shaughnessy may illustrate a significant difference by means of the discussed thought experiment, but I do not think it forces a processive view of experience on us. Before concluding this section, it is necessary to flag one last important issue. Could O’Shaughnessy argue that the case of total mental freeze is supposed to be based on our first-person introspective or otherwise phenomenological intuitions about visual experiences? That being the case, he could go on to argue that, even if it is true that perceptual experiences are states which constitutively depend on processes, my foregoing remarks would not invalidate the introspective or phenomenological data and the philosophical view of experience thereby derived. This suggestion is extremely important, for it leads to a line of reasoning which independently fuels EH. O’Shaughnessy, for example, apparently grounds his own version of Heracliteanism on the fact that perceptual experience acquaints us with the passage of time or the constant change of the present (cf. O’Shaughnessy, 2000: 49ff.). Again, Soteriou argues that phenomenally conscious processes nicely accommodate the temporal transparency and the apparent temporal continuity of perceptual experiences (cf. Soteriou, 2011, 2013). Neither line of reasoning depends on an analysis of the concept of experience or the interpretation of the mental freeze case, so they are not touched by my foregoing critique. That said, they are extremely delicate insofar as they presuppose answers to paramount and extremely difficult questions such as “What features, if any, are involved in the temporal phenomenology of perceptual experiences?,” “What and how such phenomenal features are introspectively accessible to their subjects?,” “What does the temporal structure of experiential content tell us about the temporal structure of their experiential vehicles—of experiences themselves, as it were?,” among others. These issues unfortunately escape the scope of this chapter. For the time being, all I can do here is to voice the general stance I shall pursue in chapter 6. In broad lines, I do not think that introspective or phenomenological data could determine how perceptual experiences should be ontologically classified, for those phenomenological features seem to be neutral regarding the question which ontology of perception we should endorse. To support this point, I shall show that a non-Heraclitean view could also accommodate the relevant phenomenological facts. So just as O’Shaughnessy claims at one point that the notion of experience is not a provider of ontological or categorial status to the psychological tokens falling under that concept (cf. O’Shaughnessy, 2000: 41–42), I doubt that introspecting our perceptual experiences actually tells us much about what kind of things they are. It is already controversial that introspective or phenomenological data tell us something about the worldly items we perceive. Whether
Experiential Heracliteanism
33
they unveil the ontological structure of experiential items themselves—either a processive or a stative structure—is a different, and even more controversial, issue. To sum up. In this section, I examined two key motivations behind O’Shaughnessy’s Heraclitean view of perceptual experience. Pure conceptual analysis and reflection on the case of total mental freeze seem to have a lot of pull when it comes to motivating the initial plausibility of the view that perceptual experiences are processive through and through. But, as far as I can see, such considerations are far from unproblematic. If my critical remarks are correct, they do not undermine O’Shaughnessy’s overall project; after all, as I just explained, it could still fall back on considerations concerning the temporal content of perceptual experiences. However, they do counteract considerations that may naturally predispose us to take a Heraclitean view for granted or to deem a non-Heraclitean one as a nonstarter. CHANGE AND INDIVIDUATION Soteriou’s ontology of experience is Heraclitean insofar as it relies on processes of a necessarily dynamic kind—that is, elements unique for being fundamentally dynamic or for not being describable in nondynamic terms (e.g., states, relations, or properties). As previously mentioned, I envisage at least two initial problems for such a Heraclitean account: (i) Heraclitean—that is, necessarily dynamic—processes seem to obscure rather than to clarify the dynamic character of experience; and (ii), they are also exotic entities apparently lacking straightforward conditions of individuation. Turning to (i), the worry broadly relates to Plato’s classic challenge to Heraclitus’s theory of flux. The objection springs from the thought that, no matter how it is exactly conceived, change becomes intelligible only against the backdrop of a stable or unchanging substratum; in other words, change can be predicated only of something that remains one and the same over time. As conceived by Plato, however, a Heraclitean world is one where absolutely everything changes at every instant of time, thus undermining even our ability to fix the meaning of our words (cf. Plato, 1921: 181e-182a). Thus, striving to accommodate the dynamic character of nature, Heraclitus would undermine the very linguistic or conceptual framework within which the notion of change takes shape in the first place (cf. Plato, 1921: 182d-e). Extrapolating from the previous line of criticism, one could now argue that, qua dynamic, every conscious phenomenon must be conceived against the background of something that remains stable or unchanged. By defining perceptual experience in terms of necessarily changing or dynamic items, however, EH undermines any stable background against which the dynamic
34
Chapter 2
character of experience could shine through. This point, I think, becomes specially salient when Heraclitean theorists characterize experiential processes as homogeneous, that is, as items constituted by elements of the same kind. For, whereas dynamic phenomena are internally complex in the sense that they presumably involve distinct and progressive stages, homogeneous items precisely lack that sort of internal structure (cf. Mourelatos, 1993: 386; O’Shaughnessy, 2000: 44). If experiential processes are constituted by parts which are of the exact same dynamic kind as that of the processes they constitute, they will not be capable of reflecting the internal complexity necessary to characterize conscious phenomena as dynamic. Hence, while the relevant phenomenally conscious processes were brought up to account for the occurrent or dynamic character of perceptual experiences, their radically dynamic existence might actually obscure that feature. In regard to (ii), the relevant processes are also problematic because they lack clear conditions of individuation. Although finding such criteria may not be urgent when it comes to items well entrenched in our intuitive and socially acquired perspective of the natural world, it is so when we are presented with unfamiliar, philosophically motivated entities. Heraclitean processes, I suspect, correspond to the latter category. They do not seem to be denizens of the world we know and love. How are we supposed to identify them or otherwise get a hint of their existence? Soteriou himself concedes that they are not introspectively accessible to their subjects (cf. Soteriou, 2013: 135, 139). Again, Heraclitean processes are not of a piece with ordinary processes; as previously explained, such items cannot be described in terms of the nondynamic components—that is, objects, properties, relations, among others—that constitute our more familiar physical and psychological but nonconscious world. As such, the existence of Heraclitean processes does not rest on our familiarity with instances of ordinary physical and psychological but nonconscious processes. Since the notion of process picks up on temporally extended items, one might perhaps strive to individuate the controversial items at stake by locking onto their temporal boundaries—that is, their beginnings and endings. This tactic is problematic too: for, as B.A. Farrell stressed long ago, the onset and ending of a subject’s temporally extended experiences are marked by their overt and covert behavioral responses (cf. Farrell, 1950: 178), and since physical and subpersonal responses or processes may be described along stative lines, their temporal boundaries cannot constitute those of the necessarily dynamic processes at the heart of O’Shaughnessy’s or Soteriou’s version of EH. Hence, to the extent that they lack intuitive criteria of individuation, the relevant processes constitute a mysterious or exotic category of entities. Granted: (i) and (ii) are not necessarily fatal for every form of EH. Building on a general claim he terms the Interdependence Thesis (cf. Soteriou, 2013:
Experiential Heracliteanism
35
141), Soteriou argues that phenomenally conscious processes should be individuated or specified in terms of the phenomenally conscious states they constitute, and, conversely, that phenomenally conscious states are individuated by the necessarily dynamic processes they depend on. That being the case, phenomenally conscious states might somehow provide the stable background necessary for making the dynamic character of perceptual experiences intelligible. Again, the same phenomenally conscious states could perhaps fulfill the individuating job that introspective awareness or the temporal boundaries of Heraclitean processes could not. In spite of these possibilities, (i) and (ii) nevertheless highlight that EH relies on a controversial set of entities and a potentially vicious principle of circular individuation. The non-Heraclitean proposal I outline in the next chapter, I believe, avoids those commitments. CONCLUSION Once mostly interested in debates about the intentionality and the phenomenology of mental phenomena, philosophers of mind are slowly coming to terms with the importance of heavy-duty metaphysical and ontological questions. Aiming to make a modest contribution to this tendency, this chapter introduced and put some pressure on Brian O’Shaughnessy’s and Matt Soteriou’s Heraclitean ontologies of perceptual experience. O’Shaughnessy’s view of perceptual experience is no doubt one of the most radical versions of EH available. After unpacking it, I went on to examine two initial motivations behind that position. Pure conceptual analysis and reflection on the case of total mental freeze seem to have a lot of pull when it comes to motivate the initial plausibility of the view that perceptual experiences are processive through and through. But, as far as I can see, such considerations are far from unproblematic. Soteriou in turn endorses a form of EH where experiences are occurrent states, that is, states constitutively dependent on processes of different kinds, and the relevant processes are irreducibly dynamic or processive down to instants. This proposal is once again problematic, though: on the one hand, it is unclear that the relevant processes accommodate rather than obscure the dynamic character of perceptual phenomena; and, on the other, their conditions of individuation are by no means transparent. My own suggestion is that an ontology of perceptual experiences need not rely on the controversial category of necessarily dynamic processes; instead, experiences may be conceived as the instantiation and subsequent maintenance of informational relations between a subject and her surroundings, states that in turn constitutively depend on processes of a neurobiological or otherwise subpersonal kind. At any rate, if the fine-grained import of the ontological concepts at stake—that is, state, process, and constitution—is taken seriously,
36
Chapter 2
this stance remains respectably materialist without going reductionist. I turn to a more detailed discussion of this stance next. NOTES 1. For different readings of Heraclitus’ doctrine of flux, cf. Barnes, 1982: 49–52; Graham, 2015. 2. As previously stated, even radical Heracliteanists like O’Shaughnessy conceive nonexperiential processes along these lines (cf. O’Shaughnessy, 2000: 44–47). Again, this conception of process/change echoes Russell’s “at-at” account of locomotive change: “[m]otion consists merely in the fact that bodies are sometimes in one place and sometimes in another, and that they are at intermediate places at intermediate times.” (Russell, 1917: 84) Since being located at a certain place is a state, Russell seems to suggest that locomotive processes may be understood in terms of transitions among states. 3. A worry Tom Crowther made me aware of in conversation. 4. By examining perceptual scenarios where the relevant objects of perception are static, Soteriou seeks to avoid conflating the present ontological proposal with a structurally similar account of event perception he espouses elsewhere (cf. Soteriou, 2013: 139–140). While the second explanatory model posits experiential states constitutively dependent on processes, the latter correspond to those worldly, temporally extended processes that a subject experiences. The story of experience now discussed, meanwhile, resorts to processes that literally constitute the internal stuff experiences are made of, and, as I shall presently explain, Soteriou takes such experiential processes to be unlike any physical or subpersonal process we know of. 5. My intention is not to provide counterexamples to the case of mental freeze—I have just conceded that O’Shaughnessy’s construction is internally coherent—but to put pressure on the transition from that thought experiment to a Heraclitean view. To show the latter, I shall presently argue that a condition of mental freeze could be made sense of even if perceptual experiences turned out to be mental states.
Chapter 3
Experiential Non-Heracliteanism
OF EXPERIENCE AND THE RIVER Like many other philosophical debates, the ontological discussion here outlined to a good extent hangs on how a particular metaphor is understood. As previously mentioned, Heraclitus is often held responsible for the thought that reality—or certain aspects of it, at any rate—persists and changes the way a river does. Indeed, the image of a flowing stream evokes the subtle and puzzling ways in which persistence and change interact with each other; for, while there is a natural sense in which you may bathe in one and the same river at different times, there is another sense in which you cannot bathe in the very same river twice. The “stream” metaphor and the Heraclitean thought that change is a fundamental element of reality, strongly resonate in current discussions about experience and time. The phenomenology of temporal passage, a prominent theme in contemporary philosophy of mind, refers to the apparent psychological fact that we seem to be aware of time’s flow. Philosophical views about this phenomenological datum do not challenge the apparently dynamic structure of conscious experience: they disagree on whether the fundamental ingredients of such a structure are themselves dynamic. This debate in turn bears on discussions of objective time, for metaphysicians of time explore what our sense of temporal passage tells us about the question whether objective time is dynamic or static (cf. Dainton, 2011; Skow, 2011; Frischhut, 2013). Once again, dynamic and static views of time do not argue whether the relevant explanandum is dynamic or not, but whether that feature is itself constituted by irreducibly dynamic ingredients. Neither stance in either debate is Parmenidean, then: they disagree on whether subjective or objective time is fundamentally Heraclitean or not. 37
38
Chapter 3
At a purely descriptive and metaphorical levels, the idea of temporal passage points at the apparent progression of time: given a particular point or frame of reference, what may be identified as temporally present in the physical and the psychological world changes from instant to instant; with every present event becoming a past one, our past grows like a highway we leave behind, and again, whether determined by the past and the laws of nature or not, events in the more or less distant future come to happen now so as to recede into the depths of our past. Accordingly, the relevant phenomenological datum may be metaphorically described as follows: conscious experience not only makes us aware of spatially related items but also of temporal passage. A bit more specifically, it highlights the intimate link between conscious experience and time. Building on Brian O’Shaughnessy’s suggestion that “the experiencing subject stands in a special relation to time not discoverable in those not experiencing” (O’Shaughnessy, 2000: 50), Soteriou asks what kind of distinctive conscious contact with time temporally extended sensory experiences afford us with (cf. Soteriou, 2013: 135, 136ff.).1 According to him, perceptual experience provides us with a sense of the passage of time. What sets a conscious subject apart from a temporarily or permanently unconscious one is the former’s distinctive awareness of temporal passage. Conscious sensory experience thus stands out for its ability to provide a special informational access to the flow of time, a distinctive trait that may also be expressed in terms of two conditions: on the one hand, a presentness condition, to the effect that conscious subjects are aware of subintervals within an experienced interval of time as temporally present; and, on the other, a succession condition, to the effect that the subintervals within an experienced interval of time, are experienced as successively present (Soteriou, 2013: 136–139, 141; Sattig, 2019). Both conditions seem jointly necessary for describing the phenomenology of temporal passage; for, while succession highlights the specific import of temporal passage, presentness grounds it by establishing the basic link between experience and time. A conscious subject’s special relation to time is partially spelled out by the thought that experience is bound to convey things to us as present. Indeed, we would not intuitively ascribe this trait to the dreamless sleeper’s dispositional thoughts and beliefs. But this condition is not on its own sufficient, for, on the one hand, there could perhaps be atypical cases where unconscious subjects hold dispositional states the representational contents of which are indexed to the present moment (cf. Soteriou, 2013: 136), and, on the other, awareness of several present instants—even if such instants happen to succeed each other—need not constitute a sense of passage for a conscious subject, since she could potentially represent a bundle of present yet unrelated moments of time.2 Succession comes into the picture at this point: conscious subjects stand in a special relation to time not only because they experience subintervals of time as present, but also because they experience them as successively present.
Experiential Non-Heracliteanism
39
That said, it is becoming increasingly clear that the phenomenology of temporal passage cannot be described purely in terms of experiential content or what experience consciously conveys to us (cf. Frischhut, 2013; Hoerl, 2014). To pause on one line of consideration,3 let us say that the relevant phenomenological datum could be read off from the worldly things we are directly or somehow indirectly aware of. Since it is an essential dynamic feature of experience—no experience would lack such a trait, I take it—what is experientially presented to us would always have to contain a dynamic component that went on to determine the phenomenological datum. However, we could actually think of experiential scenarios that conveyed no such elements to us—say, visual cases where the objects and conditions of perception were held fixed or perhaps episodes of extreme sensory deprivation. If coherent, this possibility thereby shows that our sense of temporal passage is not exhaustively explained by the contents of experience—that is, by what experience presents or represents to us. But if what conscious experience conveys to us does not fully determine its phenomenological features, perhaps something internal to it may do so. Putting this thought into sharper focus, new work in the ontology of mind hints at an approach that strives to throw light on our understanding of consciousness by raising the question how the temporal structure of experiences themselves could determine their temporal phenomenology.4 The underlying assumption behind this approach may be expressed in terms of the thought that conscious experience and specific aspects thereof cannot be fully described—nor, perhaps, explained—without explicitly or implicitly resorting to a story of its ontological structure over and above its purely intentional or representational dimension. Heraclitean views of experience trade on this fact in order to model perceptual experiences as fundamentally dynamic phenomena. In chapter 2, I mainly focused on putting some pressure on O’Shaughnessy’s more extreme version of EH. This chapter will, meanwhile, turn to Soteriou’s more nuanced version, as the latter view actually provides the template for the form of non-Heraclitean account I shall outline. While no doubt suggestive, it is by no means obvious what an ontological description of conscious experience as stream-like exactly comes down to. That is, how should the Heraclitean metaphor be understood in the domain of conscious sensory experience? The answer to this question is, I believe, more delicate than typically acknowledged (even by a powerful writer like O’Shaughnessy). As such, instead of addressing the sexier question how the ontology and the phenomenology of perception relate to each other, this chapter addresses the more basic but necessary task of parsing two opposing ways of fleshing out the Heraclitean metaphor in our understanding of experience. I specifically discuss Soteriou’s version of EH so as to bring out its substantial ontological cost and to introduce what seems to me a more
40
Chapter 3
ontologically conservative non-Heraclitean conception of experiences. While I do believe that EH invokes an ontologically demanding—and to that extent problematic—notion of process, this work only seeks to map different ways of understanding the dynamic dimension of our perceptual lives. Of course, specifying a less demanding way of conceiving change in experience, constitutes a necessary even if indirect step for grounding a case against EH. The next section will expand on Soteriou’s version of EH; more specifically, I parse a number of ways in which the idea of process may be understood, so as to reassert the claim that EH commits to a strong conception of the dynamic character of experience, perhaps one too strong to be endorsed— a claim I mostly voice here, as its main case has by and large been made in chapter 2. This discussion thereby leads to the subsequent section, where I outline a more conservative way of understanding the Heraclitean metaphor in the domain of conscious experience. To reinforce this alternative proposal—a version of ENH—I shall conclude addressing a worry concerning its ability to accommodate the dynamic character of perceptual experience. Hence, my overall line of reasoning by no means constitutes a definitive case for ENH or one against EH, but, if correct, it does enrich a stimulating and potentially fruitful philosophical landscape by expanding the variety of descriptive and ultimately explanatory options in what has up to now been a one-sided discussion. HERACLITEAN PROCESSES As I earlier noted, Soteriou fleshes out the Heraclitean metaphor by endorsing an ontology of perception that describes perceptual experience in terms of dynamic as well as nondynamic components (cf. Soteriou, 2013: 141). In a nutshell, perceptual experiences are occurrent states, the latter being in turn conceived as mental states constitutively dependent on processes of a phenomenally conscious kind. Processes (e.g., walking, writing, and swimming) are broadly conceived as the kind of temporally extended items that unfold through time or come into existence stage by stage, while states (e.g., being in love, having a color, believing, knowing) obtain or continue obtaining—that is, they are wholly actualized at each instant within the period of time throughout which they exist. In the present context, I take for granted a fairly widespread view of states as instantiations of properties or relations: an apple’s being red may, for example, be seen as a state that obtains when a concrete particular (i.e., an apple) instantiates a general property (i.e., redness); or again, while being married is a general relation that may be instantiated by multiple couples, John’s being married to Mary is a specific state involving John and Mary.5 As earlier mentioned, what crucially distinguishes
Experiential Non-Heracliteanism
41
both categories is not duration—states as well as processes are typically protracted—but the fact that they pick up on different ways of existing in time. As suggested by the notion of occurrent state, the previous categories are not mutually exclusive. Soteriou describes occurrent states in general as states the existence of which is underpinned by events or processes of a potentially wide variety of kinds (cf. Soteriou, 2011: 496; 2013: 104; Steward, 2018). More specifically, they are characterized in terms of two crucial features. (1) Their obtaining constitutively depends on the occurrence of events or processes. A liquid’s temperature may, for example, be understood as a state in which that substance finds itself when and because it undergoes certain processes at a molecular level. This feature entails two things in turn. First, since the relation of constitution need not be reductive, the relevant states are not explained away in terms of their constituting processes. Occurrent states depend on, but do not collapse into, their constituting processes. Second, they are occurrent precisely because of their dependence on processes or otherwise episodic items. While it is unclear whether the relevant states literally inherit the occurrent character of their constituents, the aforementioned constitutive bond at least guarantees that such states are occurrent in an important even if derivative sense, they at least entail the occurrence of certain processes. (2) Occurrent states and their constituting elements could be nonhomogeneous or heterogeneous in the sense that they need not belong to one and the same kind. Crudely put, a temporally extended item of kind S is homogeneous if its temporal parts are of the exact same (natural) kind S (cf. Vendler, 1957; Mourelatos, 1993; Rothstein, 2004: 10ff.). Being red and knowing provide more or less uncontroversial examples of the relevant feature: if an apple is red between t1 and tx, it instantiates that color throughout that whole interval t1–tx; if Mary knows that p between t1 and tx, then she knows that p at every instant throughout that period of time.6 Likewise, certain psychological phenomena seem to depend on states, processes, and events that could in principle belong to a wide variety of kinds: my back pain does not rely on pain processes, but on muscular and nervous ones of different kinds; certain forms of perceptual monitoring (e.g., watching) seem to involve processive and nonprocessive events of different kinds (e.g., spotting, looking for, staring at) (cf. Vendler, 1957: 151, 158–159; Crowther, 2009b). Against the previous backdrop, Soteriou’s version of EH characterizes perceptual experiences as temporally extended states fulfilling the following conditions: (i) they are occurrent insofar as they are constitutively—albeit not reductively—dependent on occurrent items, such as processes or events and (ii) they are heterogeneous insofar as neither they nor their constitutive elements need to coincide as to the (natural) kind they belong to. It is more or less clear that this stance displays the dynamic aspect of experience by means of those processes at the heart of experiential states. To understand how this
42
Chapter 3
ontological proposal fleshes out the Heraclitean metaphor, it is thereby necessary to expand on the notion of process at stake. For present purposes, three ways of understanding that notion are specially relevant: (P1) Processes are temporally extended, continuous changes. (P2) Processes are temporally extended, continuous changes that may be described in terms of successive series of states. (P3) Processes are temporally extended, continuous changes which cannot be described in terms of successive series of states. As its phrasing suggests, P1 captures a relatively broad notion of process. It is theoretically neutral vis-à-vis the mutually exclusive P2 and P3, in the sense that it does not impose special conditions on continuous changes. P2 establishes a link between continuous changes and states: conceived as continuous changes, processes may be described by picking up on successions of states obtaining in the subjects of change. The process of walking from point A to point B may, for instance, be described in terms of a succession of spatiotemporal states which obtain in a subject between the instant she stands on A and the instant she stands on B. That said, it is also important to appreciate that the link so established is purely descriptive: P2 does not advance an explanatory—reductive or otherwise—claim; as such, it neither entails that processes could be exhaustively described in terms of a finite series of successive states nor deny that they could be explained in terms of, say, a process ontology. The thought is simply that one could describe different parts of any given process in terms of states—that is, instantiations of properties or relations—obtaining at instants or intervals within the period of time throughout which the relevant process takes place. P3 in turn denies this descriptive possibility: that is, it rejects the possibility of describing processes in terms of nondynamic components (e.g., objects, properties, and relations). Since this denial reflects an understanding of processes as fundamentally dynamic or dynamic down to the core, it seems fair to say that P3 captures the most radical reading of the Heraclitean metaphor. In spite of expanding on the relevant notion at a purely descriptive—as opposed to explanatory—level, these two understandings of processes have different implications. Framed as a positive claim about how processes could be described, P2 does not engage in any theoretical commitment as to how they should be accounted for. In virtue of its negative edge, meanwhile, P3 has more profound consequences; for, by specifying how processes could not be described, this proposal rejects the very intelligibility of thinking about instances of one ontological category in terms of other ontological categories, and hence, presumably, the coherency of an explanatory framework along similar lines.
Experiential Non-Heracliteanism
43
With P1–P3 in place, let us turn to the question what notion of the process is involved in perceptual experience. To the extent that experiential phenomena are dynamic, EH would no doubt regard experiential processes as continuous changes. More importantly, it specifies the relevant processive items in terms of P3: that is, this ontological proposal conceives perceptual experiences as psychological states which constitutively depend on phenomenally conscious processes, where the latter processes may not be described in terms of nondynamic components such as objects, properties and relations. In other words, EH takes the processes at the heart of perceptual experiences to be fundamentally dynamic or dynamic down to the core. In O’Shaughnessy’s extreme version of EH, this understanding of experiential processes is fully manifest (cf. O’Shaughnessy, 2000: 42, 48, 61, 62, 65). Although less explicit than O’Shaughnessy, Soteriou endorses a similar stance insofar as a major goal throughout his work consists in arguing against a representationalist view of perception on the grounds of its alleged commitment to an ontology of processes along the lines of P2 (cf. Soteriou, 2013: 37ff.). One point that I have strived to highlight throughout these chapters is that EH—even Soteriou’s more nuanced version—has a significant ontological cost, it introduces processive items of a special kind into our ontological outlook. Indeed, processes—changes, in general—are a more or less obvious dimension of the natural world. That said, they are also typically conceived in terms compatible with P2: by describing processes in terms of more apprehensively obvious even if not explanatorily fundamental items (e.g., objects, properties, and relations), we typically manage to hint at ways in which processes fit into the ordinary medium-sized world or the more technical microscropic/macroscopic domains of scientific disciplines. Even a radical Heracliteanist like O’Shaughnessy concedes that physical as well as psychological but nonexperiential processes may be specified—even constitutively explained—in terms of successive transitions among states (cf. O’Shaughnessy, 2000: 44–47). Building on P3, however, EH suggests that experiential processes could not be described but in terms of further processive or otherwise dynamic items. That understanding of processes no doubt underscores the distinctive dynamic dimension of conscious experience, but it also comes at a not insignificant cost: it is not supported by a general commitment to the existence of processes, for P3 is not coextensive with the more ordinary processive notions reflected in P1 and P2; and, since it thus stands for a special category of items, it seems to obscure, at least at a descriptive level, what place (if any) experiential processes have in the natural—whether physical or psychological—world. In short, Soteriou’s phenomenally conscious processes are no ordinary processes: in virtue of their necessarily dynamic status, they seem to introduce a sui generis set of entities into our more familiar descriptive ontology. The
44
Chapter 3
crux of Soteriou’s ontological proposal lies in the fact that the phenomenally conscious processes at the heart of experiential occurrent states are not of any common kind. Such processive items are what commit him to a strong understanding of the Heraclitean metaphor: while Experiential Heracliteanists like him and O’Shaughnessy do not seem to take issue with the thought that physical and psychological, but unconscious processes are not only describable but also accounted for, in terms of continuous transitions among states; they invoke phenomenally conscious processes that can only be described, and hence, explained, in terms of further phenomenally conscious processes in order to accommodate the dynamic character of perceptual experiences. Soteriou’s experiential processes are like no other standard form of (physical or psychological) processes we know of because of their necessarily dynamic character. In a very literal sense, they fit the metaphor of Heraclitus’ river: they flow through and through. The foregoing remarks on the notion of process also aim to suggest a more positive point: EH is not the only form of fleshing out the Heraclitean metaphor. In the next section, I shall describe a proposal very similar to Soteriou’s stance, except for the fact that it relies on P2, rather than P3. Thanks to this small but significant modification, an ontology of perception could model perceptual experiences as dynamic phenomena, and also hint at potential ways of spelling out what place such phenomena have in the natural world. To reinforce this proposal, the subsequent section will then address a potential objection that Soteriou advances against an ontology of perception that does away with experiential processes along the lines of P3. A WORLD OF NON-HERACLITEAN EXPERIENCES The non-Heraclitean proposal I unpack in this section specifies Soteriou’s notion of occurrent state in terms of P2 rather than P3; in particular, it models perceptual experiences as states constitutively dependent on processes of a neurobiological or otherwise subpersonal kind. As it will be presently explained, this understanding of perceptual experiences not only avoids relying on the potentially controversial notion of process expressed in P3; it also gestures at how such experiential phenomena could fit into our natural worldview. To sketch ENH for short—let us draw some inspiration from the Vendlerian example I introduced back in chapter 1: What happens when we perceive, and what is it that makes it happen? [ . . . ] A sailor on deck looking ahead remarks, “It is pitch dark, I don’t see anything.” After a while, “Now I see a star.” We ask him, “What has happened?” “The
Experiential Non-Heracliteanism
45
cloud’s gone.” “But what else happened?” “Nothing else.” Of course things happened in the world and in the sailor. But his seeing is not one of them. (Vendler, 1957: 160)
As already explained, Vendler outlines a state-based or stative account of seeing (cf. Vendler, 1957: 154–156). By seeing, he does not mean in this context the instantaneous or near-instantaneous event by means of which a subject goes from not seeing to seeing—as he calls it, seeing in a “spotting” sense—but a temporally extended state by means of which a subject is capable of keeping visual contact with an object in particular or her surroundings in general. An example of seeing in this latter sense is a scenario where a sailor sees a star for a given interval of time, t1–tx. To shed light on the temporal or otherwise ontological nature of seeing, Vendler explicitly raises the question what happens to the subject when seeing the starry object throughout t1–tx. For sure, several extracranial and intracranial happenings or processes take place: physical processes of light transmission across space and time, neurobiological and otherwise subpersonal processes triggered by retinal stimulation, and so forth. The description of this perceptual scenario is very crude indeed, but it does highlight the processive character of visual phenomena in general, no matter how simple they might be. In spite of conceding this much, Vendler does not regard seeing as one of the aforementioned happenings: it may be constituted by the sum of such processes, but we should not commit the category mistake of counting it among the members of that sum. His commitment to a stative view seems to suggest that physical and psychological processes contribute to constitute instances of informational relations between a perceiver and her surroundings: as such, seeing would be a more or less temporally extended mental state that obtains in virtue of the instantiation of such informational transactions, where the latter relations are in turn constituted by the occurrence of the aforementioned physical and psychological processes. Like other post-Rylean philosophers at the time, Vendler is to a good extent driven by an attempt to undermine a traditional philosophical tendency to depict seeing as a special or otherwise mysterious process, and, hence, to resuscitate the ghosts of Cartesian epistemology (cf. Vendler, 1957: 159–160). As such, by classifying seeing as a state rather than as a process—or, to be more precise, as an activity conceived as a processive subcategory—he thus aims to show that we should not look for visual phenomena where they should not be looked for. While Vendler’s proposal is by no means the only historical precedent for ENH, it is my main source of inspiration here insofar as it will inform an understanding of perceptual experience that shapes occurrent states in terms of P2 and a plausible ontological line of motivation on its behalf. I expand on these two points next.7
46
Chapter 3
As previously mentioned, Vendler outlines a stative account of seeing, but his remarks may be tailored so as to hint at a stative view of visual experience. After all, at least for an intuitive and focal understanding of seeing, both phenomena are deeply intertwined, when a subject S sees a worldly object o from t1 to tx, she is consciously aware—that is, undergoes a visual experience—of o for the same or nearly the same amount of time. Like Vendler, one could raise the question what happens when S visually experiences o. Once again, a number of physical and psychological but nonexperiential processes occur, and while this experiential phenomenon is in a sense a happening or an occurrence, it need not be one of such constituting processes. Instead, it may be an overarching state that is occurrent insofar as it constitutively depends on the existence of the previous processes. To relate now Soteriou’s terminology with the notion of process reflected in P2, S’s experience may be described as a temporally extended state that instantiates an informational relation between S and o: such a state is in turn constituted by processes of potentially different kinds, and the latter processes may again be described in terms of the succession or the maintenance of states conceived as instantiations of properties or relations. When consciously aware of a star, a sailor establishes an (accurate or inaccurate) informational relation with that object; that is, her temporally extended visual experiences could be conceived as temporally extended states, as the instantiation and subsequent maintenance of an informational relation between the relevant subject and the apparently perceived surroundings. Although ENH thus incorporates an important stative element, it does not deny that perceptual experiences are dynamic, for it takes the relevant experiential states to be made of processive items. At the same time, since occurrent states could be heterogeneous, such constituting processes need not be conceived along the lines of P3. Thanks to P2, they may be described in terms of more familiar nondynamic items (e.g., objects, properties, and relations). That being the case, it seems natural to describe them as the physical and subpersonal processes unveiled by cognitive psychology and neurobiology. At this level of descriptive generality, ENH manages to place experiential states within the natural world by relating them to elements within this domain. As already anticipated, the previous thought points at a key motivation driving the present version of ENH. Influenced by Ryle’s remarks on category mistakes and the myth of the “ghost in the machine” (Ryle, 1949: ch. 1), Vendler claims that a stative account of seeing seeks to avoid the ghosts of Cartesian epistemology. What is he thus trying to avoid? Most likely, I believe, philosophical views that obscure our understanding of a given psychological feature by adopting a noncompulsory conception of the relevant explanandum. In particular, a Cartesian metaphysics accounts for our psychological lives in terms of a paramechanical realm of objects and
Experiential Non-Heracliteanism
47
events, so as to accommodate epistemic facts about our introspective access to our own selves. By taking this step, a Cartesian outlook unleashes the recalcitrant question how such para-mechanical objects and events fit into the natural world we know and love. Against this problematic backdrop, Vendler suggests that we need not look for episodes of seeing among the events and processes bridging the informational gap between perceivable items and perceivers; instead, seeing may be conceived as a temporally extended state constituted by the previous chain of physical and subpersonal events and processes. Not unlike this approach, the present version of ENH is to a good extent fueled by a similar ontological worry. For, while not amounting to a Cartesian view of experience, EH does seem committed to a noncompulsory conception of experiential phenomena, and, as a result of this, it potentially obscures our understanding of how perceptual experiences exist in the natural world. So, although not Cartesian in spirit, a Heraclitean ontology of perception would risk describing perceptual phenomena in a way that led us to look for them where they need not—and perhaps should not—be looked for. The worry is then that a notion of experiential process modeled along the lines of P3 would fail descriptively to relate perceptual experiences to its underlying physical and subpersonal basis. And since this framework is shown not to be compulsory by ENH’s alternative depiction of occurrent experiential states in terms of P2, one could thereby choose to avoid that view altogether in favor of its non-Heraclitean counterpart—that is, in favor of a framework that acknowledges the dynamic character of experience and, at the same time, anchors the processes constitutive of experiential states to our familiar physical and psychological world. The previous line of motivation may in turn be framed against a broader ontological landscape. Both Vendler’s stative account of seeing and ontological proposals like EH and ENH discharge a tacit albeit widespread causal assumption in modern discussions of the mind-body problem, that is, the general question how our minds or specific aspects thereof exist in the physical world. Mental phenomena are the kind of items that may causally produce or be causally produced by physical phenomena or other mental phenomena. Since causal relations are typically conceived as events, the previous idea naturally translates into the thought that the physical and the mental are constituted by physical and mental events, respectively, and that instances of both kinds of events causally interact with each other. This is the sort of idea that Helen Steward seems to identify by means of what she calls a Network Model of psycho-physical causation; that is, the implicit view among philosophers of mind that reality as a whole is causally interconnected and that, as such, the ontological status of mental events depends on whether such events occupy a place within that complex causal network (cf. Steward, 1997: ch. 7). As far as I can see, the Network Model perhaps derives from the metaphysical thesis
48
Chapter 3
that the identity of a given type of items hangs on whether instances of that kind make a causal difference in the world (cf. Shoemaker, 2003). But in addition to this, one may also think that the assumption is vindicated by our everyday intuitions: it is simply uncontroversial that the physical world has causal effects on our mental lives, and vice versa. When I face a desk and under normal perceptual conditions, I shall naturally undergo a visual experience of the desk. When I touch a piping hot coffee maker, I shall most certainly experience pain. Conversely, my decision to lose some weight will make me take an orange instead of the pudding at the cafeteria. These considerations would only confirm the Network Model of psycho-physical causation. Aiming to answer these questions, philosophical accounts of the mind typically end up leading to either one of two stances: the causal over-determination or the causal inefficacy of the mental (cf. Kim, 2000). The ontological views unpacked throughout this work transcend the network model insofar as they raise the possibility of bridging the gap between the physical and the mental in constitutive rather than causally productive terms. They do not a priori assume that perceptual phenomena are events causally producing or being causally produced by other (mental or physical) events, instead, they take experiential states to be constituted by physical or subpersonal states and processes. Perceptual experiences would not thereby be billiard balls in the complex table of reality, but conditions in virtue of which physical items could play that game. Or again, to use some Aristotelian terminology loosely, our understanding of the relation between the mental and the physical need not resort to the notion of efficient causality, but to that of formal causality. Thus conceived, the ontological debate about perception does not turn so much on what causally productive place perceptual experiences occupy in the natural world, as on what ingredients constitute such psychological states. EH and ENH take different stances vis-à-vis the latter question. For sure, the notion of constitution or constitutive dependence is by no means straightforward or unproblematic, but it does provide a stimulating and potentially promising conceptual framework for revisiting the otherwise stagnant mind-body problem in its more specific guises. The bearing of the present ontological proposals on our causal understanding of perceptual phenomena is explored further in chapter 4. To sum up, this section aimed to outline a version of ENH that draws inspiration from Soteriou’s work as well as that of Vendler’s. More specifically, I have sought to show that the notion of occurrent experiential state need not resort to processes like those hinted at in P3. ENH is, I believe, a more conceptually conservative way of fleshing out the Heraclitean metaphor in the realm of experience, and while my present sketch does not constitute a fullfledged defence of this non-Heraclitean proposal, I do hope it highlighted the fact that Soteriou’s version of EH is neither trivial nor a priori compulsory.
