(Download PDF) Perception First Form of Mind Tyler Burge 2 Full Chapter PDF

Perception: First Form of Mind Tyler
Burge
Visit to download the full and correct content document:
https://ebookmass.com/product/perception-first-form-of-mind-tyler-burge-2/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Perception: First Form of Mind Tyler Burge
https://ebookmass.com/product/perception-first-form-of-mind-
tyler-burge-2/
Perception: First Form of Mind Burge
https://ebookmass.com/product/perception-first-form-of-mind-
burge/
Leibniz’s naturalized philosophy of mind First Edition.

Edition Jorgensen
https://ebookmass.com/product/leibnizs-naturalized-philosophy-of-
mind-first-edition-edition-jorgensen/
Pieces of Mind: The Proper Domain of Psychological

Predicates First Edition. Edition Carrie Figdor
https://ebookmass.com/product/pieces-of-mind-the-proper-domain-
of-psychological-predicates-first-edition-edition-carrie-figdor/
Brentano’s Philosophical System: Mind, Being, Value
First Edition. Edition Uriah Kriegel
https://ebookmass.com/product/brentanos-philosophical-system-
mind-being-value-first-edition-edition-uriah-kriegel/
A Multisensory Philosophy of Perception Casey

O’Callaghan
https://ebookmass.com/product/a-multisensory-philosophy-of-
perception-casey-ocallaghan/
Evaluative Perception Anna Bergqvist
https://ebookmass.com/product/evaluative-perception-anna-
bergqvist/
Recognition and Perception of Images Iftikhar B.

Abbasov (Ed.)
https://ebookmass.com/product/recognition-and-perception-of-
images-iftikhar-b-abbasov-ed/
Undercover Wolf Paige Tyler
https://ebookmass.com/product/undercover-wolf-paige-tyler-2/
OUP CORRECTED PROOF – FINAL, 22/03/22, SPi
Perception
Perception
First Form of Mind
T Y L E R BU R G E
1
1
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Tyler Burge 2022
The moral rights of the author have been asserted
First Edition published in 2022
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2021942146
ISBN 978–0–19–887100–2 (hbk.)
ISBN 978–0–19–887101–9 (pbk.)
DOI: 10.1093/oso/9780198871002.001.0001
Printed and bound in the UK by
TJ Books Ltd
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
Acknowledgments
Chapter 1: Iris Murdoch, Existentialists and Mystics: Writings on Philosophy and Literature,
© 1950, 1951, 1952, 1956, 1957, 1958, 1959, 1961, 1962, 1964, 1966, 1969, 1970, 1972,
1977, 1978, 1986, 1997 by Iris Murdoch. Used by permission of Penguin Books, an imprint
of Penguin Publishing Group, a division of Penguin Random House LLC. All rights
reserved.
Chapter 3: Stephen E. Palmer, Vision Science: Photons to Phenomenology, The MIT Press,
1999, p. 32.
Chapter 6: W.G. Sebald, translated by Michael Hulse, from The Rings of Saturn, © 1995 by
Vito von Eichborn GmbH & Co Verlag KG. Translation © 1998 by The Harvill Press.
Reprinted by permission of New Directions Publishing Corp.
Chapter 11: William Blake, Notebook, British Library Add MS 49460, 3rd and 4th lines
from first draft of the poem.
Chapter 16: Excerpt from Vladimir Nabokov, Speak, Memory: An Autobiography Revisited
© 1947, 1948, 1949, 1950, 1951, 1967 by Vladimir Nabokov. Used by permission of
Alfred A. Knopf, an imprint of the Knopf Doubleday Publishing Group, a division of
Penguin Random House LLC. All rights reserved.
Chapter 19: Hans Moravec, Mind Children: The Future of Robot and Human Intelligence,
© 1988 by Hans Moravec. Cambridge, MA: Harvard University Press.
Contents
Preface xi
Animal Eyes xxi
Figures xxii
Abbreviations xxiv
PA RT I . P E R C E P T IO N
1. Introduction 3
Biological Function, Action, Sensing, and Perception—The Emergence of Mind 3
Principal Aims of the Book 10
The Fregean Source of Key Semantical Notions for Perception 15
2. Perception 19
Perceptual States as Sensory States 20
Representation and Information Registration 21
Representation and Veridicality Conditions 28
Representations and Representational Contents 30
The Three Fundamental Representational Constituents in Perceptual States 36
The Basic Representational Form of Perceptual States 49
Perception as Objectification 50
Perceptual Constancy—First Mark of Representational Mind 60
3. Perceptual Constancy: A Central Natural Psychological Kind 64
Scientific Practice Demarcates Perceptual Constancies from
Other Invariances 65
Two Misguided Ways of Thinking About Perceptual Constancies 70
Even Non-Perceptual Invariances Contribute to the Fitness of Individuals
that Sense 74
Efference Copy: An Example of a Non-Perceptual Invariance 76
Path Integration: Another Non-Perceptual Invariance 78
Very Simple Perceptual Color and Lightness Constancies 80
Retinal Image Contour Registration and Surface Contour Perception 82
Visual Spatial Property and Relation Constancies 90
Visual Body Categorization 91
Visual Spatial Perceptual Constancies and Body Categorization 94
Visual Temporal Perceptual Constancies 99
PA RT I I . F O R M
4. Some Basics about Perceptual Systems 109

Principles Governing Transitions Contrasted with Representational Contents 109
Perception, Computation, and the Language-of-Thought Hypothesis 111
Representational-Dependence Hierarchies in Perceptual Attribution 118
Two Methodological Points About Natural-Kind and Functional Attributives 120
Taxonomic Hierarchies in Perception 124
viii contents
5. Perceptual Reference Requires Perceptual Attribution 131

Basic Form of Perceptual Contents 131
Perceptual Reference is Partly Guided by Perceptual Attribution 133
Support for (AA1) 134
Support for (AA2) 143
General Remarks on Attributives and Perceptual Discrimination 146
Criticism of Two Attempted Rejections of (AA1) and (AA2) 148
6. Form and Semantics of Representational Contents of Perceptual States 156
Review of Basic Form of Perceptual Representation 156
Perception of Property-Instances 159
Betokening and Four Types of Perceptual Attribution 169
Perceptual Attribution of Relations 175
Scope Hierarchies in Perceptual Content 179
Scope and Modificational Attribution Hierarchy 185
Absence in Perception of Negations, Conditionals, Disjunctions,
Quantifiers 190
Perceptual Contents, Propositions, and Noun Phrases 201
7. Perceptual Attributives and Referential Applications in Perceptual
Constancies 208
Perceptual Constancies and Frege’s Sense–Bedeutung Distinction: Similarities 208
Perceptual Constancies and Frege’s Sense–Bedeutung Distinction:
Differences 213
Minimalism: Defocus and Color Constancy 214
Minimalism and Iconic Representation in a Spatial Coordinate System 219
Perceptual Units in Packages in Iconic Visual Spatial Representation 233
Linkage of Different Perceptual Attributives in Perceptual Constancies 237
The Form of Perceptual Attributives in Linkages 241
Accuracy Conditions for Perceptual Attributives in Perceptual Constancies 244
Referential Applications in Accuracy Conditions for Tracking Particulars 248
8. Egocentric Indexing in Perceptual Spatial and Temporal Frameworks 255
Egocentric Spatial Indexes in Perception 255
Egocentric Temporal Frameworks and Perceptual Representation of Motion 264
Is Temporal Representation Constitutive to Perceptual Representation? 275
9. The Iconic Nature of Perception 293
Noun-Phrase-Like Structure and Iconic Representation in Perception 296
Iconic Aspects of Perceptual Spatial Representation 304
Temporal, Qualitative, and Packaging Iconic Aspects of Visual Perception 312
Iconic Visual Perception and Maps or Pictures 314
Some Ways Not to Think about Iconic Representation 315
Iconic Perception, Iconic Concepts, Iconic Representation in Propositional
Thought 331
Part–Whole Representation in Pictures and Visual Perception 334
Compositionality in Iconic Perceptual Representation 347
Spatial Mapping in Visual Perception Again: The Non-Planar Surface
of the Scene 350
Relations Between Iconic Format and Representational Content 355
The Tractability of Iconic Attributional Complexity 360
contents ix
PA RT I I I . F O R M AT IO N
10. First-Formed Perception: Its Richness and Autonomy 367

What are First-Formed Perceptual States Like?: Three Limitative Views 367
Framework 372
Marr’s 2½-D Sketch 378
Change Detection 380
Treisman’s Binding Theory 383
Two Lines of Empirical Criticism of Treisman’s Theory 391
Philosophical Views Influenced by Treisman’s Binding Theory 401
Two Types of First-Formed Perception 405
11. Intra-Saccadic Perception and Recurrent Processing 409
Two Changes in Scientific Understanding of Perception-Formation 409
Some Main Brain Areas Involved in Visual Processing 413
Timing of Visual Processing; Some Main Types of Representation 420
Categorization and Timing 423
Levels of Specificity in Perceptual Categorization 427
Perceptual Constancies in Categorization Processing 431
12. Further Attributives: Primitive Attribution of Causation, Agency 433
Methodology for Finding Perceptual Attributives 433
Primitive Attribution of Mechanical Causation 446
Primitive Attribution of Agency 466
Attribution of Further Structural Elements of Agency 475
PA RT I V. SYS T E M
13. Perceptual-Level Representation and Categorization 483

Perceptual Categorization is Perceptual 487
Richer Perceptual Categorization and Perceptual Processing that
Contributes to It 493
14. Conation: Relatively Primitive, Perceptually Guided Action 502
Action Imperialism 502
Relatively Primitive Action 504
Form of Relatively Primitive Conative States 508
Broader Structure of Conation in Causing Relatively Primitive Action 512
Summary: Philosophical Issues 526
15. Perceptual Attention 531
Forms of Perceptual Attention 532
Attention and Accuracy 537
Sources and Levels of Attention 545
Perceptual-Level Attention Commands and Guidance of Saccades 547
The Executive Control System and Propositional Drivers 556
Supra-Perceptual Effects on Perceptual-Level Operations: An Example 565
16. Perceptual Memory I: Shorter Term Systems 567
Perceptual Memory and Consciousness 569
Priming and Memory 570
x contents
Visual Sensory Memory 572

Fragile Visual Short-Term Memory 575
Trans-Saccadic Memory 576
Visual Working Memory 584
Conceptual Short-Term Memory 599
17. Perceptual Memory II: Visual Perceptual Long-Term Memory 609
Overview 609
Ability-General Long-Term Visual Perceptual Memory 613
Episodic Visual Memory; De Re Long-Term Non-Episodic Visual Memory 619
Perceptual and Conceptual Attributives in Long-Term Memory 622
Summary of Relations Among Major Types of Visual Perceptual Memory 623
18. Perceptual Learning, Perceptual Anticipation, Perceptual Imagining 625
Perceptual Learning 625
Perceptual Anticipation 630
Perceptual Imagining 641
19. Perception and Cognition 647
The Original Epistemic Grounds for Reflecting on Cognitive Influence
on Perception 649
Fodor and Pylyshyn’s Conceptions of Modularity; The Visual System
as a Module 655
Uses and Misuses of the Term ‘Cognition’ 662
The Issue of Cognitive Penetration 677
Framework Issues 680
Conceptions of Penetration 684
The Cognitive Penetration Controversy 693
A Computational Construal of Modularity 701
Psychological Systems and Psychological Kinds 710
The Empiricist Model of Perception and Conception: Degrees of Abstraction 714
What Should Count as Cognition? 722
20. Conclusion 735
Emergence of Representational Mind 735
Empirical Characteristics of First-Formed Perceptions 738
Changes in the Science; Reading the Changes Philosophically 740
Perception: Form and Representational Content 744
Perception: The Seed of New Things to Live and Die For 746
Bibliography 749
Author Index 813
Subject Index 839
Preface
This book is not the one that I set out to write over twelve years ago. I had intended to fol-
low Origins of Objectivity with an account of what is distinctive about the main represen-
tational capacity more advanced than perception. I thought, and still think, that that
capacity is propositional representation. I took some steps in that direction in the Petrus
Hispanus Lectures, Lisbon, 2009, and the Nicod Lectures, Paris, 2010, and developed an
argument for connecting propositionality, constitutively, with propositional deductive
inference. I remain interested in that other book, and hope to complete it. But it was
pre-empted.
In writing a lead-up to discussing propositional capacities, I wanted to elaborate an
account of perception. Perception is, I think, the first representational capacity to evolve. It
is the main pre-propositional representational capacity. The lead-up was intended to be a
relatively concise refinement of the account of perception in the last chapters of Origins of
Objectivity, a refinement that now occupies approximately Chapters 1–3 of this book.
I became interested in vision just before I spent a semester teaching at the Massachusetts
Institute of Technology (MIT) in 1982. My interest resulted in part from reading David
Marr’s important book, Vision. Although Marr died before I arrived at MIT, I audited
courses in perceptual psychology by Marr’s former colleagues during the visit. I later wrote
some articles on vision and made it a central theme of Origins of Objectivity (2010).
During work on that book and later, I benefitted from discussions of vision science with
my son, Johannes Burge, who is currently a vision scientist at the University of
Pennsylvania. After 2010, I came to understand the science more thoroughly, and devel-
oped ideas about structural and semantical issues associated with perception. I did so
through giving a series of graduate seminars at University of California Los Angeles
(UCLA), and through discussing the science with various scientists—especially in person
with David Brainard, Bill Geisler, and Shimon Ullman, and via correspondence with
Jeffrey Schall and Yaffa Yeshurun—and with Ned Block, one of few philosophers who ser
iously engages with the science. In writing the refinement of the Origins account of visual
perception, I soon realized that I had too much material to present in a run-up to another
topic, in a single book. The preliminary became the whole.
The book is about first form of representational mind. I take perception to be the most
primitive type of representational mind. Relevant first form is three-fold.
One type of first form is the representational structure of perception. Here, first form is
the first representational form that emerges in the evolution of representational mind.
This form is center-stage in Chapters 1–9. I had the main ideas about this structure when I
wrote Origins, but I discovered much more in thinking through its details, applying it to
cases, developing a semantics for it, and reflecting on how it is embedded in the iconic
format of perception.
A central theme in the book, centered in Parts I and II, is developing the foundations
for a systematic semantics for perceptual states. Perceptual psychology takes perceptual
states to be accurate or inaccurate. One of its main aims is to explain, causally, how
xii preface
accurate and inaccurate perceptual states are formed. It has, however, paid little attention
to explaining what it is for such states to be accurate or inaccurate, or to reflecting on how
representational capacities combine to yield a form of representation. Nor have these
issues been related to the obvious iconic nature of perceptual representation. I take steps
toward remedying this situation. In the course of doing so, I think that I have discovered
some basic aspects of attribution, the root of predication. Predication in language and
thought has been a longstanding topic of philosophical reflection, beginning with Aristotle
and Kant, and developing mainly in Frege, but also in Tarski and Strawson. Traditionally,
the topic has been pursued by reflection on logic. I have found it enlightening to reflect on
its roots in perception.
The second type of first form of representational mind is the first-formed states in the
order of perceptual processing. What is the nature of the perceptual states that are formed
fastest? What properties in the environment do they represent? What sort of processing
leads to them? There are empirical answers to these questions, at least for mammalian
vision. These answers provide a starting point for reflecting on what the fastest-formed
perceptual states are like in lower animals and even in evolutionary history. Of course,
each species must be considered on its own. I do not much discuss lower animals, nor do I
provide an evolutionary account, although I occasionally comment on those topics. This
second type of first form is center-stage in Chapters 10–11. Issues about later-formed per-
ceptual representations, such as those used in perceptual recognition of individuals, and
perhaps for causation, agency, and functional attributes, such as mate or edible, are
touched on in Chapter 12.
The third type of first form of representational mind is a natural-kind system of repre-
sentational capacities, with perception at the representational center of the system. The
systems that I highlight are the visual-perceptual system and the visuo-motor system.
The two systems intersect and overlap. What it is to be part of these systems is the
central theme of Part IV, Chapters 13–19. I believe that perception shares its representa-
tional structure and content, outlined in Chapters 2–12, with several other representational
capacities. Generically, the capacities are conation, attention-initiation, memory, affect,
learning, anticipation, and imagining. I think that the listed generic capacities have
perceptual-level species—species that have representational structures and contents that are
essentially those of perception. These capacities differ from perception in mode (memory
vs. perception) and transition-operations, not in form or content. The notion of represen-
tational level is explained in Chapter 1, the section The Principal Aims of the
B ook, and again, more fully, at the beginning of Chapter 13.
Perceptual-level species of the listed capacities share attributional content with percep-
tion. They share attributional and iconic structure with perception. And they involve
operations or transformations that are either similar to those in perception-formation
itself, or at least not more sophisticated than they are. These sub-species join with percep-
tion to form two large, natural-kind psychological systems—the perceptual system and
the perceptual-motor system. These systems are unified (a) in sharing a function (contrib-
uting to perception in the first case, contributing to perceptually guided action in the sec-
ond), and (b) in sharing the representational structure of perception. They are also unified
(c) in using only representational attributive contents in or borrowed from perception;
and (d) in being held together by computational causal processes both within perception
and between perception and the perceptual-level species of the listed generic capacities.
preface xiii
Again, the relevant perceptual-level species are perception-guided conation, perceptual-

