U N D E R S TA N D I N G
U N D E R S TA N D I N G

S U N Y s e r i e s i n P h i lo s o ph y
George R. Lucas Jr., editor

R I C H A R D

M A S O N

understanding

understanding

S TAT E U N I V E R S I T Y O F N E W Y O R K P R E S S

Published by

State University of New York Press, Albany
© 2003 State University of New York
All rights reserved
Printed in the United States of America
No part of this book may be used or reproduced in any manner whatsoever without written permission. No part of this book may be stored in a retrieval system or transmitted in any form or by any means including electronic, electrostatic, magnetic tape, mechanical, photocopying, recording, or otherwise without the prior permission in writing of the publisher.
For information, address State University of New York Press,
90 State Street, Suite 700, Albany, NY 12207
Production, Laurie Searl
Marketing, Fran Keneston
Library of Congress Cataloging-in-Publication Data
Mason, Richard, 1948–
Understanding understanding / Richard Mason.
p. cm. — (SUNY series in philosophy)
Includes bibliographical references and index.
ISBN 0-7914-5871-7 (alk. paper) — ISBN 0-7914-5872-5 (pbk. : alk. paper)
1. Comprehension (Theory of knowledge) I. Title. II. Series.
BD181.5.M27 2003
121—dc21
2003042557
10 9 8 7 6 5 4 3 2 1

for Margie

Contents

INTRODUCTION, 1
CHAPTER ONE
WHAT WE UNDERSTAND, 7
CHAPTER TWO
HOW WE UNDERSTAND, 21
CHAPTER THREE
UNDERSTANDING AND KNOWLEDGE, 39
CHAPTER FOUR
INTELLIGIBILITY, 51
CHAPTER FIVE
FAILURES OF UNDERSTANDING, 67
CHAPTER SIX
BEYOND UNDERSTANDING, 89

viii

CONTENTS

CHAPTER SEVEN
WISDOM, 105
NOTES, 115
BIBLIOGRAPHY, 125
INDEX OF NAMES, 131

Introduction

A physicist tries to understand quantum mechanics. A parent tries to understand a child. A critic tries to understand a new style of painting. A historian tries to understand the movement of grain prices in ancient Rome. We may be baffled by other people, by ourselves, by life, by other societies, by the arts, and much more. A desire for understanding has seemed akin to a natural human instinct of curiosity. What we want may not be extra information but something—some form of understanding—that will make sense to us, or for us. What sort of definition or theory could possibly tie all this together? The scope of understanding is so wide that any general, unifying account may seem too ambitious.
This book is an investigation into understanding and how it is to be understood. An interest in understanding goes far beyond philosophy, but the subject should be central to philosophy, both in its origin and in its aims. Plato wrote that a sense of wonder is appropriate for a philosopher: that philosophy has no other foundation, in fact.1 The starting point for the philosopher’s inquiry can be a need for understanding. The aim of the philosopher can be to achieve not more knowledge, but better understanding.
The title of this book is reflexive because the subject is. Anyone presenting a theory about understanding must be aiming to understand it: surely a philosophical task. But there is a need to tread carefully, to avoid begging the question. To set off by trying to define understanding would be a poor start.
Can we assume that a definition—or a theory—offers a route to understanding? The first chapter of the book is a wide but noncommittal survey of the many areas where understanding has some bearing, to give some measure of the subject and its variety. These areas will include people, history, societies, languages, texts, the natural world, religions, and the arts. There is no reason to begin by assuming that any of these should have primacy, or that a model that makes sense for one of them should be applicable to any others. Theorists have been tempted both by diversity and by simplification.
1