Experiential Non-Heracliteanism
49
NON-HERACLITEAN EXPERIENCES: STATIVE OR STATIC? Soteriou espouses EH for its alleged ability to accommodate the phenomenology of temporal passage. Although I cannot assess this claim within the boundaries of this work, it is necessary to address one of its underpinning assumptions, namely, the negative thought that a non-Heraclitean view misrepresents the perception of change by describing protracted experiential states in terms of ordered successions of static representations—in short, the thought that, when conceived along non-Heraclitean lines, experiences turn out to be static phenomena.8 Soteriou’s critique directly bears on the plausibility of a view like the present version of ENH, for it targets any account of perception that does away with P3 on behalf of P2. In a nutshell, that line of criticism seems inconclusive to me for three reasons: first, what Soteriou actually ends up targeting is a representationalist view of experience, not a non-Heraclitean stance like ENH; second, even if ENH was somehow bound to perceptual representationalism, it is not necessarily committed to a static conception of perceptual representation; finally, Soteriou’s overall objection assumes that experiential states could answer questions of cardinality, a point I challenge here. By thus shedding light on the implications of a non-Heraclitean ontology, I hope to reinforce the previous sketch of ENH. In what exact sense are experiential occurrent states, conceived in terms of P2, supposed to be static? A static understanding of states typically concerns the previously mentioned feature of homogeneity, where a temporally extended item of kind S is homogeneous if its temporal parts are of the exact same (natural) kind S. If homogeneous, states would be static by lacking the sort of internal, complex structure that dynamic phenomena typically enjoy— for example, progressively unfolding stages in the case of processes (cf. Rothstein, 2004: 20). That said, it is relatively obvious that occurrent states are not static in this sense, for, as previously noted, one of their defining traits is precisely their nonhomogeneity: neither they nor their constituting parts have to belong to the same (natural) kind. Soteriou thinks that a non-Heraclitean conception of experience is static insofar as it fails to accommodate the perception of change over time, such states fail to capture how things dynamically appear to their respective subjects because, albeit nonhomogeneous, they are the sort of temporal items that obtain or continue obtaining over time. To unpack the objection, let us focus on an extremely crude case where a subject perceives dynamic phenomena, a subject S experiences a banana from t1 to t5, and then an apple from t6 to t10. According to Soteriou, ENH would provide only two ways of describing this basic scenario, either (i) S instantiates an ordered succession of discrete, instantaneous, or quasi-instantaneous experiential states from t1 to t10, or (ii)
50
Chapter 3
she instantiates a single, protracted experiential state for the same period of time. Both options, he thinks, fail to describe the perception of protracted change. In regard to the first possibility, Soteriou writes: [ . . . ] if the psychological states of the subject are changing in this way during the interval t1-tn—if what seems to her to be the case changes during each subinterval of t1-tn—then it looks as though we fail to capture the fact that over that interval of time t1-tn she has a psychological property that accounts for [ . . . ] its seeming to her as though she is aware of an occurrence with the temporal extension of O [that is, the perceived phenomenon spanning the interval t1-tn]. (Soteriou, 2013: 95)
A “snapshot” version of ENH—that is, a view according to which a temporally extended perceptual state is the sum of instantaneous or near-instantaneous perceptual states—would break down because a bundle of instantaneous or quasi-instantaneous experiential states could not on their own coalesce into an experience of continuous, temporally extended phenomena. Just as an ordered succession of snapshots would not necessarily constitute a single sequence of moving images, an ordered succession of experiential states does not have to constitute an experiential state of successive events. Hence, the instantiation of several momentary experiential states as of a banana from t1 to t5 and of several momentary experiential states as of an apple from t6 to t10, would not on their own accommodate the idea of S’s being aware of a banana followed by an apple from t1 to t10. To assess (ii), Soteriou examines a “one-experience” view along the lines of Michael Tye’s take on diachronic experience (cf. Tye, 2003: ch. 4). The latter proposal is prompted by a worry concerning experiential individuation: if protracted experiences are constituted by mereological sums of shorter experiences, how should we go about counting how many experiences there are within temporally extended ones? There is apparently no principled way of deciding the matter. To settle the problem, Tye advances the following principle of individuation: for each uninterrupted stream of consciousness, there is a single protracted experience. This hypothesis would allegedly be the most elegant way of dodging the aforementioned cardinality question. According to Soteriou, combining this proposal with a non-Heraclitean ontology of experience results in the view that perceivers have a single mental state with a content p throughout a whole uninterrupted stream of consciousness—say, from the moment they wake up to the instant they fall asleep. But then, a view along these lines would be problematic for the following reason: the type-individuation of a protracted state by means of a propositional content p entails ascribing to its respective subject a property concerning how
Experiential Non-Heracliteanism
51
things appear to her—crudely put, the property of it perceptually seeming to her that p (cf. Soteriou, 2013: 110, 149); but such an ascription misdescribes the way we perceive change, for it statically fixes the content of the relevant uninterrupted stream of consciousness. Soteriou formulates the previous line of reasoning as follows: Since the psychological state of the subject—its seeming to her as though she is aware of an occurrence O—is not something that unfolds over that interval of time, one might think that that perceptual state continues to obtain throughout that interval of time. In which case, we do not then capture the idea that what seemed to the subject to be the case during sub-intervals of that interval of time was different. So it looks as though we do not capture the idea that during the sub-interval t1-t2 it seemed to S as though she was merely aware of a temporal part of O, during the sub-interval t2-t3 it seemed to S as though she was aware of a different temporal part of O, and so on. (Soteriou, 2013: 94)
One could perhaps expand or highlight this line of reasoning by noting that experiential states would involve contents scoping well beyond a subject’s experiential horizon.9 For example, since S has a mental state individuated by a content including a banana followed by an apple from t1 to t10, she would stand in a banana-followed-by-an-apple state all the way from t1 to t10—that is, even when, depending on the moment of time, a banana or an apple is not presented to her. The content of that single, protracted state would present S with information about worldly items that lie in her future as well as in her past: from t1 to t5, a mental state about a banana and an apple would obtain in S even though she has not come across an apple yet; and then, from t6 onward, a state with the same content would obtain even though the banana has already left S’s sensory field. If, as it is often assumed in the philosophical literature, content-ascription at a time tx to a subject S determines how things appear to S at tx, the previous outcome is no doubt unacceptable; we are not normally aware of what lies in the past or what awaits for us in the future, but only of what exists now or within the so-called “specious” present. Hence, by statically fixing the contents of protracted experiences, a one-experience version of ENH would also fail to accommodate the distinctive dynamic way in which our perceptual experiences present the world to us across time. As previously anticipated, the previous line of criticism seems inconclusive for at least three reasons. First, it actually targets a form of perceptual representationalism, not a non-Heraclitean view of experiential states. In a nutshell, Soteriou’s two-pronged attack is that a sense of succession or protracted change is not accommodated either by a succession of representational states or by a single protracted state representing temporally extended succession. All he has done up to this point is to spell out a plausible objection as to how
52
Chapter 3
different forms of representation could account for a representation of temporally extended change. But while Soteriou goes to some lengths to show that perceptual representationalism is bound to an ontology of experiential states (cf. Soteriou, 2013: 38ff.), he does nothing to show that a view along the lines of ENH has to be representationalist. As far as I can see, a non-Heraclitean view does not entail representationalism. On the contrary, it seems natural to articulate this ontological proposal along relationalist lines; after all, a fairly straightforward way of fleshing out the thought that experience or its phenomenal character is relational consists in describing experiences in terms of states that result from the instantiation of informational relations between a subject and her surroundings. So, in spite of its potential merit as an objection against perceptual representationalism, Soteriou’s critique does not obviously target an ontology of experiential states that does away with P3. In order to do so, it still has to show that perceptual representationalism is a necessary ingredient of a non-Heraclitean ontology of experience.10 Second, the variety of perceptual representationalism targeted by Soteriou does not seem compulsory; that being the case, even a representationalist version of ENH would still be plausible. Soteriou’s line of reasoning incorporates a widespread assumption about the relationship between contentascription and perceptual appearances: the ascription of an experience with content p to a given subject S over an interval t1–tx or a subinterval thereof codifies how things appear to S during the relevant interval or subinterval (cf. Soteriou, 2013: 40).11 This chapter cannot divert into a discussion of perceptual content, whereby I shall only voice, not argue for, the plausible claim to the effect that perceptual representationalism need not commit to that assumption. Indeed, the notion of perceptual content is a philosophical device intended to codify personal-level information about what and how a subject experiences the world; as long as it fulfills this general role, the way it relates to the synchronic and diachronic way in which things appear to a subject could in principle take a wide variety of forms. The content p of a temporally extended experience E could, for instance, capture how things appear to S over the whole interval of time in which E takes place without, at the same time, fixing how things appear at any particular instant within that interval. This is, in fact, part of what Tye’s one-experience account is trying to get at (cf. Tye, 2003: 97). If sound, these remarks suggest that even a representationalist version of ENH could avoid Soteriou’s criticism. Finally, the very fact that the relevant objection is posed in terms of two allegedly implausible options—namely, (i) and (ii)—suggests that Soteriou neglects an important point about stative individuation: states are not the kind of ontological items that answer cardinality questions the same way particulars (e.g., substances and perhaps certain kinds of events) do so. Consider standard samples of property- and relation-instantiations: it would, for
Experiential Non-Heracliteanism
53
example, be odd to ask whether an apple goes through one or several instantiations of red throughout the interval of time it is ripe; or again, it does not quite make sense to ask how many instances of believing that p take place in me over the period of time I hold the belief in that p. Likewise, if perceptual experiences are conceived as states, we could not answer cardinality questions about experience for the simple reason that such questions would not arise at all. Sure, there is a broad sense in which I undergo one experience over an uninterrupted period of conscious awareness—that is, for the mere fact that my stream of consciousness has to have either definite or ambiguous temporal boundaries. But this should not lead us to think that experiential states could answer the same kind of “how many”-questions that we pose for spatial and temporal particulars. A ripe apple instantiates a particular hue of red for a given interval of time; sure enough, to the extent that such an instantiation has a beginning and an end, we could (rather awkwardly) claim that one such instantiation takes place in the apple. Of course, none of this would mean that one—as opposed to several—states of that particular hue obtain in the apple. Likewise, we could perhaps concede that one experiential state obtains in S over an experientially uninterrupted interval of time. However, this does not mean that one experiential state obtains in S in a similar way in which there is one laptop—as opposed to two or three—in this room. In short, it seems categorially mistaken to frame experiential states as countable items, that is, as items that answer cardinality questions. Soteriou’s critique thus seems controversial insofar as it is spelled out in terms that seem to assume otherwise. To wrap things up, the present section critically assessed an objection that Soteriou could potentially direct against ENH. In a nutshell, the thought is that non-Heraclitean occurrent states of perceptual awareness fail to capture our perception of change. In reply, I believe that the objection is not conclusive for three reasons: first, it actually targets perceptual representationalism, not ENH; second, it seems to rest on a very particular way of understanding the relationship between perceptual appearance and content that we need not endorse—as such, even a representationalist version of ENH could perhaps dodge it; and, finally, the objection crucially assumes that experiential states could answer cardinality questions, quod non. The foregoing remarks have thus strived to provide an initial sketch of ENH as well as to dispel an initial worry that it could face. CONCLUSION Recent philosophical and psychological work on perception is slowly steering away from debates exclusively concerned with the intentional or
54
Chapter 3
phenomenological properties as well as the cognitive function of perceptual experience, so as to address the fundamental ontological question what perceptual experiences are. On the plausible assumption that perceptual awareness exists in time and thereby involves changes in our psychological lives, the temporal and dynamic structure of experience may crucially influence how we ontologically categorize perceptual experiences. Against this conceptual background, the Heraclitean metaphor of a flowing river is no doubt suggestive when it comes to characterize the temporal or otherwise dynamic structure of experience. How should we specifically understand this metaphor when applied to the experiential domain? Very carefully, I believe. For, while experience is intuitively dynamic, there is plenty of room for debating how that thought could be cashed out. This work has focused on two ontological proposals: on the one hand, EH, a view according to which perceptual experiences depend on at least some necessarily dynamic elements; and, on the other, ENH, a stance that describes the dynamic character of perceptual experience in terms of nondynamic elements (e.g., states, properties, and relations). Building on a critical assessment of Matt Soteriou’s recent defense of EH, this chapter aimed to outline and only modestly defend a stative version of ENH. Although both ontological views conceive experiences as mental states constitutively dependent on processes of different kinds, they radically differ on the precise nature of such constituents. Soteriou endorses a form of EH insofar as he conceives the relevant processes along the lines of P3—that is, as irreducibly dynamic or as processive down to instants. My suggestion is that an ontology of perceptual experiences need not rely on the controversial category of necessarily dynamic processes; instead, experiences may be described in terms of P1 and P2—that is, as the instantiation and subsequent maintenance of informational relations between a subject and her surroundings, states that in turn constitutively depend on processes of a neurobiological or otherwise subpersonal kind.
NOTES 1. Claiming that we can be aware of instants or subintervals of time, may be read as a short-hand way of saying that we can be aware of the worldly items at such instants or subintervals. But Soteriou actually means to claim that we are aware of instants and subintervals, at least in the sense that we experience more or less protracted parts of a temporal sensory field. For present purposes, I shall remain neutral on the questions whether conscious subjects are endowed with temporal sensory fields, and, if so, how such fields should precisely be characterized. 2. Cf. Soteriou, 2013: 138–139; Fine, 2005: 287. For critical assessment of this condition, though, cf. Hoerl, 2013.
Experiential Non-Heracliteanism
55
3. The following remarks draw inspiration from a subtle yet important move made by Soteriou: when it comes to outline an Heraclitean ontology of perceptual experience, he turns to an example related to the perception of static scenes. As far as I can see, this kind of case is to a good extent chosen for strategic purposes: it avoids conflating the present ontological proposal with a structurally similar view of event perception that Soteriou previously espoused (cf. Soteriou, 2013: ch. 4 as well as pp. 139–140 for the latter stance). But, as I explain in the text, his choice of example also suggests that the temporal structure of the world we experience need not exhaustively specify our sense of passage. 4. Cf. O’Shaughnessy, 1971a, 1971b, 2000; Steward, 1997, 2011; Soteriou, 2007, 2011, 2013, 2018; Crowther, 2009a, 2009b, 2010, 2011; Lee, 2014; Di Paolo, Buhrmann, & Barandarian, 2017. 5. For a notion of states along these lines, cf. Kim, 1976; Armstrong, 1997: 20; Thau, 2002: 60–61). 6. The notion of homogeneity has a structurally similar application in the spatial domain too: gold and water provide standard examples of homogeneous substances, in the sense that, at an intuitive level of description, their constituent parts are further bits of gold and water, respectively (cf. Crowther, 2011). 7. Briefly to mention two other echoing views. At the heart of D.M. Armstrong’s doxastic or belief account of perception lies the claim that “to say that A perceives that p is to say that A comes to be in a certain state, a state which can only be described in terms of its possible manifestations” (Armstrong, 1968: 245–246). I set this view aside, however, for potentially unappealing intellectualist connotations which I have nevertheless discussed in chapter 1. Daniel Dennett, in turn, writes that “[c]onscious experience [ . . . ] is a succession of states constituted by various processes occurring in the brain, and not something over and above these processes that is caused by them” (Dennett, 1998: 136, but also cf. p. 135). While nominally describing a version of ENH very similar to the one I outline here, Dennett does not want to have anything to do with serious ontological distinctions like the one between states and processes, and most likely takes the relation of constitutive dependence to be a reductive one. As such, his claim really boils down to the idea that experiences are nothing over and above cerebral activity. By contrast, I sketch here a more ontologically robust reading of that statement: that is, one that takes the ontological import of categories like those of process, state, and (nonreductive) constitution, seriously; one that potentially delivers a general framework for understanding how perceptual experiences nonreductively depend on phenomena of our physical world. 8. In fact, I am tempted to think that this negative thought makes a greater contribution to Soteriou’s overall defense of EH than his positive reasoning from phenomenology. I can only voice this claim here, though arguing for it would require a lengthier discussion than the one this chapter could afford. 9. I am indebted to Soteriou for stressing this thought in conversation. 10. To be fair, Soteriou’s attempt to bind a non-Heraclitean ontology and representationalism together is not unmotivated. Driven by the task of exploring how ontological categories like those of process and state bear on traditional debates about the intentional and the phenomenal character of experience, Soteriou strives to show how mainstream discussions about representationalism and relationalism overlap with
56
Chapter 3
more exotic questions about the temporal or otherwise ontological structure of experience. Against this backdrop, he tends to pair ENH with representationalist views, on the one hand, and, on the other, EH with relationalist ones. Although his separate discussions of the intentional/phenomenological and the ontological structure of experience are powerful and thought-provoking, the cross-over he then attempts to drive between both discussions seems to be the chink in his project’s armor. For the time being, my main point is that Soteriou provides no reason for accepting that ENH has to be fleshed out along representationalist lines, a claim that is, I believe, essential to his criticism. 11. As previously suggested, this assumption is popular yet tacit. A relatively more explicit statement of it may perhaps be found in Charles Travis’ critical remarks on the looks-indexation of perceptual content (cf. Travis, 2004).
Chapter 4
The Individuation of Experiences
The present work addresses the ontological question what we talk about when we talk about perceptual experiences. The ontological debate unpacked throughout the previous chapters trades on how we should specify two broad understandings of perceptual experiences. As I previously explained, it would be simplistic to frame both approaches in terms of processes and states conceived as exclusive categories. For the purposes of this chapter, however, I shall reflect on the following broad claims: (P) Perceptual experiences are mental processes. (S) Perceptual experiences are mental states. One of my main goals has been to make a case for S in its non-Heraclitean specification. But I now turn to a discussion of the conditions governing the existence of perceptual experiences—for short, their identity conditions. And I shall fall back on P and S because those conditions crucially concern the distinctive—albeit not exclusive—ways in which we think about the identity conditions of processes and states. It seems natural that both proposals will provide different accounts of temporally extended perceptual experiences, for processes and states precisely vary in regard to diachronic existence: whereas processes occur or take time, states obtain or continue to obtain. A bit more specifically, the goal of this chapter is twofold: on the one hand, I challenge P’s ability to provide a story of diachronic perceptual experiences; and, on the other, I argue that one of S’s virtues precisely consists in delivering such an account. The key problem with a processive account is that it provides no guide for individuating or “counting” the temporal particulars it ultimately posits. By contrast, since mental states are not (either spatial or temporal) particulars, the stative view would preempt similar attempts to individuate or 57
58
Chapter 4
count perceptual experiences. If sound, this line of reasoning surely benefits Soteriou’s hybrid version of EH to some extent: but to the extent that the latter account also relies on Heraclitean processes, my remarks may still put some pressure on it. This chapter is divided into three main sections. First, I explain why a discussion of experiential individuation should focus on temporally extended experiences and why views like P and S have a bearing on this issue. Second, I argue that P poses the question how perceptual experiences are individuated over time, but fails to provide the necessary conceptual resources to settle it. Finally, I show that S avoids that difficulty insofar as it does not raise the same question. EXPERIENTIAL CONTENT AND EXPERIENTIAL VEHICLE To defend S, this chapter turns to the identity conditions of perceptual experiences over time. For this reason, I want to do two things in this preliminary section: first, to spell out why temporally extended experiences are so important in this context; and, second, to motivate S’s and P’s bearing on the individuation of perceptual experiences. To illustrate the forthcoming remarks, I shall once again rely on the example of a vigilant sailor’s visual experiences. Example 1 A sailor on deck, Jim, looks for a star during a cloudy night. At one point, the sky begins clearing up a bit, and our vigilant subject suddenly sees a bright star. Jim sees the star from t1 to tn, over which time there are no interruptions or conspicuous changes in his visual field, and the relevant star looks or appears a determinate way, w, to him. But now I shall also expand on this example along the following lines: Example 2 Jim sees the same bright star from t1 to tn, but now another sailor, Jack, joins him, stands right next to Jim, and sees the same star. The star looks the same determinate way, w, to both of them. Alternatively, one could try a transworld comparison. Imagine a possible world, W2, exactly like ours, W1, where Jack, not Jim, does exactly the same thing that Jim does in W1; in this case, it is natural to suppose that things appear exactly the same way, w, to Jim and Jack, each one inhabiting different possible worlds.