attention initiations (or attention commands), perceptual memory, perceptual affect, per-
ceptual learning, perceptual anticipation, and perceptual imagining.
I think that these two large, overlapping systems—the perceptual system and the
perceptual-motor system—constitute the natural-kind center of lower representational
mind. I conjecture that variants of these systems, sometimes perhaps omitting one or two
of the auxiliary capacities—such as perceptual imagining—or adding other auxiliary
capacities—like amodal mapping—occur in all animal perceivers, from insects to human
beings. These systems and the perceptual-level capacity-species are discussed in Chapters
13–19, focusing on visual perception and visually guided action.
The three notions of first form are associated with the general plan of the book. Parts I
and II discuss what perception is. Part III centers on how perception works—how it is
formed and the nature and scope of its processing. Part IV treats relations between per-
ception and satellite capacities, and wherein they form a unified perceptual system and
perceptual-motor system.
A further theme runs through nearly all parts of the book. The theme is opposition—
sometimes implicit, sometimes explicit—to a way of thinking about perception started by
the classical empiricists, Locke and Hume. In trying to understand the main features of
perception, this approach takes too seriously intuitive groupings and senses of similarity
via introspection of conscious perceptual experience. It also errs by distinguishing percep-
tion from conception primarily in terms of differences in abstraction. Perception is sup-
posed to operate toward the concrete end. Conception, toward the abstract end. The
spectrum is intuitive—not very carefully explained. My opposition is centered in reflect-
ing on the explanatory practice of perceptual psychology. It also derives from reflecting on
the form, content, function, capacities, and uses that turn up in systematic explanation—
causal and semantical.
As with language, so with perception, understanding representational form and content
depends not on introspection of intuitively salient features of the vehicle of representation
(sentence or image-like-representation), but on competencies that underlie use of the rep-
resentation. Intuitive reflection on sentential structure would treat it as linear and as simply
being composed of a string of words. Reflection, in logic and linguistics, informed by sys-
tematic consideration of sentential use, reveals a hierarchical structure with words forming
phrase units that are not immediately obvious to intuition. Finding representational struc-
ture in perception is parallel. It depends not primarily on introspection of conscious per-
ceptual images, but on systematic reflection on the capacities evinced in uses and
transformations of perceptions. This reflection must consider the representational func-
tions of perception and the structures of capacities discovered in empirical science. Most of
what there is to be understood about perception is unconscious. All operations that lead to
perceptions are unconscious, and there are many unconscious perceptions. Moreover, given
the way in which perceptual discriminative capacities, and their use in guiding action, are
molded, in evolution and learning, by perceiver-needs and by frequently unobvious causal
and statistical patterns in the environment, one cannot rely on introspection of images,
and intuitive types of image-similarity to understand perception—even the iconic, broadly
image-like character of vision. Perceptual grouping does not in general conform to intuitive
image-similarity. Yet all perceptual groupings are grounded, at least partly, in shape-size-
motion attributions, plotted iconically in an image-like way in perceptual representation.
xiv preface
The book is resolutely a work in philosophy of science—specifically philosophy of

psychology. Perceptual psychology, centrally the psychophysics of vision, has become a
mature science in the last fifty years. It gives philosophy an opportunity to understand
important features of psychological capacities at a level of depth, rigor, and empirical
groundedness that has never before been attainable. Philosophy should leap at the oppor-
tunity to make use of such a powerful and rapidly advancing science, as a basis for philo-
sophical understanding. Some philosophy of perception makes no use at all of perceptual
psychology. Much philosophy of perception makes at best decorative use. I think that it is
no longer intellectually responsible to philosophize about perception without knowing
and seriously engaging with that science. I believe that the practice of centering philo-
sophical reflection about perception on phenomenology, or on analysis of ordinary talk
about perception, without closely connecting the reflection with what is known from sci-
ence (a centering that is a residue of the early empiricist model of perception), and the
practice of allowing epistemology to guide reflection on what perception must be like, will
all soon become museum pieces of past, misdirected philosophy.
Most of the book’s claims are, of course, supported only empirically, by interpreting the
empirical results of the science. Some of the claims are, however, supported apriori. One
should not confuse apriority with innateness, certainty, obviousness, infallibility, dogma-
tism, unrevisability, or immunity from revision based on empirical considerations. To
be apriori supported, or apriori warranted, is to have support or warrant that does not
depend for its force on perception or on sensing. Most apriori warranted judgments in
this book are warranted by reflection that yields understanding of key concepts or prin
ciples used or presupposed in the science. All the relevant apriori judgments are synthetic,
certainly in the sense of being non-vacuous and the sense of not being truths of logic.
I think that the judgments are also synthetic in the sense of not being the products of ana
lysis of conceptual complexes into concepts contained in the complexes. I think that most
concepts that are central to our discussion are not complexes. They are simple. They are,
however, necessarily and apriori embedded in networks with other concepts. Reasoning
through such networks sometimes yields synthetic apriori understanding of foundations
of mind.
Apriori supported judgments can be further supported empirically, by the science. But
insofar as they are apriori warranted, they have sufficient warrant to support belief; and
the warrant derives from reasoning or understanding, independently of support from per-
ception, perceptual experience, or sensory registration. An example of an apriori war-
ranted judgment, I think, is that perceptual states can be accurate or inaccurate. Another
example is that perceptual states have a representational function—to accurately pick out
and characterize particulars via causal relations to them: perceptual states fail in some way
(representationally) if they are not accurate. I doubt that one can know apriori that any
individual has perceptual capacities. Our empirical knowledge that we do have such
capacities is, however, firm. It is more certain than some things that we know apriori about
perception. As noted, being apriori does not imply some super-strong type of support.
Apriori warrant for belief in simple arithmetical truths is super-strong. But much apriori
support is not stronger, often less strong, than strong empirical support.
Our firm empirical knowledge that individuals have perceptual states does not require a
detailed, reflective, philosophical understanding of what perception is. Knowing that indi-
viduals have perceptual states requires only a minimal understanding. One must be able to
preface xv
distinguish perception, at least by some cases, from just any sensing. And one must be able
to recognize various examples of perception. Detailed philosophical understanding
requires reflection, articulation, and elaboration of a minimal understanding of the con-
cept perception and of relations between perception and other matters—semantical, func-
tional, biological, causal, and so on. Elaboration is mainly empirical, but partly apriori. Given
an elaborated understanding of what perception is, it is possible to draw, apriori, some fur-
ther conclusions about the form, semantics, and functions of perceptual states. Such con-
clusions are abstract and limited. They are important in being basic to understanding.
Again, most of the book’s claims are empirical. For example, the accounts of how per-
ceptual and perceptual-motor systems work in Parts III and IV, and the accounts of what
these systems are in Part IV, are warranted partly by appeal to explanations in the science.
Those accounts and those explanations are certainly empirically, not apriori, warranted.
I became interested in perception partly because it promises insight into basic types of
representation of the world, and partly because it is a key factor that must be understood if
one is to understand empirical knowledge. This book shows some fruits of the first motiv
ation. In investigating the structure and semantics of perceptual representation, one inves-
tigates primitive and basic types of reference and attribution. My interest in the role of
perception in empirical knowledge remains. But I take understanding perception to owe
almost nothing to epistemology, whereas understanding epistemology absolutely requires
understanding perception. Epistemology investigates epistemic norms for capacities that
can contribute to obtaining knowledge. One cannot understand the norms without under-
standing the capacities. One understands perceptual capacities by reflecting on empirical
science and its basic commitments, not by reflecting on epistemology. Understanding per-
ception is the task of this book. Epistemic use of an understanding of perception is poster
ior. For epistemic work in this direction, see my ‘Perceptual Entitlement’, Philosophy and
Phenomenological Research 67 (2003), 503–548; and ‘Entitlement: The Basis for Empirical
Warrant’, in P. Graham and N. Pedersen eds., Epistemic Entitlement (Oxford: Oxford
University Press, 2020).
I have some slight hope, even in this specialized world, that this book will interest not
only philosophers, but at least some scientists in perceptual psychology and other areas of
psychology. The best science is informed by a breadth and depth of perspective that is
philosophical. This point is particularly relevant to perceptual psychology. A central, often
stated, aim of the science is to understand conditions in which accurate perception occurs,
and conditions under which illusions occur. (See Chapter 1, note 25.) Accuracy is a
semantical concept.
So the science is committed at its very core to there being a semantics for perception—a
systematic account of relations between perceptual representation and its subject matters.
The account must explain what it is for perception to be accurate or inaccurate. Of course,
the science is mainly concerned with causal patterns and mechanisms. Much of it, indeed
probably most of it to date, focuses on pre-representational, pre-perceptual states that
register the proximal stimulus. But the point of this scientific work is partly to build
toward understanding perception of the physical environment. Part of understanding per-
ception scientifically is to understand not only the causal patterns that lead to accurate
and inaccurate perception, but also to understand the form and content of perceptual
states, and what it is for them to be accurate or inaccurate.
xvi preface
Yet the science has paid no serious attention to these issues—specifically to semantics.
It has not developed a vocabulary or set of principles that enable it to discuss accuracy and
inaccuracy of perception with the precision and clarity of its accounts of causal, forma-
tional aspects of psychological states and processes. It provides no answers to questions
like ‘What is it for a perceptual state to be accurate or inaccurate?’, ‘What sorts of represen-
tational competencies are involved in forming a state that is accurate or inaccurate?’; ‘In
what ways can a perception be partly accurate and partly inaccurate?’; ‘What is the repre-
sentational form or structure of perceptual states?’. Such questions are addressed in Parts I
and II of the book.
Scientific understanding of perception is incomplete if it does not incorporate a system-
atic semantical understanding of perceptual states into its understanding of principles
according to which perceptual states are causally generated. Semantical understanding is
understanding of the representational contents, their forms, and their accuracy condi-
tions—the conditions for representational success. Perceptual psychology would benefit
from mastering the vocabulary necessary to think systematically about the semantics of
perception.
Philosophy is the source of modern work in semantics—first the semantics of mathem
atics and logic, later the semantics of natural language. The basic semantical concepts, in
something like their modern form, come from Gottlob Frege, about 130 years ago. In the
last section of Chapter 1, I explain some of Frege’s basic concepts. I think that these con-
cepts, with some modification, are valuable in understanding perception, even though
they were first developed for understanding much higher-level representation—represen-
tation in mathematics.
I think that parts of the science need not only a deeper grip on semantics, but a much
more rigorous terminology. Uses of terms like ‘representation’, ‘knowledge’, ‘cognition’,
‘recognition’, ‘judgment’, ‘belief ’, ‘concept’, ‘prediction’, ‘intention’, ‘voluntary’ are far from
reflective, much less standardized, in the science. Assimilating the whys and wherefores of
terminology, is often the beginning of better, more fruitful empirical inquiry. Centrally, in
Chapter 19, the section Uses and Misuses of the Term ‘Cognition’, but also
throughout the book, there is a concerted effort to emphasize sharper uses of key mental-
istic terms so as to respect basic differences in representational level. Such differences cor-
respond to important differences in representational kinds—that is, representational
capacities.
This is a long, complex book. Understanding anything well requires effort and patience.
Genuine philosophical and scientific understanding cannot be grabbed off the shelf. The
time and effort required to understand this book will be considerable. One cannot get
there in a few sittings. The key point is to read and reread carefully and slowly, noting and
reflecting on nuances and qualifications, mastering terminology, reading in context, con-
necting different contexts together, reading the footnotes, going back to earlier passages—
all the while, reflecting. Few readers outside philosophy ever read this way. Most
philosophers have, I think, lost the art. Iris Murdoch, in harmony with the marvelous
quote that heads Chapter 1, wrote: ‘In philosophy, the race is to the slow’. Too many race at
high speeds. The psychological and sociological pressures to form opinions and publish
them quickly, and often, are very strong. Academic pressures and computer fluency have
yielded much more writing, with no more time to master the increasingly complex topics
written about. Careless reading, misdirected criticism, uninformed opinions, simplistic
preface xvii
proposals abound. Perhaps it was always so. However, as knowledge grows—and grows
more complex—lack of patience in pursuing understanding is an increasingly debilitating
vice. Given that philosophical understanding of this book’s topics has become harder—
because more is known and what is known is more complex—patience is more required
than ever.
For those who may have some interest in the book, but are not initially willing to invest
large of amounts of time in it, I set out a plan for getting some of the book’s gist. My hope
is that after achieving an overview, some readers will be tempted to go back for more—not
just noting fine points, but mastering the book’s conceptual framework and conceptual
intricacies.
For vision scientists, Parts I–II deal with the least familiar ideas—principally, the repre-
sentational form and semantics of perception. Part III and Chapters 13–18 in Part IV
mainly describe matters that many vision or visuo-motor scientists are familiar with,
although these passages cast those matters in a conceptual form that may be more unified
than specialized accounts provide. Chapter 19 of Part IV, supported by Chapters 13–18,
presents a large view of the visual system and visuo-motor system that uses perhaps famil-
iar empirical materials to develop a possibly less familiar, or more sharply articulated,
view of those systems as wholes.
For those, either vision scientists or others, who simply want to get a sense for the main
lines of thought, I offer the following, tentative guide.
I recommend, in Part I, the first two sections of Chapter 1 and all of Chapter 2. In
Chapter 3, I recommend the prologue and the section Retinal Image Contour
Registration and Surface Contour Perception. This section illustrates con-
cretely how I think perceptual constancies are distinguished from other invariances in
visual perception.
In Part II, I recommend the prologue at the beginning of Chapter 4, and the third and
fifth sections of the chapter (Representational-D ependence Hierarchies in
Perceptual Attribution and Taxonomic Hierarchies in Perception). In
Chapter 5, I recommend the first two sections (Basic Form of Perceptual
Contents and Perceptual Reference is Partly Guided by Perceptual
Attribution). In Chapter 6, I recommend the first three and the last of the sections
(Review of Basic Form of Perceptual Representation; Perception of
Property Instances; Betokening and Four Types of Perceptual
Attribution; and Perceptual Contents, Propositions, and Noun
Phrase). Chapter 7 gives the central account of the semantics of perception. This may be
hard to follow if one has skipped too much of the slow development of concepts that lead
up to it. The key section is the next- to-
last one (Accuracy Conditions for
Perceptual Attributives in Perceptual Constancies). Chapter 8 centers
on the way space and time are represented in visual perception. It can be skipped by those
not interested in details of perceptual representational structure. Chapter 9 connects basic
representational structure (the structure of reference and attribution) with the iconic form
of visual perception. The key sections are the first and the last two (Noun-P hrase-L ike
Structure and Iconic Representation in Perception; Relations
Between Iconic Format and Representational Content; and The
Tractability of Iconic Attributional Complexity).
xviii preface
In Part III, I recommend the prologue at the beginning of Chapter 10 and the first and
last sections of that chapter (What Are First-F ormed Perceptual States
Like?: Three Limitative Views and Two Types of First- F ormed
Perception). In Chapter 11, I recommend the prologue and the first section (Two
Changes in Scientific Understanding of Perception-F ormation). The
second section is a brief run-through of the course that visual processing takes in the
brain. The main points will be familiar to any vision scientist. The remaining sections pro-
vide basic facts about timing and about perceptual categorization. They may be skimmed
or skipped by those seeking only gist. All these sections in Chapter 11 are mainly descrip-
tive—with some improvements, I think, in conceptual rigor—of facts familiar to many
vision scientists. Chapter 12 has a methodological discussion of considerations that bear
on whether an attributive is, or should be, considered a perceptual attributive—an attribu-
tive, or characterizer, generated in the visual system (Methodology for Finding
Perceptual Attributives). That section tries to codify and state more precisely
methods already in place in the science.
Part IV treats the visual system at a level of generality far from the focused empirical
studies that make up most of the science. The prologue for Chapter 13 is recommended. It
sets the plan of Part IV. Readers trying for an overview can pick and choose within, or
skip, Chapters 13–18. These chapters discuss categorization, conation, attention, various
forms of short- term memory, long- term memory, affect, learning, anticipation, and
imagining. The chapters develop the idea that all these capacities have functions at the
representational level of perception. As noted, the notion of level is explained briefly in
Chapter 1 and in more detail at the beginning of Chapter 13. The chapters are part of an
extended argument that the perceptual-level functions and operations of these capacities
occur within the visual system or the visuo-motor system, in a specific sense of ‘occur
within’. Chapter 19 can be read by sampling the beginning of each section, then determin-
ing how much further to read. Chapter 19 uses work in Chapters 13–18 to develop the
idea that the perceptual system—for which the visual system is paradigmatic—and the
perceptual-motor system are lower representational mind. A proposal for understanding
the unity of lower representational mind is advanced in A Computational
Construal of Modularity. Although I explore some intermediate territory in
What Should Count as Cognition?, I think that lower representational mind
contrasts most dramatically with capacities for propositional inference and language, the
more primitive of which constitute the first capacities in upper representational mind. I
believe that the most important feature of upper representational mind is competence to
produce explanations. This capacity develops into science, moral thinking and practice,
and, most broadly, into understanding. The section The Empiricist Model of
Perception and Cognition: Degrees of Abstraction brings together criti-
cisms, which recur throughout the book, of an old way of thinking about the relation
between perception and thought that, while not prevalent in psychophysics of perception,
remains widespread in other parts of psychology and in philosophy. This section might be
useful to psychologists as well as philosophers. Chapter 20 summarizes main themes, and
concludes.
For valuable input, I am grateful to: Marty Banks, John Bartholdi, Blake Batoon, Ned
Block, David Bordeaux, David Brainard, Denis Buehler, Daniel Burge, Dorli Burge,
Johannes Burge, Susan Carey, Sam Cumming, Will Davies, Frank Durgin, Chaz Firestone,
preface xix
Julian Fischer, Bill Geisler, Katherine Gluer, Peter Godfrey-Smith, Peter Graham, Gabe
Greenberg, Michael Hansen, Catherine Hochman, Gabby Johnson, Nancy Kanwisher, Ed
Keenan, Philip Kellman, Ela Kotkowska, Bill Kowalsky, Kevin Lande, Hakwan Lau, Gavin
Lawrence, Yannig Luthra, Tony Martin, Peter Momtchiloff, Bence Nanay, Peter Pagin, Alexi
Patsaouras, Christopher Peacocke, Sam Pensler, Atthanasios Raftopoulos, Jacob Reid,
Michael Rescorla, Pieter Roelfsema, Jeffrey Schall, Brian Scholl, Wesley Seurat, Houston
Smit, Sheldon Smith, Irene Sperandio, Galen Strawson, Paul Talma, Shimon Ullman,
Tamar Weber, Fei Xu, Yaffa Yeshurun. Thanks also to three anonymous reviewers. The two
complex drawings, Figures 9.2 and 9.3, are by Bill Kowalsky, as are the other drawings,
except for one by Johannes Burge.
My immediate family—sons and wife—has supported and taken interest. I am espe-
cially grateful to my wife, Dorli, for patience in understanding her often distracted hus-
band and for helping me to look more carefully at the world, especially appreciating details
and finding joy in the beauties of little things.
Tyler Burge
Animal Eye Grid
1 2 3 4 5 6
7 8 9 10 11 12
13 14 15 16 17 18
19 20 21 22 22 24
25 26 27 28 29 30
31 32 33 34 35 36
Animal eye grid credit lines: (1) Tarsier monkey, © nikpal/Getty Images; (2) Horse, © Magdalena Strakova/EyeEm/
Getty Images; (3) Cat, © Dobermaraner/Shutterstock.com; (4) Racing pigeon, © triocean/Getty Images;
(5) Red-eyed tree frog, © Mark Kostich/Getty Images; (6) Indian elephant, © Volanthevist/Getty Images; (7) Bearded
vulture, © Slowmotiongli/Dreamstime.com; (8) Robber fly, © suwich/Getty Images; (9) Blue-spotted pufferfish,
© Shahar Shabtai/Shutterstock.com; (10) Deer fly, © ekamelev/Unsplash; (11) Goat, © Tsuneya Kudoh/EyeEm/Getty
Images; (12) Iguana, © Gaschwald/Dreamstime.com; (13) Grey butterfly, © egor-kamelev/Pexels; (14) Emerald tree
boa, © George lepp/Getty Images; (15) Ant, © Visual&Written SL/Alamy Stock Photo; (16) Jumping Spider, © Razvan
Cornel Constantin/Dreamstime.com; (17) Discus Fish, © tunart/getty Images; (18) Tokay Gecko, © Spaceheater/
Dreamstime.com; (19) Octopus, © iStock.com/leventalbas; (20) Eurasian Eagle Owl, © GlobalP/Getty Images;
(21) Sumatran tiger, © ekamelev/Unsplash; (22) Human, © Tatiana Makotra/Shutterstock.com; (23) Chameleon,
© aluxum/Getty Images; (24) Bee, © ekamelev/Unsplash; (25) Chimpanzee, © Karl Ammann/Getty Images;
(26) Green tree python, © Zoran Kolundzija/Getty Images; (27) Finschs Imperial Pigeon, © Marc Dozier/Getty Images;
(28) Sparrowhawk, © Katie Kokoshashvili/Shutterstock.com; (29) California grey whale, © Michael S. Nolan/Alamy
Stock Photo; (30) Lion, © Ben Puttock/Alamy Stock Photo; (31) Northern harrier eagle, © Wildlife World/
Dreamstime.com; (32) Mantis shrimp, © Alex Permiakov/Shutterstock.com; (33) Napoleon wrasse fish, © ifish/Getty
Images; (34) Caiman crocodile, © Jonathan Knowles/Getty Images; (35) Fly, © egor-kamelev/Pexels; (36) Parrot,
© Couleur/Pixabay.
Animal Eyes
Figures
3.1. Geisler Contours 85