2

INTRODUCTION

The second chapter looks at some models of understanding. “Hermeneutics” as a label was meant to cover theories of understanding, but is too unspecific to be more than a signpost. Historically, among its critics, it has provoked justified questions about whether we should be looking at a process, a method, or a definition of understanding. Again, there is some point in setting off with an open mind. There has been a great variety of theories of understanding—almost as many as its potential objects: representational, teleological, linguistic, textual, visual, mystical, scentistic, interpersonal, conceptual, aesthetic, rationalistic, pragmatic, holistic, sub specie æternitatis, and more. On the whole, philosophers have inclined toward simplification, reduction, and order: particularly, following Plato, in terms of ranking “higher” and “lower” forms of understanding. The attraction is obvious: to set up a model of an ideal type of understanding, of which other types may then be portrayed as inadequate attempts. The most prominent example—again, following Plato—has always been mathematical intuition, which has seemed to some mathematicians so certain and so satisfyingly clear that its optimistic extension to other fields has seemed altogether natural. The truth is seen directly. The attraction of simplifying metaphor has been far more pervasive than just Plato’s use of ascent, vision, enlightenment, and liberation. We not only see but grasp, place, and connect. Understanding itself is hard to place without imagery. In the twentieth century a linguistic model of understanding seemed more attractive, both to hermeneutical and analytical writers.
It is not clear what a theory—or, more ambitiously, a “general theory”— of understanding could do. On the one hand it seems natural to hope that something can be learned by thinking about understanding. On the other, the idea of something in common, or an essence, in diverse forms of understanding can seem an antiquated philosophical myth. Once again, questions can be begged. After all, theories need to be understood. It cannot be assumed without circularity that we should look at some concept of understanding, still less the use of the English word “understanding.” Theory-making, or the development of “explanations,” can seem a natural way of producing understanding. It may be, but it is not the only way.
The first two chapters are partly descriptive, sizing up the scale of the subject and what one can expect to be said about it: why it matters. They should also be a warning against simplification. The third chapter moves from precautionary cartography to argument, in considering the priority between knowledge and understanding. Descartes placed knowledge at the head of the mainstream philosophical agenda, where it remained for three hundred years.
An alternative perspective might be to start from understanding. Instead of asking what can be known about understanding, one may ask what can be understood about knowledge. In terms of linguistic understanding—and its complement in the theory of meaning—the initiation of such a reversal has been attributed to Frege. In a wider way it formed part of the project of
Gadamer in Truth and Method.

INTRODUCTION

3

Theories of knowledge in the modern period had an overtly critical function. Their rôle was to provide some touchstone to identify genuine, legitimate knowledge and to exclude superstition or illusion, often in a religious context.
Epistemologists aimed for definitions or accounts that could be used to exclude or repudiate false or inadequate claims to knowledge. Whether or not that project was feasible, the prospects for a critical account of understanding seem extremely poor. In fact understanding seems to be unusually resistant to general theorizing, where a very general theory would be in the form: you cannot understand . . . unless. . . . If this is right, it should be bad news for comprehensive theories about language.
Understanding differs in one other important way from knowledge.
Descartes was able to launch his inquiries by asking himself: What do I know?
The question: What do I understand? seems to lack any comparable interest.
In fact it seems wholly puzzling. Why care? Why might it matter? Knowledge looks as though it might in some sense be mine. What can be understood by me may be of importance to me personally, but it is not easy to see how it could lead to any fruitful philosophical or scientific consequences. Nor is it easy to see how any systematic answers could be given. The slippery nature of understanding as a subject may be one reason why it has received much less attention than knowledge. And yet the fact that it is hard to nail down does not make it unimportant.
Chapter 4 is about intelligibility. Platonic, visual metaphor is compelling: we see with our eyes and understand with our psyche. The seeable is visible and the understandable is intelligible. And what sort of quality is intelligibility? Is it a (primary) property intrinsic to things or events, or is it (secondary) relative to those who understand? Obviously the latter in the most general sense. Even the inscriptions on Voyagers I and II, dispatched into outer space, are supposed to be intelligible to someone or something out there. What matters is what we—whoever we are—can understand. Yet it also seems reasonable to say that one situation is more intelligible than another intrinsically or in itself—apparently meaning intelligible by anyone in general. There must be some link with explanation—explanation in general, not explanation to a particular person or group. But yet again, a notion of intelligibility “in principle” is one that seems tied irremovably to its religious roots: what God could understand, from some absolutely objective standpoint.
Feelings or intuitions about intelligibility seem inconsistent. This may have an historical explanation. It is appealing to contrast an enlightenment ambition to understand the whole of nature (“rationally”) with a romantic feeling for mystery, ineffability, or opacity. In less historical terms, people may feel at the same time that they understand each other well and that they are mysterious to each other (and that this is not a problem). We may want to understand others but might not want to be totally transparent ourselves.
There is also a religious angle in that gods have been held to be intelligible to