The Individuation of Experiences
59
These cases highlight a number of points. To begin with, both examples remind us of the fact that perceptual experiences primarily play a role within a story of perception or perceptual acts. Thus, when Jim and Jack see a bright star (“seeing” referring here to a temporally protracted experiential occurrence starting at t1), the notion of visual experience is intended to capture a crucial part of their psychological lives. When a subject S perceives a certain worldly item X, and X appears a certain way to S, S is perceptually aware or has a perceptual experience of X. There is an intimate conceptual and ontological relationship between experiences and their subjects; as noted by A.J. Ayer, “[i]n the ordinary way, we identify experiences in terms of the persons whose experiences they are” (Ayer, 1963: 84; also cf. Peacocke, 2012). Again, the notion of perceptual appearance is a significant component in a description of a perceiver’s experiential life: what perceptual experiences a subject undergoes depend not only on what informational channels are established between a subject and her surroundings via her sensory systems, but also on how the relevant perceptual information is conveyed. By disregarding how things appear (i.e., look (like), taste (like), etc.) to perceivers at a time or over time, one would underdescribe their experiential lives. Jim and Jack will have different visual experiences if they see an item (say, a bright star) which looks different ways to both of them (say, like a bright star to Jim, and like a satellite to Jack). To have the same kind of perceptual experiences, subjects must be affected the same way by the items they perceive. Finally, the second example illustrates the possibility of conceiving different subjects standing in the same relations of perceptual awareness vis-à-vis their surroundings. Thus, I take it that Jack and Jim stand in a similar relation of visual awareness to the bright star: how things look to Jack is identical or very similar to how things look to Jim; Jim and Jack have visual experiences of the same type; or again, there is an intuitive sense of “seeing” in which Jack sees the same thing that Jim does. That said, I shall expand now on the importance of temporally extended experiences by stressing the ontological dependence of synchronous perceptual experiences (experiences at a time) on diachronic ones (experiences over time). Although one may uncontroversially distinguish both kinds of experiences, it is important to bear in mind that nonprotracted ones exist within a wider temporal context. Like the above examples, descriptions of perceptual experiences in general pick up on temporally protracted phenomena: for, while perceptual acts (i.e., achievements like spotting an object) get subjects in informational contact with worldly items at a time, perceptual experiences hold such informational transactions in existence; that is, perceptual experiences are temporally protracted phenomena because their role is precisely that of reflecting how a subject is perceptually related to her surroundings over an extended period of time. Perceptual experiences are not the kind of
60
Chapter 4
items that may exist for a mere instant of time; while Jim may, for example, see (as Vendler puts it, in the “spotting” sense) a bright star at an instant, he could not visually experience it during a mere instant of time. Indeed, it is possible to have very short-lived experiences of objects and properties; say, a subject may be aware of a flash for half or a quarter of a second. These are, however, not instantaneous experiences; no matter how short-lived they may be, they are temporally protracted. I am tempted to think that, whenever one refers to perceptual experiences at a time, one does not pick up on instantaneous experiential monads, but on a subject at a time at which she undergoes temporally protracted experiences. For example, although it is legitimate to ascribe a visual experience to Jim at t2, this amounts to saying that Jim visually experiences worldly items during a period of time t1–tx and that t2 lies within t1–tx. To generalize this point a bit further, one could convey the dependency of synchronous experiences on diachronic ones as follows: (D) A subject S has a perceptual experience of O at t2 only if S perceptually experiences O during t1–tx and t2 is an instant within t1–tx. This principle is compatible with the existence of statements about perceptual experiences at a time which explicitly fail to refer to a temporally protracted experience; after all, D is not a grammatical claim. The point is that, if D is true, the analysis of perceptual experiences at a time explicitly or implicitly relies on temporally extended perceptual experiences. This is the sense in which I take synchronous perceptual experiences to depend on diachronic ones. In spite of our differences, the primacy of temporally extended experiences is a point on which I coincide with O’Shaughnessy. He secures this stance by means of a phenomenological analysis where the relevant psychological phenomena capture a subject’s recent past as well as her present: thus, “[a] man staring fixedly at a chair is as directly aware of the perceptual object of a few seconds ago as is the perceiver of a movement across time. The individuation of the perception of any instant requires that it be so” (O’Shaughnessy, 2000: 60). On this view, perceptual experiences are not protracted phenomena developing out of instantaneous or discrete experiences; instead, they are fundamentally or constitutively extended in time. But, in addition to this phenomenological line of reasoning, it is also important to appreciate that O’Shaughnessy’s processive stance demands the ontological primacy of diachronic experiences. As noted in chapter 2, he holds that perceptual experiences are processive or occurrent to the core, in the sense that they can only be analyzed into process-stages—that is, processes of the same kind. Suppose now that temporally extended experiences should be conceived as bundles of instantaneous or discrete experiences, the latter could not
The Individuation of Experiences
61
be analyzed in processive terms, for processes (even short-lived ones) could not occur at single instants of time; hence, it would turn out that perceptual experiences are not processive to the core, at least in the aforementioned sense. To guarantee his processive analysis of perceptual experiences through and through, O’Shaughnessy has to subordinate synchronous experiences to diachronic ones, not the other way around; that is the only way in which he can plausibly claim that perceptual processes are always constituted by other processes of the same kind. I do think that the category of temporally extended experiences is more fundamental than that of instantaneous ones. That said, neither this ontological primacy nor D necessarily entail a processive stance. If perceptual experiences are conceived as mental states, ascribing them to a subject S at a time t also presupposes that t falls within a period of time during which S goes through a temporally extended experience. Indeed, the thought that a subject or an object could instantiate a state (say, a belief, a wish, etc.) for only an instant of time is extremely puzzling. In relation to perceptual phenomena, this feature of stative ascriptions could be accounted for by the fact that experiences are intended to set a subject in informational contact with her surrounding over a given period of time. Let us take stock. As previously stated, this work examines the question what we talk about when we talk about perceptual experiences—in other words, what kind of psychological items they are. That being the case, it should be more or less clear by now why the present discussion should say something about temporally extended experiences; they constitute central cases of the psychological category I am concerned with, and hence, a natural starting point for this project. Now, even if the foregoing remarks are along the right lines, it is unclear what precise bearing something like P or S could have on the present subject matter. This is the second preliminary point I address here. Among other things, a theory of perceptual experiences should specify how to identify or “count” perceptual experiences across time: say, if Jim visually experiences a bright star during t1–tn, such an account should be capable of determining how many visual experiences he goes though during that time; or, if the question is misguided, why it is so. I do not argue for this point, but simply take it for granted. What features of temporally extended perceptual experiences would, however, help us to individuate or count them? To address this question, I shall rely on a very general, and hence uncontroversial, distinction between content and vehicle of content; that is, I shall distinguish the things perceptual experiences are of or about from perceptual experiences themselves. Although the notion of the content may be understood in as many ways as the notion of intentionality, I expect it to be clear that the sense in which I talk about it here is not a heavy-duty one; whether
62
Chapter 4
ultimately analyzed in representational or relational terms, perceptual experiences are intuitively about or of things other than themselves. For example, there is a more or less obvious distinction between Jim’s visual experience and what that experience is about, that is, a bright star: the star is a physical object which existed light-years from Jim and probably died long ago; Jim’s visual experience is, in turn, a mental phenomenon which exists in him alone and shares none of the physical properties which a star might have—that is, it has no weight, size, degree of luminosity, and so on. Although a visual experience is no physical object, it can be about one. The content-vehicle distinction thus provides a framework to draw a number of similarities and differences: Jim and Jack go through the same kind of psychological phenomenon, namely, perceptual experiences as opposed to remembrances, beliefs, among others; in addition to this, such experiences are about the same thing, a bright star which looks the same determinate way to both subjects; again, these experiences stand apart from experiences endowed with different contents—either by concerning different physical items or by presenting the same ones in different ways; and so on. With this general distinction at hand, the previous question may be rephrased as follows: should the identification of perceptual experiences over time be accounted for in terms of experiential contents or experiential vehicles? In a nutshell, I think the content of perceptual experiences throws no lights on the issue at stake, to settle questions of individuation, it is necessary to address perceptual experiences themselves. This is the reason why S and P have a bearing on the present discussion; after all, they precisely intend to take a stance on the nature of experiential vehicles. Before diving into my positive position, let me pause on the negative point concerning experiential content. On the assumption that perceptual experiences may be analyzed into experiential content and experiential vehicles, the thought is that their diachronic identity could be fixed by the content which such psychological phenomena have: that is, to specify how many experiences a subject has over time, one only has to determine what and how worldly items are perceptually presented to her over the relevant temporal span. According to this view, what visual experiences Jim and Jack have depends on what items they are visually aware of (e.g., a star, clouds, and sea) and how such objects are presented to them, something which is in turn determined by a complex number of environmental, perspectival, and neurobiological facts. Could one thereby expect experiential content, in this broad sense of the term, to specify what experiences Jim and Jack have throughout a given period of time? I do not think so. As previously mentioned, perceptual experiences are understood here as relations of awareness between perceivers and their surroundings. If such experiences are temporally protracted, the relevant relations will also be
The Individuation of Experiences
63
indexed to periods of time. That being the case, a story of experiential individuation is accordingly bound to convey who the relevant perceiver is, what and how environmental items are presented to that subject, and how long the relevant experience lasts. Experiential content, however, precisely fails to capture at least two of those components; even if it captures what items a subject perceives and how they are presented to her, it is still silent on who the relevant subject is and how much time the relevant experience takes. For example, a description of a visual experience of the kind introduced in example 2 leaves open whether the relevant subject is Jim or Jack. Indeed, an ideal description of the relevant experiential content would incorporate facts about the perceived items and the specific mode of presentation, including perspectival facts determined by the physical location of the subject relative to the perceived scene. But even if such a description established that the relevant experience was had by some subject, it would not settle who the exact subject is. In other words, Jim and Jack could have visual experiences with the same content, where identity of content means that they have experiences of the same kind; they may have the same sort of experiences not in spite of the fact that experiential content is silent on the question which subject undergoes the relevant experience, but precisely because of that. Furthermore, the phenomenon of transparency suggests that experiential contents fail to fix the duration of its corresponding experiential vehicles. Derived from a few esoteric remarks made by G.E. Moore (cf. Moore, 1903), the general thought is that reflection on perceptual experiences is not sensitive to features of the experiences themselves, but only to features of the worldly items our perceptual experiences are of. For example, if I reflect on my current visual experiences, I cannot attend to the very mental phenomena opening the visible world to me, but only to the items I see, for example, a laptop, a table, a few DVDs,—in this sense, experiences themselves are “invisible” or “transparent” to reflection. A bit more specifically, experiential transparency involves two claims: on the positive side, that reflection on perceptual experiences refers to the worldly items such experiences are about; and, on the negative side, that reflection on perceptual experiences refers to nothing more than those items. The positive claim is fairly uncontroversial, and while it may not be immediately obvious why the negative point is correct, I think one could partially secure it on a case-by-case basis; that is, for any given feature of experiential vehicles as opposed to worldly items, one could examine whether an introspective analysis of perceptual experiences latches onto such a feature. Again, while examples of the transparency phenomenon often concern spatial items, they could also refer to temporal features; on the assumption that experiential content involves a temporal component, one could thus argue that such a component is not a feature of experiential vehicles, but of the scene perceptually presented to a subject. Suppose, for
64
Chapter 4
example, that Jim is uninterruptedly aware of a bright star for ten seconds. Let us also assume that these ten seconds are somehow built into the content of Jim’s visual experience. Now, even if all of this is the case, transparency poses a threat: given the distinction between features presented in experience and features of experiences, one could argue that the aforementioned period of time equivalent to ten seconds is a temporal feature of the scene experienced by Jim from a certain vantage point. So, even if Jim stares at a bright star for ten seconds and the content of his experience also incorporates an interval of ten seconds, the temporal properties specified by such a content are properties of the things presented to Jim, not of anything in Jim’s mind. Michael Tye, I think, presses the negative part of the transparency thought in relation to the experience of temporal features: when we perceive the world or introspect our experiences, “we are not aware of our experiences as unified or as continuing through time or as succeeding one another” (Tye, 2003: 96); instead, the positive thought goes, we are primarily aware of worldly items as unified or as continuing through time. “Continuity, change, and succession,” temporal features we are perceptually aware of, “are experienced as features of items experienced, not as features of experiences” (Tye, 2003: 97). These remarks may be naturally reinforced by the vehicle-content distinction. For example, Daniel Dennett stresses that to the extent that the temporal structure of psychological states or processes may be quite different from that of what they represent, temporal features intervene in experience only as features of what experience presents us with, not features of the experiential vehicles themselves (cf. Dennett, 1991: ch. 6). These considerations thereby suggest that, in general, experiential content does not fix the duration of temporally extended experiences, but only that of what is experienced by the relevant subject. In short, while the content of experiences may specify what type of experiences subjects have, they would fail to settle what experiences perceivers have during certain periods of time; after all, they do not really fix who undergoes the relevant experience nor how much time they take. This is why one should turn to the relevant experiential vehicles in order to individuate perceptual experiences, at which point P and S become relevant. THE INDIVIDUATING ROLE OF EXPERIENTIAL PROCESSES The individuation of temporally extended perceptual experiences cannot rest on experiential content alone; to settle this issue, it is important to understand what kind of psychological items they are. The processive and the stative conception provide different answers to this ontological question, whereas
The Individuation of Experiences
65
one view models perceptual experiences as mental processes, the other one conceives them as mental states. In the remainder of this chapter, I argue that a prominent virtue of S over P is precisely its ability to deal with the individuation of temporally extended perceptual experiences. More specifically, this section takes on the negative task of stressing P’s inability to handle, first, the very individuation of experiences over time, and, second, the relationship between experiences and the worldly items they present to a subject. Two preliminary remarks are in order, though. First, it is necessary to be clear on what kind of processes are invoked by a processive conception. I previously mentioned that P conceives perceptual experiences as processes which, once concluded, come to constitute events—that is, temporal particulars. Furthermore, the relevant processes are not any given kind of processes, but specifically psychological or mental ones, as such, they are “internal” at least in the sense that they are not publicly accessible items— for example, you cannot see a visual experience the same way you see a tree or a dog. Second, I mentioned in the previous section that the individuation of experiences depends on fixing a number of components related to each token-experience, namely, perceived items, mode of perceptual presentation, experiencing subject, and duration of experience. I shall not be concerned here with the relationship between experiences, on the one hand, and, on the other hand, subjects and modes of presentation. Processes are by definition things that a subject could do or that may happen to her, as such there seems to be a metaphysical (or at least conceptual) connection between perceivers and the processes in terms of which P defines perceptual experience. Again, I suspect that specifying how things are presented to a subject is a job for experiential contents, not experiential vehicles: that is, how things are presented in experience would, at least ideally, be captured by a description of what your experiences are of, not by a description of what your experiences are. So the most pressing task for an account of experiential vehicles—for example S or P—is that of throwing some light on the relationship between perceptual experiences, on the one hand, and, on the other, the worldly items they are about and time. As just anticipated, my negative goal is to show that a processive view does not appropriately deal with such conceptual connections. The decisive point against P, it seems to me, concerns the way in which it deals with the relationship between perceptual experiences and time. The importance of this point should be more or less clear; for, while there is a natural sense in which mental phenomena should not be spatially categorized—a belief or an emotion has no weight, it is not to the left or to the right of a physical object or another propositional attitudes, among others—they may be temporally qualified (cf. O’Shaughnessy, 1971b, 2000; Steward, 1997). After all, time seems to encompass both the physical and the mental. Since perceptual experiences are temporal items, a story of the conditions under
66
Chapter 4
which they exist should say something about their temporal structure. As I take temporally extended experiences to be ontologically primary, the present discussion shall focus on cases where a subject experiences her surroundings for a noninstantaneous period of time. As suggested by Dennett, the temporal structure of what our experiences are about does not necessarily reflect the temporal structure of the experiences themselves; in other words, the temporal features of experiential vehicles need not transpire in experiential contents. That being the case, how else could one determine the duration of perceptual experiences? If these psychological items are temporally protracted events resulting from the completion of perceptual processes, it would at least be legitimate to ask when they begin and when they end. Although he does not specifically discuss a processive view of experiences, B.A. Farrell highlights the difficulties faced by such questions of temporal individuation: Surely, that is, we can say that it [i.e., the experience in perception] stands in “temporal relations” to other events or processes? No–this will not do. For to say that “something or other happens quite frequently” is to say that the something occurs at different times. To say this is to say that this something is in principle datable. How now do we set about dating the occurrence that is X’s experience at any time? All we can do is to date X’s responses. But suppose X, as subject-observer, sets himself the task of dating the onset of a certain raw feel experience, for example, the one that is supposed to happen when he sees two changing shapes as equally elliptical. When X times himself here, say by stopping a stop watch, all that he can time in his “seeing”–e.g., his subvocal “Ah! That’s it”, his accompanying release of breath and muscular tension, and so forth. What, therefore, he dates is the onset of his seeing the shapes as equally elliptical. Difficulties only multiply if we now retreat and say “But we time the experience indirectly by timing the behaviour that it accompanies?” E.g., What sort of “accompanying” does this ghost do? (Farrell, 1950: 178)
Farrell’s reasoning moves in two stages. First, the thought that perceptual experiences are datable is, he notes, profoundly problematic insofar as it is not possible directly to trace perceptual experiences over time: in principle, one may only track those behavioral inputs and outputs related to the relevant experiential phenomena. Second, he stresses that the relationship between perceptual experiences and their behavioral correlates is quite puzzling; indeed, this is the issue at the heart of the mind-body problem. These remarks thereby suggest that a subject’s behavioral responses could hardly constitute a guide into the temporal identification of perceptual experiences. According to Farrell, these psychological phenomena are not the kind of things that can be timed. We can certainly keep track of a subject’s behavioral responses and of the things she is
The Individuation of Experiences
67
perceptually responsive to—whether the relevant subject is ourself or someone else—but the duration of the episodic experience underpinning discriminatory or locomotive behavior would be bound to remain elusive. Now, whereas Farrell focuses on a generic experiential notion, I specifically target perceptual experiences conceived as mental (more specifically, phenomenally conscious) processes. Indeed, I do think that the temporal individuation of perceptual experiences is bound to meet the kind of problems described by Farrell. However, this need not constitute an objection against the notion of perceptual experience in general, for one could still hold that experiences could not and should not be datable; that is, one could argue that it is conceptually misguided to ask question such as “When did it begin?” of items falling under the relevant experiential notion. Farrell’s line of reasoning, I think, presents a difficulty for the mental processes posited by the processive theorist: if perceptual experiences are conceived as events constituted by phenomenally conscious processes, they would have definite temporal boundaries; or, in other words, if perceptual experiences are temporally extended events, there should be a principled answer to questions such as “When did her experience begin?” At this stage, the processive theorist might opt for preserving the temporally fuzzy boundaries of perceptual processes, but it is unclear how he could support that claim. A philosophical account of perceptual experiences should be capable of illuminating when such psychological occurrences begin, or, in case that such a question could not be answered, capable of illuminating why the question has no answer. P, however, provides psychological particulars, the temporal boundaries of which we cannot by definition access, let alone individuate. Here is another worry about temporal individuation for perceptual processes: if perceptual experiences are conceived along the lines of P, it is unclear how many experiences a perceiver is subject to whenever she experiences her surroundings for a noninstantaneous span of time. Say that Jim visually experiences a bright star, uninterruptedly, from t1 to tx: if perceptual experiences are temporally extended events resulting from mental processes, one could in principle ask how many visual experiences Jim has between t1 and tx. As far as I can see, there are two lines of reply: first, that our vigilant sailor has one single experience during that period of time; and, second, that he actually undergoes a number of experiences—“how many” being, for the time being, irrelevant. Both options are, I take it, clearly incompatible. That said, my point is not that either alternative is implausible. Instead, the worry is that there is no definitive evidence in favor of either view; for all we know, both positions could be correct. This is, I submit, an unfortunate outcome, was that correct, the temporal individuation of perceptual experiences would turn out to be an arbitrary matter. I expand a bit further on this line of reasoning next.
68
Chapter 4
A prominent advocate of the “one-experience” view is Michael Tye, according to him, “for each period of consciousness, there is only a single experience—an experience that represents everything experienced within the period of consciousness as a whole (the period, that is, between one state of unconsciousness and the next)” (Tye, 2003: 97). This view has it that Jim goes through a single visual experience from t1 to tx. As previously mentioned, he stresses the phenomenon of perceptual transparency, a bit more specifically, he presses the transparency thought in relation to the spatial as well as temporal features that perceptual experience presents us with. In broad lines, the view comes down to this: just as the spatial items perceptual experience presents us with are not features of experience itself, but features of the experientially presented environment; the temporal features experientially presented to us are not features of our experiences, but features of the perceived world. According to him, the temporal structure of experiential content—that is, the temporal features presented to a perceiver in experience—would not determine the temporal structures of our perceptual experiences (cf. Tye, 2003: 98–99). Tye exploits the temporal transparency of experience precisely to defend the one-experience view: given that experiential content provides no guide into the temporal structure of experiential vehicles, the one-experience hypothesis seems to be the best stance on experiential individuation. This line of reasoning is nicely summarized in the following passage: [t]he one experience hypothesis finds further support in the general difficulty we face in individuating experiences through time. Consider an ordinary visual experience and suppose that it is exclusively visual. When did it begin? When will it end? As I write now, I am sitting in a library. Looking ahead, and holding my line of sight fixed, I can see many books, tables, people in the distance walking across the room, a woman nearby opening some bags as she sits down. Is this a single temporally extended visual experience? If not, why not? (Tye, 2003: 98)
When a subject experiences the world during a noninstantaneous span of time, she has only one experience, namely, the event constituted by the whole experiential process between her states of unconsciousness: to repeat, if Jim uninterruptedly sees a bright star from t1 to tx, he has a single experience during that period of time. The target of this rather economical framework for experiential individuation is twofold: on the one hand, the view that the temporal structure of experiential contents determines that of experiential vehicles; and, on the other, the many-experiences hypothesis or the view according to which perceptual experiences are constituted by shorter mental events. Per transparency, Tye thinks that reflection on perceptual experiences
The Individuation of Experiences
69
does not provide substantive evidence in favor of either position. The simplest hypothesis is, hence, the one-experience view. As previously mentioned, the transparency claim seems quite plausible, even if its negative component comes to be contested, I suspect it could be partially vindicated—that is, vindicated in relation to specific features presented in experience. It is, I think, extremely plausible that the temporal features perceptual experiences present us with are features of the perceived scene, not of the experiences themselves. For this reason, I provisionally conclude that Tye makes a persuasive case against his two targets. The bad news is that the same considerations concerning the temporal transparency of experiences set pressure on the one-experience hypothesis: although reflection on perceptual experiences lends little evidence in favor of the many-experiences view, it is unclear why it supports Tye’s position; after all, evidence against one position is not necessarily equivalent to evidence in favor of a competing alternative. More importantly, reflection on perceptual experiences does not show that a subject’s uninterrupted experiential stream constitutes a single mental event for the same reason it does not show that the same stream breaks into experience parts, namely, the subject’s insensitivity to the temporal features of her own experiences. Tye asks whether a subject’s uninterrupted conscious stream is constituted by a single psychological item, and, if it does not, why not, but I think one could in turn ask what evidence there is to suppose that the aforementioned experience is a single, temporally extended item. To defend the one-experience hypothesis, Tye does not rely on considerations about experiential content, but on the theoretical economy of that view relative to the many-experiences stance. It is, however, unclear to me why the one-experience stance is in any way more economical: after all, both positions resort to exactly the same type of psychological items; the main difference is purely quantitative—whereas Tye invokes only one mental event in order to account for a single, uninterrupted experiential stream, the many-experiences view invokes a number of such events over time. Granted: the latter position may be problematic, but this does not on its own constitute a positive argument or reason in favor of the one-experience view. When we experience the world or reflect about our experiences, we are only aware of worldly items, we are not aware of our experiences’ temporal boundaries, as such it is not manifest to us whether we undergo one or several experiences throughout an experientially uninterrupted period of time. A bit more tentatively, the foregoing considerations also suggest a line of criticism against P’s ability to accommodate the link between experiences and perceived items. It is natural to conceive the objects of perception as the informational sources of those interactions leading up to the occurrence of perceptual experiences, no matter how difficult it may be to specify the relevant informational-causal links.6 Thus conceived, the worldly items
70
Chapter 4
perceptual experiences are of and are just the informational sources of the perceptual achievements in virtue of which such experiences emerge. I take the claim for granted here: when Jim faces a bright star, he sees a determinate object, the source of the informational transaction (across space and time) leading up to Jim’s spotting of the star and the corresponding relation of visual awareness. If perceptual experiences are conceived as mental processes, however, it is unclear to me how they are related to our surroundings. As Farrell pointed out, the evidence is usually invoked in order to identify and ascribe perceptual experiences that would at best provide indirect access to phenomenally conscious processes: after all, recall that O’Shaughnessy and Soteriou take such processes to transcend the domain of physical processes and states, behavioral responses, and so on. As far as I can see, it is by no means obvious what the connection between phenomenally conscious processes, thus conceived, and the world, is. To address this difficulty, one might perhaps attempt to bridge the relevant gap by means of those informational-causal channels capable of linking physical and mental processes. Such a strategy is, however, problematic: on the one hand, it is not obvious how phenomenally conscious processes could interact with neurobiological phenomena in a subject’s brain; on the other, to the extent that they could interact with processes and states of a subject’s brain, the relevant processes would become items of the same neural order—but, as I previously mentioned, perceptual experiences are naturally predicated of subjects rather than brains. These considerations are by no means decisive, but it is unclear to me how a processive theorist could plausibly address them or otherwise, specify the link between perceptual experiences and the objects of perception. As such, the latter relationship seems to pose a considerable difficulty for P. A processive view thus postulates the existence of mental processes and events whose conditions of individuation over time cannot be sharply specified. The problem is not that we lack answers to questions about the existence of experiences over time, the recalcitrance of the question “When did that visual experience start?” might be justified, and I do think that perceptual experiences are the sort of psychological items about which it would be misguided to ask, “How many experiences Jim has between t1 and tx?” Instead, the problem is that P legitimizes such questions but offers no means for solving them. In short, I conclude that P does not provide adequate resources to understand how perceptual experiences are individuated: on the one hand, it is unclear how the relevant mental processes relate to the objects of perception; and, on the other hand, the processive view is bound to questions of experiential individuation over time which, at the same time, it is unable to solve.
The Individuation of Experiences
71
THE INDIVIDUATING ROLE OF PERCEPTUAL STATES To the extent that processes and states constitute different ontological categories, it is natural to expect them to have different conditions of individuation. Accordingly, by conceiving perceptual experiences as mental states, S might turn out to avoid the aforementioned difficulties faced by P. This is precisely what I argue for in this section, I hold that, where the notion of mental process fails, that of mental state manages to capture the special relationship between perceptual experiences, on the one hand, and, on the other, perceived items and time. I shall continue assuming that a description of experiential content should ideally take care of how worldly items are presented to a specific subject. Again, mental states are intimately related to their subjects such as properties, relations, and dispositions, states are identified in relation to subjects or objects, this being a point which applies to mental states no less than to physical ones.7 As far as I can see, there is no way of conceiving (physical or mental) states apart from their subjects, a token of the property of redness is not thought on its own, but as something that obtains in a subject. In other words, the idea of subjectless mental states cannot be taken seriously. As such, the numerical identity or difference of token-states crucially depends on what subject instantiates the relevant states. For example, Jim and Jack have different experiences even if the same worldly items are presented in exactly the same way to both of them—that is, even if they may have experiences of the same type, since what it means to be an experiential state depends on what subject has that state, and since Jim is numerically different from Jack, it follows that Jim’s experience is not numerically identical to that of Jack’s. That said, I think S is capable of capturing the close relationship between experiences and the items they are about. Conceived along stative lines, perceptual experiences obtain in perceivers (not their subpersonal parts), result from the interaction between those subjects and their surroundings, and dispose their subjects to behave in a complex number of ways. What this partially means is that one could not individuate a given perceptual state unless one could latch onto its subject, environmental input, and (at least partially) behavioral output. Indeed, these three components are constitutive of the relevant state. Within this framework, it is thereby clear how perceptual experiences relate to the worldly items they are about: if these psychological items are mental states, the items they present a subject with are simply constitutive elements in their definition. While it is unclear how mental processes relate to items in physical space, certain mental states are by definition related to them, for they are simply conditions resulting from the interaction between a subject and certain objects and properties in her surroundings. If perceptual
72
Chapter 4
experiences are conceived along those lines—that is, as states resulting from the causal-informational interaction between a perceiver and her environs— one should accordingly conclude that there is a deep relationship between them and the items they are about. A stative conception binds experiences to their respective objects because the worldly items presented in experience are effectively constitutive of perceptual states. More importantly, S could also deal with the question how perceptual experiences relate to time. The tension behind this delicate relationship may be expressed as follows: although psychological phenomena exist in time, it is unclear how perceptual experiences should be temporally individuated. As previously explained, P legitimizes a number of questions about temporal individuation (e.g., “When did that experience begin?,” “Does Jim have one or many experiences between t1 and tx?”), but it is incapable of settling them. S deals with them differently. To begin with, it is capable of specifying when perceptual experiences begin. But, in addition to that, it preempts the emergence of cardinality questions—although perceptual states exist across the temporal dimension, they simply cannot be individuated like processes or tokens of other ontological categories. It is, I think, more or less uncontroversial that physical as well as mental states persist: an object may be red or yellow for a number of days; I may be anxious for two weeks before my exams, or have a headache for a whole afternoon; and so on. But, at the same time, it would be conceptually misguided to track states over time the same way we track the mental processes which go on to constitute processive mental events. In the present context, the importance of states lies in the fact that they belong to a family of items (including properties, dispositions, among others), which satisfy conditions of existence without being particulars, whether spatial or temporal ones; that is, states do not have the same conditions of identity as material substances and events. Conceived as states, perceptual experiences are thereby redefined in a way that makes certain questions about their identity legitimate, and others, illegitimate. According to S, experiences have identity conditions, but the latter is not the same sort of conditions that spatial or temporal particulars (i.e., material objects or events) have. More specifically, if perceptual experiences are mental states, it would be possible to specify (i) when perceptual experiences begin and end, and (ii), why attempts to count temporally extended experiences are bound to fail. In relation to (i), Farrell pressed the impossibility of directly accessing, and thereby dating, perceptual experiences. This point is, I think, legitimate when experiences are conceived as mental processes, for the latter are neither observable nor accessible through introspection. By contrast, the same difficulty does not arise when the same psychological items are modeled along the lines of S; after all, if perceptual experiences are states, their key constitutive elements are directly accessible to philosophical analysis. To illustrate
The Individuation of Experiences
73
this point, let us return to the example I have been using thus far. Jim sees a bright star from t1 to tx: spotting the star and losing sight of it are achievements which occur at determinate instants of time. As I previously said, perceptual achievements and perceptual experiences are not identical, the instantaneous event of spotting a star is not an experience. Spotting the star should be conceived as an instantaneous event thanks to which Jim comes to be in a given mental condition from t1 onward: he stands in this state for as long as the informational channel between him and the star exists; again, he will be behaviorally responsive vis-à-vis the relevant informational source for as long as he stands in such a state. According to S, the aforementioned state is constituted by a relation of awareness between Jim and the star. Spotting the star is not the same thing as experiencing the star at t1, it only refers to the act of beginning to experience the bright object. Let us turn to the cardinality question now. Jim visually experiences a bright star from t1 to tx. How many visual experiences has he? One or many? I previously argued that P legitimizes and, at the same time, fails to settle this question in relation to those events constituted by phenomenally conscious processes. The problem is not that the question lacks an answer; after all, I think that perceptual experiences are not the kind of psychological items which may be counted. Instead, the difficulty is that the processive account throws no light on why it is misguided to attempt to solve the relevant question: it just burdens us with extremely controversial entities insofar as their conditions of identity cannot be directly addressed, let alone specified. S, by contrast, has a tactical advantage, not relying on the existence of mental processes, it explains why perceptual experiences cannot be counted. The reason is actually quite simple, states are not the kind of things that allow for questions of cardinality over time. Unlike substances and events, states instantiated throughout a period of time cannot be counted, that is, given an item instantiating a state of a certain kind from t1 to tx, it would be mistaken to ask how many tokens of that state the relevant item instantiates throughout t1–tx. States belong to a family of noncountable categories which also include properties, dispositions, and masses: by posing the cardinality question, one thereby ignores a fundamental conceptual difference between this cluster of notions and the one including concepts like substance and event. It would not be merely unconventional, but conceptually misguided to ask, for example, how many instances of being-yellowness obtain in a banana throughout the time it is ripe. This sort of question would betray a confusion about what it means for something to be in a certain state, for example, having a certain color. If perceptual experiences are conceived as mental states, it would thereby not be necessary to determine whether a subject has one or many of them across a given period of time: experiences would not be the kind of things which may be counted; accordingly, one could not pose the
74
Chapter 4
cardinality question in the present context. S does not take a stance between the one-experience and the many-experiences hypothesis, for it refuses to acknowledge the question those views attempt to address. This framework thus provides the necessary resources to understand why perceptual experiences exist in time, but does not submit to certain questions of temporal individuation. For a similar reason, a stative conception of experience is not committed to the “snapshot” view of perception. In broad lines, the latter position holds that, if a subject S stands in a relation of perceptual awareness with her surroundings during a given period of time, she instantiates numerically distinct states at different moments or subintervals throughout that period. So, if Jim visually experiences a star from t1 to tx, the snapshot view argues that he instantiates a perceptual state at t2 (i.e., taking t1 to be the instant at which he began experiencing the star), another one at t3, another one at t4, and so on, up to tx. This view, endorsed even by Armstrong himself (cf. Armstrong, 1968), multiplies the mental processes underpinning perceptual phenomena. For one reason or another, this bit of a stative conception has been regarded as unappealing. Although it is unclear to me why the snapshot view is incorrect, what one should do here is to highlight the ontological significance of mental states as opposed to categories of countable items: if perceptual experiences are states, they cannot be counted across time; for exactly the same reason, it is mistaken to analyze temporally extended perceptual experiences into temporally discrete states following each other across time. In the present context, it should be clear that the snapshot view is just a specific version of a general conception according to which experiences should answer questions of cardinality. The best antidote against this line of reasoning is, I think, to insist on the ontological differences between states and spatial or temporal particulars. To sum up, I think that S does a better job than P when it comes to individuate perceptual experiences over time: more specifically, I argued that a stative conception neatly deals with the relationship between perceptual experiences, on the one hand, and, on the other, perceived objects and time. The worldly items presented in experience are effectively constitutive of perceptual states: as such, a stative conception ties experiences to their respective objects. Again, S addresses at least two questions concerning the temporal individuation of experiences: first, when they begin; and, second, whether they can be counted. If the stative view is correct, both problems can be disposed insofar as mental states have clear temporal boundaries (i.e., startingpoints and end-points) but, at the same time, cannot be counted over time. In relation to all these issues, I think S fares much better than P. Accordingly, the previous considerations constitute a partial reason to favor the stative view over a processive one.
The Individuation of Experiences
75
CONCLUSION Whereas chapters 2 and 3 were mainly concerned with formulating a stative conception of perceptual experiences, this one aimed to defend it. In particular, I argued that S provides an elegant framework, first, to get a grip on the way in which perceptual experiences relate to their objects, and, second, to understand the delicate relationship between experience and time. I broke this task into three parts. First, I focused on temporally extended experiences and explained why their contents throw little light on them. Second, I argued that a processive view does not fare well with the identification of experiences over time, and then, in the final section, I turned to S in order to show how it deals with the problems faced by P in a more elegant way.