7.1. Drawings conveying the different modes of representation for the top edge and
side edges and their lengths in two orientations of the same surface, or two
orientations of two different but similarly sized and shaped surfaces 211
9.1. A sample surface made up of six smallest-discriminable cells 337
9.2. A model of the iconic representational content of a perceptual state representing
the 6-cell surface. 339
9.3. Depiction of a scene showing different points of view 353
10.1. Illustration of a simple example of the Rapid Serial Visual Presentation (RSVP)
experimental paradigm. Source: M. Potter, B. Wyble, C. Hagmann, and E. McCourt,
‘Detecting Meaning in RSVP at 13 ms per Picture’ 393
11.1a. Diagrammatic section through the head, sketching major features of the main visual
pathway linking the eyes to the striate cortex (V1). Source: J.P. Frisby and J.V. Stone,
Seeing, second edition, figure 1.5, p. 4, © 2010 Massachusetts Institute of Technology
by permission of The MIT Press 413
11.1b. Illustration of two visual pathways underneath the cortical areas of the brain.
Source: J.P. Frisby and J.V. Stone, Seeing, second edition, figure 1.6, p. 5, © 2010
Massachusetts Institute of Technology, by permission of The MIT Press 414
11.2. Some main visual areas of the brain 416
12.1. The launching (top row) and entrainment (bottom row) effects, discovered
by Michotte 448
12.2. Habituated launching sequence and the same sequence in reverse 451
12.3. William Ball experimental paradigm 456
12.4. Causation and overlap 463
14.1. Important states in Relatively Primitive Action-Motor System 523
15.1. Sequence of display in each trial of flashed discs 538
15.2. A visibility map of a 15-degree-circumference-sized part of a retinal image.
Source: J. Najemnik and W. Geisler, ‘Simple Summation Rule for Optimal
Fixation Selection in Visual Search’, figure 2a, p1288, © 2009, with
permission from Elsevier 555
15.3a. Flow chart for computational program of Tsotsos and Kruijne for visual
discrimination task. Source: Adapted from J. Tsotsos and W. Kruijne, ‘Cognitive
Programs: Software for Attention’s Executive’, Figure 1, p. 4, open-access distributed
under the terms of the Creative Common Attribution License (CC BY) 559
15.3b. Abstract diagram of the structure of the functional components necessary to
support the executive control of attentive processing with information-passing
channels indicated in red arrows. Source: Adapted from J. Tsotsos and W. Kruijne,
‘Cognitive Programs: Software for Attention’s Executive’, Figures 5–6, p8,
open-access distributed under the terms of the Creative Common
Attribution License (CC BY) 559
figures xxiii
15.3c. Diagram of model for system of processing of visual executive-control system’s

(vAE’s) attention commands for visual task execution. Source: Adapted from
J. Tsotsos and W. Kruijne, ‘Cognitive Programs: Software for Attention’s
Executive’, Figures 5–6, p8, open-access distributed under the terms of the
Creative Common Attribution License (CC BY) 560
16.1. A typical Sperling test display 574
18.1. Examples of figures used in mental rotation experiments. Source: Adapted from
Roger N. Shepard and Lynn A. Cooper, Mental Images And Their Transformations,
figure, page 495, © 1982 Massachusetts Institute of Technology by permission
of The MIT Press 643
19.1. Causal analogy 727
Abbreviations
CSTM conceptual short-term memory

FEF frontal eye fields
FFA fusiform face area
FFE fast field echo
fMRI functional magnetic resonance imaging
FVSTM fragile visual short-term memory
IT inferior temporal cortex
LGN lateral geniculate nucleus
LOT language of thought
LTM long-term memory
MST medial superior temporal area
MT middle temporal area / medial temporal cortex
ms millisecond
OFC orbital frontal cortex
PFC pre-frontal cortex
PHC para-hippocampal cortex
RC retrosplenial cortex
RDS random dot stereograms
RSVP rapid serial visual presentation
SM simple model
SOA stimulus onset asynchrony
TSM trans-saccadic memory
V1–V5 visual cortical areas
VLTM visual long-term memory
VWM visual working memory
WM working memory
PART I
PE RCE P T ION
1
Introduction
In philosophy, if you aren’t moving at a snail’s pace, you aren’t moving at all.
Iris Murdoch, Existentialists and Mystics
In the grand array of occupants of the universe, from sub-atomic particles through higher
animals, the animals with minds stand out as special. Some of the specialness derives from
our being in this group. We interest ourselves. This interest extends to animals like us.
The natural interest is grounded in more than self-interest. It is grounded in a deep
joint in nature—the joint between the minded and the mindless. The joint may be ragged.
It may have borderline cases. But it is real. It is relevant to matters of great value and
importance—science, understanding, morality, art. Without mind, none of these pursuits
or goods exist. They seem good independently of whether we engage with them. They
depend for being good on realizing functions of certain types of minds. The minded-
ness and the type of mind ground the goodness, not their being ours. My project is
understanding—understanding some central aspects of mind.
Understanding the mind–mindless joint requires understanding mind in its most basic
forms. Of course, mind depends on the mindless and makes use of it. Without the regularities
of organic chemistry, minds could not be minds. Without the regularities of the broader
macro-physical world, minds could not navigate through it. In many ways, the non-minded
physical world stamps itself into the very natures of mental states. In understanding mind,
however, it is important to understand what is new and different about it, at its most
elementary levels. Such understanding aids understanding richer forms that guide the
listed pursuits. Those pursuits are not possible where mind begins. They are not possible for
the first forms of mind. However, they depend on and employ these forms. A central aim of
this book is to understand forms of mind at this initial juncture—forms of perception.
Biological Function, Action, Sensing,

and Perception—The Emergence of Mind
Before mind emerged, another deep and interesting joint in nature had already developed.
Brief attention to this joint is valuable in understanding the joint on which mind hinges.
Most of the universe is fire, rock, or emptiness. Already with life, there is a momentous
difference. Life occupies small crevasses in the universe. However, it marks a large change
from the chemical mixes from which it emerged. Although the living share a material
basis with the non-living, the living comprise a genuinely new and different group of the
universe’s occupants.
The point is not just intuitive. It shows up in new terms and methods in the scientific
study of life. Notions of function, growth, reproduction, natural selection, adaptation,
Perception: First Form of Mind. Tyler Burge, Oxford University Press (2022). © Tyler Burge.
DOI: 10.1093/oso/9780198871002.003.0001
4 Introduction
life-cycles, ecology have no place in physics or chemistry. They are central to biology. The
historical study of evolution and a lesser emphasis on law, are foreign to physics and
chemistry. These scientific differences signal subject-matter differences. The living are
very different from the rest.
A key aspect of life, lacking in the lifeless natural world, is function. Rock and fire have
no functions in themselves. Biological functions are patterns of operation whose existence
derives from their contributions to success in reproduction.1 The function of photo-
synthesis is to convert light energy into chemical energy that subsequently is transformed
in a way that feeds an organism’s other operations. The process exists in plants and other
organisms because it contributed to their reproduction. It was naturally selected.
Like other types of function, biological function is conceptually linked with doing well,
being successful. These types of goodness are not moral, or products of plans or purposes.
They are good for the system, or individual, or species, because they aid survival long
enough to reproduce. Being a biological function is an objective matter: functions are
what they are independently of whether anyone recognizes them. They are open to object
ive evaluation, even scientific evaluation, by rational standards. Whether and to what
degree a process fulfills one of its functions can be empirically assessed. Either a process
functions well or it does not.
Some functions are attributable to plants and animals as whole organisms. Others are
attributed only to subsystems or parts of a plant or animal. For example, functions to grow
and reproduce are functions of the whole plant. By contrast, production of pollen grains is
a function of the anther in a plant’s flower. Photo-synthesis is an operation in each indi-
vidual cell. The whole plant, its sub-systems, and its parts can succeed or fail in realizing
their functions.
Functional processing in plants responds in ways sensitive to the environment. Photo-
synthesis depends on features of the plant that are specialized to be sensitive to light.
Photo-synthesis yields responses, such as directional growth, that depend on chemical
reactions in the plant that transform the light’s energy into chemical energy. These are
antecedents of sensing and action that occur in animals. I think it well to follow common
sense in thinking that these are not strictly cases of sensing or acting. I say that plants are
sensitive to the environment, whereas animals sense it. Animals, but not plants, act.
Directional growth and pollination are not actions. A plant’s absorption of water and
nutrients is not drinking or eating. The Venus Fly Trap’s engulfing of visitors is not eating.
I conjecture that this is so because the relevant changes in the plant can be too easily
1 This conception of biological function derives from L. Wright, ‘Functions’, The Philosophical Review 82
(1973), 139–168. There are other, compatible notions of function that figure in biology. I use Wright’s because
it is familiar, explanatory, and teleological. Other conceptions of function in biology also take teleology ser
iously and are not as centered in evolutionary history—so-called organizational conceptions. A revision of the
Wright account of function, which I think of as an expansion of it, rather than purely a correction of it, cen
ters on any contributions to reproduction within a living system, rather than purely on original-etiological
contributions. For excellent work on this conception of function, see G. Schlosser, ‘Self-Reproduction and
Functionality: A Systems- Theoretical Approach to Teleological Explanation’, Synthese 16 (1998), 303–354;
W. Christensen and M. Bickhard, ‘The Process Dynamics of Normative Function’, The Monist 85 (2002), 3–28;
M. Mossio, C. Saborido, and A. Moreno, ‘An Organizational Account of Biological Functions’, British Journal
of Philosophy of Science 60 (2009), 813–841; M. Mossio and L. Bich, ‘What Makes Biological Organization
Teleological?’, Synthese 194 (2017), 1089–1114. See also J. Garson, A Critical Overview of Biological Functions
(Dordrecht: Springer, 2016). Although I use Wright’s notion, the notion of biological function that I rely on does
not need to be specific enough to choose between these conceptions. Any scientifically based explanatory notion
of teleology will suffice.
Biological Function, Sensing, and Perception 5
explained in terms of summations of changes that occur in plant cells. There is no need to
postulate a central locus of conation, as we do for agents.
Some speak metaphorically, even poetically, of sensing and acting by plants. Some sci-
entists like to say that plants communicate with one another. Perhaps there is a broad
enough notion of communication to allow such talk to be non-metaphorical. I believe that
any such communication is not action. Calling functional, cross-individual patterns of
sensitivity and response among plants’ “communication” serves advertisement more than
understanding. True understanding depends more here on exploring differences than on
engaging in assimilation.
I think that serious conceptual and scientific investigation of these matters will confirm
some variant of what common sense assumes. Plants may communicate. They do not act.
Some or all animals, and perhaps other organisms that are neither plants nor animals, do.
I think that action has to do with a coordination among central capacities of an
organism—typically, but not necessarily, endogenously causing movement by the individual.2
As noted, I conjecture that plants do not act, because their changes derive not from a central
coordinating capacity, but from a mere aggregation of changes in plant parts. Photo-synthesis
occurs in every plant cell. Growth is not action partly because it is a summation of aggre-
gate increases in various cells. Directionality in growth stems from the fact that more
stimulated cells multiply faster. Plants are sensitive to light. Relevant stimulation for
growth is often from light. Plants grow toward the light. Directional growth is an aggregate
response of changes in the plant’s most stimulated cells. This is not action. Similarly, for
absorption of water and nutrients. Such absorption is not drinking or eating. These points
are at best a gesture at a position on a complicated topic.
I take the notion of sensing, as distinct from that of sensitivity, to be tied to action.
Plants are sensitive to stimuli. Animals sense stimuli. Sensing and action emerge
together.3
The distinction between plant responses and animal (or other organisms) action is not
central to the present project. It is background. What I have pointed to is a broad analogy
between plant sensitivity and functional growth-like responses, on one hand, and animal
sensing and functional behavioral responses that include action, on the other.
The sensing-action nexus is very old, older than the emergence of mind. Organisms
that surely lack minds—paramecia, simple worms, snails—act. They eat, swim, or crawl.
They depend on elementary sensing capacities in fulfilling these activities’ functions.
Here again, biological functions of the organism are to be distinguished from biological
functions of organs and operations within the organism. For example, eating, mating,
swimming, crawling, and navigating are biological functions of the whole animal. All
meet earlier-discussed conditions for being a biological function. All depend for being
functional on the well-functioning operation of biological systems or organs within the
organism. For example, eating’s fulfilling its function depends on the well-functioning
operation of a digestive system. The whole-animal functions are relevant to understanding
success and failure for the whole animal, not just sub-systems or sub-parts of the animal.
Acting and sensing are functional pursuits at the level of the whole animal. They form the
womb out of which basic forms of mind are born.
2 See T. Burge, Origins of Objectivity (Oxford: Oxford University Press, 2010), chapter 8.
3 I advance this conjecture in Origins of Objectivity, 376–377; see also 331–334.
6 Introduction
Philosophical tradition has come to a broad consensus on the most general marks of
mind. They are consciousness and representation.4 Thought and perception—both of which
are types of representation—can be conscious or unconscious. Consciousness, I think,
can be either representational or not. A representational state like a perception can,
obviously, be conscious. A feel of a pain or a tickle can, I think, be distinguished from a
proprioceptive representation of its location. Such feels are conscious, but not in them-
selves representational in the ordinary sense of representation—the sense that I will refine
and develop.
This book focuses on representation, not consciousness. I take perceptual representa-
tion to be a basic mark of mind and a mark of nature’s mind–mindless joint. This view
does not compete with the idea that consciousness is also a basic mark of mind. There may
be two joints in nature between minds and the mindless. Many minds are both representa-
tional and at some times conscious. But it may be that there are conscious beings that do
not represent and representational beings that are not conscious. I focus on the first mark
of mind—representation. Vastly more is known about it. The science of consciousness is in
its gestation stage. The science of perceptual representation is in its early maturity.
There is no consensus about how consciousness and representation are related in being
marks of mind. The issue is complicated by the fact that there are importantly different
historical understandings of the putative subject matter here—mind. The notions of mind,
psyche, soul, psychological system, and so on, each has different historical associations.
I ignore nuance here, in the interests of providing a broad-brushed setting for the main
project. I think that having consciousness and having representation are each sufficient for
having a mind.
Neither is by itself necessary for having a mind. An animal that feels pain—and hence is
conscious—has a mind. It may or may not have a capacity to represent, in the sense
of ‘represent’ that will occupy us. For example, it may or may not have perceptual states.
An animal that perceives, and hence represents, has a mind. It may or may not be capable
of consciousness.
So I think that representation and consciousness are in principle separable. Each is a
mark of mind.5 It follows that there could be two paths to mind in the evolution of
animals—one through consciousness, one through representation. If one wants to distin-
guish mind—marked by consciousness—from psychology—marked by representation—,
I have no strong objection. Then there may be minds without psychologies, and psycholo-
gies without minds. I do not, however, write in these ways. I do sometimes write of con-
scious mind or representational mind.
I assume that the two marks of mind—consciousness and representation—are each suf-
ficient for having mind. Having at least one is necessary. Neither is by itself necessary.
Of course, many animals that are conscious are capable of representation, and many
animals capable of representation are conscious. All higher animals, certainly all
4 What I call ‘representation’ is often called ‘intentionality’. I think that the latter term is historically associated
with unclear thinking about representation, and is best discarded. I mean ‘representation’ here in a specific
restricted sense, developed in Chapter 2.
5 One could qualify these points. Consciousness is historically more closely tied to the notion of mind.
Representation is more closely tied to the notion of psychology. Some may find saying that an always uncon-
scious animal has a mind is harder than saying that it has a psychology. Some may find that saying that an ani-
mal, that feels pain but represents nothing, has a psychology is harder than saying that it has a mind, though not
much of a mind. One could talk of conscious mind and representational mind to ease the rub. I sometimes do so.
mammals, have representational capacities and are (often) conscious. So if there are, in
evolutionary history, separate streams into mind—animals that are conscious but do not
represent and/or animals that represent but are not conscious—, these two streams flow
back together in more complex animals.
The point about the separability of consciousness and representation is a very general,
conceptual point. There is no necessary connection between consciousness and
representation.
The point is not just of general conceptual importance. It bears on understanding the
kind of representation that figures in this book. I have said that, for all we know, there may
be animals that have perception and lack consciousness. Bees and other arthropods have
perception. They may lack consciousness. We do not know enough about consciousness to
settle the question. One day, unconscious robots might be produced so as to have visual
perception.
I think that perception without even a capacity for consciousness is epistemically, meta-
physically, and nomologically possible. Epistemically: I think that nothing that we know,
either empirically or apriori, rules out perceptual representation without consciousness
(or vice-versa). Metaphysically and nomologically: I think it a real possibility that an ani-
mal have perceptual representation and lack any capacity for consciousness. Representation
is primarily a functional matter. It hinges on what an individual or the individual’s sub-
systems can do. Consciousness is not a functional matter; it hinges on an individual’s
material basis.
Psychophysical explanations posit human perceptual states that are not and cannot
become conscious.6 Much of the science of perception is carried on without specifying
whether a perception is conscious. These points form the ground for the conjecture that
there is nothing in the nature of things that requires some association between conscious-
ness and representation.
Of course, conscious perceptions are an interesting topic. They have different
psychological-representational, as well as phenomenological, properties from uncon-
scious ones. But consciousness itself is not yet a central scientific topic. Perceptual science
has been spectacularly successful without theorizing much about it. Eventually, more will
be known.
Some philosophers claim that perception must be conscious. Some claim that percep-
tion that picks out bodies, or perception that is not guesswork and could support know
ledge, requires consciousness. None of the arguments that I know of for such positions has
any force. I discuss some in Chapter 10, the section Philosophical Views
Influenced by Treisman’s Binding Theory. Several are incompatible with what
is known from science. Issues about consciousness are not central in our story, but they
arise recurrently. I mention them here both to acknowledge their natural interest and to
motivate not centering on them. My primary focus is representation.
6 For one of many reviews of psychological work that posits unconscious perceptual states, see S. Kouider and
S. Dehaene, ‘Levels of Processing During Non-Conscious Perception: A Critical Review of Visual Masking’,
Philosophical Transactions of the Royal Society of London B 362 (2007), 857–875. When I generalize about vision
science or perceptual psychology, I usually intend psychophysics. Sometimes when I make a point about vision
science, I do not cite specific articles. In such cases, I take the point to be uncontroversial in the science’s main-
stream. Then, I encourage novices to become acquainted with the mainstream, and to check my judgment. I take
the point footnoted in this case to be mainstream-uncontroversial. But since there has been some dissent in
philosophy, I give the nod, above, to the scientific literature.
8 Introduction
More specifically, I focus on perceptual representation. Perceptual representation is