4

INTRODUCTION

some degree but unintelligible in others. What we want to be intelligible is not so clear. The hiddenness of some gods has been significant.
Chapter 4 will argue that discussions about “the intelligibility of nature” have something badly wrong with them. It is not evident what might be meant by a suggestion that some or all of nature might be unintelligible. On the other hand, this need not imply some rationalist attribution of an objective property of intelligibility.
The next chapter looks at failures and breakdowns in understanding.
There have been many differing versions of the thought that understanding may be blocked or limited in some way. Philosophical skepticism was a general theory along such lines. Its earliest modern versions rested on the belief that our minds were not made by God to grasp everything (or, more radically, anything) about nature. Such incapacity could have been a consequence of original sin: of a general human failing in contrast with the angelic and the divine. There have been many modernized versions: for example, the idea that the intention behind an utterance or a text can never be entirely reconstructed in a purely objective way. There are other possible barriers: the space between one person and another, for example, might be seen as interestingly fundamental, as might the difference between genders. There is also the perpetually elusive suggestion of relativism, that differing societies or cultures or sects cannot understand something of each others’ ways of life in some radical way.
It is simple enough to see how intelligibility can be used by definition to insulate contexts, cultures, or theoretical frameworks. “They just can’t understand each other” often seems to offer a convincing barrier. And yet the implied relativism appears almost indefensible.
Notions of what cannot be understood are connected in an important way to concepts of possibility and necessity. The basic project of Descartes made use of the idea that there may be ways of understanding that you could not understand, as it were, in principle. An evil genius, whose workings you cannot understand might be subverting your understanding at this very moment.
We need to ask what senses of cannot and might these could be.
Alleged barriers or blockages to understanding raise once again the question of the standard that may be assumed. Someone who tells me that I can never understand another person as I understand myself—as if this is meant to suggest some sort of limitation—has an obviously partial perspective that can be questioned with good reason. Why, for example, not say that I can never understand myself like I can understand other people (or even as they understand me)? What difference is implied by the changed order of priority?
I may never understand another culture as I understand my own, but is that a problem, a failing, or perhaps an advantage?
One special barrier in understanding is provided by the asymmetry of time. Features of understanding noted in the platonic Seventh Letter included its suddenness and its irreversibility. “Now I understand!” would have been a

INTRODUCTION

5

characteristic feeling to Plato the geometer. It is common to see a proof suddenly. Once it has been seen, you can’t see how you could not see it. This may be impressive, as it was to Plato and as it has been to mathematically inclined thinkers more recently; or it might just be a quirk of mathematical understanding that we would do well not to generalize. You do not suddenly understand a foreign language, and that sort of understanding is easily enough forgotten.
Another special barrier to understanding is the subject of chapter 6:
Beyond understanding. A unique failing in understanding would be implied by the notion of being unintelligible in principle. Critically minded thinkers have hoped that some limit can be drawn to understanding, beyond which must lie nonsense or ineffability (both, in the case of the early Wittgenstein). Once more, religious models from the past have had a powerful influence. Job’s problem, he came to see, as he said, was that “I spoke without understanding of things beyond me, which I did not know.”2 As late as the eighteenth century, human understanding may have seemed partial or finite in contrast with the infinite understanding of God. A barrier between the finite and the infinite or the ineffable may have remained attractive even after the religious framework had ceased to be attractive.
The final chapter, Wisdom, looks at understanding as an aim. These days, philosophers, despite the etymology of the title of their subject, tend to be embarrassed by any suggestion that they might be searching for wisdom, still less offering it. On the other hand, philosophy does seem to deal in achieving insights, making connections, attaining clarity, and providing general explanations rather than in (“merely”) acquiring information. This may be a further reflection of a contrast between understanding and knowledge, reframed as an opposition between Geistes- and Naturwissenschaft. But if philosophy is supposed to be about understanding, there seems to be some sense of paradox if philosophers do not theorize with much success about it themselves. One might imagine that there might be some general understanding of what understanding is, how and when it might be attained, what its value was, and so on: but no, these are scarcely to be found.
The rhetoric that surrounds wisdom—depth, proportion, penetration, vision—may sound suspiciously vague to hardboiled thinkers. Yet the thought that there might be only knowledge is also disconcerting: a recollection of positivism. The thought that understanding might be different or even (in some way) better than just knowledge is disconcerting as well. One of the reasons why philosophers have had a lot to say about knowledge is that perhaps a good deal can be said about what it is like, where it comes from, and how to get it.
Understanding, regrettably, is far more elusive. One modern strategy for deferring discussion of wisdom has been the thought that, philosophically, it may be as useful to travel as to arrive. But what is gained along the way, and what would be attained at the destination if we ever reached it? Illumination?