Chapter 5
Experience and Causation
A GAP IN THE MIND-WORLD CONTINUUM Throughout earlier chapters, I have introduced, compared, and partially assessed different ways of addressing the question what we talk about when we talk about perceptual experiences. A number of increasingly popular ontological proposals—all of them encompassed here under the umbrella term of Experiential Heracliteanism or (EH)—characterize experience as fundamentally dynamic: that is, they argue that perceptual experiences to a greater or lesser extent involve dynamic elements that could not be described—let alone accounted for—in nondynamic terms. Against this backdrop, this book has introduced an alternative approach to the ontology of perception, namely, Experiential Non-Heracliteanism or (ENH)—the view (or family of views) according to which experience could ultimately be described in terms of nondynamic ingredients (e.g., states, properties, and relations). What is at stake in the dialectic between both stances is not whether experiences are dynamic: they are naturally so. Perceptual experience is a phenomenon that modifies a perceiver’s cognitive economy in a way that allows her successfully or unsuccessfully to interact with her surroundings. The relevant dialectic turns on whether such a dynamic phenomenon could or could not be described and ultimately explained in nondynamic terms. As explained throughout this book, different conceptual, phenomenological, and ontological considerations support either stance. Again, the debate is complicated even further by a far from harmless metaphor, namely, one where either experience or its temporal structure is characterized as stream-like. But metaphors are not theoretically innocent: in what may be termed their theory-constitutive role (cf. Boyd, 1979), they hint at ways of intelligibly framing phenomena within a larger representation of the world, and while initially illuminating, they are 77
78
Chapter 5
potentially controversial to the extent that they could support conservative as well as radical interpretations of a given subject matter. Metaphors are indeed necessary devices of scientific and philosophical reflection, but should be handled with care. Aiming to move the line of discussion even further toward the crossroads between the philosophy and the sciences of the mind, this chapter explores the general question how mental phenomena like perceptual experiences fit into a fairly broad intuitive and scientific understanding of the natural world. Since contemporary philosophical discussions about the mind tend to frame the place that mental phenomena occupy in nature in terms of the causal roles they fulfill therein, the previous task could be rephrased as that of exploring how perceptual experiences could be causally related to the natural world. When assessing this question, a key difficulty faced by contemporary philosophers and scientists of the mind is the apparent existence of an explanatory gap between the mental and the physical. Edgar Adrian, a seminal figure in the field of neurobiology, clearly stated this line of concern: “[ . . . ] whatever our views about the relation between mind and body, we cannot escape the fact that there is an unsatisfactory gap between two such events as the sticking of a pin into my finger and the appearance of a sensation of pain in my consciousness. Part of the gap is obviously made up of events in my sensory nerves and brain” (Adrian, 1928: 11–12). Little has changed since then: contemporary cognitive scientists are intensely engaged in working out our underlying neural and otherwise subpersonal architecture, so as to shed light on the emergence and the very existence of conscious organisms. In the case of psychological phenomena like perceptual experiences, the same concern would be expressed in terms of what is known as the perceptual process, that is, the series of events that take place between the moment a person encounters—for example, looks at—objects in her surroundings and the moment she perceives it (cf. Goldstein & Brockmole, 2014: 5). While the sciences of the mind have systematically examined the links between environmental stimuli and neural processing over the past hundred years, they have by and large neglected or downplayed as philosophical the question how our neural architecture specifically constitutes personal-level perceptual phenomena.1 An explanatory divide between the mental and the physical thereby keeps haunting scientific and philosophical thinking about the mind up to the present day. In fact, this issue lies at the heart of what philosophers systematically assess as the mind-body problem. A sense of a gap between the physical and the mental pervades the philosophy as well as the sciences of the mind, and it is certainly worth asking why we could be inclined to believe in its existence. For sure, the way I approach the question here is not intended to uncover the psychological or historical motives that led philosophers and scientists to take the divide for granted:
Experience and Causation
79
instead, it concerns the conceptual conditions that make the emergence of that gap possible. As a general working hypothesis, this chapter will explore the thought that the relevant divide is to a greater or lesser extent determined by our causal assumptions at work: more specifically, the suggestion is that the widespread sense of a radical divide between the physical and the mental is intimately bound to an exclusively efficient understanding of causation in the modern philosophy and sciences of the mind. I suspect that another suggestive yet problematic metaphor is at work here. When it comes to specify the causal place of mental phenomena in nature, this is typically done so in terms of individuation within a complex network of mutually interacting forces, where the image of billiard balls impacting and pushing each other specially comes to mind. Metaphors like this are of consequence, for they tend to predefine the causal assumptions at play in our pretheoretic and scientific understanding of the physical, the mental, and their mutual relationship. By reflecting on the causal metaphor of mutually interacting billiard balls or—to use Alicia Juarrero’s terminology—a system of mutually interacting “pull-push” forces, for better or worse, these illustrative devices simplify the way we think of psycho-physical causal interactions: in particular, they tend to depict the natural world as a complex mosaic of events, where the causal interactions among said events are exclusively conceived in terms of efficient causation. This chapter will suggest that the aforementioned causal assumption is not compulsory. To do so, I draw on a blooming literature of neo-Aristotelian inspiration in order to show that our understanding of the place of perceptual experiences in nature may be fleshed out in causal terms other than the modern notion of efficient causation. More specifically, I suggest that the causal structure of perceptual experiences may be fleshed out in terms of the complementary notions of matterers and constraints. To flesh this suggestion out, I shall first resort to Helen Steward’s notion of matterers, for it introduces an elegant means for thinking about relations of (nonefficient) causal determination. That said, this notion does not specify on its own what we would be thinking about when thinking about such relations. This is why I then resort to constraints, that is, to structural properties which obtain when a set of elements interact with each other, but which are in turn capable of determining the behavior of the set it originally resulted from. I connect both causal notions in order to argue that perceptual experiences may be conceived as higher-order structural states that, on the one hand, emerge from the complex interactions among those neural and otherwise subpersonal components relevant to our sensory systems, and, on the other, constrain the ways in which a perceiving organism relates to her environment. This proposal not only avoids the stagnant predicament in which contemporary debates on the mind-body problem typically find themselves: it also specifies—and hence
80
Chapter 5
operationalizes—the concept of constitutive dependence at the heart of a non-Heraclitean ontology of experience. The present discussion thus complements the foregoing ontological survey by describing how non-Heraclitean experiential states could constitutively depend on subpersonal events, states, and mechanisms—in short, to the natural world scientists of the mind know and love. The remainder of this chapter is divided into three parts. First, I set up some necessary preliminary groundwork by introducing two crucial ways of thinking about causation in the context of the philosophy of mind and stressing a widespread modern tendency to think of about causation in its purely efficient variety. This general trend is problematic to the extent that, as the debate on the causal closure of the physical has made it clear, a natural world conceived as an incredibly complex mosaic of efficiently related events leaves little or no room for perceptual experiences, among other types of mental phenomena. By fleshing out the relationship of experience to the natural world in terms of efficient causes, an unbridgeable gap seems bound to open between both domains. Second, I start paving the way for an alternative causal understanding of perception by introducing a framework in terms of which Steward conceives perceptual experiences as facts and their causal role in terms of the causal yet nonefficient category of matters. Since Steward’s proposal is incomplete on its own, the third and final section brings up Juarrero’s key contribution to this discussion, namely, a conception of causation in terms of constraints. Against this backdrop, I suggest that perceptual experiences could well be conceived as constraints, that is, as structural properties that result from neurobiological or otherwise subpersonal events and states, but that, at the same time, determine what kinds of mental and physiological phenomena a given organism could subsequently undergo. At this stage, I also make a point of explaining how this causal understanding of experience specifies the notion of constitutive dependence at the heart of a non-Heraclitean ontology of perception. To circumscribe what could easily become an overambitious and thereby unmanageable discussion, I shall simplify it in three respects. First, the following remarks focus on perceptual phenomena. Second, a full-fledged assessment of the relevant gap would not only involve discussing its underpinning notion of cause but also that of explanation. The latter notion cannot be tackled here, as it involves delicate issues in the philosophy of science that transcend the boundaries of this work. Since the present project primarily seeks to outline a framework for thinking about the ontological structure of perceptual experiences—as opposed to, say, one for thinking about our epistemological access to them—this chapter only addresses the causal assumptions at the basis of our thinking about the place of perceptual experiences in nature.2 Third, since it concerns well-known portions of recent philosophical
Experience and Causation
81
history, I beg the reader to forgive the extreme generality of the following survey. TWO CAUSAL PARADIGMS As just anticipated, this preliminary section introduces two causal paradigms around which I frame the subsequent discussion. The contrast I aim to draw may be briefly stated as follows: whereas a modern Cartesian outlook reduces our understanding of causation to its widely known efficient variety, an Aristotelian paradigm rejects that reduction. The present comparison will no doubt be sketchy, but it does provide the necessary background for introducing the suggestion that perceptual experiences may be causally yet nonefficiently related to the natural world. To begin with, let us turn to a classic Aristotelian framework that dominated philosophical and scientific thinking about causation across ancient and medieval times. A basic—albeit not the only—goal of this paradigm was to account for accidental and substantial change in the natural world. In particular, it invoked four mutually irreducible causal notions: material causes, or the set of material conditions that contribute to the constitution of natural objects and phenomena; formal causes, or those elements that capture the internal essence or structure of such natural item; final causes, the ends toward which objects and phenomena are naturally projected; and efficient causes, those items—either events or objects, as the case may be— that induce change in other objects or phenomena. For Aristotle, these four notions comprise different explanatory dimensions the joint statement of which constitutes a full account of natural change. In fact, Aristotle did not only seem to apply this explanatory model to dynamic phenomena but also to natural objects. Although artifacts fall beyond the scope of these explanatory principles to the extent that they fail to be natural objects, they do provide intuitive means of illustrating the previous causal framework. Consider a plain wooden chair. Our intuitive understanding of causation is most likely captured by the notion of efficient cause: an artisan’s action is the efficient cause of a chair’s coming into existence. This particular chair is made out of particular pieces of wood, but there are chairs made of other pieces of wood or even made of completely different materials. The material that constitutes a given object thereby makes a crucial difference for this chair being the chair it is instead of another one. In the case of an artifact, the formal cause of a chair is projected by the artisan, who structures the relevant materials so as to build a working chair instead of, say, a small coffee table or some other piece of furniture. She also provides the purpose or the end for which this artifact will exist and be used. Hence, these four causal notions strived
82
Chapter 5
to capture different dimensions in terms of which an object may be defined as the object it is. As Aristotelian physics met its demise at the hands of the new observational and experimental sciences of the fifteenth and the sixteenth centuries, its accompanying causal framework also fell out of favor. By endorsing a worldview of mechanistically governed corpuscles, scientists and philosophers increasingly committed to an approach that delivered—or promised to deliver, at any rate—better ways of predicting how the natural world behaves but that, at the same time, left little or no room at all for explanatory notions such as those of formal and final causes. Since material causes simply collapsed into material conditions of existence, only efficient causes remained as explanatorily relevant principles in modern sciences and philosophy. For sure, this paradigm shift in our understanding of causation and its bearing on the divide between the mental and the physical is salient in Descartes’ work on epistemology and the philosophy of mind. Against the backdrop of a mechanistic physics unpacked throughout a number of writings—for example, his Meteorology and The World (cf. Descartes, 1965, 1988)—Descartes systematically criticized the explanatory need of formal and final causes within our scientific worldview. As such, when it comes to specifying the relationship between the physical and the mental, the only relevant causal notion he is left with is that of efficient causes. The move quickly proved to be problematic, for the relation of efficient causation typically ranges over spatially extended items, whereas Descartes himself strived to prove that mental phenomena and their subjects are not endowed with extension. Pressed by Princess Elizabeth of Bohemia on how mind and body interact with each other, for example, Descartes draws on the notion of forces—that is, one that characteristically captures the efficient variety of causation—in order to hint at the relevant explanatory relation. Of course, she remained unimpressed: for that notion suggests the existence of motions and impacts among extended objects—that is, items endowed with mass and spatial boundaries—whereas Descartes’ Meditations famously argued that mental subjects and their attributes are defined for their lack of extension. To make things worse, Descartes himself noted that the notion of force becomes somewhat blurry when extended from the purely physical world to a psychophysical domain of study, whereby it is unclear that it holds water when it comes to specify the relationship between physical and mental phenomena. Ultimately, he tries to reassure his correspondent that whatever difficulties we might face when thinking about the union of mind and body, they derive more from limitations on our philosophical imagination than from the internal coherency of his psycho-physical worldview. Princess Elizabeth may, Descartes suggests, conceive the relationship between mind and body in a two-step process: first, by thinking of psychological
Experience and Causation
83
phenomena as extended; and, secondly, by acknowledging the existence of a fundamental distinction between the mental and the physical. Of course, this reply was controversial. With some hindsight, Descartes could alternatively have avoided accounting for the relationship between mental and physical phenomena in terms of forces or otherwise causally efficient notions. But given his systematic allegiance to a mechanistic worldview, that option was naturally unavailable. The Cartesian paradigm has certainly been influential over time—if nothing else, its disdain of causal yet nonefficient categories (e.g., formal and final causes) made itself felt well into the twentieth century. That said, its sole reliance on efficient causality has proven to be controversial over the past few decades. On a purely practical note about scientific methodology, for example, it is far from clear that the different natural, social, and formal sciences actually use the notions of explanation and causation in a purely efficient sense (cf. Cartwright, 1999). More specifically relevant for present purposes, I also believe that this causal paradigm limits the ways in which certain philosophical and empirical problems may be thought. Indeed, this is the claim I go on to explore regarding our understanding of how perceptual experiences exist in nature. To begin doing so, the next section introduces suggestions in the spirit of what I previously termed an Aristotelian approach to causation. Was it not for the conceptual evisceration that the notion underwent in modern times, we could more easily appreciate that our causal understanding of mental phenomena like perceptual experiences need not exclusively rely on a relation of the efficient kind. The new conceptual possibilities opened by a richer causal framework by no means close the recalcitrant gap between mind and world, but they do suggest novel ways of reassessing a paramount yet unmanageable subject. STEWARD’S PLURALIST UNDERSTANDING OF CAUSATION As previously mentioned, a venerable line of Aristotelian thought suggests that causation need not be conceived in univocal terms, a claim that Aristotle himself fleshed out by distinguishing four types of causes. In more recent times, the same general thought has been rehearsed in a number of different ways. To illustrate this approach, let us examine Helen Steward’s discussion of causation in the ontology of perception. While deeply inspiring, her proposal is, I believe, only partially persuasive, for, while it makes a strong case for the claim that our understanding of perceptual experiences need not rely on the notion of efficient causation, it seems to specify how we think about perceptual experiences rather than what they actually are. This qualified
84
Chapter 5
endorsement will in turn set the ground for introducing the outlook I shall favor. A given causal framework is intuitively expected to take a stance on causal relations as well as causal relata, for what kinds of causal relations may hold in nature is bound to what kinds of entities are so related. Reflecting on everyday ways of referring to causal relations—say, linguistic constructions of the form “C because E”—Steward draws a distinction between three notions: first, movers or those substances that ordinary people—as opposed to philosophers—take to be causes; second, makers-happen or the events that philosophers traditionally take to be causes and effects; and, third, matterers or features that contribute to the production of effects (cf. Steward, 2011: 154). When a match is lit, for example, different causal concepts come into play: a person (a mover) strikes a match; a match’s being struck (a makers-happen) causes it to ignite, and matterers intervene to the extent that a match cannot be lit without certain conditions in place—for example, dryness or lack of moisture, the presence of enough oxygen for combustion to take place, among others. It is worth noting that movers and makers-happen are (either spatial or temporal) particulars, whereas matterers need not be so. As such, these notions not only expand our understanding of causal relations but also that of their corresponding relata. Against this backdrop, Steward goes on to model perceptual experiences as facts, a view that clearly departs from a more traditional outlook of experiences understood as temporal particulars—that is, mental events—standing in relations of efficient causation vis-à-vis other mental and physical events. Of course, she is able to thread this alternative path by modeling the relevant psychological phenomena as matterers rather than as maker-happeners: more specifically, she conceives them as facts relating a psychological subject. Although I do not examine Steward’s view of perceptual experience at length here, it is worth pausing on its rationale. Disjunctivist or relationalist views of perception highlight the constitutive link between veridical cases of perceptual experiences and the corresponding perceived objects. Although currently popular, however, they potentially clash with a widespread framework of philosophical and scientific understanding known as the causal theory of perception: for, to the extent that the latter outlook conceives perceptual experience in terms of the event at the end-side of the previously mentioned perceptual process, it suggests that experiences could take place independently of the worldly items they present to their subjects. Steward’s causal ontology for perception is precisely motivated as an attempt to dissolve that tension: on the one hand, it naturally accommodates the constitutive link between veridical experience and its objects, for facts are complex entities that could in principle stand as the reference of constructions like “s visually experiences/sees o,” and, on the other, it dodges the potential conflict with the
Experience and Causation
85
causal theory of perception, since matterers—as Steward would classify perceptual experiences—do not need to fit within the chain of efficiently related elements that constitutes the perceptual process. By thus avoiding an understanding of perceptual experiences as mental events standing in relations of efficient causation vis-à-vis other mental and physical phenomena, Steward also steers clear from the rather recalcitrant question how to individuate the relevant experiential events within the landscape of the perceptual process. As previously suggested, it is problematic to think about perceptual experiences in terms of events efficiently or productively related to other physical or psychological occurrences. Albeit natural, this conception is deeply intertwined with a more general picture of the physical and the mental riddled by a recalcitrant explanatory gap. That said, by identifying different ways of thinking about causation, Steward hints at a radical yet promising suggestion: if our causal understanding of the natural world is not thereby limited to a complex representation of efficiently intertwined (either physical or mental) particulars, neither do we need to think of perceptual experiences in terms of particulars—that is, as substances or events—nor their causal structure in terms of efficient causation. How could we think about experiential phenomena, then? Steward evidently sets the notion of makers-happen aside, for it refers to the obtaining of relations of efficient causation among events. She does not say much about movers, but I believe it is safe to say that they are not suitable for the present task: after all, they refer to causal—albeit nonefficient—relations holding between spatio-temporal objects, on the one hand, and, on the other, the causal properties or events they instantiate. If nothing else, however, perceptual experiences are not causally related to other physical or mental phenomena the same way that subjects and their causal powers or attributes do so. Steward herself models experiences as facts, and their corresponding causal relations to other physical and mental items in terms of the notion of matterers. And while I do take issue with Steward’s proposal—as explained next—I do believe there is an important insight behind the thought that the causal structure of experience should be fleshed out in terms of matterers. While deeply attractive, Steward’s approach seems incomplete for at least two reasons. First, it does not spell out how perceptual experiences specifically latch onto the natural world. For example, makers-happen are intended to refer to events that interact with each other via relations of efficient causation. Again, Steward models perceptual experiences as facts, which in turn opens a path for categorizing them as matterers: that being said, one could reasonably infer that, as causal antecedents, experiences could constitute relevant factors for the formation of perceptual beliefs, among other propositional attitudes; as causal consequents, the emergence of perceptual experiences surely depends on antecedent neurobiological or otherwise subpersonal
86
Chapter 5
events, states, and mechanisms. But none of these claims actually throw light on how experiences so conceived relate to the makers-happen that constitute the perceptual process. Hence, although broadly sympathetic toward it, I do not believe that Steward’s proposal fully addresses the problematic gap between perceptual experience and the world. By detaching such psychological phenomena from our understanding of nature as an efficiently interwoven network of events or makers-happen, she also raises the question how they would relate back to the perceptual process at all. The expanded causal framework she relies on does not provide any help here: after all, even if the categories of matterer and makers-happen intuitively intervene in the way we think about natural objects and phenomena, it is by no means obvious how they relate to each other. Second, and perhaps more importantly, Steward’s causal understanding of perceptual experiences in fact addresses their intelligible rather than ontological structure. Facts may be conceived in a number of different ways, but it is relatively clear that Steward resorts to them in order to avoid categorizing experiences as tokens or particulars—that is, as items that could, among other things, answer cardinality questions such as “how many X’s are there?.” This point becomes salient in her rejection of D.H. Mellor’s understanding of facts as facta—that is, as items that are fact-like and countable at the same time (cf. Steward, 2011: 152–153). This line of reasoning certainly involves claims I agree with: for example, Mellor’s proposal is suspicious insofar as it tailors the notion of fact so as to make its instances accommodate the relation of efficient causation; or again, it is by no means obvious that perceptual experiences should be thought of as countable items—a claim I come back to later on. However, up to this point, Steward has only provided resources for thinking about perceptual experiences, not for spelling out what kind of items they are. To appreciate this, consider the following paragraph that follows her critical assessment of Mellor’s view: The thing to recognise is that causation is not a force, but a category in terms of which we think. As such, it can encompass a wide range of ontologically different sorts of relationship. This does not make causal claims any less true, and it does not imply that causation is a creature only of the mind, or anything similar. It only means that the correct account of what makes all causal claims causal (not, note, what makes them true) may be something which has more to do with us than it has to do with mind-independent reality. (Steward, 2011: 156)
Steward’s causal framework is not intended to import new entities and relations into the natural world: it is intended to reflect “a category in terms of which we think” about the world. The suggestion seems to be that reality may be conceived as a complex mosaic of items that interact with each
Experience and Causation
87
other in “a wide range of ontologically different sorts of relationship.” The notions of makers-happen, movers, and matterers, meanwhile, hint at ways in which we could and do actually think about those relationships. For sure, this stance would not amount to a deflationary understanding of causation as a mind-dependent feature: the previous causal notions are intended to capture different ways of thinking about the world, not to determine what makes the corresponding thoughts true. This being the case, however, Steward’s proposal now faces the following question: what makes such causal thoughts true? What kinds of relationship in the natural world ground the relevant causal propositions? The notion of makers-happen does not pose a major difficulty, for it intuitively reflects the relation of efficient causation that typically holds among worldly occurrences. But what about the other two notions? Since matterers are specifically relevant here to the extent that they are intended to capture the way we think about perceptual experiences, what kind of worldly relationship is expressed by them? On pain of collapsing into an idealist view of causation, the very notion of matterer cannot solve this question. Neither is the notion of fact any useful here, as it only specifies what kind of relata matterers involve. Hence, Steward’s causal ontology of perception is at best incomplete: it addresses how we could think about experiences, not what kind of items they are. In spite of the previous concerns, it is undeniable that Steward makes a significant contribution to the line of thought I strive to outline here. By revisiting and fleshing out the Aristotelian thought that the ontological structure of the mental need not be conceived in terms of efficient causation alone, she thereby suggests how we could expand the space of intelligibility when it comes to think about the causal role of experiences in the natural world. Against this basic yet fundamental backdrop, the next section goes on to describe how, if not in terms of efficient causation, the causal structure of experience may be understood. PERCEPTUAL EXPERIENCES AS CONSTRAINTS By introducing different ways of thinking about causation, Steward inspires new ways of thinking about the causal place of psychological phenomena in the natural world. This line of thought may be deemed Aristotelian in spirit, as it challenges a widespread assumption in the modern sciences and philosophy of mind, namely, that efficient causation is the main or primary causal category when it comes to account for the ontological structure of psychological phenomena like perceptual experiences. Against this backdrop, Steward suggests that perceptual experiences may be conceived as matterers—or, more specifically, as causally relevant facts. Although the proposal
88
Chapter 5
is suggestive, its underpinning framework only specifies different ways of thinking about causation, not the actual causal relations that bind different kinds of items in the natural world: as such, it does not completely settle the ontological question how perceptual experiences are causally related to the natural world. Although Steward’s framework thus provides a necessary backdrop of intelligibility for questions about the causal structure of perceptual experiences, it does not actually specify their ontological structure. To address the ontological question at stake, I pick up on two questions raised throughout my critical assessment of Steward’s causal framework: first, which causal relations the notion of matterers exactly refers to—or, to put the same point in a slightly different way, what ontologically grounds the notion of matterers; and, second, how perceptual experiences fit into the mosaic of efficiently related natural events when conceived in terms of matterers and their ontological counterparts. Regarding the first question, I draw on Alicia Juarrero’s provocative yet somewhat neglected work on the causal structure of intentional action. By combining the notion of constraint Juarrero brings up with the non-Heraclitean account of perception outlined in chapter 3, I go on to describe what causal place perceptual experiences could have in the natural world. Turning to the first question, how could the notion of matterers be ontologically specified? Although her philosophical project exhibits considerable differences vis-à-vis that of Steward’s, Juarrero also parts company with a Cartesian paradigm of causation when it comes to think about psychological phenomena. After describing how the latter causal notion—alongside what is widely known as the nomological-deductive model of explanation—shaped much of the modern understanding of scientific as well as philosophical explanations, she undertakes the ambitious task of providing a model of intentional action where neither actions nor their respective agents are necessarily conceived in terms of the “push-pull” impact of external forces between material objects or temporal particulars. To address this challenge, Juarrero resorts to constraints, a notion intended to pick up on structural or relational properties of complex systems. Originally derived from mechanical dynamics but later on applied to a number of other fields (e.g., chemistry, biology, neurobiology, and information theory), these structural features constrain a system in the sense that they make it to behave in ways that its single constituting parts could not on their own. A simple pendulum, for example, has a determinate architecture that forces its weights to move in ways they would not otherwise move on their own. For sure, constraints do not constitute efficient causes, that is, the relevant structural features do not cause a system to exhibit new forms of behavior in the “push-pull” way in which events efficiently interact with each other, for they are not (either spatial or temporal) particulars—they are not the kind of items that could constitute
Experience and Causation
89
efficient causes or effects. That said, they are still causal insofar as they determine events and states that would not otherwise occur or obtain without them: when different elements come to interact with each other, they bring about complex structures individuated by their ability to exhibit new patterns of behavior; thanks to these structures, in turn, the constituting components develop forms of behavior they would not exhibit on their own or when conceived as mere aggregates. Constraints thus specify bidirectional relations of causal dependence: bottom-up ones, insofar as ontologically independent elements unite to constitute complex structures; and top-down ones, insofar as said structures determine the future behavior of their constituting parts. Of course, efficient causation does not account for these relations. Since they highlight how an overarching self-organizing structure determines its constituting elements (and vice versa), the notion of constraints specifies or “operationalizes” the interrelated Aristotelian notions of formal and material causes. Juarrero’s framework not only shares the Aristotelian undertones of Steward’s causal ontology of perception but also complements it by ontologically grounding matterers on the notion of constraints. This point may be addressed in terms of the following question: when our causal thinking concerns the relations which Steward brings together under the notion of matterer, could it actually refer to the interlevel relations Juarrero outlines in terms of constraints? In short, could matterers refer to constraints? I believe they could. On the one hand, both categories suit each other as far as their respective relata are concerned: matterers are not intended to refer to relations holding among events, a condition that, to the extent that they pick up on relations of interlevel dependence involving structural features rather than episodic items, constraints do fulfill. On the other, matterers could also accommodate the relations of inter-level dependence between structured wholes and their constituents. After all, the latter relations are intuitively expressed in terms of the nonefficient notion of causal determination at the heart of Steward’s matterers: on the bottom-up direction, a number of constituting elements determine the obtaining of the structuring network that will in turn encompass such elements; on the top-down direction, meanwhile, a structuring network will determine the future behavioral patterns of the elements that originated it in the first place. Since none of the relata at stake are near-instantaneous or protracted events, it is fairly clear that the relevant dependence relations are not of the efficient kind. To the extent that constraints could thereby provide a plausible answer, the ontological question what we think about when we think about matterers, the previous remarks help fleshing out a nonefficient notion of causation that could in principle reframe how we pose and go about solving questions concerning the causal role of psychological phenomena (e.g., perceptual experience, and intentional action).