where representational mind begins. I reflect on the joint in nature between mind and the
mindless by reflecting on differences between perceptual representation and those non-
perceptual sensory capacities that underlie it. I center on perception—and related capaci
ties like perceptual memory and perceptual anticipation—because it is functionally the
most basic representational capacity. No other representational capacity evolved earlier.
If other representational capacities (perceptual memory, for example) evolved equally
early, they depend functionally on perception.
Perception is, evolutionarily, the first known manifestation of representational mind.
Arthropods—bees, praying mantises, and certain spiders—are known to have visual per-
ception. Visual perception is distinct, in ways to be discussed, from other types of light-
based sensing. Snails, molluscs, and tapeworms sense light, even light-direction. Ants
respond to light-produced templates that correspond to surface shape.7 These capacities
for visual sensing are not perception. Snails and molluscs are not known to have visual
perception. I think it unlikely that animals that evolved much earlier than the arthropods
had perceptual representation. So we have a rough sense of where representational mind
begins. It begins among the arthropods.8
I do not center on the evolutionary emergence of representational mind, despite its
great interest. My topic is different. Given that perception is the earliest form of represen-
tation, and that other forms develop from it, what can be learned about representation and
the earliest form of representational mind by reflecting on perception? I center on what
perception is—its structure and function—not on how it evolved. Evolution will, however,
recurrently come up.
7 On snails and other molluscs, see P. Hamilton and M. Winter, ‘Behavioural Responses to Visual Stimuli by
the Snail, Littorina irrorata’, Animal Behaviour 30 (1982), 752–760; V. Zhukov, M. Bobkova, and A. Vakolyuk, ‘Eye
Structure and Vision in the Freshwater Pulmonate Mollusc, Planorbarius corneus’, Journal of Evolutionary
Biochemistry and Physiology 38 (2002), 419–430. On these animals and tapeworms, see M. Land and D. Nilsson,
Animal Eyes (Oxford: Oxford University Press, 2002), 4. On ants, see K. Basten and H. Mallot, ‘Simulated Visual
Homing in Desert Ant Natural Environments: Efficiency of Skyline Cues’, Biological Cybernetics 102 (2010),
413–425; V. Aksoy and Y. Camlitepe, ‘Behavioural Analysis of Chromatic and Achromatic Vision in the Ant,
Formica cunicularia (Hymenoptera: Formicidae)’, Vision Research 67 (2012), 28–35; M.-C. Cammaerts, ‘The Visual
Perception of the Ant Myrmica ruginodis (Hymenoptera: Formicidaxe)’, Biologia 67 (2012), 1165–1174. The visual
template capacity mentioned in the text is not perception. But ants are known to have visual perception, with
perceptual constancies. They mainly act not on vision but on olfaction.
8 Bees and other arthropods have visual perception—color, location, and size constancies, for example.
See T. Collet, ‘Peering: A Locust Behavior for Obtaining Motion Parallax Information’, Journal of Experimental
Biology 76 (1978), 237–241; R. Wehner, ‘Spatial Vision in Arthropods’, in H. Autrum ed., Comparative Physiology
and Evolution of Vision in Invertebrates: Invertebrate Visual Centers and Behavior (Berlin: Springer Verlag, 1981);
M. Lehrer, M. Srinivasan, S. Zhang, and G. Horridge, ‘Motion Cues Provide the Bee’s Visual World with a Third
Dimension’, Nature 332 (1988), 356–357; G. Horridge, S. Zhang, and M. Lehrer, ‘Bees Can Combine Range and
Visual Angle to Estimate Absolute Size’, Philosophical Transactions of the Royal Society of London B 337 (1992),
49–57; M. Lehrer, ‘Spatial Vision in the Honeybee: The Use of Different Cues in Different Tasks’, Vision Research
34 (1994), 2363–2385; R. Foelix, Biology of Spiders (Oxford: Oxford University Press, 1996), 87–92; C. Neumeyer,
‘Comparative Aspects of Color Constancy’, in V. Walsh and J. Kulikowski eds., Perceptual Constancy (Cambridge:
Cambridge University Press, 1998); R. Wilcox and R. Jackson, ‘Cognitive Abilities of Araneophagic Jumping
Spiders’, in R. Balda, I. Pepperberg, and A. Kamil eds., Animal Cognition in Nature (San Diego: Academic Press,
1998); M. Lehrer, ‘Shape Perception in the Honeybee: Symmetry as a Global Framework’, International Journal of
Plant Sciences 160 (1999), S51–S65; K. Kral, ‘Behavioral-Analytical Studies of the Role of Head Movements in
Depth Perception in Insects, Birds and Mammals’, Behavioral Processes 64 (2003), 1–12. For further discussion of
this issue, see Burge, Origins of Objectivity, 419–420; T. Burge, ‘Origins of Perception’, Disputatio 4: 29 (2011),
1–38; and T. Burge, ‘Perception: Where Mind Begins’, Philosophy 89 (2014), 385–403, reprinted in T. Honderich
ed., Philosophers of Our Times (Oxford: Oxford University Press, 2015). Whether arthropods are conscious is not
known. Those squeamish about taking arthropods to have minds, if they lack consciousness, can say that they
have representational psychologies.
I also want to understand relations between perception and closely, almost inevitably
associated capacities—perceptual attention, primitive perceptually guided action, per-
ceptual memory, perceptual anticipation, perceptual imagining, perceptual learning.
I argue later that these capacities participate, with perception, in a psychological system
all of whose capacities share representational form and representational content with
perception itself.
Philosophy is fortunate to be able to reflect on science. A science of perception, particu-
larly vision science, has bloomed into a rigorous enterprise over the last fifty years. It is by
far the most impressive psychological science. It is more advanced in mathematization,
and in predictive and explanatory power than many biological sciences. Especially in the
second half of the book, I make extensive use of what is known from this science. Doing
so is part of gaining a philosophical understanding of issues attaching to perception and
perceptual-level representation.
Such understanding is inevitably affected by scientific change. A lot of detail that I dis-
cuss will turn out to be mistaken and superceded. Some of it is probably already known by
someone to be mistaken. The science is so vast that no account can keep up with every
relevant discovery. At worst, my use of empirical work is a challenge to do better.
Some of my most general points are, I think, safe. The referential-attributional structure
of perception, its iconic format, and the distinction between perceptual-level representa-
tion and propositional representation, for example, will weather changes in the science,
such as details of timing, of what perceptual-level attributives there are, and so on.
I focus on vision. I discuss hearing, touch, proprioception—the other main human per-
ceptual senses—only occasionally. This approach inevitably evokes complaints, especially
from those interested in other types of perception, and in relations between perceptual
modalities.
Several answers to such complaints seem to me apropos. First, understanding visual
perception is a huge task. Vision is the most complex of the senses. I plead human limita-
tion. I invite others to do similar work on other types of sense perception.
Second, vastly more is known about vision than about other types of perception. The
number of vision scientists and the amount of work done in visual psychology dwarfs the
number and amount in all the sciences about other perceptual modalities put together.
Third, vision is by far the most important sense for most humans and apes for guiding
their lives. It is the main basis for the development of empirical science by human beings.
Although I think it important to remember that human perception presents only one
suite of capacities, and understanding perception in other animals is of great importance,
I focus on humans because I am interested in situating perception in relation to higher
representational powers. As far as is known, only humans have some of these higher
powers—linguistically expressed thought, reflective deliberation, and scientific reasoning,
for example.
Fourth, many visual capacities have definite analogs in other perceptual modalities. So
understanding vision can provide a boost in understanding them. The sciences of hearing
and touch derive much of their success from using ideas and methods that come from
vision science.
Fifth, many of the most basic features of visual perception—its most basic representa-
tional structure and its iconic format, for example—are present and basic in the other
perceptual senses.
10 Introduction
In no way do I deny the import of non-visual perceptual modalities. Different modal

ities are always connected in perceivers that have more than one. Connections are causally
reciprocal. Perceptual-state formation in one modality can always be influenced by input
from another. Inter-modal influence occurs among touch, hearing, proprioception, and
vision. To limit complexity, I touch on inter-modal relations only glancingly.
My work on vision should not be seen as detracting from work on other modalities, or
on inter-modal influences, in any way. It should be seen as a possible resource in under-
standing inter-modal relations and relations between visual perception and perception in
other modalities.
My focus on perception is also not to be construed as minimizing the huge role of
action and reaction in forming and using perceptual capacities. Perception evolved, fun-
damentally, not because it was accurate, but because it contributed to action and reaction
that, in turn, contributed to fitness for reproduction. Action evolved before perception
did. Its flexibility and precision is enhanced by perception. Perception’s main use is in
guiding action. Its guidance is, of course, sensitive to feedback from states that set action’s
targets and feedback from action itself.
Perception and perceptual-level conative states—psychological states that function to
produce action guided by perception—form a large psychological system, the perceptual-
motor system. I discuss such conation in Chapter 14. I am acutely aware that more could
be said to provide a more balanced account of interaction between perception and con
ation. I believe, however, that perception is the main, or entire, source of representational
resources in primitive conation. That is, perceptual attributives that guide conation are the
main or entire repertoire of representation in primitive action. In primates, action guid-
ance derives primarily from vision and proprioception. Again, visual capacities do not
operate alone, or independently of relations to other perceptual modalities. Still, visual
perception looms large enough and constitutes such a formidably complex phenomenon
that I do not apologize for focusing on it.
Principal Aims of the Book
Five principal aims inform this book.

The first is to develop a detailed understanding of the core fundamental representa-
tional features of visual perception. In Part I of the book, Chapters 2–3, I explain what
I mean by perception. I distinguish it from non-perceptual sensing. I discuss the key mark
of perception—perceptual constancies. In Part II, Chapters 4–9, I develop the form, func-
tion, and content of visual perception. I set out the primary types of representation in
visual perception—such as referring (or picking-out) types and attributive (or character-
izing) types. I explain ways in which these types of representation combine to yield a
representational form—on an analogy to logical forms of propositional representation.
I discuss the basic semantics of visual perception. Finally, I explicate and explore the
iconic, map-like or picture-like format of visual perception.
The second aim, pursued in Part III, Chapters 10–12, is to situate these structures in
the processing of visual perceptual representations. I sketch some of what is known about
how the representational forms discussed in Part II are caused to occur by stimulation.
I discuss processing and time courses of some of the main types of visual perceptual
representation—representation of size, shape, position, motion, speed, color, surface,

body, body-parts, and so on. I consider whether attention is necessary for the formation of
perceptual states. I discuss types of representation that may extend this list of core types of
visual attributives. Possible extensions include attributives for various types of function
and attributives for causation and agency.
The book’s third aim is to understand some relations between visual perception and
closely related psychological capacities—perceptually guided action, perceptual attention,
perceptual memory, perceptual affect, perceptual learning, perceptual anticipation, and
perceptual imagining. I devote special attention to the perceptual level within each of the
generic capacities—conation, attention, memory, affect, learning, anticipating, imagining.
I caution against thinking of the generic capacities as monolithic. I discuss wherein sub-
species are at the same representational level as, and part of the same computational
system as, perceptual representation.
As rough approximation, higher-level, supra-perceptual-level, processes involve types
of representation or transformation that are more sophisticated and more knowledge-like
than those in perceptual-level representation. Higher-level processes use representations
that have forms and functions that perceptual-level representations lack. In most cases
higher-level capacities evolved later than perceptual-level capacities. Roughly, perceptual-
level capacities include perception and other capacities typed by the representational con-
tent of perceptual states. Such capacities do not enter into operations that are more
sophisticated or more knowledge-like than perception. Fuller explication of the notion of
a perceptual-level capacity occurs at the beginning of Part IV. Chapters 13–19 focus on
such capacities.
A fourth aim is not centered in one part of the book. It runs through almost the whole. It
is systematic opposition to an old empiricist way of thinking about perception, and its rela-
tion to thought, inherited from classical empiricists—mainly Locke and Hume. This view
correctly takes perceptions to have an image-like character. However, it conceives the con-
tents of images in terms of what seems salient in perceptual experience. Such a view tends
to ignore, doubt, or underestimate the prevalence of unconscious perception and percep-
tual operations that are not part of conscious experience. More importantly, it neglects the
capacities, functions, and uses that give perception the form and content that it has. (I see
Kant as an early opponent of such neglect.) For example, it takes perception to group
aspects of images by what, to the empiricist, seem to be intuitive types of similarity or
intuitive relations that image-parts bear to one another. Such groupings are taken to be
concrete, in accord with the idea that a perception is an image that is to be understood by
introspection of intuitively salient aspects of experience. All thought is taken to contrast
with perception by being more abstract and perhaps initially derived by a process of
abstraction from perception. So thought cannot be as concrete or imagistically specific as
perception. And perception cannot be abstract, because it is imagistically and intuitively
concrete. All these points are supposed to be derived from reflection on phenomenally salient
features of perceptual images. The empiricist distinction between perception and thought
lies along an intuitive, but not-very-well-articulated, spectrum of concrete-to-abstract. Few
explicitly develop and elaborate the model nowadays. It retains a surprising unreflective
resilience in philosophy and parts of psychology outside mainstream psychophysics.
The view elaborated in this book follows not introspection or intuitive senses of
image-similarity. It focuses on capacities to form and use perception that are discovered in
12 Introduction
perceptual psychology. Some of these capacities produce intuitively familiar features of