6

INTRODUCTION

Too much was written on methodology in the twentieth century. This may have been the last gasp of a tradition begun three hundred years before, when it seemed to Descartes and his successors that the right method could light the way on the search for truth. That itself was a view about the place and nature of understanding: it had to be methodical to deliver the goods. The way to understand was to follow the approved method: in that case, a geometric, mechanical one. Styles of philosophy defined themselves in terms of their characteristic method: criticism, analysis, linguistic description, hermeneutic investigation, deconstruction. The extent to which practice matched such characterizations was much less clear.
No particular method is adopted or implied in this book. Any study of understanding could (by one definition) be called hermeneutic; but that label has come to be used to cover a specific tradition that cannot be taken for granted. History must not be ignored. It would be foolish to ignore the fact that different models or styles of understanding have seemed appealing at different periods—geometrical in the seventeenth century, aesthetic in the nineteenth century, linguistic in the twentieth—but of course the idea that the understanding of understanding can only be historical is itself from a particular period, presupposing a particular relativism. It may be an unconvincing pretense, but the socratic assumption that we know nothing at all may well be the best starting-point.

Chapter One

What we understand

What is it that we understand, or hope to understand? This first chapter contains an outline map, showing the objects of understanding, not its methods or styles. The listing is not meant to be either exclusive or exhaustive—only to take as wide a view as possible. Some of the territories surveyed may overlap, but it would make a bad start to assume any order of priorities or importance.
Some carry a long history of argument or interpretation. Others have attracted much less interest.
(a) I may feel sure, or not, that I understand myself, though I may be uncertain about what this means. I may believe that my motives and intentions are directly or infallibly accessible to myself, possibly as their owner, and possibly in some unique way. The exact object of my understanding will itself have intricate ramifications. It might be natural for me to assume that this will depend on my philosophy of mind which, presumably, would include my understanding of the nature of people: myself or others.1 “There are some philosophers,” wrote Hume, “who imagine we are every moment intimately conscious of what we call our SELF; that we feel its existence and its continuance in existence; and are certain, beyond the evidence of a demonstration, both of its perfect identity and simplicity”2—and there is no need to take up a position on that. To believe that I understand myself is definitely not to assume that I possess some object called a self that I understand (or that understands itself ).
There must be few areas where historical intuitions have varied so widely.
On one side, my sight of myself could be taken as the most direct, unmediated perception, a benchmark for other types of understanding (or knowledge). This might be taken either as a starting-point or as a desirable ambition. On the
7