90
Chapter 5
Against the previous backdrop, let us turn to the more specific question how perceptual experiences could fit within our causal understanding of the natural world. Contemporary treatments of the mind-body problem—that is, the metaphysical question how mental phenomena exist in the natural world—typically strive to navigate a middle course between two unappealing fates: on the one hand, endorsing an epiphenomenalist view of the mental; and, on the other, accepting that mental phenomena causally overdetermine the natural world. Either predicament is avoided in Juarrero’s causal framework, that is, one where the physical and the mental are conceived as domains at different levels of evolutionary or otherwise biological complexity, and where the causal links between both domains need not be of the efficient kind. Whereas Juarrero is pretty much interested in applying this general framework to a philosophical account of intentional action, I invoke it here to shed light on our causal understanding of perceptual experience. In a nutshell, my suggestion is that, from a causal point of view, perceptual experiences may be conceived as constraints, that is, as structural features that stand in relations of causal determination vis-à-vis the subpersonal and behavioral elements—whether inputs or outputs—of the perceptual process. The perceptual process is traditionally supposed to comprise a chain of efficiently connected physical and neurobiological events underpinning our perceptual lives, where conscious phenomena like perceptual (say, visual) experiences are the personal-level effects at the very end of the process. On the alternative causal framework at play here, perceptual experiences should be thought of as detached from the aforementioned chain of efficiently interconnected events. This simple yet radical move should probably be justified on independent grounds lest it turns out to be ad hoc. For the time being, I voice it here as a plausible thought in light of at least two facts: first, that perceptual experience is a conscious, personal-level phenomenon, whereas the perceptual process is constituted by subpersonal events; and, second, that experiences need not be conceived as the type of episodic item that constitutes the perceptual process. The second fact is supported by a non-Heraclitean ontology of perception as well as the suggestion that experiences may be conceived as constraints. The general thought to the effect that experiences are fundamentally subjective states of a phenomenal kind, could indeed be specified in terms of the view that experiences consist in the relevant subject’s instantiation of structural or relational properties. Perceptual experiences may be understood as cognitive structures emerging from perceptual processes, where the relevant structures endow a particular organism with new possibilities of behavior that the same constituting elements would not otherwise endow it with. That said, subpersonal perceptual events do not efficiently cause experiences so conceived, that is, they do not cause experiences in the “pull-push” way in which interacting forces do so. Instead, the former causally determine the
Experience and Causation
91
latter: the events constituting the perceptual process are matterers vis-à-vis perceptual experiences. Likewise, experiences are not simple epiphenomenal off-shots of said processes: they causally determine the subject’s larger cognitive economy. In general, perceptual experiences allow their subjects to display environmentally sensitive behavior: thanks to them, a perceiver becomes capable of skillfully navigating or interacting with her surroundings as well as informing the contents of her propositional attitudes with environmental information. Hence, the causal framework laid out throughout this chapter suggests at least one way in which perceptual experiences could causally fit into the natural world, namely, as not only instantiations of structural properties which are causally—yet not efficiently—determined by an underpinning perceptual process but also capable of determining a subject’s cognitive economy. To sum up, the notion of constraint is a crucial element for spelling out interlevel causal relations between complex overarching structures and their constituting elements: as elements within an aggregate start interacting with each other, they establish conditions on their mutual interaction to the point of developing an overarching higher-order structure; the latter framework is not as it were a mere epiphenomenon or être de raison, for it sets limitations on how its constituting elements interact; and so, by constraining such elements, novel structural properties open up new ways in which such constituent items may behave—without the former, the latter would remain a primitive, unorganized aggregate. Building on this notion, I suggested that perceptual experiences may be conceived as higher-order structural states that, on the one hand, emerge from the complex interactions among those neural and otherwise subpersonal components relevant to our sensory systems, and, on the other, constrain the ways in which a perceiving organism relates to her environment. For sure, this proposal accommodates the relation between experience and the world in terms of a causal understanding that owes more to the classic Aristotelian idea of formal cause than to the modern notion of efficient causation. This proposal, I believe, enriches the ontological view of experience I outlined in chapter 3, for it operationalizes or specifies the concept of constitutive dependence at the heart of ENH. CODA: LINKS TO BEHAVIORISM AND FUNCTIONALISM This chapter addressed the question how the causal structure of perceptual experience may be conceived within a framework along the lines of ENH. Throughout this assessment, I have strived to highlight how assumptions about causation bear on our understanding of this problem and its potential solutions. By departing from the suggestion that modern thinking about
92
Chapter 5
mental phenomena tends to rely on a purely efficient notion of causation, I have sought to explore a causal yet nonefficient way of conceiving the relationship between perceptual experiences and the natural world. In particular, I have drawn on Steward’s and Juarrero’s neo-Aristotelian works in order to suggest that experience may be fleshed out in terms of constraints, that is, structural features that reflect the self-organization and evolving behavior of complex, dynamic systems—in the case at stake, those systems being subjects of experiences like us. The notion of constraint is ideally suited to shed some light on the question at hand, for it specifies bottom-up as well as top-down causal relations among phenomena lying at different levels of ontological complexity—a pattern that precisely mirrors the overall kinds of relationships holding between mental and physical phenomena. As a result of the previous line of reasoning, the present survey perhaps stands apart from modern assessments of mental causation: since I do not take a Cartesian causal paradigm for granted, perceptual experiences are not necessarily expected to fit into the natural world via relations of efficient causation. Again, although my remarks owe much to Steward’s and Juarrero’s writings—indeed, this chapter could be read as an extrapolation of Juarrero’s causal account of intentional agency to our thinking about perceptual experience—I do believe that it makes a modest contribution to the existing literature. For, as it has already been made clear throughout the previous sections, Steward’s causal ontology of perception seems insufficient to the extent that it addresses how we could think about experience rather than what they are. While my proposal falls more squarely in line with that of Juarrero’s, I do make an effort to expand on the latter by outlining one of its further applications and complementing its ontological framework with a story about the intelligibility of the relevant category at play. To close this chapter, I shall briefly pause on the potential connection between the picture unpacked throughout the last couple of sections and two contemporary views of mental phenomena—namely, behaviorism and functionalism.3 It is, I believe, worth pointing out that, by conceiving mental phenomena in dispositional or otherwise relational terms, both views share two broad themes: on the one hand, stressing the intimate link between our mental lives and (overt or cover) behavior, and, on the other, doing so by specifying mental phenomena in nonepisodic terms. This latter theme is certainly problematic when psycho-physical causation is modeled along the lines of efficient interaction among events. Against the backdrop of the foregoing discussion, however, both approaches reemerge as inspired even if somewhat misunderstood: like few other systematic accounts, they stressed that certain elements of the natural world—for example, mental as well as physical states, properties—need not be causally understood in terms of the place they occupy within a mosaic of efficiently related events. Since mainstream philosophers and scientists of the mind by and large neglected to address their own
Experience and Causation
93
underpinning causal assumptions or simply took a Cartesian understanding of causation for granted, it is not surprising that even behaviorists and functionalists failed fully to appreciate the previous insight at the time. Again, this general diagnosis could potentially address a curious historical question, namely, why the behaviorist and the functionalist proposals refuse totally to abandon the scientific and philosophical landscape even long after their heyday (cf. Eliasmith, 2003). The resilience that both accounts have exhibited throughout recent decades, probably reflects a tacit conflict between their core insights and a causal framework against which they cannot be made fully intelligible. To wrap things up, this book has strived to outline and highlight the relevance of two opposing ontological approaches to perceptual experiences: Experiential Heracliteanism (EH) and Experiential Non-Heracliteanism (ENH). I adopted a largely critical stance toward the former in order to highlight that, although increasingly popular, it is by no means uncontroversial. At the same time, I presented ENH in a more favorable light than that adopted by the philosophical mainstream. This chapter is part of the latter effort: turning to the question how perceptual experiences relate to the natural world, I have suggested that our somewhat neglected causal understanding of perceptual experience may be specified in terms of the notion of constraint, a proposal I find particularly attractive insofar as it does not presuppose thinking about experiences in a purely Heraclitean way. Of course, none of this amounts to a definitive critique of EH or a definitive defense of EHN. From the very outset of this endeavor, I have only aimed to provide a general framework for discussing different issues concerning the ontology of perception, hoping that it may prove itself useful for other players, for other battles. NOTES 1. “The mind-body problem is asking how the flow of sodium and potassium ions across membranes that creates nerve impulses becomes transformed into the experience we have when we see a friend’s face or when we experience the color of a red rose. Just showing that a neuron fires to a face or the color red doesn’t answer the question of how the firing creates the experience of seeing a face or perceiving the color red. Thus, the physiological research we have been describing in this book, although extremely important for understanding the physiological mechanism responsible for perception, does not provide a solution to the mind-body problem” (Goldstein & Brockmole, 2014: 89). 2. In order to outline an alternative understanding of intentional action, Juarrero, 1999 makes sure of addressing the causal as well as the explanatory assumptions relevant to that task. As such besides being a key source of inspiration for this book, her work also suggests ways in which the present proposal may be expanded. 3. I return to these views and the general question they address—that is, the mindbody problem—in chapter 6.
Chapter 6
Applications beyond the Ontology of Perception
Drawing on the work of writers such as Zeno Vendler, Helen Steward, Matt Soteriou, among others, I have outlined an ontological framework for assessing the categorial structure of perceptual experience. As Heraclitean views gain traction in the contemporary literature, this book has made a very modest attempt at outlining and defending a non-Heraclitean approach. Later on, chapter 7 will suggest that pure linguistic analysis does not reveal too much about the ontological structure of experience, a conclusion that should—even if indirectly—highlight the importance of engaging in the sort of ontological reflection I pursue in this book. But what is the point of experiential ontology if it does not improve our lives in other philosophical subdisciplines? Besides its inherent ontological interest as part of a more general project of casting some light on the basic building block of our conscious lives, assessing the dynamic profile of perceptual experiences also has a bearing on a wider array of philosophical issues. For instance, understanding whether and, if so, how experiences are spread out in time, seems relevant for assessing how objects are perceptually represented as distinct from their backgrounds (cf. Burge, 2010). Neuroscientists and perceptual psychologists are also engaged in debating whether perception is a process that unfolds over time or a series of snapshot-like states. Or again, considerations about the temporal structure of experience also transpire in current philosophical debates on how we should think about the perceptual representation of time and whether it somehow depends on the temporal features of perceptual experiences (e.g., Foster, 1982; Dennett & Kinsbourne, 1992; Dainton, 2000; Phillips, 2014; Lee, 2014). Finally, the above ontological debate might provide new perspectives on the perceptual version of the classic mind-body problem. This chapter will specifically focus on examining three subject matters in relation to which the 95
96
Chapter 6
categorial identification of perceptual experiences could be significant: the traditional mind-body problem; their assertive character of experience; and the experiential sense of temporal passage. Contemporary work on the mind-body problem has turned on three forms of psycho-physical materialism: behaviorism, the mind-brain identity theory, and functionalism. In spite of their mutual differences, these views share two key elements: on the one hand, a systematic rejection of Cartesian dualism; and, on the other, the assumption that mental and physical phenomena must be somehow related within a causal framework. Although apparently harmless, the latter assumption is implemented in a number of different ways by the relevant views depending on how they conceive mental phenomena. What is perhaps even more surprising is that the second assumption has been the subject of relatively little systematic attention. Bearing in mind the ontological view I outlined in chapter 3 and my remarks on experience and causation in chapter 5, I sketch how the ontological framework discussed in previous chapters bear on the mind-body problem. In a similar vein, I explore the relevance of that framework for an epistemological feature of perceptual experience that may be termed its assertive character. Experiential assertivity hints at the fact that, whether veridical or hallucinatory, perceptual experiences assert or tell us how the world is. This feature is phenomenologically as well as epistemologically significant, for it determines not only part of what it is like to undergo perceptual experiences, but also part of what it means for perceptual experiences to justify beliefs, among other propositional attitudes. As a number of writers have pointed out, however, the assertive character of perceptual experience—or perceptual assertivity for short—poses an important challenge: what does it literally amount to claiming that perceptual experiences are assertive? After all, the notion of assertion is primarily applied to sentences, utterances, or perhaps intentional agents—not to mental states like perceptual experiences and such. So, no matter how intuitive the thought that perceptual experiences are assertive might be, it seems to be a metaphorical or analogical one in need of further specification. In the light of this challenge, I outline a more or less literal story of perceptual assertivity along functionalist lines, and then briefly describes how it bears on an ontology of experience like ENH, and vice versa. Finally, I move in the opposite direction, as I outline ways in which the relevant ontological framework need not interact with what may otherwise seem thematically adjacent discussions. For this purpose, I examine Matt Soteriou’s suggestion to the effect that the temporal phenomenology of experience hints at the type of temporal profile which experience itself should exhibit. Indeed, Soteriou’s brand of EH is to a good extent driven by its alleged ability to accommodate certain phenomenological facts about
Applications beyond the Ontology of Perception
97
temporally extended experiences. This line of reasoning crucially rests on two general thoughts: (i) that the previous facts are nonnegotiable features of perceptual experiences, or at least of the way we think about them and (ii) that ontological views of perceptual awareness have a bearing on such facts. Without (i), it would be unclear why the relevant facts should be accommodated by a philosophical account at all. Without (ii), it would be unclear why an ontological exploration of perceptual experiences should take those facts into account, even if they are nonnegotiable features of what it means to be perceptually aware of the world. These two ideas jointly ground the thought that an ontology of perception should be able to accommodate a particular phenomenological picture of temporally extended experiences. Although I do believe that (i) and (ii)—specially the first tenet—may be challenged, this section proceeds on the assumption that they are correct. Against this backdrop, I critically assess Soteriou’s phenomenological motivation by suggesting that perceptual states do not necessarily accommodate the relevant phenomenal features. This line of criticism is intended to ground the idea that the temporal phenomenology of experience—the phenomenological aspects Soteriou focuses on, at any rate—may be silent on its corresponding ontological structure. Of course, this list is far from exhaustive. But it does, I believe, convey a sense of how views in the ontology of perception could affect what might otherwise seem unrelated discussions in the general metaphysics of mind, the epistemology of perception, and the phenomenology of temporal experience. Likewise, I come nowhere near providing a specific take regarding any of the previous subject matters: instead, the present remarks only strive to give a rough taste of the ways in which different types of discussion could be reshaped by an ontological framework where perceptual experiences are not necessarily individuated in terms of their intentional or phenomenal features. PHILOSOPHY OF MIND SANS ONTOLOGY Contemporary discussion of the mind-body problem—that is, the metaphysical question how mental phenomena exist in the natural world—has been driven by a systematic reaction against a Cartesian doctrine of psycho-physical interaction up to that point shared not only by philosophers but also by scientists of the mind.1 Although it is unlikely that Descartes himself regarded the mind-body problem—(MBP), from now on—as a serious issue in his philosophical agenda at all, he did inspire a profoundly influential stance on that issue when undertaking the ambitious project of justifying human knowledge as a whole. In extremely general terms, such a stance involves two theses:
98
Chapter 6
1. Mind and body are sui generis substances. 2. Mind and body interact with each other via relations of efficient causation. (1) is a concise way of expressing that our mental lives are constituted by substances or objects the nature of which is radically different from that which defines material bodies and their constitutive parts—for example, the central and the peripheral nervous system. This idea captures what is traditionally known as the dualist component of the Cartesian view. (2), meanwhile, captures the thought that our mental lives and the spatio-temporal world mutually interact with each other via relations of efficient or productive causation. This thesis captures the causal component of the relevant doctrine. Cartesian dualism has thus featured in contemporary discussions of MBP as a theory according to which mind and body are substances of different kinds, but nevertheless capable of causally interacting with each other. Modern work on MBP has by and large turned on three forms of psychophysical materialism: behaviorism, the mind-brain identity theory, and functionalism. In spite of their mutual differences, these views share two key elements: on the one hand, a systematic rejection of Cartesian dualism, and, on the other, the assumption that mental and physical phenomena must be somehow related within a causal framework. Although apparently harmless, the latter assumption is implemented in a number of different ways by the relevant views depending on how they conceive mental phenomena. To expand on this point, let us briefly examine how each stance approaches the causal structure of mental phenomena. Behaviorism encompasses a family of theories that share the common claim the mind may exhaustively be conceived in behavioral and dispositional terms (cf. Farrell, 1950; Skinner, 1953). For example, a visual experience of grass and plants would not be an internal state or process that is logically or ontologically independent of its triggering physical stimuli or its ensuing behavioral effects; instead, it would be constituted by the set of such (overt and covert) behavioral responses. Or again, the pain ensuing from touching a hot coffee maker would not simply be an introspectively accessible sensation, but the set of behavioral responses that follow touching that object. Behaviorism thus denies that our mental lives are sui generis particulars that causally interact with the spatio-temporal physical world that we know and love; instead, the mind would best be conceived in terms of the behavioral or otherwise dispositional outcomes we do or could manifest in virtue of our informational interaction with the environment. The behaviorist stance is no doubt driven by methodological considerations; by specifying mental phenomena in behavioral and dispositional terms, behaviorism
Applications beyond the Ontology of Perception
99
models our psychological lives as observable—and hence, scientifically quantifiable—phenomena. A behaviorist response to MBP is radical not only because it rejects a particularist understanding of mental phenomena; it also downplays the belief, shared by other materialist views as well as a fairly intuitive understanding of the mind, that mental phenomena are causally efficacious. In other words: if correct, behaviorism seems to imply that (2) cannot be true. The behaviorist for sure concedes that the natural world has a causal impact on our mental lives; our behavioral responses and dispositions result from the world’s causal action on our bodies in general and our nervous systems more specifically. However, to the extent that mental phenomena are not particulars—whether substances or events—on the behaviorist outlook, it becomes unclear how they could in turn cause physical effects. Of course, the sole fact that behavioral responses and dispositions are not efficient causes or effects, does not imply that they cannot be causally relevant in some other sense. But since neither the behaviorist nor other materialist players seriously considered that other causal notions could be at play in an assessment of MBP, denying the causal efficacy of the mental actually amounted to denying its causal relevance altogether. This is a route that even materialists found unappealing, whereby subsequent proposals sought alternative ways of articulating a non-Cartesian view of the mind that incorporated (2) in some shape or form. Facing the crossroad thus opened by behaviorism, psycho-physical materialism evolved in two different directions: whereas one strand reinforced the conceptual link between mind and brain, another one went on to put some distance between our mental lives and their underpinning organic basis. The mind-brain identity theory—also known as central-state materialism or Australian materialism after its better-known champions, J.C.C. Smart and U.T. Place—is a view of the first type and the immediate historical response to the behaviorist stance (cf. Feigl, 1967; Place, 1956; Smart, 1959). Although identity theorists hold behaviorism in high esteem for avoiding a Cartesian mental ontology as well as stressing the intimate link between mind and behavior, they ultimately part company with it for diluting our understanding of mental phenomena as internal, causally efficient events. To accommodate the latter element, the identity theory conceives mental processes and states as brain processes and states. While intuitively materialist, this account also manages to save the thought that mental phenomena are internal, causally efficient events; neural phenomena are caused by the external informational inputs, but are also capable of producing subsequent behavioral outputs. Hence, the identity theory takes our mental lives to be internal without resorting to the mysterious ghosts of Cartesian dualism—mental phenomena are
100
Chapter 6
internal simply in the sense that they are intra-cutaneous or intra-cranial. And while such events are not so easily observable, they can be objectively studied by disciplines such as cognitive psychology and neurobiology. The second materialist strand is famously instantiated by functionalist views of the mind, a family of approaches that strived to avoid binding our general understanding of the mind to its corresponding neurobiological basis (cf. Putnam, 1967; Lewis, 1980). For sure, functionalists do not deny that the existence of mental phenomena is materially grounded on a given neural or otherwise organic system. The relevant point is that the way we conceive such phenomena is relatively independent of the particular neural system they happen to be implemented in. This suggestion is not only grounded on the intuitive fact that our familiarity with mental concepts well preceded the historical appearance of systematic sciences of the brain (e.g. neurobiology and cognitive psychology) but also on what is known as the multiple realizability of the mental. In a nutshell, the latter thought may be expressed in terms of a crude yet illustrative comparison: although the neural constitution of octopi is striking insofar as it follows a different evolutionary path than ours, their behavior is sufficiently sophisticated for allowing the ascription of psychological features such as perceptual experiences, sensations, among other cognitive phenomena.2 Cases along these lines suggest that there is an important sense in which psychological attributes are not bound to a specific neural setup; in principle, neurally heterogeneous species like human beings and octopi could nevertheless share aspects of mentality. Plausible as it is, this suggestion naturally raised the following challenge for the identity theory: how could our minds be identical to our nervous systems if organisms endowed with very different neural architectures are also subjects of psychological attributes? Functionalists themselves went on to specify mental phenomena in terms of their causal profile or structure rather than in terms of their material basis. For example, pain could be defined across different species in terms of types of stimuli that typically cause it as well as the overt or covert responses that tend to follow it. After all, just as the same software may be run on different hardwares—a metaphor functionalism owes much to—those causal structures that define psychological processes and states could also be instantiated by different neural mechanisms. This stance would by no means imply that mental phenomena could exist apart from a given material basis; it only avoids tying them down to a specific kind of neural system. To the extent that functional views reject a Cartesian view of the mind but simultaneously avoid reducing our mental lives to a particular neurobiological narrative, they are suitably characterized as examples of nonreductive materialism—a general trend of thought that, in spite of functionalism’s controversial history, has remained popular in the mainstream literature on the nature of the mind.
Applications beyond the Ontology of Perception
101
Like its behaviorist counterpart, functionalism departs from the more traditional paradigm of mental causation; after all, it conceives mental phenomena in general as states defined by a set of causal relations within a cognitive landscape. So conceived, they do not intuitively fit the category of events, that is, the kind of entities that could stand in relations of efficient causation with other protracted or quasi-instantaneous events. A couple of analogies help illustrating this point. By relying on a purely efficient understanding of causation, accounts like the identity theory conceives mental and physical events as dancers or perhaps billiard balls capable of interacting with each other. For the functionalist, meanwhile, a better analogy for expressing the interaction between mental phenomena and the nervous system is the relationship between the act of driving and its underpinning players—that is, the driver herself, the engine, and so on. In other words, the functionalist does not conceive of mental and physical phenomena as items of the same specific ontological category; as such, it is unclear how this could causally interact with each other when the only available resource to think about such a connection is the notion of efficient causation. To sum up, the identity theory is practically the only option that fleshes out the relation between the physical and the mental in terms of efficient causation. Since the latter causal notion typically ranges over events, however, it does not carry over to the behaviorist and the functionalist approaches quite so smoothly. Behaviorism conceives mental phenomena as sets of inclinations (overtly or covertly) to behave in complex ways, whereas functionalism models our mental lives in terms of structural, multiply realizable states. To the extent that neither inclinations nor states are episodic, it is unclear how, according to a behaviorist or a functionalist proposal, mental phenomena could stand in relations of efficient causation to other elements of the natural world. Be that as it may, the overall debate devoted virtually no attention to the question how our background causal assumptions bear on our understanding of mental phenomena. Instead of pausing to examine the ontological scaffolding that holds MBP in more detail, the mainstream literature simply moved on to address a different set of problems. Interest in the themes at the heart of the mind-body problem has actually persisted thanks to more recent debates about our understanding of mental phenomena vis-à-vis the causal closure of the physical (e.g., Kim, 1997, 2000). But again, in spite of its apparent emphasis on causation, the latter discussion did very little besides taking the notion of efficient causation for granted, an assumption that in turn seems to fate mental phenomena to become redundant vis-à-vis the causally closed chain of physical events. Since physical events are in principle exhaustively interconnected to each other via relations of efficient causation—as per the principle of the causal closure of the physical—and efficient causation is
102
Chapter 6
the only explanatorily relevant notion brought to bear in the debate, mental phenomena unavoidably struggle to find a place within the causal mosaic of natural events. In a complete, closed causal framework of physical events, mental phenomena quickly become additional and perhaps ultimately unnecessary ingredients. In line with the thoughts unpacked throughout chapter 5, these general remarks suggest that contemporary philosophy of mind has paid very little attention to questions regarding the ontological structure of our mental lives. In a general but relatively familiar sense of ontology as the philosophical study of reality in its most general terms, such an ontological indifference already becomes salient in the motivations at play throughout modern discussions of MBP. Whereas a writer like Descartes is invested in developing a metaphysics of mind in order to ground his more ambitious epistemological building, behaviorism and the identity theory were driven by methodological rather than ontological considerations. For instance, it is quite telling that Skinner took neurobiological states and processes to be as objectionable as immaterial mental substances; his resistance toward such items had more to do with the challenges that their quantifiable or otherwise objective study posed at the time, than with their potentially problematic ontological status—the very existence of the central nervous system being uncontroversial, after all. Again, the identity theory was inclined to reject Cartesian dualism on methodological grounds concerning its underpinning and understanding of philosophy as a discipline aiming to clarify the conceptual scaffolding that holds our scientific knowledge of the natural world—philosophy of mind being accordingly understood here as an attempt to shed some light on the conceptual framework at play throughout the modern cognitive sciences. Against this backdrop, the previous rejection does not have so much to do with Ryle’s classic critique of the “ghost in the machine,” as with a naturalist reluctance to allow for “nomological danglers” into a respectable scientific world-view—that is, items that resist integration into such a worldview via psycho-physical laws. In short, the identity theory does not seem to take issue with the Cartesian tradition on account of ontological considerations: instead, its driving considerations concern the methodological relationship between science and philosophy, on the one hand, and, on the other, the apparent absence of psycho-physical laws that latched a Cartesian metaphysics of mind onto the natural spatio-temporal world we know and love. The aforementioned sense of ontological indifference perhaps has shaped for the worse not only the standard way of understanding MBP but also philosophical discussions about more specific aspects of the mind, for example, perceptual experiences. With the previous depiction of MBP in place, I believe there is a relatively straightforward way of relating that debate to an ontological framework like
Applications beyond the Ontology of Perception
103
ENH via the remarks on experience and causation unpacked back in chapter 5. As previously pointed out, behaviorist and functionalist accounts conceive mental phenomena in dispositional or otherwise relational terms. In spite of their specific differences, both strategies share at least two important themes: on the one hand, stressing the intimate link between our mental lives and (overt or cover) behavior, and, on the other, doing so by specifying mental phenomena in nonepisodic terms. This latter theme is certainly problematic when psycho-physical causation is modelled along the lines of efficient interaction among events. Against the backdrop of the foregoing discussion, however, both approaches reemerge as inspired even if somewhat misunderstood, like few other systematic accounts, they suggest that certain elements of the natural world—for example, mental as well as physical states, properties, among others—need not be causally understood in terms of the place they occupy within a mosaic of efficiently related events. Since mainstream philosophers and scientists of the mind by and large neglected to address their own underpinning causal assumptions or simply took a Cartesian understanding of causation for granted, it is not surprising that even behaviorists and functionalists failed fully to appreciate the previous insight at the time. Again, as I pointed out at the end of chapter 5, this general diagnosis could potentially address a curious historical question, namely, why the behaviorist and the functionalist proposals refuse totally to abandon the scientific and philosophical landscape even long after their heyday. The resilience that both accounts have exhibited throughout recent decades, probably reflects a tacit conflict between their core insights and a causal framework against which they cannot be made fully intelligible. This conflict does not arise in the ontological framework this book has introduced, as it accommodates the thought that mental phenomena like perceptual experiences could exist in the natural world without being conceived as events that should be efficiently related to other mental and physical events. EXPERIENTIAL ASSERTIVITY AS A MARK OF OCCURRENT STATES Let us join our friendly sailor Jim again. On a cloudy night, he looks for a star from the ship’s main deck. At one point, the sky clears up, and our vigilant subject suddenly spots a bright star. Jim stares at the star from t1 to tx, an interval throughout which the starry object asserts its existence in Jim’s temporally extended visual experience. Again, since his current visual experience tells him what the world is like, Jim will be disposed to form certain beliefs and behave in ways related to the information his experience is about. Like Jim’s visual experience, perceptual experiences are typically assertive,
104
Chapter 6
whether veridical or hallucinatory, they tell us or assert how our perceivable surroundings are. Saying that experience tells us what the world is like is naturally a bit of poetic license. But it does intuitively point at a notion of assertivity or assertive character widely featured in the epistemological literature. A few sample formulations. Richard Heck claims that perception “purports to represent how the world is [ . . . ]. Even when the world appears to be a way I know it not to be—when a stick I know to be straight looks to be bent when I partially immerse it in water (to use a tired example)—–it still looks as if the stick is bent” (Heck Jr, 2000: 508). In the thriving debate on phenomenal conservatism, assertivity has been brought up as a key feature of seemings, the latter being understood as a general category of mental states—one that embraces perceptual experiences, beliefs, recollections, and intuitions—capable of justifying our beliefs about the perceivable environment. Michael Huemer, a prominent phenomenal conservatist, describes perceptual seemings (i.e., perceptual experiences) as assertive mental representations (cf. Huemer, 2001: 53–54), in the sense that “perceptual experiences represent their contents as actualized” (Huemer, 2001: 66; also cf. 2007, 2013; Tucker, 2010: 530; Chudnoff, 2012; Moretti, 2015; and most articles in Tucker, 2013b).3 Again, voicing one way in which the coherentist might think of the justifying role of perceptual experience, James Pryor states that “[t]here are some ways of representing the proposition P that purport to be saying how the world is, and other ways that don’t. When a state represents that P in the first way, we can say that the state assertively represents that P” (Pryor, 2014: 209). Finally, Kathrin Glüer has stated that experiences “represent the world as actually being the represented way, as actually fulfilling their condition of correctness or truth. Such states have been called “committal” (Burge, 2003: 452) or “stative” (Martin, 2002: 386f.), but the maybe most suggestive metaphor characterizing these attitudes is that of assertion” (Glüer, 2018: 4; also cf. 2009: 306). It is relatively clear that assertive character is an epistemologically significant feature because it partially captures the justificatory and reason-providing role of perception. To justify beliefs about our immediate surroundings, it may be necessary though not sufficient for perceptual experiences to have propositional or otherwise representational content (cf. McDowell, 1994); after all, mental states like desires may be propositional without playing any justificatory or reason-providing role vis-à-vis belief and action. But a story of the rational role of perceptual experience has to invoke an additional ingredient which concerns not so much what worldly items perceptual experiences represent, but how they represent them—that is, an ingredient concerning experiential attitudes rather than experiential contents. The notion of assertive character comes into play at this stage, for it purports to stand for the
Applications beyond the Ontology of Perception
105
relevant missing ingredient, perceptual experiences would justify perceptual beliefs and actions not only on account of heir propositional content but also on account of their assertive component. So although its nature is rarely discussed, assertive character is a notion of consequence in the epistemology of perception. That said, the problem is that it seems unclear what the aforementioned feature of assertive character exactly amounts to. Michael Tooley expresses the general difficulty as follows: The term assertive, as ordinarily used, applies only to people and is definable in terms of the verb “assert.” The only things that can literally assert anything, then, are things that can have intentions and that can perform actions. Mental states, then cannot literally assert anything, and if one characterizes a mental state as assertive, that is a metaphor that needs to be cashed out in terms of some literally true characterization. (Tooley, 2013: 310; also cf. Heck Jr, 2000: 509n.29; Briesen, 2015: 2243, 2258; Glüer, 2018: 4)
In principle, only intentional subjects and perhaps sentences say—and thereby assert—how thing might be. Mental states like perceptual experiences and beliefs, however, constitute very different types of items from those that usually say or bear witness about anything in the world. In particular, the sense in which Jim’s visual experience asserts how the world is, does not seem to be the same as that in which Jim himself or a sentence assert something. More generally, mental states are not the sort of things that report or assert anything at all—or at least not in the same sense in which people do so. As such, extrapolating features deriving from a theory of speech acts to mental states seems a controversial move. Sure there may be a metaphorical or analogical sense in which perceptual experiences tell us what the world is like. But such a metaphorical understanding does not specify what it exactly means for perceptual experiences to be assertive. Since assertive character does not primarily scope over mental states, it is far from clear how it should be used in that context. Hence, it seems legitimate to press the question what it literally means for perceptual experiences to be assertive. To draw on what Glüer has called the “Attitude Problem,” the same question may be phrased as follows: in virtue of what feature of the way in which mental states, like perceptual experiences or beliefs, present or represent the world to us, do we claim that they assert what things are like?4 This question presses the very intelligibility of ascribing assertive character to mental states like perceptual experiences. That said, let us specify the notion of perceptual assertivity a bit further. In line with a metaphorical understanding of assertive character, it seems possible to get a general grip on the notion of perceptual assertivity by
106
Chapter 6