conscious perception. But many others produce aspects—especially categorizing aspects—
of perception that depend on specific species-needs or individual-needs and on statistical
patterns in the environment that perceptual systems have capitalized upon. Some of these
do not line up with intuitive senses of image-similarity. In focusing on capacities and use
not introspection, the account also finds perception to engage in levels of abstraction that
can easily seem not to be capable of being represented directly by an image. Yet, I believe
that in being iconic, visual perception is always grounded in image-like representation.
Developing this combination of ideas is a central part of this initiative in the book.
I reject the empiricist way of distinguishing perception and thought (or cognition, or
conception)—which cites differences in levels of abstraction. Both perception and propos
itional thought occur at highly concrete and highly abstract levels. Thought can rely on
perception in its modes of presentation, and be every bit as concrete and detailed as per-
ception can be. Conversely, perceptual representation extends from very concrete and
finely discriminated to very generic groupings. Thoughts share various levels of concre-
tion and abstraction with perceptions by constitutively relying, partly, on perception for
representing. Of course, some thoughts are abstract in ways that no perceptions are.
Thoughts can be about unobservables. Some thoughts are amodal, independent of a sen-
sory modality. Some represent logical functions or numbers that lack instances in space or
time. But since perceptions can be very abstract and thought can be very concrete, the
empiricist way of distinguishing them is misdirected.
The difference between perception and thought resides not in what can be intuitively
introspected in an image. It resides in different forms, functions, uses, and capacities
(or competencies). Although thought shares perception’s basic representational capacities—
specifically reference and attribution—, it constitutively involves further capacities, func-
tions, and forms not present in perception. Use and competence ground constitutive
differences between perception and thought, not image-like format and not (primarily)
levels of concreteness or abstraction.
These themes run throughout the book. They emerge first in Chapter 2, Perception as
Objectification. They become much more prominent in Chapter 4, Taxonomic Hierarchies
in Perception. They run strongly through Chapters 8–10 and 12. They mark all discussions
of categorization. They culminate in the last two sections of Chapter 19.
Replacing the traditional empiricist introspective approach to perception aids the fifth
main aim of the book: developing notions of perceptual system and perceptual-motor sys-
tem. This aim occupies Chapter 19. The aim is guided by the attempt to understand differ-
ences between perceptual- level and supra- perceptual-level capacities, which has as
background Chapters 13–18. I think that the deepest divide in the mind is between
perceptual-level states and propositional-level states. Otherwise put, the divide is between
the level of perception and the level of propositional reasoning or thought. I explore pos
sible non-propositional, supra-perceptual capacities. But the key target is understanding
the large perceptual and perceptual-motor systems. These systems mark, I think, the terri-
tory of lower representational mind.
What immediately follows is an unnuanced forecast of some ideas to be developed
about relations between perception and conception. More detailed discussion of exactly
how I understand relations between higher-level representational resources and percep-
tion, or perceptual-level ones, occupies Part IV, especially Chapter 19.
I believe that one feature of the relation between perceptual and perceptual-level cap
acities, on one hand, and conceptual and propositional capacities, on the other, is that the
former have a kind of self-sufficiency. I believe that no type of perceptual state constitu-
tively, or by computational law, depends for its formation on conceptual influence, or on
influence from propositional states. Any perceptual state can be generated computation-
ally without any concept’s or propositional state’s figuring in the computation. Formation
of perceptual representations does not depend essentially on conceptual resources.
Similarly, relations between perceptual states and other perceptual-level states do not
depend computationally on input from conceptual or propositional resources, and could
in principle occur without such resources. As far as is now known, computational explan
ations of how perceptual memories and perceptual anticipations are formed from percep-
tual states, the perceptual-level initiations of attention and their effect on perception, the
task dependence of perception on conative states, and so on, can be computationally
explained without reference to conceptual or propositional input. Computational explan
ations in the psychophysics of perception do not include reference to conceptual states.
None of this is to say that concepts cannot influence what perceptions are formed.
Instructions in propositional form by scientists influence perceptual search and make per-
ceptual processing take courses that it would not have taken, apart from the instructions.
They thereby influence perceptual processing. I think that there is a sense in which, even
in these cases, they do not enter the process itself. Causal computational sequences that
lead from non-conceptual input into a perceptual system always suffice to explain the
formation of perceptual states.
Similarly, certain types of conceptually initiated attention affect perceptual-level atten-
tion commands, and thereby affect attention during perceptual processing. The attention
affects the processing. So perceptual processing can be causally affected by conceptual
states, even as the processing occurs. However, causal-computational sequences that begin
with the perceptual-level attention commands, or with attention within perception, omit
inclusion of the conceptual antecedents, yet are explanatorily satisfying. In principle, the
same attention commands and attention effects could have been formed without concep-
tual influence. Although not always the whole causal story, explanation by reference to the
concept-free causal computational sequence is, in a sense to be discussed, sufficient to
provide systematic explanation of the formation of a perception. I think that the concept
of perceptual system, and distinctions between perceptual-level and conceptual-level
psychological kinds can be partly understood by reference to these facts.
In the last two decades, scientific understanding of visual processing has substantially
changed. It used to be thought that processing was much more “bottom up”. It is now
known to have substantial recurrent processing–processing that proceeds from more ana-
tomically downstream visual areas back upstream toward initially early areas. Neural pro-
cessing begins with the onset of a proximal stimulus and proceeds to early processing
areas, such as the lateral geniculate nucleus (LGN), the primary visual cortex (V1), and
the middle temporal area (MT). It then proceeds to visual areas, V2–V4, the infero-
temporal cortex (IT), parietal areas, and the pre-frontal cortex. From the beginning, hori-
zontal processing takes place, for example, among areas within LGN or V1. Later, recurrent
processing begins, from V2-V4 and beyond, back to V1 (and similarly for each later
stage—V2 . . .). So processing is much more multi-directional than was thought as recently
as the 1990s. The neurological sequence has psychological analogs.
14 Introduction
The change in understanding how complex visual processing is has encouraged philo-
sophical and scientific discussion of “cognitive penetration” into visual processing. To
what extent and in what ways do representational capacities that are not perception enter
into visual processing? Are some of the capacities, like attention, memory, and so on, “cog-
nitive”? Does background propositional-level intention, belief, or knowledge affect visual
processing? If so, how? This discussion requires detailed reflection on the psychological
processing, and the specific character, of generic capacities like attention, memory, and so
on—insofar as they enter into visual processing. It also requires serious discussion of the
terms ‘cognitive’ and ‘penetration’. The term ‘cognitive’ is used in many, often thoughtless,
ways. The term ‘penetration’ is also used in different ways. In both cases, undisciplined
uses and insufficient reflection obscure how perceptual-level processing is explained in
perceptual psychology.
The philosophical interest of these issues, for me, is that they bear on understanding
joints in nature that border perceptual representation. It is arrogant to think that we
“carve” joints. We do not do anything to them. When we try to carve, we make a bloody
mess. We discover them. At least, sometimes. Discovery is more delicate and less intrusive
than carving. Perceptual psychology has advanced to a stage where it can help guide dis-
covery. I argue that visual perceptual processing forms a system with several capacities
that are not themselves perception—including some types of attention, action-guidance,
anticipation, and memory. I argue that these capacities are perceptual-level in that they
derive all or many of their representational contents from perception and are not in any
way more sophisticated than perception. The basic representational structure and the
sophistication of transitions or operations in these systems are not more advanced or fun-
damentally different from the structure and transitions in perception.
As noted, these perceptual-level capacities do interact with propositional, “cognitive”
capacities. Yet, in a sense to be discussed, propositional capacities are not part of a certain
system with perception, or its most closely associated perceptual-level capacities.
The third and fifth aims bear on the upper border of perception and perceptual-level
processes—their relation to higher level representational capacities, especially propos
itional capacities. In the early chapters of the book, I discuss the lower border of percep-
tion. This border divides perception and perceptual-level capacities from non-perceptual
sensing and from action that is not guided by perception or any other representational
capacity. This discussion mostly reviews accounts already provided in Origins of Objectivity.
It also refines these accounts. Parts II and III of the present book discuss the heartland—
visual perception itself. In Part IV, I try to understand some central aspects of perception’s,
and perceptual-level capacities’, upper border—their border with more advanced kinds of
representation.
Propositional-level intention, belief, and knowledge lie on the other side of this
upper border. I discuss in Chapter 19 whether there are non-propositional types of repre-
sentation that are more “advanced” in some deep principled way than perception—thus
perhaps a kind of intermediate territory between the perceptual level and the level of
propositional thought.
The book concludes with reflections on the importance of the border between perceptual-
level representation and propositional representation. The latter is necessary for science,
understanding, morality, art. Perceptual-level states are an indispensable ground for these
enterprises. This project aims at understanding the capacities that formed this ground.
Fregean Source of Semantical Notions for Perception 15
The Fregean Source of Key Semantical Notions for Perception
I wrote in the Preface that Gottlob Frege fashioned central theoretical semantical concepts
that are relevant to understanding perception.9 These concepts have been refined and sys-
tematized by philosophers and mathematical logicians through the twentieth century. The
concepts have been applied to the semantics of natural language and thought.10 Linguists
appropriated the concepts from philosophy and mathematical logic in the 1970s.
Frege was more concerned with mathematical thought than with language, let alone
perception. Perception is not thought or language. It differs from both in fundamental
ways. Still, four of his concepts are directly applicable to perceptual representation.
I develop them, as they apply to perception, in Part II. I make some general points here.
First, Frege’s understanding of predication, as functional application, took the key step
in gaining a modern understanding of characterization, or attribution. Predication in lan-
guage and conceptual attribution in thought are rooted in more primitive attribution in
perception. Perceiving something as brown is perceptually characterizing it, making an
attribution to it.
As I use the term, predication is attribution that functions to contribute to a propos
itional structure. So predication is conceptual or linguistic attribution. Attribution in per-
ception is not predication. Perceptual states are not propositional. They do attribute
properties, kinds, and relations. Properties, kinds, and relations are attributes. Perceptual
states characterize entities that they function to represent by representing them as having
attributes. I use ‘attribution’ to cover both predication and the purely reference-serving
characterization in perception (and in many uses of pictures). So attribution, but not
predication, occurs in perception.
Second, Frege took the key step in achieving a modern understanding of reference—in
particular the “picking out” of a particular entity. The key step was understanding how
singular reference fits with predication to produce a complex representation. Frege’s
notion of Bedeutung, one form of which is a type of reference, was a highly theoretical
notion, fitted to the semantics of logic and mathematics. However, close examination of
his informal explications of relevant forms of Bedeutung (forms for names) shows that,
in forming his technical notion, he reflected on the common-sense notion of singular
reference—picking out an individual entity. He was aware of connections between his
9 G. Frege, Begriffsschrift (1879), in J. van Heijenoort ed., From Frege to Gödel (Cambridge, MA: Harvard
University Press, 1970); G. Frege, ‘Funktion und Begriff ’ (1891) and G. Frege, ‘Über Sinn und Bedeutung’ (1892),
both translated in P. Geach and M. Black eds., Translations of the Philosophical Writings of Gottlob Frege (Oxford:
Blackwell, 1966), also in M. Beany ed., The Frege Reader (Oxford: Blackwell, 1997). For extensive discussion of
Frege’s work, see T. Burge, Truth, Thought, Reason: Essays on Frege. Philosophical Essays, Volume 1 (Oxford:
Oxford University Press, 2005).
10 A. Tarski, ‘The Concept of Truth in Formalized Languages’ (1931, 1933), translated in Logic, Semantics,
Metamathematics (Cambridge: Cambridge University Press, 1956); R. Carnap, Meaning and Necessity (1947)
(Chicago: Chicago University Press, 2nd edition, 1967); A. Church, ‘A Formulation of the Logic of Sense and
Denotation’, in P. Henle, H. Kallen, and S. Langer eds., Structure, Method, and Meaning (New York: Liberal Arts
Press, 1951); A. Church, Introduction to Mathematical Logic, Volume I (Princeton: Princeton University Press,
1956); A. Church, ‘A Revised Formulation of the Logic of Sense and Denotation’, Nous 7 (1973), 24–33, Nous 8
(1974), 135–156. All of the relevant Church material occurs in T. Burge and H. Enderton eds., The Collected
Works of Alonzo Church (Cambridge, MA: MIT Press, 2019); W. Quine, Word and Object (Cambridge, MA: MIT
Press, 1960); D. Kaplan, Foundations of Intensional Logic (Dissertation, University of California, at Los Angeles,
1964); D. Davidson, ‘The Logical Form of Action Sentences’ (1967), in Essays on Action and Events (Oxford:
Clarendon Press, 1980).
16 Introduction
notion of singular Bedeutung and perception.11 Perceptual states function to refer to—or
pick out—particular entities in the environment. Their attributive, or characterizing,
function is systematically associated with their referring, or picking-out, function.
Frege’s technical analog of singular reference, singular Bedeutung, was context-free. He
wanted to understand context-free names, like numerals or diagrams of geometrical fig-
ures. Because of his interest in mathematics and logic, he neglected detailed reflection on
context-dependent reference. His conception of how context-free singular reference com-
bines with attribution can, however, be naturally extended to contextual-dependent,
demonstrative-like singular reference. Reference via perceiving something, which depends
on causal relations to what it picks out, depends on context in that it depends on occurrent
stimuli to pick out anything.
Philosophical and mathematical extensions of Frege’s ideas about singular denotation
(Bedeutung) were applied mainly to context-free languages, like mathematics and pure math-
ematical logic, until the 1970s. Philosophers extended Frege’s idea of context-free denotation
to demonstrative uses and other context-dependent devices in language and thought.12
I think that context-dependent demonstrative reference in language and thought are
literal outgrowths from the referential relation between a perceptual state and perceived
entity. Of course, demonstratives in language and much demonstrative-like reference in
thought are applied in individuals’ acts. Perceptual reference is mostly not active. However,
demonstrative-like reference in thought and demonstrative-like reference in perception
are structurally and functionally the same. Although perception is the root of reference,
understanding reference began with understanding language and thought. That under-
standing can be fruitfully applied back to the root source—perception.
Third, Frege’s notion of sense, as a way in which entities are presented to the mind, is a
further semantical idea that he applied to language and thought, but that has clear applica-
tions to perception.13 That notion is the antecedent of modern notions, certainly my
notion, of representational content. In Chapter 2, I explain the notion of representational
content in some detail. For now, think of it as a way of representing that sets, or contrib-
utes to setting, conditions for accuracy or truth. Different thoughts can represent the same
entity in different ways. Different kinds of perceptual states can, and commonly do, repre-
sent a given particular and given attributes of the particular in different ways. Differences
in “ways” constitute different kinds of psychological competence or capacity, different
representational contents. Perceptual reference to a particular, and perceptual attribution
11 Frege, ‘Über Sinn und Bedeutung’; G. Frege, ‘Der Gedanke’ (1921), translated in Beany ed., The Frege
Reader. See Frege’s use and discussion of Kant’s notion of intuition, a form of singular reference, in Begriffsschrift,
section 8; and G. Frege, Grundlage der Arithmetik (1884), J. Austin trans., The Foundations of Arithmetic
(Evanston, IL: Northwestern University Press, 1968). See also I. Kant, Critique of Pure Reason, P. Guyer and
A. Wood eds. (Cambridge: Cambridge University Press, 1998), B33–B34. Frege used Kant’s notion to account for
reference in geometry, focusing on Kant’s sub-species of sensible intuition: apriori sensible intuition. Frege was
aware that Kant’s other sub-species—empirical sensible intuition—is equivalent to sense perception. Frege
understood both species to be theoretical notions that function to do the work of the common-sense notion,
singular reference (or singular denotation).
12 See, among others, T. Burge, ‘Demonstrative Constructions, Reference, and Truth’, Journal of Philosophy 71
(1974), 205–223; T. Burge, ‘Belief De Re’, The Journal of Philosophy 74 (1977), 338–362, reprinted in T. Burge,
Foundations of Mind. Philosophical Essays, Volume 2 (Oxford: Clarendon Press, 2007); D. Kaplan,
‘Demonstratives: An Essay on the Semantics, Logic, Metaphysics, and Epistemology of Demonstratives and
Other Indexicals’ (1977), in J. Almog, J. Perry, and H. Wettstein eds., Themes from Kaplan (Oxford: Oxford
University Press, 1989); D. Kaplan, ‘On the Logic of Demonstratives’, Journal of Symbolic Logic 8 (1979), 81–98.
13 Frege, ‘Funktion und Begriff ’; Frege, ‘Über Sinn und Bedeutung’; Frege, ‘Der Gedanke’.
Fregean Source of Semantical Notions for Perception 17
of a property or relation, in a specific way that hinges on a specific stimulus and perspec-
tive on the particular, property, or relation, lies at the heart of perception. The psycho
logically specific way is the way of characterizing the entities that are purportedly referred
to in a perceptual state. That way is integral to perceptual constancies.
The modern notion of representational content is an heir of Frege’s notion of sense.
As will become evident in discussion of representational content, my notion drops Frege’s
Platonic ontology of sense, and grounds representational contents in psychologies. In fact,
I take representational contents to be psychological kinds of representational states, events,
or competencies. Representational contents can include psychological events in time
(what I call context-bound referential applications). This idea, too, departs from Frege’s
notion of sense, which was context-free. However, representational contents retain key
functions that Frege assigned to sense. Both sense and representational content set, or
contribute to setting, veridicality conditions, and constitute specific ways in which repre-
sentata are presented to minds.14
A fourth Fregean idea is a corollary of the first three. Since Frege took senses to be or
contribute to thought contents, he took them to have logical form.15 He took thoughts’
logical forms to be structures that ground systematic semantic explanation of truth and
falsity in terms of denotation and predication. Analogs of logical forms also underlie
propositional inference—for him, specifically deductive inference.
Frege’s notion of logical form has an analog and root in perception. The simplest logical
forms involve reference and predication. Reference occurs in perception: perception of
something is a kind of reference to it. Predication is a sub-species of attribution.
Attribution occurs in perception: perceiving something as being a certain way involves
attribution of a property, kind, or relation. Perception attributes properties, kinds, or rela-
tions in perceptually characterizing what is perceived—what perception referentially picks
out. Combination of aspects of perception that function to refer, on one hand, with aspects
that function to attribute, on the other, yields complexes that are either accurate or inaccurate.
These complexes are analogs of thoughts or sentences that have logical form. The com-
plexes are literally representational structures of perceptual states. The combinations of
reference and attribution can be evaluated for perceptual accuracy or inaccuracy. Logical
form is a structure for propositional thoughts and sentences. Logical forms go well beyond
anything present in perception. Perception contains no logical connectives (or, if–then),
no quantifiers (all), and no modal operators or modal attributives (is necessary). I argue
these points in Chapter 6, the section Absence in Perception of Negations,
Conditionals, Disjunctions, Quantifiers. They are intuitively plausible
enough to be stated here. Perception exercises reference and attribution. Reference and
attribution are psychological, representational capacities. Understanding ways in which
these semantical factors combine to yield semantically structured perceptual states, which
are accurate or inaccurate, is part of understanding perception.
14 Burge, ‘Belief De Re’, section 4; and T. Burge, ‘Postscript: “Belief De Re” (1977)’, in Foundations of Mind.
Philosophical Essays, Volume 2.
15 Strictly, for Frege, the notion of denotation (Bedeutung) applied to a relation between symbols and represen-
tata, not a relation between senses and representata. His term for the latter relation was ‘determination’
(Bestimmung). Determination is structurally analogous to denotation. In fact, Frege thought of determination
as more basic than denotation, because he thought that the contents of thoughts (senses) are more basic than
linguistic expression of them.
18 Introduction
Since the science of perception aims at understanding accuracy and illusion in percep-
tion, it should think systematically about the semantics of perception. Although this
semantics is not nearly as rich and varied as the semantics of language and thought, it is,
as we shall see, much richer and more complex than one might imagine from off-the-cuff
reflection.
All four key Fregean semantical notions, then, have roots in perception. Perceiving an
object, or other particular, is a type of “picking out” or contextual reference. Perceiving
something as being a certain way is a type of characterization or attribution. Perceiving
something as having a given property from one perspective, via one kind of perceptual
state—as distinct from perceiving it as having the property from another perspective,
via another kind of perceptual state—just is attributing the property through one kind of
representational content rather than another. All perceptual states can be evaluated for
accuracy. All accurate and all inaccurate perceptual states are combinations (a) of seman-
tical elements (representational capacities) whose representational function is to refer
(to pick out), with (b) semantical elements whose representational function is to attribute
some property, relation, or kind to what is referred to. Such combinations systematically
yield an analog of logical form.
These ideas are, I think, central to obtaining systematic understanding of accuracy and
inaccuracy in perception. The terminology and traditions drawn on here can deepen
understanding of basic aspects of the psychology of perception. They thereby deepen
understanding of perception itself, specifically its representational aspects. That is a reason
why perceptual psychologists interested in understanding general features of their subject
matter would do well to master the technical vocabulary and ways of thinking that mark
parts of this book.
Although the basic notions of reference, attribution, representational content, and
representational form are all applicable to perception, there are differences between
their applications to perception and their applications to language and thought. An aim
of this book is to tailor application of these and other semantical notions to perception.
Since the most basic forms of reference, attribution, representational content, and repre-
sentational form are perceptual, semantical reflection on perception can illumine the roots
of these notions.
A corollary is that one of the deepest contributions that thinking about perception
by using tools from the logical and semantical traditions can make is to clarify what it
is about perception that differentiates it from thought—and vice-versa. Reflecting on
differences between the representational function and representational form of percep-
tion, on one hand, and the representational function and representational form of belief,
on the other, is a route into understanding differences in psychological kinds. Whereas
perception shares reference and attribution with propositional thought, it utilizes these
representational capacities in more limited ways. Belief is propositional. Perception is not
propositional. It is nominal. Understanding the difference will, I think, help in under-
standing the most basic joint in the mind—the joint between perception—and perceptual-
level capacities—, on one hand, and propositional thought, on the other. The structures of
perception that I discuss are not linguistic or logical. They are more basic than and prior
to the structures of language and propositional thought. I think that noun-phrase struc-
tures governed by contextual, referential determiners in language and thought derive from
perception. Perception has the most basic, and evolutionarily earliest, semantics.
2
Perception
…while a sense is what is receptive of sensible [or perceptible] forms without