8

UNDERSTANDING UNDERSTANDING

other hand, self-understanding might be taken as strenuous or impossible, in a
Socratic or Freudian sense. Again, opinions have varied on whether perfect self-understanding is admirable or not. In one way it might be seen as a form of integrity or wisdom. In another it might be shallowness or simplicity.
So what I have when I have an understanding of myself remains elusive.
It could be framed in narrative terms—a coherent, orderly story about my aims and position—or in terms of vision—a true vision. An overtly linguistic model would seem less promising. To understand myself is surely not the same as understanding some set of statements about myself, at least in some purely linguistic sense. And, as with other objects of understanding, understanding myself would seem to be something that, in some important way, I have to do for myself. It is easy to imagine a sense in which someone else could understand me better than I do myself, and even explain myself to me, to my surprise. Nevertheless, I not only have to understand the explanation for myself, but recognize and understand myself in terms of it.
It would be reasonable to complain that understanding “myself ” sounds overly simple. Understanding my capacities or limitations, physical or intellectual, may be wholly different from understanding my wishes, fears, or dreams. Again, there might be implications for any supposed “structure” of the personality or the self, and the degrees of expected success may be completely different in differing areas. How, and how far, I can understand myself will take me immediately into imagery of transparency and depth, as well as murkier metaphors of levels or structures.
The priority given to self-understanding is important. “I know plainly that I can achieve an easier and more evident perception of my own mind than of anything else”: Descartes started by regarding knowledge as his knowledge.3 Equally, to understand others as well as I understand myself may seem a reasonable aspiration; but it presupposes that I do understand myself to some degree.
(b) Understanding other people shares many of the difficulties in understanding myself. There is the same uncertainty about how far we want our understanding to be successful, or complete. There may be the same conflicting intuitions about how well I understand another person.4 That may be a feature of this personal understanding and of its corollary misunderstandings.
“I thought I understood her until . . . and then I realized. . . .” Everything, it seems, can change suddenly. More than with self-understanding, there can be a temptation to absorb the understanding of other people into a form of judgment, or, still more coldly, into the acceptance of assertions or propositions: “I thought that she was truthful until she told me a brazen lie and then I realized that she was dishonest.” So first I believed (the proposition) that she was truthful and then I believed (the proposition) that she was dishonest? Then there might be some wish to understand the understanding of other people in terms of the acceptance of lists of judgments about them. That could be sen-

WHAT WE UNDERSTAND

9

sible; but so could a pervasive visual interpretation: “Suddenly I saw that she was a liar.” So could an understanding as sympathy, where a particular philosophical brutality is needed to insist on a reductive linguistic interpretation, for example, of: “I understood his grief only when I felt it later myself.”
Understanding myself looks as though it can only be particular or individual. After all, there is only one of me, at least as far as I am concerned.
Understanding other people (and their motives and intentions) may not be the same. It may be that I understand others as instances of whatever they are: members of a family, colleagues, scientists, politicians, strangers. Just as it is plausible that any identification of anything has to imply judging something as whatever it is, so any understanding of a person might imply understanding as a member of some class or category. Or again, it might not. Understanding another person might be as primary or as direct as understanding myself (whatever that means). There is no need to get into any abstract dispute (over knowledge by direct acquaintance against knowledge by description) to accept that a grasp of individuals and their actions may be firmer than agreement on a language to describe them, or a set of concepts by which to judge them. If understanding does imply some framework or context, there arises the further question of whether any such framework has to be inescapably moral, as Charles Taylor has argued.5 Any informative description has to contain language that can never be aridly non-evaluative.
Philosophers who have written about “the problem of other minds” must have been wrestling with a question about understanding as well as knowledge. The “problem” was supposed to be whether “I” could “know” with as much certainty about the contents of other minds as I know about the contents of my own. In the background was some skeptical doubt—never framed with any concern by any historical thinker but nevertheless debated in the twentieth century at length—that such certainty might not be available to me.
Missing from such discussions was the prior question of what I was supposed to expect. Was it some complete insight? Do I want other people to be transparent to me in some way? How likely is this? Also assumed, and rarely discussed, was the order of priority. The “problem” was whether I understood others as well as I understand myself. Few philosophers (Levinas was a striking exception) saw this as odd. Why not ask whether I understand myself as well as I understand others? After all, it is not difficult to find acute perceptiveness about others combined with complete incapacity for self-understanding. Biographies of novelists provide a number of examples.
(c) Understanding people in the past presents different, maybe distinct, questions. This has been one of the classic areas for hermeneutical and other methodological discussions, with well-established battle lines. How far, if at all, is it possible to claim any understanding of past motives, desires, or intentions, especially where any firsthand testimony is past recall? Must we appeal to some version of sympathy, empathy, or identification, or to something