means of comparisons and contrasts between intuitively ostensible examples of assertive and nonassertive attitudes. For example, it seems natural to align experiences and beliefs along the same side of the assertive/nonassertive divide (cf. Heck Jr., 2000: 508n.29; Martin, 2002: 386–387; Glüer, 2009: 307). Sure both psychological categories also differ in a plethora of respects. To the extent that they are events where one takes a stance toward the world—as opposed to states where a stance on worldly items has already been taken—experiences are perhaps more like judgements than like beliefs. Again, perceptual content may not need be propositional or conceptual, whereas doxastic content is concept-dependent. In general, perceptual experiences and beliefs have different causal sources: in a fairly intuitive sense, experiences are caused by our surroundings, while beliefs are typically caused by perceptual experiences or other propositional attitudes. But whatever their differences, they do seem to share at least this much— asserting how things are. When a subject S believes that p experiences a given state of affairs X, her belief or experience asserts p or X as being so. Whereas some propositional attitudes may be such that their respective subject merely entertains what is presented to her, perceptual experiences and beliefs commit their corresponding subject to take the world the way it appears to be.5 In the previous general yet intuitive respect, perceptual experiences and beliefs stand apart from propositional attitudes such as thinking, desiring, and hoping: when the latter attitudes represent p, the relevant subject need not take p to be true; in fact, certain types of attitudes (e.g., desiring) may presuppose that the relevant subject do not take p to be so. A subject may desire or hope there to be a bright star in a certain area of the sky, but none of these instantiated attitudes implies, as far as the respective subject is concerned, that there is actually a star at the distance. Jim may spot a bright star at to, after which he experiences the luminous object and may acquire beliefs the content of which could be expressed as “That is bright,” “There is a star,” “Lo and behold, a star!”. Before t0, Jim could only relate to that object by looking for it, hoping to spot it, imagining a bright star to be roughly where he expects to find it, and so on. After t0, Jim’s visual experiences and beliefs will assert the star’s presence in his environs. Unlike hopes and desires, beliefs and experiences assert the existence of the items they (re)present. Hence, one could pick up on perceptual experiences and beliefs as paradigmatic cases of assertive attitudes, so as to contrast them against intuitive samples of mental states which do not tell us how the world is (e.g., desires, imaginings, assumptions). By a vulgar notion of assertive character, I understand here the combination of the aforementioned metaphorical conception plus our intuitive ability to classify mental states as assertive or as nonassertive. The possibility of thus
Applications beyond the Ontology of Perception
107
distinguishing samples of assertive and nonassertive states may constitute an important reason for thinking that perceptual assertivity is unproblematic. For ordinary purposes, a vulgar understanding of that feature may be all we need. However, if the notion of assertive character is supposed to fulfill a heavierduty role in epistemological and phenomenological stories of perception, Tooley’s demand for a more literal philosophical understanding of assertive character is legitimate; the vulgar notion of assertive character will not do. Now, while the vulgar notion seems insufficient for philosophical purposes, it does impose an important constraint on potential philosophical accounts of perceptual assertivity: an account of perceptual assertivity should deliver the right psychological classifications (cf. Briesen, 2015). That is, a literal story of assertivity should not only aim to make sense of what it means for perceptual experiences to be assertive; it should also suitably extrapolate the relevant criterion to those psychological kinds which we would intuitively classify as assertive (e.g., beliefs, pains), and fail to do so for intuitively nonassertive kinds (e.g., desires, hopes, imaginings). Consider the following functionalist criterion outlined by Jochen Briesen: (1) A propositional attitude Ap of a subject S (where p is a proposition concerned with the world around us) assertively represents the world iff–in the absence of counter-evidence with respect to p—Ap tends to be accompanied by the disposition of S to judge p more likely to be true than p. (Briesen, 2015: 2247)6 The thought is that perceptual experiences and beliefs with content p are assertive iff, in the absence of relevant counterevidence, the relevant attitudes result in dispositions to judge or believe that p is the case. So while Jim’s visual experience of a star is assertive because it could incline him to believe that a certain bright object is thus presented to him, his wish for a bright star to become visible need not prompt him to believe that one will do so. (1) thus provides a less metaphorical understanding of assertive character, and, at the same time, delivers a number of correct psychological classifications. While the previous proposal seems plausible to me, it should probably be modified in two respects. First, it is far from clear that a characterization of the assertive character of mental states ought to resort to the ways in which evidentially relevant mental states could affect the allegedly assertive attitudes. A person may, for example, assert p no matter how much evidence or counterevidence there be for or against p. Or again, the assertive status of a statement need not change on account of existing counterevidence—although whether it would be rational to believe in the truth of that statement in spite of counterbalancing evidence is a whole different question. As for perceptual experiences, their belief-independence suggests that they may assert
108
Chapter 6
how things are no matter what evidence there might be against what they present a subject with. Likewise, a belief that p need not be any less assertive—although it would be an irrational one—in circumstances where there is persuasive counterevidence against p. Second, by exclusively focusing on the causation of dispositions to form some specific beliefs, (1) seems too restrictive regarding the informationally significant ways in which assertive attitudes could affect our epistemic economy. Although a functionalist description certainly has to invoke the causal impact that the relevant states could have on doxastic dispositions, it should also take into account how they bear on (linguistic and nonlinguistic) behavior as well as other mental states. After all, beliefs are not typically instantiated for their own sake, but for the role they play guiding behavior—in some special cases, an assertive attitude with content p might even determine behavior without raising a corresponding belief with content p (e.g., Heck, 2000: 508n.27). In the light of the previous qualifications, I propose to revise (1) along the following lines: (2) A propositional attitude Ap of a subject S (where p is a proposition concerned with the world around us) assertively represents the world iff Ap tends to be accompanied by the dispositions of S to judge/believe p more likely to be true than p, or (linguistically and nonlinguistically) to behave accordingly. According to (2), the import of being assertive is causally to determine our mental states and behavior: what it means for a psychological attitude about p to be assertive is for it to dispose a subject to instantiate mental states (e.g. beliefs, desires, fears) and act on the explicit or implicit assumption that p is the case. What a visual experience of a bright star does for Jim is to dispose him to think, feel, and act taking into account information about the star he was looking for. In principle, this is all perceptual assertivity might amount to. (2) seems more plausible for a number of reasons. First, it accommodates the intuitive classifications underpinning the vulgar notion of assertivity. Second, it injects general but literal meaning to the ascription of assertive character to mental states—so it meets Tooley’s challenge from intelligibility. And third, it provides a straightforward way of understanding the relevant feature.7 That is, on the natural assumption that mental states may be characterized by specifying the causal role they play in a subject’s wider psychological life, the present proposal outlines the simplest way of understanding the relationship between the assertive character and the functional role of such mental states: they are identical. To the extent that it does not rely on conceptual resources other than the notions of assertive character and
Applications beyond the Ontology of Perception
109
functional role, a functionalist story of perceptual assertivity is thereby quite austere. By contrast, any other proposal turns out to be less economical: to begin with, it has to concede that perceptual experiences play some causal role within our epistemic or psychological lives, but in addition to that step, it has to invoke an extra ingredient in order to make literal sense of perceptual assertivity.8 This second step naturally constitutes a challenge that the functionalist need not address. Building on a vulgar or relatively intuitive notion of perceptual assertivity, one could thus argue that perceptual experiences are assertive in the sense that they affect other mental states and behavior in informationally relevant ways. While modest, this philosophical notion of perceptual assertivity accommodates a number of intuitive psychological classifications that underpin the vulgar notion and models assertive character in a way that different types of mental states could actually implement. But how is a discussion about the assertive character of experiences related to the ontological debate I have unpacked throughout earlier chapters? The impact is, I believe, mutual. Let me unpack this point a bit. To begin with, perceptual assertivity belongs to a family of epistemological and phenomenal features that are typically fleshed out as aspects of perceptual content. As heated debate in this tradition carries on, the ontological framework outlined here opens up a different landscape that could enrich the previous landscape of competing views. To the extent that it approaches perceptual experiences in a way that does not demand individuating them in terms of their contents or their phenomenological traits, a study of perceptual assertivity could strive to describe—if not account for—such an epistemological feature as an aspect of experiential states. How would such a story run? This is where the aforementioned functionalist story proves itself relevant, for it specifies what kind of features a given type of mental states would implement in order to classify as assertive. In other words, while writers like Vendler and Soteriou provided the categorial resources to categorize perceptual experiences as mental states of a yet to be specified kind, statements like the functionalist description of perceptual assertivity provide the necessary specification; that is, (2) could be taken to specify the mental states in terms of which the ENH categorizes perceptual experiences. While such a functionalist story like (2) could thus complement an ontological view like ENH, the ontological framework provided by the latter stance could also provide a horizon of intelligibility that functionalist views are typically in dire need of. After all, standard criticism against functionalist views has less to do with problems of internal coherency than with an intuitive inclination to question why a description of mental phenomena, in all their subjective depth, should be shoehorned into the sort of causal, structural story that the functionalist typically provides. In other words, standard
110
Chapter 6
objections against functionalism do not usually hint at some deep-rooted form of internal coherency, but at the apparently implausible consequences following from conceiving mental phenomena according to its depiction. This is what I mean by saying that functionalist views seem to lack a framework of intelligibility; they seem to lack a prior understanding of mental phenomena capable of helping us to understand why a structural characterization is good enough to resist classic objections raising from philosophical zombies or Chinese multitudes. The ontological scaffolding introduced by the present book could provide just that. How so? To address this question, bear in mind that standard objections against functionalism rely on the possibility of separating a description of the functional profile of mental phenomena from descriptions involving their intentional and, more importantly, phenomenal features. That is, they seem to rely on the possibility of comparing functionally similar phenomena—say, F and F*—that nevertheless varied in the following sense: whereas F has certain phenomenal or otherwise subjective features we hold dear to the relevant type of mental phenomena, F* fails to instantiate them. This way of thinking about mental phenomena probably reflects a way of understanding their ontological or categorial profile, though. And while my sketchy remarks by no means provide a robust picture of the underpinning ontological framework at play here, the latter probably involves framing the relevant mental phenomena as potential subjects of properties like the aforementioned functional and phenomenal features. At any rate, I cannot see how else the relevant objections could even begin taking off; an ontology of mental phenomena as items or entities that may share potentially matching or mismatching functional and phenomenal properties seems to underpin those objections. That said, the framework provided by ENH suggests that perceptual experiences are not the type of mental phenomena that answer straightforward cardinality questions. Accordingly, experiences may be thought of in a way that constitutively bind them to the relevant functional or structural traits: the relevant functional profile would not pick up on certain properties that experiences could own or lack; it would capture a specification of the relevant occurrent states. ENH would thereby provide a framework for intelligibly binding perceptual experiences to their functional profiles, in the sense that they would clarify why a zombified subject of experiences could be imagined, but not actually conceived. EXPERIENTIAL ONTOLOGY AND TEMPORAL PHENOMENOLOGY The previous sections described how an ontological framework along the lines of ENH could bear in such diverse philosophical issues as the
Applications beyond the Ontology of Perception
111
mind-body problem and the assertive character of experiences. Going now in the opposite direction, the present section will illustrate ways in which that framework need not interact with what may otherwise seem thematically adjacent discussions. For this purpose, I examine Matt Soteriou’s suggestion to the effect that the temporal phenomenology of experience hints at the type of temporal profile which experience itself should exhibit. Indeed, Soteriou’s brand of EH is to a good extent driven by its alleged ability to accommodate certain phenomenological facts about temporally extended experiences. This line of reasoning crucially rests on two general thoughts: (i) that the previous facts are nonnegotiable features of perceptual experiences, or at least of the way we think about them; and (ii), that ontological views of perceptual awareness have a bearing on such facts. Without (i), it would be unclear why the relevant facts should be accommodated by a philosophical account at all. Without (ii), it would be unclear why an ontological exploration of perceptual experiences should take those facts into account, even if they are nonnegotiable features of what it means to be perceptually aware of the world. These two ideas jointly ground the thought that an ontology of perception should be able to accommodate a particular phenomenological picture of temporally extended experiences. Although I do believe that (i) and (ii)—specially the first tenet—may be challenged, this section proceeds on the assumption that they are correct. Against this backdrop, I critically assess Soteriou’s phenomenological motivation by suggesting that perceptual states do not necessarily accommodate the relevant phenomenal features. Rather than supporting ENH, this line of criticism is intended to ground the idea that the temporal phenomenology of experience—the phenomenological aspects that Soteriou focus on, at any rate—may be silent on its corresponding ontological structure. This line of response will thus set up the stage for chapter 7, where I argue at greater length for a structurally similar claim—namely, that considerations deriving from the way we talk about perceptual experiences do not necessarily determine our ontological understanding of such phenomena. The phenomenological facts Soteriou is concerned with are the following: first, that the temporal properties of perceptual experiences are transparent; second, that certain objects of perception appear to be temporally continuous; and, third, that a subject’s perceptual experiences apparently share the temporal continuity exhibited by the objects she is perceptually aware of. I elaborate on each aspect next. The general thought behind the idea of transparency is that reflection on perceptual experiences is not primarily sensitive to features of the experiences themselves, but only to features of the worldly items or states of affairs our perceptual experiences are of. For example, if I reflect on my current visual experiences, I do not attend to the very mental phenomena opening the visible world to me, but to the items I see, for example, a laptop,
112
Chapter 6
a table, a few DVDs—in this sense, experiences themselves are “invisible” or “transparent” to reflection. Again, when Jim introspects his visual experiences during t1–t10, he can only attend to a star, clouds, among other worldly features: the experiences or the experiential properties in virtue of which the relevant scene is presented to him, however, remain elusive to his attentional capacities. As illustrated here, the idea of transparency involves two specific claims: first, that reflection on perceptual experiences refers to the worldly items such experiences are of or about; and, secondly, that it refers to nothing more than those items (cf. Martin, 2002: 378; Crane & French, 2017). Thus understood, the idea of experiential transparency is controversial in virtue of the second, negative claim. Sensitive to this point, Soteriou identifies a “weaker version” of the claim, “when one attempts to attend introspectively to what it is like for one to be having a perceptual experience it seems to one as though one can only do so through attending to the sorts of objects, qualities and relations one is apparently perceptually aware of in having the experience” (Soteriou, 2011: 488). This formulation is weaker precisely because it constrains the negative claim behind the idea of transparency: instead of holding that experiential features are wholly inaccessible to our attentional capacities, it only states that we have indirect access to such features, for it subordinates the introspective awareness of mental items to the possibility of attending to the relevant experienced items. This subtle qualification is crucial, for, as I shall explain in a moment, it carries over into Soteriou’s formulation of the temporal transparency of perceptual experiences. Given the previous considerations, Soteriou goes on to unpack the idea of temporal transparency as follows: [ . . . ] the temporal location of one’s perceptual experience seems to one to be transparent to the temporal location of whatever it is that one is aware of in having that experience. When one introspects one’s experience, the temporal location of one’s perceptual experience seems to one to be transparent to the temporal location of whatever it is that one is aware of in having that experience. Introspectively, it doesn’t seem to one as though one can mark out the temporal location of one’s perceptual experience as distinct from the temporal location of whatever it is that one seems to be perceptually aware of. (Soteriou, 2011: 589)
This formulation does not deviate from a traditional understanding of experiential transparency: when we introspect or reflect on our own perceptual experiences, the latter’s temporal properties are not obviously manifest to us; when we exercise our introspective capacities, the only temporal features which are immediately salient to us belong to the worldly items or states of affairs we are thus acquainted with. For example, when we decide
Applications beyond the Ontology of Perception
113
to introspect our current visual experiences of a person running across the park, it is natural to think that the temporal features we can attend to in virtue of attending to these experiences correspond to the perceived runner or her current activity. Of course, this does not mean that we or our experiences would be immune to temporal modifications; the point is that, although we and our perceptual occurrences are no doubt subject to the rule of objective time, it does not follow that we can access any of such modifications via introspection. So far, so good. Immediately after the above quote, however, Soteriou relates the temporal transparency of perceptual experiences to a somewhat different thought: Furthermore, it seems to one as though the temporal location of one’s experience depends on, and is determined by, the temporal location of whatever it is that one’s experience is an experience of. So, for example, when one perceives an unfolding occurrence (e.g., the movement of an object across space), it seems to one as though one’s perceptual experience has the temporal location and duration of its object, and it seems to one as though the temporal location and duration of each temporal part of one’s experience is transparent to the temporal location and duration of each temporal part of the unfolding occurrence one seems to perceive. (Soteriou, 2011: 589)
This remark constitutes a substantive addition to the previous line of reasoning: for, while the initial thought about transparency presses the inaccessibility of perceptual experiences via introspection, Soteriou now draws a link between the temporal properties of perceived objects and those of the corresponding perceptual experiences; in particular, he claims that the “temporal location” of the items we perceive governs the “temporal location” of our perceptual experiences. This claim qualifies the negative component behind the idea of transparency, for it suggests that reflection on what we are perceptually aware of reveals something about the temporal character of our perceptual experiences. Although Soteriou does not make the point explicit, I think this deviation should be read in the light of his “weaker version” of experiential transparency: if the background thought is not that perceptual experiences are inaccessible through first-person introspection, but only that the possibility of so accessing them depends on attending to the worldly items we experience; it would be reasonable to hold that the temporal structure of perceptual experiences is introspectively revealed to oneself by attending to the temporal structure of what is thus experienced. The second phenomenological aspect Soteriou picks up on concerns “the apparent temporal extent of the objects of perceptual experience” (Soteriou, 2011: 489). To illustrate this point, he turns to the perception of temporally
114
Chapter 6
extended events, related to the fact that such happenings have temporal parts intimately related to each other, subjects may only experience them by being aware of their parts. True: it is possible to perceive duration-less events or parts of temporally extended ones. But what Soteriou intends to reject here is the possibility of perceiving temporally extended events without perceiving temporally extended parts of them. As he puts it, “‘[i]f one tries just to attend to an instantaneous temporal part of the occurrence, without attending to a temporal part of the occurrence that has temporal extension: then one will fail” (Soteriou, 2011: 489). The third phenomenological aspect is closely related, for it also concerns the idea of temporal extension; whereas the second fact highlights the continuity or temporal extension of perceived items, Soteriou next invokes “the apparent continuity of conscious experience over time” (Soteriou, 2011: 490). Just as certain objects of perception seem to have temporal parts intimately related to each other, subjects undergo perceptual experiences constituted by closely connected temporal parts: “when one undergoes a conscious perceptual experience that fills an interval of time, each sub-interval of that interval of time is filled by some successive phase of that experience, and each successive phase of the experience shares a temporal part with some prior phase of experience” (Soteriou, 2011: 490). For example, when a subject stares at the second hand of a clock during a period of time, she also observes a temporally extended event or at least part of one. But, in addition to that, the continuity of the perceived event also seems to transpire into the temporal structure of the very experience she goes through, when she observes the second-hand, her visual experiences do not seem to bundle together as a set of temporally discrete items, but to coalesce into one continuous experiential stream. The remaining two phenomenological facts are thus closely related: while one highlights the temporal continuity of perceived items, the other stresses that of perceptual experiences themselves. With the previous features in place, it is relatively easy to spell out Soteriou’s phenomenological case for EH. The phenomenological picture emerging from the conjunction of these three facts—that is, (i) temporal transparency, (ii) the apparent continuity of perceived objects, and (iii) the apparent continuity of experiences themselves—allegedly grounds the hybrid form of EH he espouses. After all, Soteriou takes a pure form of EH—that is, something along the lines of O’Shaughnessy’s stance—to accommodate (i) and (iii), whereas ENH would only be equipped to incorporate (ii). Neither ontology of perception, he thinks, accommodates (i)–(iii) on their own. A hybrid stance would solve this problem; a view of perceptual experiences as states constituted by phenomenally conscious processes would allegedly incorporate the necessary resources to accommodate the aforementioned phenomenological triad.
Applications beyond the Ontology of Perception
115
Pace Soteriou, it is unclear to me what bearing phenomenally conscious processes have on the temporal transparency and the apparent continuity of perceptual experiences. According to him, perceptual processes could accommodate the first phenomenological fact because “one might maintain that each momentary temporal part of an experiential occurrence that unfolds over some interval of time presents some aspect of the environment as concurrent with it” (Soteriou, 2011: 492). Conceived along the lines of pure EH, perceptual experiences are temporally extended happenings of a processive kind; they are temporally structured events insofar as they have a beginning, a middle, and an end, and such parts are in turn realized by the different parts of processes of a phenomenally conscious kind. Soteriou’s point seems to be that if perceptual experiences are thus conceived, one could save the idea of transparency by taking each subinterval of a perceptual process to be transparent; that is, one could argue that each part of such a process presents a subject with different aspects of a temporally unfolding world, and hence, makes such worldly items accessible to introspection. If I get Soteriou’s line of reasoning right, it seems sound but somewhat thin. It does not, for example, unveil a necessary link between temporal transparency and irreducible processes. More generally, it fails to clarify how such processes implement that phenomenological feature; that is, EH fails to account for how perceptual experiences implement the feature of transparency. As far as I can see, Soteriou only shows that perceptual processes and temporal transparency are not mutually incompatible. But he fails to show how phenomenally conscious processes would specifically incorporate the transparency thought into our understanding of perceptual experiences. Granted: mutatis mutandis, ENH does not throw light on the link between perceptual experiences and temporal transparency either. From an explanatory point of view, perceptual states have no comparative advantage over perceptual processes. At this point, the relevant phenomenological fact does very little to incline our allegiance toward either ontological framework. Soteriou also resorts to the apparent continuity of temporally extended perceptual experiences. He writes as follows: This view [i.e. the hybrid form of EH he espouses] accommodates the idea that when one undergoes a conscious experience that fills an interval of time, each sub-interval of that interval of time is filled by some successive phase of that experience, and each successive phase of the experience shares a temporal part with some prior phase of experience—e.g., it can accommodate the idea that experience one undergoes from t1 to t10 can share a temporal part with experience one undergoes from t5 to t15, and it can do so without needing to commit to the idea that there is some one perceptual state of the subject that continues to obtain from t1 to t15. (Soteriou, 2011: 491)
116
Chapter 6
The background thought is that in virtue of their distinctive temporal structures, temporally extended processes and states have subparts which relate to each other in different ways. Process parts are intimately connected insofar as they jointly constitute a whole—that is, a determinate process or event— and necessitate each other; for Soteriou, each successive phase of a process shares a temporal part with a prior phase of the same process. Meanwhile, states exist wholly at each moment of time they obtain: hence, there is a sense in which their parts are not internally related to each other; each state part is complete on its own. The previous quote simply applies that distinction to our philosophical understanding of perceptual experiences: conceived as processes, these psychological items would be such that their temporal subparts are intimately intertwined; successive experience parts are mutually connected—that is, each successive phase of an experience shares a temporal part with some prior phase of the same experience—so as to constitute an organic whole. As far as I can see, Soteriou takes this ontological categorization to ground the apparent continuity of perceptual experiences, that is, the fact that temporally extended experiences seem continuous to their own subjects. Having said that, it is unclear to me how the previous remarks help to motivate EH. Soteriou has only established that, conceived as processes, the constituting temporal parts of perceptual experiences would behave as process parts—in particular, they would be intimately related to each other. But this ontological point has not automatically any phenomenological implications. For example, it does not automatically show that perceptual experiences seem to be temporally continuous to their respective subjects. One could imagine a world where temporally continuous experiences are presented to their owners as successive snapshots, that is, as temporally discrete items. To rule out this line of reply, it is necessary to invoke a controversial principle along the lines of Hume’s “all the actions and sensations of the mind [ . . . ] must in every detail appear to be what they are, and be what they appear” (Hume, 173940/2000: 1.4.2.7, SBN 190).9 However, Soteriou has not secured a principle along these lines, and as long as this issue remains pending, he cannot assume that a processive ontology automatically accommodates the apparent continuity of perceptual experiences. To wrap things up, Soteriou’s reliance on perceptual processes is to an important extent fueled by a phenomenological motivation: a processive ontology allegedly accommodates the temporal transparency and the apparent continuity of perceptual experiences. In relation to the first phenomenological aspect, Soteriou outlines how a story combining a processive view and temporal transparency could run, but does not show how or why perceptual processes exactly implement such a phenomenological feature. Concerning the apparent continuity of perceptual experiences, he only secures the ontological
Applications beyond the Ontology of Perception
117
point that the successive parts of perceptual processes are intimately related to each other: unless he proves that such processes have to appear to their owners the way they are—something which is by no means obvious—it is unclear how his ontological considerations show that a processive view accommodates the relevant feature. In short, I do not think that Soteriou’s phenomenological case for perceptual processes has enough traction. NOTES 1. Cf. Descartes, 1986, Meditations II & VI; Shapiro, 2007. For a historical survey of Descartes’s influence on early twentieth-century’s neuroscientists and cognitive psychologists, cf. Bennett & Hacker, 2012. 2. The same point is often made in terms of more exotic entities (e.g., robots or angels), but the latter raise the question whether they are logically conceivable at all. 3. Throughout this assessment, I keep my distance from seemings for at least two reasons. First, I find the ontological status of such attitudes a controversial but rarely addressed theme; for exceptions, cf. Tucker, 2013a; Tooley, 2013. For all I know, there might not be anything to seeming-statements beyond the fact of being qualified predicative statements about what we perceive or think (cf. Sellars, 1963; Hacker, 1987). Second, this chapter does not aim to deliver a story of epistemic justification, but only to address the challenge of fleshing out the feature of perceptual assertivity. While I am not certain what to make of debates turning around seeming-attitudes, I assume that the functionalist proposal here sketched could be incorporated into such discussions. 4. For Glüer, the attitude problem is the question “what precisely it is that makes an attitude into a propositional justifier—without by the same token making it into a belief” (Glüer, 2018: 4). On the assumption that belief constitutes a paradigmatic form of justifying or reason-providing mental state, the question turns out to be how one could conceive the justifying role of other psychological attitudes (e.g., perceptual experiences) without reducing them into beliefs. While Glüer herself thinks that conceiving perceptual experiences as beliefs is not such a bad idea (cf. Glüer, 2009), she does regard the attitude problem as an urgent one. Since she explicitly links the notion of propositional justifier to the idea of assertive attitude, her attitude problem pretty much overlaps with the question I address here. If any, the main difference between my question and her problem is that I remain neutral about a propositional or intentionalist understanding of perceptual content; for, while Glüer states that the “attitude problem is an unsolved worry for intentionalists thinking of experience as a propositional justifier” (Glüer, 2018: 4), I think that that problem is pressing no matter what stance you take on the content of perceptual experience. The questions what it means for a mental state to be assertive and whether a justifying or reason-providing mental state must have propositional content, are different (even if related) issues that a philosophical story of the rational role of perception has to address.
118
Chapter 6
5. Of course, other forms of intuitively assertive states come to mind, for example, somato-sensory experiences. I focus on perceptual experiences and beliefs here only for the sake of simplicity. 6. Although I shall stick to it for the sake of simplicity, Briesen’s terminology might raise two issues about which I remain neutral: on the one hand, whether, and in what sense, perceptual content is propositional; and, on the other, whether (1) and the variant I shall presently introduce are supposed to be reductive or constitutive claims. 7. For statements of assertive character inadvertently but naturally invoking functionalist terminology, cf. Martin, 2002: 388ff. and Glüer, 2009: 308. 8. For this reason, the burden of the proof seems to be on the nonfuncionalist’s side: she has to explain why we should prefer a more complicate story of perceptual assertivity rather than something like (1) or (2). Accordingly, rather than defending a functionalist view from nonfunctionalist alternatives, the truly pressing task is to provide a charitable formulation of what seems to be the most natural way of thinking about perceptual assertivity. 9. For a more recent formulation of a principle along these lines, cf. Phillips, 2010.
Chapter 7
Perceptual Experience and Language
TACKING STOCK Thus far, I have framed Heraclitean and non-Heraclitean views of experience in purely ontological or metaphysical terms. That said, some of the relevant figures in this debate have also brought linguistic considerations from the study of aspect—that is, the study of the “different ways of viewing the internal temporal constituency of a situation” (Comrie, 1976: 3)—to bear on the ontological question how perceptual experiences persist over time (cf. O’Shaughnessy, 1971, 2000; Soteriou, 2007, 2011, 2013; Crowther, 2009a, 2009b). While the relevant philosophical debate is primarily ontological, one could not deny that it borrows concepts and distinctions that linguists have systematically examined when discussing the notion of aspect—that is, a pervasive but widely varying feature from natural language to natural language. My worry is that, apart from rare exceptions (e.g., Gill, 1993), relatively little attention has been devoted to assess whether the way we talk about perceptual phenomena has any bearing on the ontological nature of perceptual experiences. The goal of this chapter is to argue that aspectual considerations are silent on how we should think about the temporal structure of perceptual experiences. I divide the present task into three parts. The present section sets the necessary background machinery to proceed with the set task: taking off from Zeno Vendler’s seminal work on verbs and their temporal structures, I briefly explain how elements deriving from the study of aspect have become relevant for the ontological analysis of perceptual experience. Then, I argue that grammatical tests for aspectual classification throw little light on the ontological classification of experiences, to focus the discussion, I specifically address what is widely known as the continuous test. Finally, I flesh out the thought 119
120
Chapter 7
that a number of key aspectual features—viz. atelicity, progression, and heterogeneity—fail to shape a distinctive ontology of perception. To begin with, then, let us recall that Helen Steward describes the intimate link between the notions of mind and time in the context of MBP as follows: “There is room for dispute about whether or not, and in what sense, mental phenomena are physical, whether they are spatially located, and whether they have subjects, and if so, what those subjects might be. [ . . . ] But there is no controversy about the temporality of mental phenomena–about the fact that they take place in, or persist through, time” (Steward, 1997). The temporal strategy, to borrow Steward’s terminology, constitutes a fresh and promising approach for clarifying the temporal structure—and hence, the nature—of psychological phenomena, for example, perceptual experiences and thoughts, even if controversial, thinking about such phenomena in terms of their temporal properties seems the safest starting-point for describing our mental reality in the most general ontological terms. That being the case, this book may be seen as an attempt to rehearse Steward’s temporal strategy so as to throw light on perceptual experiences. Intuitively albeit loosely understood, perceptual experiences are those conscious happenings at the heart of perceptual episodes such as seeing or hearing: when I see a bright star, I typically mean that I am visually aware of that distant object; when I hear or listen a loud noise or a symphony, I typically undergo an auditory experience of the relevant sounds; and so on. Since a subject may perceive (i.e., see, hear, among others) an object for a shorter or longer period of time, it is natural to think that the conscious or experiential phenomena at the heart of those episodes are temporally protracted too. What temporal structure do perceptual experiences specifically have? Although he was concerned with seeing rather than experiencing, Vendler anticipated something along the lines of a non-Heraclitean view of experience, a view that, perhaps thanks to the influence of property-instantiation view of events (cf. Kim, 1976), is echoed in later characterizations of perceptual experiences. Neglecting the ontological significance of the relevant notions at stake, for example, Daniel Dennett writes: “[c]onscious experience, in our view, is a succession of states constituted by various processes occurring in the brain, and not something over and above that is caused by them” (Dennett, 1998: 136). Or more recently, and following a more ontologically sensitive approach, Geoffrey Lee claims that “experiences are events, and that these events are instantiations of experiential properties by subjects” (Lee, 2014: 3). What plays the role of the relevant subject and properties/ relations, meanwhile, is a major philosophical question. For the time being, I shall simply assume that experiences are crucially constituted by persons (not their brains or otherwise subpersonal components) and by the instantiation of informational relations between persons and their surroundings.1
Perceptual Experience and Language
121
This stance, widespread but somewhat implicit throughout the literature, stands apart from a number of recent attempts to model perceptual experiences along Heraclitean lines. Brian O’Shaughnessy, one of its most provocative champions, conceives perceptual experiences as necessarily processive psychological items, that is, necessarily processive in the sense that such items (i) unfold over time or protractedly progress into existence, and (ii) do so down to their instantaneous or near-instantaneous parts. What drives a crucial tension between both approaches lies in (ii), a key modal qualification in virtue of which the relevant experiential processes are only constituted by further processive elements. For someone like O’Shaughnessy, perceptual experiences are necessarily processive in the sense that they are processive through and through, processive to the core. A Heraclitean proposal seems problematic to me precisely because such special processes obscure how experiences are supposed to be related to their respective subjects as well as to the relevant experienced objects and properties/relations. The conceptual anatomy of experiences that the Heraclitean performs only reveals further processes of a special modal kind. As I just mentioned, this chapter will not directly address the ontological dispute between a processive and a stative view of experience, but a tactical maneuver underpinning that debate, viz. bringing a number of considerations derived from the study of lexical aspect on what ontology of perception we should endorse. That is, while I do think that the application of the temporal strategy to ontological discussions concerning perceptual experiences is a sound move, I shall challenge one way of implementing that strategy, namely, that of deriving ontological views from considerations regarding the aspectual structure of perceptual claims. My general stance is that the linguistic evidence at hand is of little use to settle the relevant ontological issue. For sure, I do not mean to insinuate that, in the recent philosophical literature, someone claimed that pure analysis of linguistic constructions could definitively settle the relevant ontological debate. But there are, I believe, relatively clear signs in the literature to the effect that, to a greater or a lesser extent, aspectual considerations are supposed to bear on the ontological issue at stake. More specifically, I shall argue for two claims: (i) that grammatical tests developed in the aspect literature and then tailored for the ontological classification of occurrence-types are by and large silent on how we should conceive our perceptual experiences; and (ii), that well-known pairs of aspectual features—viz. telicity-atelicity, progressionnonprogression, and homogeneity-heterogeneity—provide no support for a non-Heraclitean or a Heraclitean view of experiences. My scope is thereby very narrow, but I believe it is necessary so as to bring out and critically address a controversial methodological move on which a number of ontologists of mind have relied.