the matter, just as the wax receives the seal-imprint of the ring without the
iron and the gold, but takes the golden or the brazen imprint, but not qua gold
or bronze, yet in a similar way too the sense is affected by each thing that has
color or flavor or sound, but not qua each of those being called, but qua of
such-a-sort [paralleling qua as golden or as brazen], and according to propor-
tion [or as a measurement]
Aristotle, De Anima, Book 2, Chapter 12, 424a17–24
translation in consultation with Gavin Lawrence
In Origins of Objectivity, chapter 9, I proposed a rough explication of what perception is.

Here I set out that account with minor refinements. The explication is not a definition. It
does not purport to give necessary and sufficient conditions. I think that the notion of
perception is a primitive notion that cannot be defined in other terms in an illuminating
way. The explication functions to provide orientation and background.
Here is the explication:
Perception is sensory, objective representation—paradigmatically by individuals—

that is generated from current stimulation and normally (except in unusual circum-
stances) represents a roughly present subject matter as roughly present.
I say ‘paradigmatically by individuals’. I doubt that all perception is necessarily attributable

to an individual perceiver. Perhaps there are cases in which it is attributable only to an indi-
vidual’s sub-system. I know of no clear cases. Certainly, all cases of perception occur in
individuals and serve perception attributable to individuals. Paradigmatically, individuals
perceive.16 Here, the key concepts in this explication are sensory, representation, and objective.
16 Origins of Objectivity, 369, where I write:

I do not claim that all perceptions are perceptions by an individual. I claim that necessarily and
constitutively, some perceptions in an individual’s perceptual subsystem are perceptions by the
individual. And I claim that all perceptions, including any that are not attributable to the individual,
serve perception by individuals.
I do not push cases of perception that are not imputable to an individual. My reason for allowing for them is not that
I have specific cases in mind. It is that I know of no apriori reason for disallowing them. I do not think that a percep-
tion’s being inaccessible to consciousness, in itself, renders it a perception that is not attributable to the individual.
I lay aside perception by groups. If there is such a thing, parallel points would apply. Paradigmatically, percep-
tual states are attributable to the macro-perceiver, whether the individual or the group. Of course, all perceptual
states attributable to perceivers are also attributable to their sub-systems. For criticism of a debate about uncon-
scious perception that misuses my views on perception “by the individual”, see T. Burge, ‘Entitlement: The Basis
for Empirical Warrant’, in P. Graham and N. Pedersen eds., Epistemic Entitlement (Oxford: Oxford University
Press, 2020), 69n45. (cont.)
Perception: First Form of Mind. Tyler Burge, Oxford University Press (2022). © Tyler Burge.
DOI: 10.1093/oso/9780198871002.003.0002
20 Perception
Perceptual States as Sensory States
Perception is sensory. It is a certain capacity or competence for discrimination as a result

of current stimulation of psychological states formed from causal impact.17 The discrimin
ation is discrimination of causes. It need not, and commonly does not, discriminate them
as causes. A sensory state discriminates entities that are among its causes. Since a percep-
tual state is a type of sensory state, particular entities that it discriminates are its causes.
Discriminating a type of cause is co-varying with and responding differentially to the
type; the discrimination must have a function for the individual or for the sensory system
of which it is a part. Thus sensory states are information registrational states. Discriminating
a particular—a non-repeatable—is responding to the causal impact of that particular,
where the particular is of a type that is discriminated. I conjectured earlier that the larger
functioning system within which these discriminations occur supports a capacity for action.
Sensory states in a sensory system—including a sensory perceptual system—are not
generated by operations attributable to the individual. The states are produced in the
individual. Although the individual perceives, the individual does not produce the

perceptions.
Sensory states are generated from current stimulation. Here I intend ‘generated’ to
exclude ‘propositionally inferred’. One might make propositional inferences caused by
current stimulation. Neither the premises nor the conclusions of the inferences would be
sensory states.
A sensory state has a basic discriminatory function—to discriminate via causal sensi-
tivity. Perceptual states discriminate via causal sensitivity. Unlike non-perceptual sensory
states, they discriminate representationally. This difference will be discussed shortly.
Perceptual beliefs also have this discriminatory function. Unlike sensory states, including
perceptual states, perceptual beliefs have other functions as well—centrally, contributing
capacities for propositional inference.
Like perception, perceptual memory, perceptual anticipation, and perceptual imagining
are broadly sensory. Their states are not sensings, or hence perceptual states. They are not
generated from current sensory stimulation. Perceptual memory is memory that functions
to retain perceptual contents. The contents of perceptual memory are perceptual con-
tents—prominently, perceptual attributives—that function to retain contents of percep-
tual states already generated. Perceptual memory represents its subject matter as in the
past. Perceptual anticipation is usually mediately related to perception via perceptual
memory. Like perceptual memory, it utilizes perceptual contents, but is not produced by
current sensory stimulation. It does not function to discriminate causes of present sensory
effects. Unlike perceptual memory, it functions to represent future, anticipated states or
events. Perceptual imagining produces states with perceptual content. They are not caused
by current sensory stimulation. Unlike perception, perceptual memory, and perceptual
anticipation, states of perceptual imagining do not function to represent veridically.
For just one of countless examples of scientific discussion of unconscious perception, see V. Axelrod, M. Bar,
and G. Rees, ‘Exploring the Unconscious Using Faces’, Trends in Cognitive Sciences 19 (2015), 35–45. See also note
6 above and note 827 below.
17 Here and throughout, I use the terms ‘capacity’ and ‘competence’ interchangeably. I use them in the sense
that Chomsky articulated—distinguishing competence from performance: N. Chomsky, Aspects of the Theory of
Syntax (Cambridge, MA: The MIT Press, 1965).
Another random document with
no related content on Scribd:
epithet, the “semper-august” character of Virgil, and innumerable
other things of the kind, disturb us not. Scaliger’s idol has spoken
Scaliger’s doom in Qui Bavium non odit—not, of course, that Hero
and Leander is itself by any means Bavian, but that it is so in
comparison with Homer. Nearly a hundred pages are given up to this
main comparison of Homer and Virgil. The others are shorter, but
always result in the same dogged maintenance of the superiority of
Latins to Greeks—that is to say, the same involuntary confession of
Scaliger’s preference of Rhetoric to Poetry. It is interesting, however,
to find him conducting his comparisons in a way in which, as in most
other cases, posterity for two centuries thronged to follow him—the
assemblage, that is to say, of passages on the same subject from
different poets.
Still less can we abstract the curious and invaluable survey of the
Hypercriticus. Not a little of it is actual review of actual
contemporaries or very recent predecessors, and review of the
ancients takes the same form, reinforced constantly by discussed
quotations. Sometimes, as in the case of Juvenal, these are
arranged into a little anthology of “jewels five words long,” strung
together with acute et hoc, illud valde festivum, and the like
appreciative interjections. His preference of Juvenal to Horace is
seasoned with a characteristic fling at Erasmus (p. 876).
Book VII.: Lastly comes an Epinomis or Codicil, which is
Epinomis. divided into two parts, and takes up some of the
special points of poetical or dramatic criticism then most interesting
—the relative importance of action and character, the parts of
tragedy, the Chorus, the metres most appropriate to the stage, and
the like, ending with a sort of “gratillity” or bonus in the shape of an
examination of a codex of Terence, which we could spare, at least in
this place. More piquant, at least, are the diatribes de negligentia aut
inscitia professorum, directed (with a show of respect) against
Erasmus once more; the occasional flights, such as “Variety is the
tirewoman of poetry”[103] (p. 906); the amusing references to mea
poemata, which in some parts of the book he has obligingly, and
once more with a fearlessness drawing nigh to rashness, exposed to
the arrows; and other things which are perhaps here all the more
numerous because the Book is an avowed Appendix, and, as it
were, omnium gatherum. They are, however, plentiful everywhere;
and if it were possible to revive the old periodical Literary
Miscellanies of commonplace-book character—a thing which will
have to be done sooner or later, if the accumulations of the last few
centuries are not to became mere Nineveh-mounds, as yet
unexplored—I should like to compose a florilegium of memorabilia
out of Scaliger.
For in this great space, occupied with equal method and erudition,
it could not be but that remarkable pronouncements on the more
general questions of literary criticism, whether given obiter or in
definite reference to argued questions, should emerge. Scaliger is,
General ideas indeed, less set than most of his predecessors in
on Unity and Italian criticism, and than some at least of his
the like. successors, on these general pronouncements.
“The disinterested and philosophic treatment of æsthetic problems
wholly aside from all practical considerations,” as the tendency of
Italian criticism has been rather unguardedly characterised,[104] does
not seem to have had the first attraction for him. Yet he could not, in
the wide sweep of his net, have avoided such questions if he would;
and, with his fearless temper and eager literary interests, there is no
reason to suppose that he would have avoided them if he could. He
did not explicitly enjoin the Three Unities,[105] but he did more than
any other man had done to inculcate that unfortunate notion of
“verisimilitude”[106] from which, much more than from Aristotle, they
were deduced. Not many words need be wasted (especially as the
point will recur only too often during the volume) on the absurdity of
this wresting of Incredulus odi. The whole arrangements of the
theatre are invraisemblables, no matter whether you have electric
light or cross-shaped laths with candles on them, marquises sitting
on the stage or millionaires in stage-boxes, elaborate scenery or
directions to the audience, “Here is Thebes.” You do not murder, or
(if you can help it) make love, in real life, before a miscellaneous
audience who have paid to see you do it; in real life you do not talk in
any regular stage lingo that has hitherto been invented, whether the
outward form of it be senarii, or fourteeners, or complicated rhymed
stanzas, or doggerel, or couplet, or blank verse, or stage prose. The
sixteenth century Globe, and the twentieth century Lyceum, are alike
unlike any place in which one habitually performs any action of life
from birth, through marriage, to death. That there is a stage
verisimilitude, which it is dangerous or fatal to break, need not be
denied. But neither Scaliger nor any of his successors in purism has
proved that we are, or ought to be, any more shocked by Æschylus
when he shifts from Delphi to Athens than by Thackeray when he
transports us from Flanders to Chelsea.
We may venture indeed to suspect that Scaliger “had more wit
than to be here.” One may frequently differ with him; but he seldom
runs mad on mere theory. It is he, for instance, who, while, as we
have seen, he lays down uncompromisingly that the material of
poetry is verse, instances the Æthiopica as a perfect epic. Instead of
confusing poetry and learning, as some have done, he holds the
much more sensible position that learning is useful to a poet. He
takes the hard-and-fast ethical view of the ends both of tragedy and
of all poetry, and he believes firmly in the type. But he does not
bemuse himself, as some had done and more were to do, in the
explanation of katharsis, and the definition of the tragic hero.
His greatest and also his most pervading critical fault is that
“deification of Virgil,” whereof, though by no means the inventor, he
was the chief prophet to the best part of three centuries. Let it be
His Virgil- admitted (with every possible emphasis on the fact
worship. that it is no mere extorted admission but a genuine
and spontaneous opinion) that anybody is free to admire Virgil or any
one else as much as he likes. “She that is fair to him” is so, and
there’s an end on’t. But if any one proceed, not merely to intimate
indifference to other fair ones, but to find positive fault with them
because they are not like her, then he becomes at once uncritical:
still more so if he erect her qualities, features, style, into abstract
virtues and positive truths, all opposites to which are sin and
vileness. He may call “Simula Silene, nervosa et lignea Dorcas,” to
take two only out of the famous list in the classic place of this matter.
But he must not declare that a girl who has a straight Grecian nose
is therefore ugly, or that softness and plumpness are not excellent
things in woman. Scaliger does this. For him Virgil is, at once, the
standard of excellence and the infallible touchstone of defect. Nay,
he is actually a better Nature; a wiser but more perfect Creation,
whereby you may save yourself the trouble of outside imitation,
inasmuch as everything worth imitating is there better done than by
Nature herself. It is impossible to exaggerate or caricature Scaliger’s
Maronolatry: as the Highfliers did in the case of Defoe’s Shortest
Way, he would cheerfully accept and indorse the most outrageous
statement of it.
Grave, however, as is this fault, and seriously as it vitiates
Scaliger’s attitude as a critic, there is no doubt that it served in itself
as the backbone of that attitude, and gave it the stiffness which
enabled it to resist at once argument and time. A cause of disquiet to
some critics themselves, and a rallying-cry to most enemies of
criticism, has been constantly found in the apparently floating and
uncertain character of the completest critical orthodoxy. Longinus
himself, perhaps the best exponent of that orthodoxy, has been and
is charged with vagueness; and all those who follow him must lay
their account with the same accusation. In the last resort we often
cannot give a clear, definite, cut-and-dried reason for the faith that is
in us, and we still oftener had better not try to do so. Scaliger and
Scaligerism are in no such plight. Their Sortes Virgilianæ are ex
hypothesi decisive, and of universal application. What is found in
Virgil is good, is the best; what is different from Virgil is bad or
mediocre; what is like Virgil is good in direct proportion to the
likeness. This of itself gives confidence both to the critic and to his
disciples.
Again, Scaliger, though he has no more right to arrogate Reason
and Nature as on his side than the rest of his school, possesses, like
all of the best of them, a certain sturdy prima facie common-sense. It
His solid is this which dictates his theory of dramatic
merits. verisimilitude; this which palliates some of his
Homeric and other blasphemies. Though uncompromisingly moral,
and by no means illogical (when you have once granted his bundle
of postulates), he is not in the least metaphysical. The wayfaring
man, with tolerable intelligence and a very little trouble, can
understand him perfectly.
Still more unmixed praise can be given to him from other points of
view. To any scholar his scholarship is singularly refreshing in its
thoroughness and range; he really neglects nothing proper to his
subject, though he may define that subject with a somewhat arbitrary
hand. Agree with him or differ with him as we may, it is an infinite
comfort to be brought thus in contact and confrontation with the
actual texts—to exchange the paper symbols of “the poet,” “the
dramatist,” “the satirist” in the abstract, for sound ringing coin of
actual poetry, drama, satire, told down on the counter, and tested by
file and acid if required. The literary atlas of the Hypercriticus is, as
has been said, the first attempt at a complete thing of the kind since
Quintilian, and of necessity far more complete than his. In fact,
Scaliger taught the school opposed to him—the school which after
many a generation of desultory fighting at last worsted his own—the
way to conquer. History and Comparison—the twin lights of criticism,
the only road-makers across the abyss—are resorted to by him
fearlessly. That he loses the best of their light, and twists the road in
the wrong direction, by following Will-o'-the-wisps like his Virgil-
worship, matters in detail but not in principle. He has practically
come back to the safe way which Aristotle entered, but was
precluded from treading far enough, which Quintilian and Longinus
trod, but on which most of the ancients would not set foot. He has
not found the last secret—the secret of submitting to History and to
Comparison; he still looks upon both as instruments to be used
merely under the direction of, and in subordination to, the purposes
of a priori theory. His neglect of the vernaculars is not only wrong,
but by his time absurd. His minor prejudices (as against Erasmus)
are sometimes contemptible. His actual taste, as has been said, was
probably neither delicate nor versatile. But he has learning, logic,
lucidity within his range, laborious industry, and love of literature. The
multitude which followed him followed him partly to do evil; but it
would have been a surprise, and almost a shame, had so bold and
capable a leader lacked a multitude of followers.
As has been said, Lilius Giraldus also refers to Lodovico
Castelvetro, who at least resembled Scaliger in the characteristic
Castelvetro. Ishmaelitism of the Renaissance critic. His quarrel
with Caro, also already referred to, was unluckily, we
must not say distinguished, but marked, by unfair play on the part of
his adversary, who “delated” him to the Inquisition for heresy; and
Castelvetro had to fly the country. His most important work appeared
late, the famous edition and translation, with commentary, of the
Poetics[107] not being published till a year before his death. “He was
of his nature choleric,” says his biographer; and he bestowed a good
deal of this choler not merely upon Caro, but upon the majestic
Bembo and others. Yet Castelvetro was a very remarkable critic, and
perhaps deserved the ascription of actual critical genius better than
any man who has yet been mentioned in this volume. It is but for
chequered righteousness that his practically certain formulation of
the Three Unities can be counted to him; but, as we shall see, he
has other claims, from which it is not necessary to write off anything.
His impartial attachment to both classical and vulgar tongues
ranks him, of itself, in a higher sphere than that of Scaliger; and a
certain impetuous, incalculable, prime-sautier genius puts him higher
still. Even contemporaries seem to have recognised this in him,
though they sometimes shook their heads over its pronouncements.
[108]
It may, indeed, sometimes seem that these pronouncements are,
if not inconsistent, difficult to connect by any central tie-beam of
critical theory. But this is almost inevitable in the case of a critic
whose work takes the form, not of regular treatises on large subjects,
nor even of connected essays on separate authors and books, but of
commentaries and adversaria, where the passage immediately
under consideration is uppermost in the writer’s mind, and may—not
illegitimately in a fashion—induce him to display a facet of his
thought which does not seem logically connected with other facets.
This peculiarity is perhaps the only excuse for the depreciation of
Dacier, who, reinforcing his native dulness with the superciliousness
of a Frenchman in the later years of Louis XIV., accused Castelvetro
of ignorance, and even of contradiction of Aristotle. The fact is, that
Castelvetro is first of all an independent critic, and that, though there
are few less common, there are no more valuable critical qualities
than independence, even when it is sometimes pushed to the verge
of eccentricity, providing only that it is sincere, and not ill-informed. It
seems to me uncharitable, if not flagrantly unjust, to deny
Castelvetro sincerity, and either impudent or ignorant to deny him
information.
But he had also acuteness and taste. I do not know a better
example in little of the latter quality at the time than his short and
scornful description[109] of a preposterous comparison by another
critic, Bartolommeo Riccio, between the “Sparrow” of Catullus and a
pretty but commonplace poem of Navagero on a dog. One may sigh
The Opere over the ruling passion, not to say the original sin, of
Varie. critical man, on passing from this to a tangle of
recrimination and “that’s my thunder” which follows with reference to
Riccio and Pigna and Cinthio. But this passes again into a solid
discussion on the material and form of poetry, and on the office of
the Muses. Many of these animadversions are, as we should expect,
purely verbal, sometimes not beyond the powers of the
grammaticuccio, of whom Castelvetro himself not unfrequently talks
with piquant scorn. But the comfort of finding annotations on Virgil
alternating with discourses on Dante, like that of placing a quarto on
Petrarch side by side with one on Aristotle, more than atones for any
occasional hair-splitting. We are at last in the Jerusalem of general
Literature which is the mother of us all, which is free and universal;
not in this or that separatist Samaria or exclusive Hebron. The
Platonic annotations, which are numerous, are important, because
they show just the other side of Castelvetro’s talent from the merely
verbal one—almost the whole of them being devoted to the
exposition and illustration of meaning. It is a great pity that he did not
work his notes[110] on the Gorgias (which he regards expressly as
Plato’s Rhetoric) into a regular treatise of contrast and comparison
on this subject between Aristotle and Plato. But all these notes show
us the qualification of the commentator to deal with so difficult a
subject as the Poetics.
The stout post quarto, with its vignette of an exceedingly
determined-looking owl standing on a prostrate pitcher and hooting
The Poetica. Kekrika, is dedicated to Maximilian II. It is arranged
on a system equally simple and thorough. First
comes a section of the Greek Text; then a short Italian summary of
its contents; then the Italian translation; and then the spositione—the
Commentary—which may be long or short as circumstances require.
Often, on a Greek text of a few lines, it will run to as many quarto
pages, full-packed with small print. Not the least advantageous part
of this quadripartite arrangement is that the summaries—being,
though very brief, to the point—are capable of being put together as
a table of contents. This, however, but partially applies to
Castelvetro’s commentary, which is often not a little discursive from
the text. The defect was, however, supplied in the second edition by
an elaborate index specially devoted to the Spositioni, and
consisting, not of mere words or names with page references, but of
reasoned descriptions of the subjects, as thus—
"Allegrezza.
“Come nasca dalla tristitia, che si sente del male del giusto, e del
bene del malvagio.
oblica, che si prende dalla miseria, o dalla felicita altrui qual sia,”
&c. &c.
This is a great help in tackling Castelvetro’s text, the book containing