10

UNDERSTANDING UNDERSTANDING

apparently more objective? In terms of the objects of understanding (rather than the means), the issue is less contested. Historical explanations need not look different from explanations in the present. “What were the causes of
Hitler’s anti-Semitism?” need not be seen as different in form from “Why are you angry with me?” or “Why am I angry now?” despite the radical disparity in contents. In one case the evidence to support an explanation is simply in the past, and may always remain controversial. In the others an immediate and full response may be available, but it may not be, and it may be just as controversial. There are also parallels in the uniqueness or generality of understanding. “I could never understand Mozart’s fluency in composition” appeals to a sense in which Mozart was unique, with abilities beyond normal imagination.
This is reasonable, but may not be different in shape from any incapacity to see the abilities of someone in a room with me now.
It seems that one way to understand the past may be to reduce it, in effect, to the present, but this may not be either significant or even interesting.
Alexander the Great may be understood as ruthless or ambitious, where these are our terms from today. You can learn ancient Greek and ancient history and try to understand him through his own contemporary vocabulary, but still, tritely, this will be you now trying to understand him then. It is not clear how removal in time creates obstacles that differ importantly from normal distance in space or degrees of evidence. There is one step of removal in that the conceptual framework of the past may need to be understood, as well as an individual being understood through a framework; but that can happen in the present, too . . .
(d) . . . particularly in the case of understanding other cultures or societies.
In fact, that may well be a characterization of whatever problem may be thought to exist: where a background of explanation for actions or motives is itself in need of explanation. (In social anthropology this used to go under the crude, if readily intelligible, title of The Savage Mind Problem.) Moving further away from immediate personal understanding seems to have a temptingly simplifying effect, where “to understand” starts looking helpfully indistinguishable from “to possess an explanation.” And here, an explanation can become dangerously close to just a satisfying answer. The problem familiar to ethnographers is that the notion of explanation itself may be culturally loaded. Only certain styles of answer seem appropriate to questions like “Why are you killing that goat?”—perhaps answers that match certain standards of intention, supposed rationality or causation. Chapter 5 will look at the view that there can be barriers to understanding of this kind. For the moment, the focus is not on whether or how understanding can be attained, but on its possible objects. As Pierre Clastres wrote: “it is often in the innocence of a half-completed gesture or an unconsciously spoken word that the fleeting singularity of meaning is hidden, the light in which everything takes shape.”6

WHAT WE UNDERSTAND

11

One conclusion might be that if understanding is related to wonder or curiosity, then these in turn must have some connection with unfamiliarity, and this may be haphazardly subjective. The most unbridgeable-looking cultural chasms may be unworrying in practice for those who are culturally or literally bilingual.
(e) A central type of understanding—or rather non-understanding—has always been seen to apply to religions. A set of practices and beliefs can be immunized against understanding within a code of mutually reinforcing terminology and symbolism. Religions may seem to offer archetypal frameworks of rationality and explanation that may be inaccessible to each other. There may even be intimations of obstacles to understanding that may only be surmounted by participation or initiation: credo ut intelligam. It has not always been obvious to thinkers from Christian backgrounds how far such problems may be specific to Christianity, or rather to a religion that has been exhaustively defined in terms of specified and overt tenets that its adherents are supposed to accept. Understanding against such a background can be understood more readily, to some extent, in linguistic or propositional terms. What you do or do not understand may be doctrines that have been formulated with some specific care to exclude misinterpretations or alternatives. These doctrines may embody some element of mystery, but not too much to make them accessible to some degree. This may well be a predicament entirely unique to Christianity, but it is one that has been massively influential on thinking about belief and understanding. Understanding is meant—up to a point—to be modeled on a certain lucidity. Where there is opaqueness its scope is to be defined and contained. Few, if any, other religions embody practices that are buttressed by elaborate systems of explicit beliefs which are meant to serve to some degree as justifying reasons.
It might appear attractive to regard this situation as specifically religious, and to regard the understanding of other kinds of rituals or practices as (merely) anthropological in contrast. But the threat of a vicious circle is evident. There is a dilemma about how far religions can or should be understood in particularly religious terms. The view that they should seems selfdefeating, leading to relativism or ineffability. The view that they should not seems reductionist.
(f ) One of the first fields to interest hermeneutically minded philosophers was law. The interpretation of legal codes and precedents provides a clear, self-contained model for vaguer objects of understanding. It shares with personal and social understanding a reading of hidden or lost intentions and meanings, and also a characteristic indeterminacy about correctness. Not only is a right interpretation often uncertain, but the criteria for deciding and accepting a right interpretation may also be negotiable. Law offers a useful model because there can always be a reasonable presumption that something— and usually something clear and specific—had been intended in the past. It is