122
Chapter 7
That said, how do linguistic constructions bear on the ontological structure of psychological phenomena like perceptual experiences? Vendler distinguished four general categories intended to clarify the behavior of certain verbal constructions: activities, accomplishments, achievements, and states. He thinks that, while these categories may be characterized in a number of different ways, time is a relevant criterion of distinction—at least “important enough to warrant separate treatment” (Vendler, 1957: 143). Activities as well as accomplishments take time or unfold over time, but while activity verbs do not incorporate a climax or terminal point within their definition, accomplishment verbs do. Achievements stand for punctual or momentary occurrences; they are over as soon as they have begun. States, meanwhile, obtain or continue obtaining: unlike achievements, they may persist over time; unlike activities or accomplishments, they do not persist in a progressive or unfolding way. Since these categories codify the temporal schemata or structures of different verbal constructions, Vendler thereby provides an elegant framework in terms of which Steward’s temporal strategy could be implemented. In the following table, Susan Rothstein provides examples of each category (cf. Rothstein, 2004: 6): States
Activities
Achievements
Accomplishments
know believe have desire love understand be happy
run walk swim push a cart drive a car
recognize spot/notice find/lose reach die
paint a picture make a chair deliver a sermon draw a circle recover from an illness build a house
Before expanding on this picture, two minor remarks are in order. First, I take Vendler’s categories to identify different ways of understanding events, occurrences, or situations (Casati & Varzi, 2020: section 2.1). While this stipulation is not entirely uncontroversial, I take it to be only a terminological one in the present context. Second, even though Vendler himself used the previous framework to analyze action verbs, I follow the subsequent tradition in applying the previous categories to all sorts of occurrences, whether agential or nonagential (cf. Mourelatos, 1978, 1993; Gill, 1993: 366; Steward, 1997: 82–88). To throw further light on Vendler’s framework, it is helpful to bear in mind a familiar distinction between those features which define or characterize the previous aspectual categories, on the one hand, and, on the other, those grammatical tests in virtue of which specific situation- or occurrence-types (e.g.,
Perceptual Experience and Language
123
running, writing, knowing, seeing) may be classified into a given aspectual category. I spell out this distinction a little bit further next. In relation to the relevant aspectual features, Vendler’s framework seems to result from the combination of two crucial elements and their opposites: (a) telicity, or the property of having a climax or terminal point; and (b) progression, or the property of coming into existence by stages or parts. With these features at hand, the previous categories may be redescribed as follows: activities are atelic and progressive; accomplishments are telic and progressive; achievements are telic and nonprogressive; and states are atelic and nonprogressive. In connection with telicity and progression, Vendler, among many others, also discusses a further aspectual property, namely, that of homogeneity. In a nutshell, this feature may be expressed as follows: if “S is x-ing” refers to a homogeneous occurrence x of type X and is true from t1 to t10, then “S is x-ing” is true at every instant between t1 and t10. Homogeneity thus traces a relationship between a temporally protracted occurrence and its internal parts; it tells us that certain temporally extended occurrences may be constituted by occurrences of the exact same type as themselves.2 Although this feature has attracted a great deal of attention in the linguistic and philosophical literature, it is not wholly clear whether it is more closely related to telicity or to progression. On the one hand, a wide number of writers have assumed that homogeneity is related to telicity in the sense that activities and states are atelic occurrences, and homogeneity was precisely thought to set activities and states apart from accomplishments and achievements. However, on the other hand, it has more recently been observed that, while activities may well be atelic, they are not strictly speaking homogeneous—or at least not homogeneous in the same sense states seem to be (cf. Mourelatos, 1978; Dowty, 1979; Rothstein, 2004: 11–15). If John loves Mary for twenty years, he loved her at each and any instant throughout that span of time. By contrast, Vendlerian activities—that is, atelic processes—will reduce into occurrences of a different kind beyond a contextually fixed point: just as a fruitcake will ultimately divide into sultanas, among other ingredients; activities like walking or waltzing are constituted by events of a different kind, for example, single steps or even more discrete leg-movements. As such, it is not obviously true that homogeneity is related to the telic-atelic contrast. For present purposes, I do not aim so much to adjudicate the role of homogeneity within the network of Vendler’s aspectual features, as to stress its relevance in the study of lexical aspect. In line with the critical stance of this work, I shall eventually argue that homogeneity also fails to support either of the competing ontological views of perception I shall survey here. As to the relevant grammatical tests, they classify occurrences into one of the above aspectual categories by picking up on patterns in the ways speakers use the expressions referring to those occurrences. The literature on
124
Chapter 7
this aspect in the English language includes linguistic tests concerning the possibility of construing expressions in the continuous or the progressive, the possibility of combining verbal constructions with certain temporal or otherwise adverbial modifiers (e.g. “at x time,” “for x time”), the possibility of complementing linguistic constructions with certain aspectual verbs, for example, stopping and finishing, among others (cf. Rothstein, 2004: ch. 1). In the present context, one should be specially careful about not confusing these grammatical tests for features which define a given aspectual category: for example, while the continuous or the progressive is often a distinctive mark of expressions referring to activities and accomplishments, one should not thereby suppose that the continuous or the progressive constitutes a defining trait of activities and accomplishments—instead, compatibility with the continuous is only indicative, at the level of our ordinary linguistic practices, of such a trait. Bearing in mind this distinction, the next section argues that grammatical tests for aspectual classification—with special focus on the continuous test—fail to throw any light whatsoever on the ontological classification of perceptual experiences, whereas section 3 contends that no particular ontological view of perception is suggested by the above aspectual features. To the extent that Vendler’s framework picks up on different temporal schemata or structures—that is, to the extent that they distinguish different ways in which an occurrence may exist or persist through time—it suggests at least two ways of conceiving the temporal structure of perceptual experiences. As previously noted, O’Shaughnessy conceives the relevant experiential items as processes of a special kind, where processes are minimally understood here as items which, independent of whether they have a terminal point (Vendler’s accomplishments) or fail to do so (his activities), persist by unfolding through time. A key element of this view is that the relevant processes may only be analyzed into processes of the same special kind. O’Shaughnessy himself is not blind to this peculiarity, for he explicitly compares his modally strong account of experiential processes with a more traditionally stative understanding of physical as well as psychological albeit nonexperiential processes: “A particular state always lies at the heart of nonexperiential process. I can think of no exception to this rule” (O’Shaughnessy, 2000: 47). In this specific respect, his experiences stand sharply apart from the rest of the physical and psychological-nonexperiential realm: “No experiences merely obtain or merely exist. No experiences are states” (O’Shaughnessy, 2000).3 By contrast, one could argue that perceptual experiences do not unfold, but obtain at a time or continue obtaining over time. As such, this stance sets the notion of state at the heart of our understanding of conscious perception. Since he was specially interested in the conscious or experiential component of visual perception, Vendler’s own temporal analysis of the perceptual verb seeing came close to constitute a stative view of perceptual experiences (cf.
Perceptual Experience and Language
125
Vendler, 1957: 154–160). On the face of it, the spirit of this account seems to run directly against O’Shaughnessy’s Heraclitean account. While the previous stances by no means exhaust the realm of possible approaches to the ontology of perception, I believe it is extremely helpful to think of other theoretical alternatives against the background of these opposing options. Between both extremes, it is indeed possible to find hybrid views; for example, drawing on Vendler as well as O’Shaughnessy, Matthew Soteriou argues for the thought that perceptual experiences are psychological states that dependent on the processes of a phenomenally conscious kind. This qualified stance, he thinks, not only avoids the excesses of a purely stative or a purely processive view but also accommodates the temporal phenomenology of perceptual experience (cf. Soteriou, 2011, 2013; Crowther, 2009a, 2009b). While these remarks are no doubt sketchy, I hope they manage to illustrate how a conceptual framework like Vendler’s is helping to shape the current ontological debate concerning what type of entities perceptual experiences are. That said, the specific issues I shall focus on in the next two sections are the following: (i) Are the aforementioned grammatical tests for aspectual classification relevant for the temporal categorization of perceptual experiences? (ii) Are the above aspectual features (e.g., telicity, progression, homogeneity) relevant for the temporal categorization of perceptual experiences? I address (i) next. Then, I shall turn to (ii). PERCEPTUAL EXPERIENCES AND THE CONTINUOUS TEST Do grammatical tests for aspectual classification have any bearing on the temporal classification of perceptual experiences? A number of philosophers engaged in the relevant ontological discussion seem to think so. Let us consider two representative cases. Vendler himself used what is now known as the continuous test in order to hold that “notwithstanding the fact that one might see something for a long period, it does not mean that he “is seeing” that thing for any period [ . . . ]” (Vendler, 1957: 155)—that is, to hold that seeing is not a process, whether of the activity type or of the accomplishment type. While this quote might initially be read as making a point about the way in which English speakers typically use the perceptual verb “seeing,” Vendler subsequently makes it clear that the claim is intended to have epistemological and psychological
126
Chapter 7
ramifications: “[w]e have no reason to fear that seeing, for example, as it is not always an achievement, might turn out to be an activity after all, a mysterious process, reviving thereby all the ghosts of epistemology” (Vendler, 1957: 159–160). It is relatively clear that the mysterious processes and ghosts thus targeted by Vendler are those that fit the Cartesian paradigm criticized by Ryle (cf. Vendler, 1957: 154, 159). Should we blame Vendler’s move on the historical context of Rylean and Wittgensteinian ordinary-language philosophies within which he was working? Perhaps, but that sort of move resurfaces even in the writings of philosophers who have long superseded that historical tradition. Soteriou has recently used the continuous test to address the question whether we should “regard the perceptual/experiential aspects of mind that are individuated in terms of their representational contents as states, accomplishments, activities, or achievements” (Soteriou, 2007: 551). Since locutions of the general form “it seems to S that p,” “it appears to S that p,” or “it looks to S that p” are commonly used to pick up on perceptual experiences, and since they do not naturally take the continuous tense, it apparently follows that “such phrases pick out mental states.” (Soteriou, 2007: 551) Conversely, he suggests that perceptual constructions capable of passing the continuous test lead up to a processive stance: “We can say ‘S was having an experience such that it seemed to him that p’ and here it looks as though we are picking out an unfolding occurrence in terms of its representational content” (Soteriou, 2007: 551). It is thus fairly common to find applications of the continuous test—a test initially suited for verbal, predicative, or aspectual classification—in a drastically different ontological context concerning the temporal structure of experiential items. The test for aspectual classification at stake here concerns the possibility of expressing occurrence types with the continuous or the progressive. Across different languages, popular wisdom has it that the continuous—the progressive being often regarded a subclass of the continuous (cf. Comrie, 1976: 24ff.)—expresses the gradual coming into existence of certain situations (cf. Comrie, 1976; Landman, 1992). In other words, the use of the continuous is commonly associated with reference to temporally protracted changes. Accordingly, it is natural to envisage a continuous test along something like the following lines: if a given occurrence-type x allows for constructions of the form “S is/was x-ing at t,” then x should be classified in terms of the dynamic and temporally protracted category of processes—that is, in terms of Vendler’s activities or accomplishments. A given occurrence-type fails to pass the test for one of two reasons: either because it refers to occurrences which, to the extent that they are punctual, do not take time; or because it refers to persistent but nondynamic items like states. Thus understood, the continuous test might seamlessly (but controversially, I believe) apply to
Perceptual Experience and Language
127
perceptual statements so as to classify the psychological items those statements refer to: if a given perceptual statement passes the continuous test, it would be natural to think of the referred perceptual phenomenon as processive; conversely, if such a statement fails to satisfy the continuous test, surface grammar would thus fail to support a processive understanding of the relevant psychological item. That said, I do not think that something like the continuous test provides much support for an ontological, let alone a processive, view of perceptual experiences. As previously anticipated, my general worry with the previous line of reasoning is that grammatical tests for aspectual classification are, in a very intuitive sense, silent on the question how we should ontologically think of perceptual experiences; instead of providing evidence for ontological theorizing, what these linguistic criteria seem to do for us is to identify patterns in linguistic usage so as to classify types of expressions into semantic or lexical kinds. More specifically, the limitations of the continuous test transpire in two different points. First, the import of such a grammatical test may be seriously altered by pure linguistic stipulation, which in turn suggests that, as a criterion of temporal classification, that test codifies the linguistic structure of perceptual judgments, rather than the ontological structure of perceptual experiences. For example, while Vendler and Soteriou used it to argue for a stative view, this move could be blocked with a relatively simple reply: “Although seeing can be formulated in terms of the continuous tense, one could stipulate a new sense—call it seeing* th—for which a construction of the form ‘S is/was seeing* O’ would be quite legitimate, or perhaps a technical notion of perceptual awareness capable of expressing the exact same perceptual situation in a more natural way—as in a statement of the form ‘S is visually aware of O’ or ‘S is visually experiencing O.”4 Depending on what constructions one uses to speak about one and the same psychological phenomenon, exactly the same grammatical test could be used to motivate a stative or a processive view of perceptual experiences. But on O’Shaughnessy’s assumption that states and processes are mutually exclusive categories within the realm of experiences, that conclusion is inadmissible: if perceptual experiences are mental states of a certain kind, they cannot be, in the exact same sense, processes. Although a possible reply against this point is that O’Shaughnessy’s assumption is thoroughly misplaced—a claim I am actually sympathetic with—a provisionally more conservative diagnosis is that the continuous test does not provide definite support for a particular ontology of perception because it is not intended to tell us anything about the true ontological structure of perceptual experiences—instead, it only indicates how we do or could talk about such psychological items. Second, one should also bear in mind that the application of the continuous test to perceptual verbs provides different results in different natural
128
Chapter 7
languages. This is a point neatly made by Bernard Comrie in his now classic textbook on lexical aspect: [ . . . ] different languages in fact have different rules for determining when explicitly progressive forms can be used. [ . . . ] In English, it is not in general possible to use progressive forms of verbs of inner perception like see, hear [ . . . ], i.e. not *I am seeing you there under the table, or *you aren’t hearing. In Portuguese, however, such forms are perfectly acceptable: estou te vendo lá embaixo da mesa; você não está ouvindo. In these particular cases, the relevant factor seems to be that it is possible to view raining, seeing, hearing, etc., either as states or as non-states (dynamic situations): different psychological theories differ as to just how active a process perception is, and there is no reason to suppose that language presupposes the answer by uniquely classifying perception as either a state or a dynamic situation. Different languages are free to choose, essentially as an arbitrary choice, whether such verbs are classified as stative or not. (Comrie, 1976: 35)
Comrie acknowledges the existence of linguistic uses that speakers in English, among other languages, are familiar with, as well as the aspectual distinctions on which linguists ground these usages. The existence of such distinctions and their relevance for linguists as well as philosophers are not at stake here. But Comrie also stresses that considerations about the stative or the dynamic (processive, shall we say) character of expressions concerning perceptual occurrences do not necessarily inform our ontological understanding of perceptual experiences: at most, they only indicate what aspectual features certain occurrence types instantiate, based on how expressions referring to those occurrences are typically used within one or different natural languages. That is, even if the continuous test could be used to classify a particular occurrence type (e.g., seeing-occurrences) within one aspectual category in a given language, this would not guarantee that the same classification would carry over into other languages. Of course, it would be rather arbitrary to avoid this point by claiming that the way we should think about perceptual occurrences is shaped by the way we talk about them in one particular natural language—say, English. A quick overview of how perceptual locutions in different languages fare with the continuous tense thus suggests that the relevant grammatical test does not provide a clear-cut answer to the question whether we should think of perceptual experiences as progressive (that is, as processive) or as nonprogressive (more specifically, as stative). To account for the fact that Portuguese perceptual verbs pass the continuous test while English ones fail to do so, I would not hold that different ontologies of perception underpin perceptual talk in both languages:
Perceptual Experience and Language
129
instead, it seems much more natural to claim that the continuous test delivers different results in different languages because its only purpose is to trace or codify linguistic conventions which vary from language to language. In other words, the variety of results in applying the continuous test across languages would not reveal different ontologies or worldviews, but, much more modestly, only different linguistic practices. Although the analysis of our linguistic practices may often be a good way of studying how we think about reality, it does not automatically provide evidence in support of one ontological picture or another.5 To wrap this up, I do not think the continuous test (nor, for similar reasons, other grammatical tests derived from the literature on lexical aspect) is very helpful when it comes to adjudicate between different ontologies of perception. Of course, I am not thus challenging the linguistic relevance of such criteria or their limited interest in ontological debates. All I intend to show here is that they do not help to settle the relevant ontological issue on their own. The continuous test may tell us something about our perceptual locutions, but, on its own, it provides as much or as little support for either a Heraclitean or a non-Heraclitean view. In this particular sense, the test is by and large silent about perceptual ontology. PERCEPTUAL EXPERIENCES AND ASPECTUAL FEATURES Vendler characterizes activities, accomplishments, achievements, and states, in terms of telicity and progression: while a given situation is telic iff its very description or definition involves reference to a climax or terminal point, a situation is progressive iff its temporally extended existence takes place through the succession of temporal stages or parts. In connection to these features, linguists and philosophers also discuss a third feature, namely, homogeneity: if a temporally protracted occurrence is homogeneous, it will be constituted by occurrences of the exact same type taking place at every instant of the relevant interval or period of time. Could the above aspectual features throw any light on the ontological classification of perceptual experiences? In this section, I shall argue that the existing aspectual evidence is by and large silent on the relevant ontological question: it is silent in the sense that even what I take to be the most natural attribution of aspectual features for perceptual experiences—viz. experiences as atelic, progressive, and nonhomogeneous—fails to provide suggestive evidence for a stative or a processive view of perceptual experience. For each relevant aspectual feature, I shall briefly spell out how experiences are related to them, and then discuss what that tells us about the ontology of perception.
130
Chapter 7
Although the very notion of experience is a term of art in contemporary philosophy of mind, it seems to me that there is something like a minimal sense in which perceptual experiences are conceived throughout the philosophical literature, namely, as conscious events or happenings which are typically extended over a period of time. To the extent that this notion is extremely general, it does not automatically fix the internal temporal structure of the relevant happenings; this further point depends on philosophical theorizing. My aim here is to evaluate whether the abovementioned aspectual features cast some light on the question how perceptual experiences, in that minimal sense, should be temporally classified. Turning first to the alleged relevance of the telicity-atelicity for the relevant ontological debate, I shall argue that the question how perceptual experiences should be temporally classified does not turn on whether they are telic or atelic. Occurrences are telic insofar as they are partially defined by terminal or climax points. For example, describing an occurrence as one of walking to the corner fixes when it comes to an end, viz. when the relevant subject actually reaches the corner. As a matter of fact, obstacles could intervene in the development of this event, in which case it will not reach its natural completion—say, being hit by a car before reaching the corner. Whether the relevant occurrence is complete or incomplete will depend on whether the subject reaches the corner. Atelic occurrences are, meanwhile, complete no matter when they conclude. On the basis of the previous distinction, it seems natural to classify perceptual experiences as atelic: unlike active tasks of perceptual search or monitoring, having perceptual experiences is not something in the need of reaching a certain terminal point. For example, when I stare at the Pacific Ocean, I will undergo a certain visual experience for as long as I am able to hold visual contact with that expanse of sea and sky—nothing in the definition of this conscious phenomenon forces it to conclude at some moment. When a subject sees O or undergoes visual experiences of O from t1 to t10, the concept of this perceptual occurrence does not seem to refer to any particular terminal point at which the occurrence must end: practical limitations aside, a subject could in principle be engaged in that situation indefinitely. That said, even if it is clear that perceptual experiences are atelic, this does not throw light on the questions whether perceptual experiences are ultimately stative or processive through and through. After all, most writers in the literature on aspect will concede that states as well as processes are atelic. As such, ascribing the aspectual feature of atelicity to perceptual experiences will not get us any closer to adjudicate on the relevant ontological question. We would require further aspectual evidence, derived from the features of progression and homogeneity, to settle that issue.
Perceptual Experience and Language
131
Moving on, I shall take it for granted here that perceptual experiences, in the minimal sense above described, are progressive. This claim is often advanced by processive theorists aiming to stress one or another dynamic component crucial to our understanding of experiential items. Building on the thought that “experience guarantees a direct confrontation with the passage of time” (O’Shaughnessy, 2000: 65), O’Shaughnessy highlights the fundamentally dynamic dimension of the temporal content of experience, as well as the thought that the very temporal structure of what we experience shapes the temporal structure of our experiences (cf. O’Shaughnessy, 2000: 59–67). Likewise, Soteriou invokes an experiential progressive component in order to accommodate two phenomenological features of experiences: on the one hand, temporal transparency, the fact that awareness of the temporal features of perceptual experience is dependent on the awareness of the temporal features of experienced items, and, on the other, temporal continuity, the fact that temporally protracted perceptual experiences are continuous, not a mere succession of discrete experiential parts (cf. Soteriou, 2011, 2013: ch. 6). Even though he reflects on aspectual considerations elsewhere, Soteriou is primarily trying to accommodate dynamic aspects of perceptual experience by conceiving the latter as items which come about temporal stage by temporal stage. For O’Shaughnessy and Soteriou, perceptual experience has to persist by stages or in a progressive way because representational or phenomenological aspects of perceptual experience do so. Although this thought depends on heavy-duty commitments concerning the relationship between experiential content and experiential vehicle, it is safe to assume (a) that certain dynamic features of experience are neatly accommodated or at least suggested by the notion of progression, and (b) that, as such, any ontology of perception—either Heraclitean or non-Heraclitean—should try to incorporate this aspectual feature. With the notion of experiential progression in place, the question is whether EH or ENH is vindicated by the progressive character of perceptual experience. Since O’Shaughnessy and Soteriou apparently think that the relevant aspectual feature leads up to a Heraclitean stance, I shall mainly focus on whether experiential processes, in O’Shaughnessy’s sense, follow from experiential progression. My negative answer to this question is not, however, intended to score a point in favor of a stative view, but to suggest that experiential progression is silent on what ontology of perception we should endorse. It does not seem to me, then, that the progressive aspect of experience plays a supportive role vis-à-vis a Heraclitean view for at least two reasons. First, it actually fails to have such a role in an account like O’Shaughnessy’s. Recall that he and possibly Soteriou conceive perceptual experiences as progressive or as stage-constituted because they conceive the content and
132
Chapter 7
the phenomenology of experience in very specific ways, not the other way around. For them, perceptual experiences have a progressive character—that is, consist in temporal stages—because they are in some sense dynamic: their progression is not something one reads off linguistic usage. In this context, experiential progression is not a piece of evidence which supports a processive view of experience, but a follow-up of and ontological view endorsed on phenomenological or representational grounds.6 Second, even if one conceded that experiential progression hinted at the notion of experiential processes, the latter would not be the idiosyncratic items posited by O’Shaughnessy. For it would only provide discreet support for processes broadly understood as changes, which may in turn be conceived along the lines of instantiations of properties/relations or the maintenance thereof. O’Shaughnessy’s processes are not analyzable in terms of such instantiations, though. So even if progression provided some sort of support for EH, it would at least suggest the existence of processes compatible either with a Heraclitean or with a nonHeraclitean account of perceptual experiences.7 If these remarks are along the right lines, I believe they generally support the thought that considerations from aspectual classification fail on their own to throw light on how perceptual experiences are temporally structured; in the case at point, the progressive-nonprogressive pair, as a purely aspectual distinction, provides no clue as to how perceptual experiences should be ontologically or temporally classified. Finally, let us turn to homogeneity. If a subject is engaged in a homogeneous occurrence for a given period of time, she will be engaged in an occurrence of that type at every instant of that period of time. Or again, using a very minimal notion of temporal part, one could also say that a given type of temporally protracted occurrences is homogeneous if its component temporal parts are in turn occurrences of the same exact type. O’Shaughnessy thinks that processes are homogeneous in the sense that their constituting parts are of the same kind as the process which they constitute (cf. O’Shaughnessy, 2000: 44). But while it is relatively clear that he is thus stressing the necessarily processive character of experience, there is an intuitive sense in which his claim fails to hold water. On the traditional assumption that content partially or exhaustively individuates psychological attitudes/vehicles, it would simply be puzzling how, if a subject stares at a number of different objects throughout t1–tx, she would undergo a homogeneous experience during that period of time. Again, to the extent that he takes them to be states constituted by processes of different kinds, Soteriou also regards perceptual experiences as nonhomogeneous. Either way, O’Shaughnessy’s remark on the homogeneity of experience is problematic. If one then assumed that perceptual experiences are heterogeneous, what would that tell us about their temporal structure? Not much, I gather. If
Perceptual Experience and Language
133
perceptual experiences are atelic, only two Vendlerian categories could account for them, namely, activities and states. In the linguistic literature, states (e.g., believing, loving) have traditionally been regarded as homogeneous (cf. Rothstein, 2004). As such, if perceptual experiences are atelic but heterogeneous, they could not be states. So far, so good; for this result could in principle suggest that perceptual experiences should be classified as Vendlerian activities—that is, as processes. The problem is that activities are also supposed to be homogeneous. As such, the feature of heterogeneity would not provide evidence in support of EH or ENH. True: Vendler’s original claim that activities were thoroughly homogeneous was soon qualified. In his influential work on aspect, for example, David Dowty persuasively established the distinction between strong homogeneity or homogeneity down to instants and weak homogeneity or homogeneity down to minimal parts while strongly homogeneous situations are those the ultimate temporal parts of which are situations of the same type, weakly homogeneous situations may be divided into parts of the same type only down to certain contextually fixed intervals (cf. Dowty, 1979; Taylor, 1977; Mourelatos, 1978).8 According to Vendler, a given occurrence of, say, running could be divided into its temporally punctual or near-instantaneous stages, and those stages would still be activities of running. But, as stressed by Dowty, this is not quite right: for most activities one can think of, the relevant situations will be constituted by parts which are not situations of the exact same type. Thus, the run I took this morning has parts which are not events of running themselves, for example, the single steps I take in order to carry my body across my neighborhood. Or again, the danced waltz is composed of several events of a three-step sequence which are not themselves occurrences of waltzing. Even if activities fail to be strongly homogeneous, they are still homogeneous down to minimal parts. And the problem is that the examples which illustrate the heterogeneity of perceptual experiences rule out weak homogeneity no less than its strong counterpart: as far as I can see, the differences between a whole temporally extended experience and its constituting experiential (temporal) parts may be appreciated, whether the relevant constituting parts be near instantaneous or noticeably protracted. Furthermore, in the above quoted lines, O’Shaughnessy seems to be interested in something closer to strong homogeneity rather than its weak counterpart when it comes to set experiences apart from other physical and psychological albeit nonexperiential occurrences; the import of his remarks to the effect that experience is homogeneous is that a temporally extended experiential process is constituted by experiential processes down to its quasi-infinitesimal parts (cf. O’Shaughnessy, 2000: 48).9 O’Shaughnessy needs strongly homogeneous processes because they provide the only means for articulating the idea of experiential items which are processive to the core; weakly homogeneous
134
Chapter 7
processes will not do, for they could be ultimately constituted by items of different kinds, for example, instantiations of properties and relations. As such, historical qualifications in linguistics aside, O’Shaughnessy must rely on a notion of process capable of resisting the incorporation of states at its heart. In short, both activities and states are in some sense homogeneous categories; since perceptual experiences are neither strongly nor weakly homogeneous, they cannot be accommodated by either Vendlerian category. Pace O’Shaughnessy states that the foregoing considerations are built on the assumption that perceptual experiences are most intuitively thought of as heterogeneous. It should also be clear by now why no ontology of perception would be vindicated even if, suspending the previous assumption, perceptual experiences were conceived as homogeneous. Atelic processes and states fall on the same side of the homogeneity-nonhomogeneity divide: if processes are (strongly or weakly) homogeneous, so are states; if processes fail to be homogeneous, similar considerations lead up to the same conclusion regarding states. As such, no matter which aspectual feature be ascribed to perceptual experiences—either homogeneity or nonhomogeneity—no evidence would be provided for a Heraclitean or a non-Heraclitean ontology of perception yet. All I have intended to address here is whether there is suggestive linguistic evidence—that is, evidence exclusively derived from reflection on the aspectual study of perceptual claims—in support of ENH or EH. I have argued that traditional grammatical tests for aspectual classification throw little light on the relevant ontological debate. This section aimed to show that a number of aspectual features also fail to throw light on the temporal or otherwise ontological structure of perceptual experiences; even if there is some sense in which perceptual experiences may be said to be atelic, progressive, and heterogeneous, this does not take us any closer to one or another ontology of perception—that is, either to an ultimately non-Heraclitean ontology or to an ontology of items which are processive through and through. To bear on the relevant ontological debate, these aspectual considerations necessarily need to rely on extra-linguistic claims. CONCLUSION What does language tell us about the ontology of perceptual experiences? If the foregoing considerations are persuasive, the answer to this question is not that much. In recent years, conceptual frameworks derived from Vendler’s influential work on lexical aspect have been put to extensive use in a number of ontological debates. The goal of this chapter has been critically to examine what bearing, if any, a framework of that sort has on the recent debate concerning the question whether perceptual experiences are states or processes.
Perceptual Experience and Language
135
In section 1, I sketched the relevant aspectual framework and the ontological debate I am concerned with. Section 2 challenged the relevance of traditional grammatical tests for the aspectual classification of specific situations. Finally, I picked up on aspectual features related to telicity, progression, and homogeneity: even if there is a way of thinking about perceptual experiences in terms of such properties, the resulting conception would not unveil a processive or a stative view of the relevant psychological phenomena. In short, our aspectual intuitions are more or less silent on matters of perceptual ontology. As far as I can see, the latter issues should be addressed in the battlefields of epistemology and metaphysics, not that of linguistics.
NOTES 1. For discussion about the proper subjects of experience, cf. Bennett & Hacker, 2003 and Crane, 2015. 2. In the present context, I am not using the concept of temporal part or constituent as a heavy-duty one. For instance, when I claim that states are homogeneous, I am not implying that states have temporal parts or constituents in the same way processconstituted events have temporal parts. If homogeneity is phrased here in terms of the whole-part relationship, the latter should be understood, as it were, as a distinction of reason: by claiming that states are homogeneous, all I mean is that, if a state obtains for a period of time, it is possible to pick up on any instant within that period in which that state obtains—the state’s obtaining at that instant being conceived here as part of the state’s obtaining over the relevant period of time. 3. In a somewhat self-serving way, I want to stress here that, for present purposes, I don’t bear the burden of spelling out what sort of exclusive distinction O’Shaughnessy has in mind. To begin with, my primary goal in the present chapter is not to argue for or against either a Heraclitean or a non-Heraclitean view of experience: for present purposes, I only aim to assess whether aspectual considerations provide useful evidence for that debate. More importantly, it is not I who has to prove anything here: O’Shaughnessy is the one who takes a hard, revisionist take on our understanding of processes and states: broadly speaking, processes may be understood as changes constituted by the obtaining or maintenance of successive states. As such, O’Shaughnessy thus faces the challenge of clarifying why an account of experiential processes in the nondynamic terms of property/relation-instantiations—that is, states—won’t do. 4. Which is what Soteriou himself seems to do in the above quoted remarks. 5. Bluntly put, the background thought is that the temporal structure of a given linguistic construction need not mirror the temporal structure of the psychological item referred to. This is a methodological point that even O’Shaughnessy seems to endorse at one point (cf. O’Shaughnessy, 2000: 174) 6. Of course, this remark does not engage with the question whether O’Shaughnessy or Soteriou had good phenomenological or representationalist reasons in the offing.