some seven hundred pages, of perhaps as many words each.
No analysis of a book of such a size, so necessarily parasitic or
satellitic on another in general run, and yet branching and winding
with such a self-willed originality of its own, is possible. One might
On Dramatic easily write a folio on Castelvetro’s quarto. Here we
conditions. can only, as in most other cases now, except those
of books or parts of books at once epoch-making in character and
moderate in bulk, give an idea of the author’s most important views
on general and particular points. It was necessary, since Castelvetro
is revolving round Aristotle, that the greater part of his treatise should
deal with the drama: and perhaps nowhere is that originality which
has been praised more visible than here, whether it lead him wrong
or right. He has undoubtedly made a step, from the mathematical
towards the æsthetic view of literature, in conditioning, as he does,
his view of the Drama by a consideration of the stage. To literary a-
priorists this is of course horrible; to those who take the facts of
literature, as they take the facts of life, it is a welcome and
reconciling discovery. The conditions of the Greek stage were
admittedly such as can never be naturally reproduced, and therefore,
however great and perfect the Greek Tragedy may be in its own way,
it cannot usurp the position of “best in all ways”; and can still less
pretend to dictate to other kinds that they shall not be good at all in
ways different from its own.
If the details of Castelvetro’s theory do not always correspond in
excellence to the sense and novelty of the general view, this is
because he adulterates his notion of stage requirements with that
unlucky “verisimilitude” misunderstood, which is the curse of all the
neo-classic critics, and which comes from neglect of the Aristotelian
preference of the probable-impossible to the improbable-possible.
On the Three The huge Mysteries of the Middle Ages, which
Unities. ranged from Heaven to Hell, which took weeks to
act, and covered millennia in their action, did at least this good to the
English and some other theatres—that they familiarised the mind
with the neglect of this verisimilitude. But Castelvetro would have
none of such neglect. His play must be adjusted, not merely in
Action, but in Space and Time, as nearly as possible to the actual
capacity of the stage, the actual duration of the performance.[111] And
so the Fatal Three, the Weird Sisters of dramatic criticism, the
vampires that sucked the blood out of nearly all European tragedy,
save in England and Spain, for three centuries, make their
appearance. They “enter the critical literature of Europe,” as Mr
Spingarn has very truly laid it down,[112] “from the time of
Castelvetro.”
But to balance this enslaving of the Drama (in which he far
exceeds Aristotle), Castelvetro frees the Epic from Aristotelian
On the restrictions in an almost equally important manner.
freedom of From his references in the Opere Varie to Cinthio
Epic. and Pigna, it would appear that he claimed, if not
priority, an even portion with them in the consideration of the subject
of Epic Poetry. And though not agreeing with them altogether, he
certainly agrees with them in enlarging the domains of the Epic.
Poetry, he says in effect,[113] may do anything that History can do;
and, like the latter, it may deal, not only with one action of one man,
but with his life-actions, or with many actions of many men.
With Castelvetro, however,—and it is probably the cause why
pedants like Dacier undervalue him,—both the character of his
compositions, and probably also the character of his mind, draw him
much more to independent, though by no means always or often
isolated, critical aperçus and judgments, than to theoretical
discourses, with or without illustration. To put it differently, while there
His eccentric is usually a theory at the back of his appreciations,
acuteness. the appreciation generally stands in front of the
theory. But however this may be, that quality of “unexpectedness,” in
which some æsthetic theorists have found such a charm, belongs to
him as it does to few critics. One might, for instance, give half-a-
dozen guesses to a tolerably ingenious person without his hitting on
Castelvetro’s objection to the story of Ricciardetto and Fiordispina in
the Orlando.[114] That objection is not moral: not on the ground of
what is ordinarily called decorum: not on that of digression, on that of
improbability generally, on any other that is likely to occur. It is, if you
please, that as Fiordispina was a Mahometan, and Ricciardetto a
Christian, and as Christians and Mahometans do not believe in the
same kind of Fauns and Fairies, as, further, Fauns do not eat ladies
or goddesses, whether alive or dead, Ricciardetto’s explanation of
his alleged transformation of sex is not credible. In a modern writer
this would look like an absolute absence of humour, or like a clumsy
attempt at it; and I am not prepared to say that humour was a strong
point with these Italian critics as a rule. But Castelvetro strikes me as
being by no means exceptionally unprovided with it: and such a
glaring lapse as this is probably due to the intense seriousness with
which these critical questions, new as they were, presented
themselves to him and to his class.
They get, as was once said, “into logical coaches”; and are
perfectly content to be driven over no matter what minor precipices,
and into no matter what sloughs of despond, so long as they are not
actually thrown out. Yet Castelvetro at least is never dull. At one
time[115] he compares the “somnolent indecorum,” the
sconvenevolezza sonnachiosa, of Homer to the practice of German
innkeepers (whether observed by himself in his exile, or taken from
Erasmus, one cannot say) in putting the worst wines and viands on
the table first, and the best later. Elsewhere[116] he gives a very
curious reason against that other sconvenevolezza (this sonorous
word is a great favourite with him) which he too saw in the use of
prose for tragedy—namely, that in reciting verse the speaker
naturally raises his voice, and so makes it more audible to the
audience. He has been blamed for adopting the notion of rank being
necessary to tragic characters, but on this see ante (p. 61).
His irreverent independence in regard to Virgil is noticeable in a
critic of his time, and of course especially so if one comes to him
Examples: Homer’s straight from Scaliger. It would not be fair
nodding, prose in to represent him as a “Virgiliomastix,” but
tragedy, Virgil, minor his finer critical sense enables him to
poetry.
perceive the superiority of Homer, in
respect of whom he goes so far[117] as to say that Virgil “is not a
poet.” But this—per se, of course, excessive—had been provoked by
the extravagance of Maronolatry from Vida downwards: and
Castelvetro does not scruple to praise the Mantuan for his grasp, his
variety of phrase, and other good things. He has an extremely
sensible passage—not novel to us, but by no means a truism to his
contemporaries or to a good many poets still—on what he who
publishes miscellaneous poetry has to expect. By the publication,
says this other Messer Lodovico, of a thing which nobody asked him
for (cosa non richiesta) without any necessity, he publishes at the
same time his confidence in himself, and affirms that the thing is
good. “Which thing,” goes on Castelvetro in his pitiless critical
manner, “if it be found to be faulty (rea) and blameworthy, it convicts
him who publishes it either of malice or of folly.” Alas! for the minor
bard.
His attitude[118] to the everlastingly vexed question of the
connection of verse and poetry is very sensible, and practically
anticipates, with less reluctant circumlocution, that of Coleridge, who
in more things than one comes close to Castelvetro, and who
The medium probably knew him. He does not here contradict
and end of Aristotle by denying that verse is un-essential to
Poetry. poetry. But he insists—and points out the undoubted
truth that Aristotle’s practice, whatever his theory may do, admits this
—that Verse is a kind of inseparable accident of poetry,—that it is the
appropriate garb and uniform thereof, which cannot be abandoned
without impropriety. And he takes up this attitude still more
emphatically in regard to the closely connected, and still more
important, question of the end of Poetry. Here, as we have seen, the
Uncompromisi great Master of Criticism temporised. He did not
ng doubt that this end was Delight: but in deference to
idols, partly of the Cavern, partly of the Market-
championship place, he yokes and hampers this end with moral
of Delight. improvement, with Imitation, itself for itself, and so
on. Castelvetro is much more uncompromising. One shudders,
almost as much as one rejoices, at the audacity of a critic who in
mid-sixteenth-century calmly says, “What do beginning, middle, and
end matter in a poem, provided it delights?”[119] Nay, Castelvetro has
reached a point of view which has since been attained by very few
critics, and which some who thought they had gained this peak in
Darien first may be mildly chagrined to find occupied by him—the
view that there are different qualities of poetry, suited to delight
different qualities of persons and of mind.
How seldom this view has been taken all critics ought to know, if
they do not. Even now he who climbs the peak must lay his account
with stone-throwing from the garrisons of other points. That Burns
administers, and has a right to administer, one delight to one class of
mind, Shelley another to another; that Béranger is not to be denied
the wine of poetry because his vintage is not the vintage of Hugo:
that Longfellow, and Cowper, and George Herbert are not to be
sneered at because their delight is the delight of cheering but not of
intoxication; that Keble is not intrinsically the less a poet because he
is not Beddoes, or Charles Wesley because he is not Charles
Baudelaire—or vice versa in all the cases—these are propositions
which not every critic—which perhaps not very many critics—will
admit even in the abstract, and which in practice almost every critic
falsifies and renounces at some time or other.[120] But they are
propositions which follow fairly, and indeed inevitably, from
Castelvetro’s theory of the necessary end, Delight, and the varying
adjustment of the delighting agent to the patient’s faculty of being
delighted.
He is perhaps less sound in his absolute condemnation of
“knowledge” as material for poetry. He is right in black-marking
Fracastoro from this point of view: but he is certainly not right in
extending the black mark to Lucretius. The fact is, that even he could
not wrench himself sufficiently free from the trammels of old time to
see that in the treatment lies the faculty of delighting, and that
therefore, on his own scheme, the treatment is the poetry.
There are few writers to be dealt with in this volume—none, I think,
already dealt with—to whom it would be more satisfactory to devote
His the minutest handling than to Castelvetro. He has
exceptional been called by Mr Spingarn “revolutionary.” The
interest and term, in an American mouth, probably has no
importance.
unfavourable connotation; but waiving that
connotation altogether, I should be inclined to demur to it. Even the
Vehmgericht (if one may rely on the leading case of Vgr. v. Philipson,
reported by Sir Walter Scott) acquitted of High Treason those who
had spoken evil of it in countries where its authority was not
acknowledged, and indeed its name hardly known. Now, Castelvetro
was dealing—as we must, for his honour as well as for our
comprehension of him, remember that he dealt—with modern as well
as with ancient literature at once, and instead of adopting the
injudicious though natural separation of Minturno, or the one-sided
treatment of Scaliger, was constantly exploring, and always more or
less keeping in view, territories not merely in which Aristotle’s writ did
not run, but which in Aristotle’s time were No Man’s Land and terra
incognita. He can no more be regarded as a revolutionary or a rebel,
in framing new laws for the new facts, than a man could be regarded
in either light for disregarding the Curfew Law at the North Pole, or
for disobeying sumptuary regulations as to the use of woollen in the
tropics. His ethos is really that of the self-reliant, resourceful, and
adventurous explorer, as he has been called—of the experimenter in
new material and under new conditions. That the paths he strikes out
sometimes lead to culs-de-sac—that the experiments he makes
sometimes fail, is nothing more than is natural, than is inevitable in
the circumstances.
More generally his value is great, and we may forgive him
(especially since he did us little or no harm) the binding of the Unities
on the necks of Frenchmen and Italians, in consideration of the
inestimable service which he did in standing up for Epic—that is,
Romantic—Unity of a different kind, and in formulating, in a “No
Surrender” fashion, the doctrine of Delight as the Poetic Criterion. By
doing this he not merely fought for the freedom of the long narrative
poem (which, as it happens, has been a matter of minor importance,
save at rare intervals, since his time), but he unknowingly
safeguarded the freedom of the long narrative prose romance or
novel, which was to be the most important new contribution of
modern times to literature. Nor may it be amiss once more to draw
attention to a more general merit still, the inestimable indifference
with which he continually handles ancient and modern examples.
Only by this—the wisest “indifference of the wise”—can true criticism
be reached. It is an indifference which neglects no change of
condition, which takes count of all features and circumstances, but
which, for that very reason, declines to allow ancient literature to
prescribe unconditionally to modern, or modern to ancient, or either
to mediæval. As to this last, Castelvetro has, and could be expected
to have, nothing to say: as to the others, he is more eloquent in
practice than in express theory. But his practice speaks his
conviction, and it is the practice by which, and by which alone, the
serene temples of the really Higher Criticism can be reached.
The last third of the century provides only one author who
deserves (though he has seldom received) at least equal attention
with Scaliger and Castelvetro: but it has, like the second, a crowd of
Tasso and the minor critics who must not be wholly passed over.
controversies Moreover, it boasts—if such a thing be a subject of
over the boasting—one equally famous and weary
Gerusalemme.
controversy, that over the Gerusalemme. This,
which expects the critical historian as its prey, and will test his
powers to the utmost if haply he may wrestle free of it at once
without inadequacy and without tedium, we may dare first: may take
the interesting single figure of Patrizzi or Patrici second, and then
may sweep the rest into a conclusion, which will itself leave not a
little summarising to be done in the Interchapter succeeding this
Book.
Torquato Tasso was, in more ways than one, fated to the ordeal of
controversy. His work would, in the already unfolded state and
temper of Italian criticism on the subject of the “heroic poem,” have
invited it in any case; but he had, in a manner, inherited the
adventure. His father, Bernardo, as has been briefly recorded above,
had himself taken much interest in critical questions; and after being
at first a classicist, had come round to the position of Cinthio. It was
Torquato’s object, by argument and example alike, to reconcile the
combatants. His Discorsi did not appear till late in 1587;[121] but they
are said to have been written some twenty years earlier, after the
appearance of Minturno’s Italian book. His plan is as simply obvious
—shall we say as obviously defective?—as that of the immortal
contributor to the Eatanswill Gazette. He, too, “combined his
information.” Some kind of Unity is to be imposed on the Romantic
Variety; and though this Unity cannot possibly be the Aristotelian, it
need not be quite such a different kind as that of Castelvetro. It is to
be organic, but may permit itself the organs of a complex animal
system.
Nor did Tasso stick to generalities; nor did he shrink from giving
hostages to fortune, and his enemies, by embodying his ideas in
practice. These ideas we have already seen floating in various
critical minds from Fracastorius to Castelvetro. The “heroic poem”—
for his theory and his example alike consecrated that word for use,
instead of either “epic” or “romance,” for nearly two centuries—must
not be pure invention, but must avail itself of the authority of history.
It must be animated by religion, true religion—that is to say,
Christianity. It must have the supernatural. The hero must be a pious
and moral, if not necessarily faultless, character. It must not be too
dogmatic—that the poet may be free. It must deal with ancient or
modern history so as to be neither absolutely unfamiliar, nor too
familiar in its atmosphere and manners. The persons, things, and
scenes must be noble and stately. It will probably strike every one
that this is an admirable receipt for a historical novel; and thus do we
constantly find blind strivings at things that cannot yet get
themselves born. But whether it is an equally good receipt for a
poem may be doubted. Some of us, at least, have no doubt that the
Gerusalemme, which is faithfully constructed in accordance with it, is
not nearly so good a poem as the Orlando, for the graceless graces
of which it was expressly devised to substitute something more
orderly and decent.
The extensive and execrable controversy which followed did not,
however, turn wholly, though it very largely turned, on the actual
case of Ariosto v. Tasso. But, as usually happens, the partisans of
the latter provoked it by unadvised laudations of him, and worse-
advised attacks on his great predecessor. The Florentines had not,
as such, any special reason for championing the “turnip-eating”
Ariosto; but Tasso had offended the coteries of the Della Crusca, and
a Della Cruscan chief, the Salviati already mentioned, took the field
against the author of the Gerusalemme. He sallied forth in turn; and
the bickering became universal. Five mortal volumes of the standard
edition of Tasso appear to be occupied with an incomplete collection
of the documents on the subject—a collection which I have not read
and do not intend to read, but which whosoever rejoices in such
things may, if he likes, supplement with all the Histories of Italian
Literature from Tiraboschi downwards, and all the Lives of Tasso,
especially those of Serassi in the eighteenth century and Solerti in
the nineteenth.
The most important upshot of the controversy is not itself in
dispute. The impregnable historical position of Cinthio was strangely
neglected by both sides (except by Ishmaelite outsiders like Bruno
and Patrizzi); nor was even the modified Aristotelianism and
“Unitarianism” of Castelvetro, as a rule, attempted. Both sides swore
fealty to Aristotle, and all debated what Aristotle meant—what Unity
was. And, in spite of the exceptions, this was the condition in which
the question was left to the next century.
The controversy, like that between Caro and Castelvetro, and (I
fear it must be said) like literary controversies in general, did not
pass off without a muddying of the waters. Salviati, Tasso’s chief
adversary, and author of the dialogue L’Infarinato against him, had at
first been a great admirer and almost flatterer of the Gerusalemme,
had offered the author his friendship, had praised his scheme, and
had actually proposed to celebrate it in that very commentary on the
Poetics which Mr Spingarn (who has read it in MS.) describes as
actually devoted to “undermining Tasso’s pretensions.” Exactly by
what personal, or cliquish, or patriotic offences he was induced to
take the opposite line, belongs to the obscure, dull, and disgusting
history of these literary squabbles generally, and we need not
concern ourselves with it. The points “for us” in the whole matter are,
first, that the controversy shows the strong hold which a certain
conception of criticism (whether the right one or not) had obtained of
the Italian mind; and, secondly, that the main question on which it
turned—“What sort of Unity heroic poems must have?”—“In what
manner must the precepts of Aristotle be interpreted and
adjusted?”—shows more than the shadow of coming Neo-
Classicism. The path of safety and truth which Giraldi and Pigna had
opened up many years earlier, and which even Castelvetro,
Unitarian as he was, had been careful to leave open—the path
starting, that is to say, from the positions that Aristotle had not all
literature before him, and that the kinds of literature which he had not
before him could not, therefore, be subject to his dicta—was now
ignored or barred. Apparent diræ facies, the faces of the Unities, and
there is nothing left to do, in the general opinion, but to wrangle
about their exact lineaments.
The critical work of Tasso is far from inconsiderable, and only a
sense of duty prevents the consideration of it here at greater length.
Tasso’s It consists[122] of the Discorsi which, as noted above,
Critical appeared at Venice (with divers Lettere Poetiche) in
writings one of the thin small parchment-covered quartos for
which the student of this literature begins, after a time, to feel a
distinct affection. The much longer and later Discorsi del Poema
Epico partly repeat, partly correct, partly expand, the earlier work;
and sometimes stand in a curious relation to it.[123] But this by no
means exhausts the tale. Tasso, nothing if not conscientious,
appears to have taken his art in general, and his work in particular,
very seriously indeed. He makes extracts from Castelvetro; writes on
the Allegory of his own Gerusalemme, an Apology for it in dialogue,
a formal Reply to the strictures of the Della Cruscans, a tractate in
answer to Patrizzi’s defence of Ariosto, another on Poetical
Differences, a long “Judgment of the Conquistata,” a discourse on
the Art of the Dialogue. Also he has some curious considerations on
three Canzoni of Pigna’s entitled Le Tre Sorelle, written in honour of
Lucrezia Bendidio, and dealing with Sacred and Profane Love.
These considerations have the additional interest of being addressed
to Leonora d’Este, and of breathing a peculiar blend of that half-
sensual, half-Platonic Renaissance rapture of which the great locus
is the discourse assigned to Bembo at the end of Castiglione’s
Courtier, with the religiosity which we more specially think of in
Tasso. He has an elaborate lecture on a single sonnet of La Casa,—
a great favourite of Tasso’s, and deservedly so as far as his serious
poetry goes,—and some minor matter of the kind.
To the writing of this not inconsiderable corpus of criticism Tasso
brought, besides his own genius and the interesting association of
his creative power, really wide reading, and, as has been said, an
and position. indefatigable interest in the subject. He exercised a
good deal of influence in the time to come—both
Milton and Dryden, for instance (the latter again and again), refer to
his critical work. Yet it may perhaps be said without presumption that
this criticism is rather more interesting to a student of Tasso, or to
one who wishes to obtain at famous hands some knowledge of the
Italian sixteenth century ethos in this kind without going any further,
than to the student of criticism itself. Tasso is very fairly
representative of it in its combination of Plato and Aristotle, in its
anxiety to get general notions of poetry and poetic kinds, in its
respect for the ancients, in its ethical turn. But he is rather more
representative than original or distinct; and his criticism is not
perhaps improved by the very natural fact that sometimes avowedly,
and probably in most cases really, it is less a disinterested
consideration of Poetry in general than an apologetic of the poetry of
Torquato Tasso. And as that poetry itself, beautiful as it often is, is
notoriously something of a compromise between the Romantic and
the Classical, so the criticism which is connected with it is
compromising and compromised likewise. Tasso has many
interesting observations, intelligent aperçus, just remarks: he is a
link, and a very early link, in the apostolic succession of those who
have held and taught the great doctrine that poetry makes the
familiar unfamiliar, the accustomed strange and new.[124] But he has
not shaken himself free enough to gain the standpoint of his friendly
antagonist Patrizzi, and to recognise, even imperfectly, that the
secret of poetry is treatment poeticamente, and that only the historic
method unfettered by rules will tell you what poeticamente has been
and is, even thus leaving unknown what it will be.
At about the same time, however, a last, and the most vigorous, if
not altogether the best informed, attempt was made to put the matter
Patrizzi: his on this true historical basis. A year (1586) before the
Poetica. publication of Tasso’s Discorsi, and of his Apologia,
though long after the writing of the first, and not without reference to
himself and the dispute between his partisans and those of Ariosto,
there had been printed at Ferrara, in two parts, one of the most
important and original of the numerous treatises which appeared
during this half-century or more, under the title of Della Poetica. It
was the work of Francesco Patrizzi (as he is generally cited in books,
though both in the title-pages of this work, and in the signature of his
Dedication, it is spelt Patrici). The inspiration of the book was, at
least partly, due to the violent anti-Peripateticism of which Patrizzi
was at this time the twin champion with Bruno;[125] and while we must
no doubt thank this party spirit for being in great part the cause of the
volume, there may be room for objecting that it somewhat obscures
the pure critical value of the treatises. That value, however, remains
great, and would be great even if there were nothing in the book but
an ill-carried-out idea. For its idea is the basing of the inquiry into
poetry, not on a priori discussion of the nature of the thing, and of its
exponent the poet,—not on previous authority as to these questions,
—but on a historical examination of extant poetical composition. It is,
of course, true that an examination of the kind was ready at hand in
Scaliger’s book. But nothing was further from Scaliger’s mind than to
base his inquiry on this: on the contrary, it comes late, and is merely
intended to supply illustration and texts for verbal criticism.
Patrizzi’s plan is quite different. His book consists of two parts or
“decades”—La Deca Istoriale and La Deca Disputata; and though in
some copies (my own is an instance) the cart is perversely put
before the horse, this is evidently a mere stupidity of the binder, due
to the fact that both books, which are separately paged and title-
paged, are of the same year (1586), and perhaps to the other fact
that the Dedication of the Disputata to Don Ferrando Gonzaga,
Signor di Guastalla, is dated, while that of the Istoriale to Lucrezia
d’Este, Duchess of Urbino, is not. But the very first line of the
Disputata makes references to the other as already done.
That the “History of Poetry” of il gran Patricio, as his
commendatory sonneteers love to call him, should be either
completely exhaustive or impeccably methodical, it would be
The Deca unreasonable to expect. There are indeed some
Istoriale. surprising touches,[126] both of knowledge and of
liberality, in his admissions of the Architrenius and the
Anticlaudianus, of Marbod and Bede. But for the most part he
confines himself to classic and scriptural authors; and his notices are
rather those of a classical dictionary maker, or hand-list man, than of
a critical historian in the best sense. Still, all things must have
beginnings; and it is a very great beginning indeed to find the actual
documents of the matter produced and arranged in any orderly
fashion, even if we do begin a little in the air with Giubale and
Giafeto, and end a little in the dark with Gaufredo and Guntero.
Only when he has spent 150 pages on this arrangement does
Patrizzi pass to his Second Book, in which (once more in the true
logical order) he arranges the productions of his poets in kinds, of
which he is a generous and careful distributor. The much shorter
Third deals with the kinds of verses; and the Fourth with the festivals
and spectacles at which poetry was produced, the Fifth continuing
this with special reference to Games and Contests. The Sixth deals
with the singing of ancient poetry; the Seventh with its accompanying
Music; the Eighth with Rhythm; the Ninth with the Chorus; and the
Tenth with the persons who produced ancient poetry—rhapsodists,
priests, actors, &c.
It is, of course, to be observed that all this is strictly limited to
Ancient Poetry; indeed Patrizzi repeats the very words religiously in
The Deca the title of every Book. To support his examination
Disputata. with a further one of modern or even Italian “vulgar”
poetry does not seem to have occurred to him. Perhaps, indeed—
since he refers, as has been said, in the very first line of his second
part to la lunga e faticosa istoria delle cose a poeti, a poemi, e a
poetica spettanti as “condetta a fine” with a sort of sigh of relief—he
may have thought that his readers would not stand it. But it is
noteworthy that in this Decade he constantly cites Italian writers, and
that the last forty pages of his Tenth Book consist of a Trimerone of
controversy with Tasso himself, amicable (they were actually
friends), but by no means unanimated.
The First Book of the Disputata is given up to the cause of poetry,
which Patrizzi, again in accordance with Bruno, decides to be
Enthusiasm (Furori[127]), relying much on Plato, especially on the
Tynnichus passage (v. supra, vol. i. p. 20), and even a little on