12

UNDERSTANDING UNDERSTANDING

usually known who—legislators or judges—meant it. The only problem is to understand or interpret what it was, in a situation where some answer has to be given, for pragmatic reasons. Law is not a field in which a philosophical view about the indeterminacy of meaning could cut much ice. It might well be that all meanings are indeterminate, but courts have to produce rulings or verdicts anyway. This can be so even where interpretations seem to be genuinely endless, as with rabbinical law, where commentaries on commentaries on commentaries are commonplace and where the historical deposit of accumulated understanding is itself recognized as only a foundation for further efforts in the future.
The simplicity of a legal model is tempting. (Gadamer went as far to write that legal hermeneutics offers “the model for the relationship between past and present that we are seeking.”)7 Understanding appears to be almost measurable in terms of practice. So a pragmatic understanding of understanding might take law as a paradigm. A court may reach a view that legislation is so badly drafted as to be senseless, or that precedents are entirely inconsistent, but may still need to conclude a case one way or another, crystallizing an understanding for the time being. There is also the practical notion of an understanding, in an apparently objective sense. The way in which a court reads the law is how it is understood, despite any differences of opinion or sentiment among those in the courtroom. The meaning is unambiguously there, even if overturned immediately by another ruling. It is in no sense subjective. To say that there is a difference in understandings is not to say that lawyers think differently in a subjective sense—which is true though irrelevant—but that different readings may be advocated and accepted. Further, although debates and appeals are the essence of law, it is hard to imagine a legal system that did not contain some procedure for reaching final decisions, however temporarily and however controversially. Cases may be left open or not proven, but these too are specifiable outcomes reached as decisions. Legal understanding must be attainable.
(g) A more popular target in the twentieth century was the understanding of texts, either in one’s own or in another language. Interest in understanding began from the study of the interpretation of the bible as an archetypally controversial historic text (possibly with Augustine’s On Christian Teaching).
Extreme claims can be made in opposite directions. One way, the understanding of (say) pre-Socratic fragments is contentious enough to make anyone accept that a retrieval of an author’s intended meaning can be a hopeless task.
That experience can be generalized to a wider skepticism about understanding.
But, in another way, a written text may be a paradigm of objective clarity. The whole aim of the style of scientific reports is to minimize ambiguities, subjectivities, and cultural distortions, letting the content speak, as it were, as much as possible for itself. Neither extreme is ultimately defensible, though the former proved surprisingly fashionable in fin de siècle literary circles.

WHAT WE UNDERSTAND

13

The understanding of texts may seem a natural model because reading, like seeing, is one of the metaphors that seems almost inescapable.
Wittgenstein’s discussion of “the mental process of understanding” in the
Philosophical Investigations rested on a long interpolation about reading.8
Wittgenstein strove to illustrate the variety of activities that could be covered by the use of the word “read.” The same—he may have meant—might apply to “understand,” including the thought that some, but far from all, types of understanding can be likened to reading. Similarly, he would have been cautious to separate the understanding of texts in one’s first language from the translation of foreign language texts or the deciphering of codes.
Much less cautious about generalization was Gadamer, who wrote that
“Every work of art, not just literature, must be understood like any other text that requires understanding.”9 One of several natural objections might be that a text in itself is not very much, if it is taken to be literally the words on the paper. Even—literally—understanding a text goes so far beyond merely linguistic understanding that it should give some pause to any ambition to make textual understanding basic or paradigmatic. This is a lesson from Quentin Skinner:
The understanding of texts . . . presupposes the grasp of what they were intended to mean and of how that meaning was intended to be taken. To understand a text must at least be to understand both the intention to be understood, and the intention that this intention be understood, which the text as an intended act of communication must have embodied.10

(h) Still more generally, and almost universally in the twentieth century, the central image for understanding has been taken to be the understanding of language, or meanings. This appeared most grandiosely in the claims made by Gadamer, for example, that “Man’s relation to the world is absolutely and fundamentally verbal in nature, and hence intelligible.” This was the premise for his conclusion that hermeneutics is a “universal aspect of philosophy, and not just the methodological basis of the so-called human sciences.”11 Certainly, there seems to be a convenient interrelation between notions of understanding and meaning in language, as with Michael Dummett: the notions of meaning and understanding are very closely related, as is shown by the intuitive equivalence between “to understand A” and “to know what A means,” whether or not, in the latter phrase, the verb “to know” is to be taken seriously. Meaning is correlative to understanding: meaning is, we may say, the object (or, alternatively, the content) of the understanding.