136
Chapter 7
All I aim to do here is to suggest that, even if perceptual experiences could be said to be progressive, this would not constitute (partial or definitive) evidence for a processive view of experience. My first critical claim is that even an extreme Heraclitean theorist like O’Shaughnessy seems to grant no evidential value to the progressive aspect of experience: for him, that feature is a corollary of a more complex ontological and phenomenological story. 7. I think that, mutatis mutandis, similar considerations would show that experiential progression fail to support a stative view. 8. Also cf. Rothstein, 2004: 11: “The distinction between homogeneity down to instants and homogeneity down to small parts is crucial in distinguishing between states (which are homogeneous down to instants) and activities (which are homogeneous down to small parts) [ . . . ].” 9. It is indeed a nice question whether the processive and the homogeneity strains of O’Shaughnessy’s stance are compatible with each other. As Alexander Mourelatos noted, there seems to be a tension between the progressive character of processes and the homoeomerous character of Vendler’s activities (cf. Mourelatos, 1993: 386). One could, I suspect, exploit this tension to mount a challenge against the internal coherency of O’Shaughnessy’s stance. Since this chapter is not primarily concerned with criticizing a Heraclitean stance, I set the critical suggestion aside here.
Bibliography
Adrian, E. (1928). The Basis of Sensation. Christophers. Armstrong, D. M. (1962). Bodily Sensations. Routledge & Kegan Paul. Armstrong, D. M. (1968). A Materialist Theory of the Mind. Routledge. https://doi .org/10.4324/9780203003237 Armstrong, D. M. (1997). A World of States of Affairs. Cambridge University Press. Ayer, A. (1963). Philosophy and Language. In The Concept of a Person (pp. 1–35). St Martin’s Press. Barnes, J. (1982). The Presocratic Philosophers. Routledge & Kegan Paul. Bartlett, G. (2018). Occurrent States. Canadian Journal of Philosophy, 48(1), 1–17. https://doi.org/10.1080/00455091.2017.1323531 Bennett, M. R., & Hacker, P. M. S. (2003). Philosophical Foundations of Neuroscience. Blackwell Publishing. Bennett, M. R., & Hacker, P. M. S. (2012). History of Cognitive Neuroscience. In History of Cognitive Neuroscience. https://doi.org/10.1002/9781118394267 Bermúdez, J. L. (1998). The Paradox of Self-Consciousness. The MIT Press. Boyd, R. (1979). Metaphor and Theory Change. In A. Ortony (Ed.), Metaphor and Thought (pp. 481–532). Cambridge University Press. Briesen, J. (2015). Perceptual Justification and Assertively Representing the World. Philosophical Studies, 172(8), 2239–2259. https://doi.org/10.1007/s11098-014 -0407-8 Burge, T. (2010). Origins of Objectivity. Oxford University Press. Byrne, A. (2009). Experience and Content. The Philosophical Quarterly, 59, 429– 451. https://doi.org/10.1002/9781444343915.ch4 Cartwright, N. (1999). The Dappled World: A Study of the Boundaries of Science. Cambridge University Press. https://doi.org/10.1017/CBO9781139167093 Casati, R., & Varzi, A. (2020). Events. In The Stanford Encyclopedia of Philosophy (Summer 202). Metaphysics Research Lab, Stanford University. Chisholm, R. M. (1996). A Realistic Theory of Categories: An Essay on Ontology. Cambridge University Press. 137
138
Bibliography
Chudnoff, E. (2012). Presentational Phenomenology. In S. Miguens & G. Preyer (Eds.), Consciousness and Subjectivity (pp. 51–72). De Gruyter. https://doi.org/10 .1515/9783110325843.51 Comrie, B. (1976). Aspect. Cambridge University Press. Craig, E. (1976). Sensory Experience and the Foundations of Knowledge. Synthese, 33(1), 1–24. http://www.jstor.org/stable/20115122 Crane, T. (2015). The Mental States of Persons and their Brains. Royal Institute of Philosophy Supplements, 76, 253–270. https://doi.org/10.1017/S1358246115000053 Crane, T. & French, C. (2017). The Problem of Perception. In E. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Spring 2017 Edition). https://plato.stanford. edu/archives/spr2017/entries/perception-problem/ Crowther, T. (2009a). Perceptual Activity and the Will. In L. O’Brien & M. Soteriou (Eds.), Mental Actions (pp. 173–190). Oxford University Press. https://doi.org/10 .1093/acprof:oso/9780199225989.003.0009 Crowther, T. (2009b). Watching, Sight, and the Temporal Shape of Perceptual Activity. The Philosophical Review, 118(1), 1–27. https://doi.org/10.1215 /00318108-2008-027 Crowther, T. (2011). The Matter of Events. The Review of Metaphysics, 65(1), 3–39. Crowther, T. (2010). The Agential Profile of Perceptual Experience. Proceedings of the Aristotelian Society, 110(2pt2), 219–242. https://doi.org/10.1111/j.1467-9264 .2010.00285.x Crowther, T. M. (2006). Two Conceptions of Conceptualism and Nonconceptualism. Erkenntnis, 65(2), 245–276. https://doi.org/10.1007/s10670-006-0001-3 Dainton, B. (2000). Stream of Consciousness: Unity and Continuity in Conscious Experience. Routledge. https://books.google.com.tr/books?id=ZnCPhpxBEaIC Dainton, Barry. (2011). Time, Passage, And Immediate Experience. In C. Callender (Ed.), The Oxford Handbook of Philosophy of Time (pp. 382–419). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199298204.003.0013 de Vignemont, F. (2011). Bodily Awareness. In E. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Fall 2011 Edition). https://plato.stanford.edu/archives/ fall2011/entries/bodily-awareness/ Dennett, D. (1991). Consciousness Explained. Penguin. Dennett, D. C. (1998). Brainchildren. In Brainchildren. The MIT Press. https://doi .org/10.7551/mitpress/1663.001.0001 Dennett, D. C., & Kinsbourne, M. (1992). Time and the Observer: The Where and When of Vonsciousness in the Brain. Behavioral and Brain Sciences, 15(2), 183–201. https://doi.org/10.1017/S0140525X00068229 Descartes, R. (1965). Discourse on Method, Optics, Geometry, Meteorology (P. Olscamp (ed.)). Bobbs-Merrill. Descartes, R. (1986). Meditations on First Philosophy (J. Cottingham (ed.)). Cambridge University Press. Descartes, R. (1988). Descartes: Selected Philosophical Writings (J. Cottingham, R. Stoothoff, & D. Murdoch (eds.)). Cambridge University Press. Di Paolo, E., Buhrmann, T., & Barandarian, X. (2017). Sensorimotor Life: An Enactive Proposal. Oxford University Press. https://doi.org/10.1093/acprof:oso /9780198786849.001.0001
Bibliography
139
Dowty, D. (1979). Word Meaning and Montague Grammar. Reidel Publishing Co. https://doi.org/http://doi.org/10.1007/978-94-009-9473-7 Eliasmith, C. (2003). Moving beyond Metaphors: Understanding the Mind for What It Is. The Journal of Philosophy, 100(10), 493–520. Evans, G. (1982). The Varieties of Reference. Clarendon Press. Farrell, B. A. (1950). Experience. Mind, 59, 170–198. https://doi.org/https://doi.org /10.1093/mind/LIX.234.170 Feigl, H. (1967). The Mental and the Physical (NED-New). University of Minnesota Press. http://www.jstor.org/stable/10.5749/j.cttts4rf Fine, K. (2005). Modality and Tense: Philosophical Papers. https://doi.org/10.1093 /0199278709.001.0001 Foster, J. (1982). The Case for Idealism. Routledge & Kegan Paul. Frischhut, A. (2013). What Experience Cannot Teach Us About Time. Topoi, 34, 143–155. https://doi.org/10.1007/s11245-013-9211-x Gill, K. (1993). On the Metaphysical Distinction Between Processes and Events. Canadian Journal of Philosophy, 23, 365–384. https://doi.org/10.1080/00455091 .1993.10717326 Glüer, K. (2009). In Defence of a Doxastic Account of Experience. Mind & Language, 24(3), 297–327. https://doi.org/10.1111/j.1468-0017.2009.01364.x Glüer, K. (2018). Defeating Looks. Synthese, 195(7), 2985–3012. https://doi.org/10 .1007/s11229-016-1186-x Goldstein, E. B., & Brockmole, J. (2014). Sensation and Perception. Cengage Learning. Gomes, A. (2017). Perception and Reflection. Philosophical Perspectives, 31, 131– 152. https://doi.org/10.1111/phpe.12093 Graham, D. (2015). Heraclitus. In E. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Fall 2015 Edition). https://plato.stanford.edu/archives/fall2015/ entries/heraclitus/ Hacker, P. M. S. (1987). Appearance and Reality: A Philosophical Investigation into Perception and Perceptual Qualities. Basil Blackwell. Heck Jr, R. G. (2000). Nonconceptual Content and the “Space of Reasons.” The Philosophical Review, 109(4), 483–523. https://doi.org/http://doi.org/10.2307 /2693622 Heidegger, M. (2010). Being and Time. SUNY Press. Heil, J. (1984). Perception and Cognition. University of California Press. Hoerl, C. (2013). ‘A Succession of Feelings, in and of Itself, is Not a Feeling of Succession.’ Mind, 122(486), 373–417. https://doi.org/10.1093/mind/fzt070 Hoerl, C. (2014). Do We (Seem to) Perceive Passage? Philosophical Explorations, 17, 188–202. https://doi.org/10.1080/13869795.2013.852615 Huemer, M. (2001). Skepticism and the Veil of Perception. Rowman & Littlefield. Huemer, M. (2007). Compassionate Phenomenal Conservatism. Philosophy and Phenomenological Research, 74(1), 30–55. https://doi.org/10.1111/j.1933-1592 .2007.00002.x Huemer, M. (2013). Phenomenal Conservatism Über Alles. In C. Tucker (Ed.), Seemings and Justification: New Essays on Dogmatism and Phenomenal Conservatism (pp. 328–350). Oxford University Press. https://doi.org/10.1093/ acprof:oso/9780199899494.003.0015
140
Bibliography
Hume, D. (n.d.). A Treatise of the Human Nature. Oxford University Press. Juarrero, A. (1999). Dynamics in Action: Intentional Behavior as a Complex System. The MIT Press. Kahn, C. (1979). The Art and Thought of Heraclitus: An Edition of the Fragments with Translation and Commentary. Cambridge University Press. Kant, I. (2007). Critique of Pure Reason (H. Caygill, G. Banham, & N. Kemp Smith (eds.)). Palgrave Macmillan. https://doi.org/http://doi.org/10.1007/978-1 -137-10016-0 Kim, J. (1976). Events as Property Exemplifications. In M. Brand & D. Walton (Eds.), Action Theory (pp. 310–326). D. Reidel. Kim, J. (1997). The Mind-Body Problem: Taking Stock After Forty Years. Philosophical Perspectives, 11, 185–207. http://www.jstor.org/stable/2216130 Kim, J. (2000). Mind in a Physical World: An Essay on the Mind-Body Problem and Mental Causation. The MIT Press. Landman, F. (1992). The Progressive. Natural Language Semantics, 1(1), 1–32. Lee, G. (2014). Temporal Experience and the Temporal Structure of Experience. Philosophers’ Imprint, 14(3), 1–21. http://hdl.handle.net/2027/spo.3521354.0014 .003 Lewis, D. (1980). Mad Pain and Martian Pain. In N. Block (Ed.), Readings in the Philosophy of Psychology, Vol. I (pp. 216–222). Harvard University Press. Lowe, E. J. (2005). The Four-Category Ontology: A Metaphysical Foundation for Natural Science. Clarendon Press. https://doi.org/10.1093/0199254397.001.0001 Lowe, J. (1996). Subjects of Experience. Cambridge University Press. Martin, M. G. F. (1995). Bodily Awareness: A Sense of Ownership. In J. L. Bermúdez, A. Marcel, & N. Eilan (Eds.), The Body and the Self (pp. 267–289). The MIT Press. Martin, Michael G. F. (2002). The Transparency of Experience. Mind and Language, 17, 376–425. https://doi.org/10.1111/1468-0017.00205 McDowell, J. (1994). Mind and World. Harvard University Press. Moore, G. E. (1903). The Refutation of Idealism. Mind, 12(48), 433–453. Moretti, L. (2015). Phenomenal Conservatism. Analysis, 75(2), 296–309. https://doi .org/10.1093/analys/anu153 Mourelatos, A. (1978). Events, Processes, and States. Linguistics and Philosophy, 2, 415–434. Mourelatos, A. P. D. (1993). Aristotle’s Kinesis/Energeia Distinction: A Marginal Note on Kathleen Gill’s Paper. Canadian Journal of Philosophy, 23(3), 385–388. Nagel, T. (1974). What Is It Like to Be a Bat? The Philosophical Review, 83(4), 435–450. O’Shaughnessy, B. (1971a). Processes. Proceedings of the Aristotelian Society, 72, 215–240. https://doi.org/10.2307/4544825 O’Shaughnessy, B. (1971b). The Temporal Orderings of Perceptions and Reactions. In F. Sibley (Ed.), Perception: A Philosophical Symposium. Methuen. https://doi .org/10.2307/2218319 O’Shaughnessy, B. (2000). Consciousness and the World. Oxford University Press. https://doi.org/10.1093/0199256721.001.0001
Bibliography
141
Papineau, D. (2021). The Metaphysics of Sensory Experience. Oxford University Press. https://doi.org/10.1093/oso/9780198862390.001.0001 Peacocke, C. (2012). Subjects and Consciousness. In A. Coliva (Ed.), The Self and Self-knowledge (pp. 74–101). Oxford University Press. https://doi.org/10.1093/ acprof:oso/9780199590650.003.0004 Phillips, I. (2010). Perceiving Temporal Properties. European Journal of Philosophy, 18(2), 176–202. https://doi.org/10.1111/j.1468-0378.2008.00299.x Phillips, I. (2014). Experience of and in Time. Philosophy Compass, 9, 131–144. https://doi.org/10.1111/phc3.12107 Pitcher, G. (1971). A Theory of Perception. Princeton University Press. Place, U. T. (1956). Is Consciousness a Brain Process? British Journal of Psychology, 47(1), 44–50. https://doi.org/10.1111/j.2044-8295.1956.tb00560.x Plato. (1921). Theaetetus. Sophist. (H. N. Fowler (ed.)). William Heinemann. Plato. (2010). Meno and Phaedo (D. Sedley & A. Long (eds.)). University Press Cambridge. Pryor, J. (2014). There is Immediate Justification. In M. Steup, J. Turri, & E. Sosa (Eds.), Contemporary Debates in Epistemology (2nd ed., pp. 202–222). Wiley Blackwell. Putnam, H. (1967). The Nature of Mental States. In W. H. Capitan & D. D. Merrill (Eds.), Art, Mind, and Religion (pp. 37–48). University of Pittsburgh Press. Robinson, H. (1994). Perception. Routledge. Rothstein, S. (2004). Structuring Events: A Study in the Semantics of Lexical Aspect. Blackwell. https://doi.org/10.1002/9780470759127 Roxbee Cox, J. W. (1971). An Analysis of Perceiving in Terms of the Causation of Beliefs. In F. Sibley (Ed.), Perception: A Philosophical Symposium (pp. 23–62). Methuen. Russell, B. (1917). Mysticism and Logic. George Allen & Unwin. Ryle, G. (1949). The Concept of Mind. Hutchinson. Ryle, Gilbert. (1937). Categories. Proceedings of the Aristotelian Society, 38, 189–206. Sanhueza Rodríguez, S. (2020). The Ontological Importance of Being a Perceptual Attitude. Organon F, 27(1), 2–28. https://doi.org/10.31577/orgf.2020.27101 Sattig, T. (2019). The Flow of Time in Experience. Proceedings of the Aristotelian Society. https://doi.org/10.1093/arisoc/aoz014 Sellars, W. (1963). Science, Perception and Reality. Routledge & Kegan Paul. Shapiro, L. (2007). The Correspondence between Princess Elisabeth of Bohemia and René Descartes (L. Shapiro (ed.)). The University of Chicago Press. Shoemaker, S. (2003). Identity, Cause, and Mind: Philosophical Essays. Clarendon Press. Sibley, F. (1971). Analysing Seeing. In F. Sibley (Ed.), Perception: A Philosophical Symposium (pp. 81–132). Methuen. Skinner, B. F. (1953). Science and Human Behavior. The B.F. Skinner Foundation. www.bfskinner.org/newtestsite/wp-content/uploads/.../ScienceHumanBehavior .pdf Skow, B. (2011). Experience and the Passage of Time. Philosophical Perspectives, 25, 359–387. https://doi.org/10.1111/j.1520-8583.2011.00220.x
142
Bibliography
Smart, J. J. C. (1959). Sensations and Brain Processes. The Philosophical Review, 68(2), 141–156. https://doi.org/10.2307/2182164 Soteriou, M. (2016). Disjunctivism. Routledge. https://books.google.cl/books?id =ZwneCwAAQBAJ Soteriou, Matthew. (2007). Content and the Stream of Consciousness. Philosophical Perspectives, 21(1), 543–568. https://doi.org/10.1111/j.1520-8583.2007.00135.x Soteriou, Matthew. (2011). Occurrent Perceptual Knowledge. Philosophical Issues, 21(1), 485–504. https://doi.org/10.1111/j.1533-6077.2011.00213.x Soteriou, Matthew. (2013). The Mind’s Construction: The Ontology of Mind and Mental Action. Oxford University Press. https://doi.org/10.1093/acprof:oso /9780199678457.001.0001 Soteriou, Matthew. (2017). The Ontology of Emotion. In H. Naar & F. Teroni (Eds.), The Ontology of Emotions (pp. 71–89). Cambridge University Press. https://doi.org /https://doi.org/10.1017/9781316275221 Soteriou, Matthew. (2018a). Experience, Process, Continuity, and Boundary. In R. Stout (Ed.), Process, Action, and Experience (pp. 82–101). Oxford University Press. https://doi.org/http://doi.org/10.1093/oso/9780198777991.003.0005 Soteriou, Matthew. (2018b). Process, Action, and Experience. In R. Stout (Ed.), Process, Action, and Experience (pp. 82–101). Oxford: Oxford University Press. Speaks, J. (2005). Is There a Problem about Nonconceptual Content? The Philosophical Review, 114(3), 359–398. https://doi.org/10.1215/00318108-114-3-359 Spener, M. (2012). Mind-Independence and Visual Phenomenology. In D. Smithies & D. Stoljar (Eds.), Introspection and Consciousness (pp. 381–404). Oxford University Press. https://doi.org/http://doi.org/10.1093/acprof:oso/9780199744794 .003.0014 Steward, H. (1997). The Ontology of Mind: Events, Processes, and States. Oxford: Clarendon Press. https://doi.org/10.1093/acprof:oso/9780198250647.001.0001 Steward, H. (2011). Perception and the Ontology of Causation. In R. Johannes, L. Hemdat, & E. Naomi (Eds.), Perception, Causation, and Objectivity (pp. 139– 160). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199692040 .001.0001 Steward, H. (2018). Occurrent States. In R. Stout (Ed.), Process, Action, and Experience (pp. 102–119). Oxford University Press. https://doi.org/10.1093/oso /9780198777991.003.0006 Strawson, P. F. (1970). Categories. In O. Wood & G. Pitcher (Eds.), Ryle (pp. 181– 211). Palgrave. https://doi.org/10.1007/978-1-349-15418-0 Taylor, B. (1977). Tense and Continuity. Linguistics and Philosophy, 1, 199–220. https://doi.org/https://doi.org/10.1007/BF00351103 Thau, M. (2002). Consciousness and Cognition. Oxford University Press. Tooley, M. (2013). Michael Huemer and the Principle of Phenomenal Conservatism. In C. Tucker (Ed.), Seemings and Justification: New Essays on Dogmatism and Phenomenal Conservatism (pp. 306–327). Oxford University Press. https://doi.org /10.1093/acprof:oso/9780199899494.003.0014 Travis, C. (2004). The Silence of the Senses. Mind, 113, 57–94. https://doi.org/10 .1093/mind/113.449.57
Bibliography
143
Tucker, C. (2010). Why Open-minded People Should Endorse Dogmatism. Philosophical Perspectives, 24, 529–545. https://doi.org/10.1111/j.1520-8583 .2010.00202.x Tucker, C. (2013a). Seemings and Justification: An Introduction. In C. Tucker (Ed.), Seemings and Justification: New Essays on Dogmatism and Phenomenal Conservatism. Oxford University Press. https://doi.org/10.1093/acprof:oso /9780199899494.003.0001 Tucker, C. (2013b). Seemings and Justification: New Essays on Dogmatism and Phenomenal Conservatism. Oxford University Press. https://doi.org/10.1093/ acprof:oso/9780199899494.001.0001 Turing, A. (1950). Computing Machinery and Intelligence. Mind, 59(236), 433–460. Tye, M. (1995). Ten Problems of Consciousness: A Representational Theory of the Phenomenal Mind. The MIT Press. Tye, M. (2003). Consciousness and Persons: Unity and Identity. The MIT Press. Vendler, Z. (1957). Verbs and Times. The Philosophical Review, 66, 143–160. https://doi.org/10.2307/2182371 Westerhoff, J. (2005). Ontological Categories: Their Nature and Significance. Clarendon Press. https://doi.org/10.1093/acprof:oso/9780199285044.001.0001 Wolgast, E. H. (1960). The Experience in Perception. The Philosophical Review, 69(2), 165–182. https://doi.org/10.2307/2183502 Zimmerman, A. (2012). Introspection, Explanation, and Perceptual Experience: Resisting Metaphysical Disjunctivism. In D. Smithies & D. Stoljar (Eds.), Introspection and Consciousness (pp. 353–380). Oxford University Press. https:// doi.org//10.1093/acprof:oso/9780199744794.003.0013
Index
accomplishments, 4, 18, 122–26, 129 achievements, 4, 18, 59, 70, 73, 122–23, 126, 129 activities, 4, 10, 18, 45, 55, 113, 122–26, 129, 133–36 Armstrong, D.M., 8–9, 14, 17–18, 21, 27, 55, 74 assertivity, 16–17, 49, 53, 96, 103–9, 111, 117–18
78, 81, 107, 126, 132, 135. See also dynamic phenomenon Comrie, Bernard, 119, 126, 128 constitution, 7, 21, 24, 35, 41, 48, 55, 81, 100. See also constitutive dependence constitutive dependence, 48, 55, 80, 91. See also constitution constraint, 79–80, 88–93, 107 continuity, 119, 125–29
behaviorism, 92–93, 96, 98–99, 101–3 Briesen, Jochen, 105, 107, 118 Cartesian epistemology and metaphysics, 45–47, 81, 83, 88, 92–93, 96–100, 102–3, 126 causal framework, 81–84, 86, 88, 90–91, 93, 96, 98, 102–3 causal ontology, 84, 87, 89, 92 causal structure, 16, 79, 85, 87–88, 91, 98 causal understanding, 48, 80, 83, 85–86, 90–91, 93 causation, 15–16, 47–48, 77, 79–89, 91–93, 96, 98, 101, 103, 108 Network Model of, 47–48 understanding of, 79, 81–82, 87, 93, 101 change, 2, 12–13, 19–20, 22–24, 30–33, 36–38, 40, 42–43, 49–54, 58, 64,
Dowty, David, 123, 133 dynamic phenomenon, 20, 77. See also change efficient cause, 16, 48, 79–92, 98–99, 101, 103. See also causation Evans, Gareth, 7–9, 11, 17–18 event, 2, 4–5, 11, 16, 18, 21–24, 26–29, 36, 38, 41, 45, 47–48, 50, 52, 55, 65–70, 72–73, 78–81, 84–86, 88–92, 99–103, 106, 114–16, 120, 122–23, 130, 133, 135 experiential content, 32, 39, 62–64, 68–69, 71, 131 Experiential Heracliteanism, 1, 5, 12–16, 19–21, 23, 25–27, 29, 31–36, 39–41, 43–44, 47–49, 51–52, 54–56, 58, 77, 80, 90, 93, 96, 111, 114–16,
145
146
Index
120–21, 129, 131–35. See also necessarily dynamic processes experiential individuation, 50, 58, 68, 70 Experiential Non-Heracliteanism, 1, 5, 13–16, 19, 20, 30–33, 35, 37, 39–41, 43–57, 64, 72, 74, 75, 77, 80, 88, 90–91, 93, 95–96, 103, 109–11, 114–15, 119–21, 129, 131, 133–35 snapshot version of, 50, 74, 95, 116 experiential process, 47, 68, 133 experiential state, 48, 50, 53, 71 experiential vehicle, 58, 131
mental state, 5, 13, 23–24, 31, 36, 40, 45, 50–51, 54, 57, 61, 65, 71–74, 96, 104–9, 117, 126–27 mind-body problem, 16, 47, 48, 66, 78, 79, 90, 93, 95–97, 101, 111 mind-brain identity theory, 96, 98–102 movers, 84, 85, 87. See also makershappen; matterers
Farrell, B.A., 4, 34, 66–67, 70, 72, 98 final cause, 81–83. See also efficient cause; formal cause formal cause, 16, 81, 91. See also efficient cause; final cause functionalism, 92–93, 96, 98, 100–101, 103, 107–10, 117–18
occurrence, 23–24, 28–29, 41, 45–46, 50–51, 66–67, 69, 85, 87, 113–15, 121–24, 126, 128–30, 132–33 occurrent state, 11, 14, 16, 24, 35, 40–41, 44–46, 49, 53, 103, 110 O’Shaughnessy, Brian, 5, 12–14, 19–23, 25–36, 43–44, 55, 60, 61, 65, 70, 114, 119, 121, 124–25, 127, 131–36
Glüer, Kathrin, 17, 104–6, 117–18 Heck Jr., Richard, 11, 17, 104–6, 108 homogeneity-heterogeneity distinction, 49, 55, 121, 123, 129, 130, 132–36 identity conditions, 31, 57–58, 72 informational relations, 14, 35, 45–46, 52, 54, 120 interlevel causal relations, 89, 91 Juarrero, Alicia, 79–80, 88–90, 92–93 lexical aspect, 119–36 makers-happen, 84. See also matterers; movers matterers, 79, 84–89, 91. See also makers-happen; movers mental events, 5, 12, 16, 47, 68–69, 72, 84–85 mental freeze, 13, 23, 25, 30–33, 35–36 mental process, 5, 13–14, 20, 23, 25–26, 29–33, 35–36, 57, 65, 67, 70–74, 99
necessarily dynamic processes, 1, 5, 19–20, 25, 33–35, 43–44, 54. See also Experiential Heracliteanism
perception: doxastic views of, 7–11, 17–18, 55, 106, 108 ontology of, 7, 32, 40, 44, 77, 80, 83, 89–90, 92–93, 95, 97, 99, 101, 103, 105, 107, 109, 111, 113–15, 117, 120–21, 125, 127, 129, 131, 134 perception of change, 49, 53 perceptual awareness, 19, 21–24, 26, 29, 53, 59, 74, 97, 111, 127 perceptual content, 4, 6–10, 17, 52, 56, 106, 117, 118 perceptual experience: cardinality questions about, 52–53, 72, 86, 110 dynamic character of, 5, 12, 15, 19–20, 23, 25, 30, 33–35, 40, 44, 47, 54 dynamic structure of, 1–5, 7, 9, 11, 13, 15, 17, 37, 54 individuation of, 58, 64–65, 67, 74 ontology of, 20, 24–25, 33, 52, 80, 96 phenomenal/ phenomenological character of, 4, 5, 10, 17, 55 phenomenology of, 6–9, 17, 25, 32, 37, 39, 96–97, 111, 132 temporal features of, 11, 16,
Index
29, 63–64, 66, 68–69, 95, 113, 131 temporal phenomenology of, 17, 32, 39, 96–97, 111 temporal structure of, 7, 13, 16, 23, 29, 32, 39, 55, 64, 66, 68, 77, 95, 113–14, 119–20, 124, 126, 131–32, 135 perceptual occurrence, 113, 128, 130 perceptual process, 14, 21–22, 29, 61, 66–67, 78, 84–86, 90–91, 115–17 processive view of experience. See Experiential Heracliteanism the progressive, 34, 49, 120–21, 123–26, 128–32, 135–36 property instantiation, 13–14, 24, 35, 40, 42, 45–46, 50, 52–54, 90–91, 120, 132, 134–35 psycho-physical materialism, 96, 98–100 relationalism, 25, 55 representationalism, 25, 49, 52–53, 55 Rothstein, Susan, 41, 49, 122–24, 133, 136 Soteriou, Matthew, 5–6, 11, 13–15, 19–20, 23–25, 31–36, 38–41, 43–44, 46, 48–56, 58, 70, 95–97, 109, 111–17, 119, 125–27, 131–32, 135
147
stative view of experience. See Experiential Non-Heracliteanism Steward, Helen, 5, 11–12, 19, 21, 24, 41, 47, 55, 65, 79–80, 83–89, 92, 95, 120, 122 telic-atelic distinction, 120–21, 123, 125, 129–30, 133–35 temporal individuation, 66–67, 74 temporal particulars, 21, 53, 57, 65, 72, 74, 84, 88 temporal parts, 16, 21, 41, 49, 51, 53, 57, 65, 72, 74, 84, 88, 113–16, 132, 133, 135 temporal passage, 25, 37–39, 49, 96 temporal stages, 129, 131–32 the temporal strategy, 11–12, 120–22 temporal succession, 38, 42, 46, 49–51, 55, 64, 120, 129, 131 temporal transparency, 32, 112–16, 131 Tooley, Michael, 105, 107–8, 117 Tye, Michael, 3, 6, 50, 52, 64, 68–69 Vendler, Zeno, 2–5, 11–15, 18, 22, 24, 28–29, 41, 44–48, 60, 95, 109, 119–20, 122–27, 129, 133–34, 136
About the Author
Sebastián Sanhueza Rodríguez is an assistant professor in the Department of Philosophy, University of Concepcion (Concepcion, Chile). After receiving his Philosophy Licentiature from the University of the Andes (Santiago, Chile), he completed his MLitt in mind and metaphysics at the University of Glasgow and his PhD at University College London. His work primarily focuses on perceptual experience and its relation to issues in ontology and philosophical psychology. He has published articles in a number of international academic journals, including Grazer Philosophische Studien, Acta Analytica, and Pacific Philosophical Quarterly.
149