(Download PDF) Perception First Form of Mind Tyler Burge 2 Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Download PDF) Perception First Form of Mind Tyler Burge 2 Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Perception: First Form of Mind Tyler

Perception: First Form of Mind Tyler Burge

Perception: First Form of Mind Burge

Leibniz’s naturalized philosophy of mind First Edition.

Pieces of Mind: The Proper Domain of Psychological

A Multisensory Philosophy of Perception Casey

Evaluative Perception Anna Bergqvist

Recognition and Perception of Images Iftikhar B.

Undercover Wolf Paige Tyler

4. Some Basics about Perceptual Systems 109

5. Perceptual Reference Requires Perceptual Attribution 131

10. First-­Formed Perception: Its Richness and Autonomy 367

13. Perceptual-­Level Representation and Categorization 483

Visual Sensory Memory 572

Again, the relevant perceptual-­level species are perception-­guided conation, perceptual-­

The book is resolutely a work in philosophy of science—specifically philosophy of

Animal Eye Grid

3.1. Geisler Contours 85

15.3c. Diagram of model for system of processing of visual executive-­control system’s

CSTM conceptual short-­term memory

Biological Function, Action, Sensing,

Biological Function, Sensing, and Perception 5

Biological Function, Sensing, and Perception 7

More specifically, I focus on perceptual representation. Perceptual representation is

Biological Function, Sensing, and Perception 9

In no way do I deny the import of non-­visual perceptual modalities. Different mo­dal­

Principal Aims of the Book

Five principal aims inform this book.

Principal Aims of the Book 11

representation—representation of size, shape, position, motion, speed, color, surface,

perceptual psychology. Some of these capacities produce intuitively familiar features of

Principal Aims of the Book 13

Fregean Source of Semantical Notions for Perception 15

The Fregean Source of Key Semantical Notions for Perception

Fregean Source of Semantical Notions for Perception 17

…while a sense is what is receptive of sensible [or perceptible] forms without

In Origins of Objectivity, chapter 9, I proposed a rough explication of what perception is.

Perception is sensory, objective representation—paradigmatically by individuals—

I say ‘paradigmatically by individuals’. I doubt that all perception is necessarily attributable

16 Origins of Objectivity, 369, where I write:

Perceptual States as Sensory States

Perception is sensory. It is a certain capacity or competence for discrimination as a result

This is a great help in tackling Castelvetro’s text, the book containing

You might also like

10. First-Formed Perception: Its Richness and Autonomy 367

13. Perceptual-Level Representation and Categorization 483

Again, the relevant perceptual-level species are perception-guided conation, perceptual-

15.3c. Diagram of model for system of processing of visual executive-control system’s

CSTM conceptual short-term memory

In no way do I deny the import of non-visual perceptual modalities. Different modal