This can be solidified into a basis for reductive theorizing. Insofar as any diverse objects of understanding can be presented in propositional forms (“I understand that . . .”) and reduced to the model of a (linguistic) text, it may

14

UNDERSTANDING UNDERSTANDING

seem plausible to treat linguistic understanding as fundamental.12 Discussions about the relation between thought and language can have the same result.
Naturally, it is possible to limit discussion of understanding explicitly to the understanding of language. Dummett took the view that “a theory of meaning is a theory of understanding” without expressing a converse view.
Even so, his work on the theory of meaning tended to dwell in much greater depth on conditions for language capacity or mastery than on the much vaguer topic of conditions for understanding. At times there seemed to be an assumption that these might be symmetrical: once we are clear about what it is to know the meaning of an expression, then questions about whether, in such-and-such a case, the meaning of a word has changed can be resolved by asking whether someone who understood the word previously has to acquire new knowledge in order to understand it now.13

Yet the view that any particular knowledge is needed to understand, even in a narrow linguistic sense, could be hard to sustain. As Andrew Bowie has written: “there is an asymmetry between the production of grammatically wellformed sequences of words according to rules, and the ability to interpret such sentences in ever new situations without becoming involved in a regress of background conditions.”14
Although language has been widely taken to be a fine exemplar of an object for understanding—not least because rules for its correct use seem temptingly attainable—in some ways it is not an obvious candidate. After all, for most of the time in most circumstances, our own native languages are fairly transparent to us. It is easy to dwell on misunderstandings and misconstruals, but for most of the time we hardly regard understanding ourselves or others as much of a linguistic challenge. Nor do we feel the need for a theory to help us. In fact linguistic misunderstanding or confusion can be perfectly straightforward to sort out, and the necessary form of explanation is of no philosophical interest. The answer to the query “What do you mean by . . . ?” is normally in a familiar, uncontroversial form: “The explanation of meaning can remove every disagreement with regard to meaning. It can clear up misunderstandings.” The understanding here spoken of is a correlate of explanation.
This was Wittgenstein, who also warned against taking the understanding of a second language as a model for the understanding of a first.15 (And an understanding of another language is not a productive topic. If I have never learned Chinese, then I cannot understand it, on the whole. That sort of ignorance must be the most common and least interesting barrier to understanding.)
(i) It is controversial whether the understanding of a language is narrower or wider than the understanding of meaning or significance in some other sense, maybe applied to life or some activities within it. In one direction, it can be argued that the meaningfulness of words is only grasped because we have

WHAT WE UNDERSTAND

15

a wider conception of symbolic or ritual significance. More precisely, it can be argued that a representative or propositional use of language can only work within a wider conception of expressiveness.16 In the other direction, it can be argued that phrases like “the meaning of life” are simply an illicit extension of meaningfulness from language, where it belongs, to life, where it does not.
One of the first lessons for apprentice analytical philosophers used to be that
“what ‘mother’ means to me” should not be confused with “the meaning of
‘mother’”: significance and color (or tone) are supposed be detachable from sense. The elements of meaning that were not thought to be relevant to consistency or inference might be considered to be secondary.17 That could be innocuously circular where the aim was only to start a discussion on logic; less so if the hope was to make logic essential to meaning.
This is an important debate about meaning, but it does not need to be addressed now. In any event, there can scarcely be any doubt that many people at many times, justifiably or not, have sought to find significance in all or part of their lives, and that this search has been framed in terms of understanding. So here is one of the fields for possible understanding. It is typically vague, in that we cannot say what kind of understanding is sought, in what context, or what kind of satisfaction it might bring. There may be an association—some would say an illicit association—with a notion of purpose, where “understanding the significance” might be taken to mean “finding the purpose.” In some cases that might be trivially correct. In others it might be circular or unsatisfying. Someone worried about the meaning of their life might be consoled to be told that it was to glorify God, for example, though this might just as well raise as many problems as it resolves. Philosophers who have been opposed to teleology or final causes may also take a restricting view on what can be understood. Spinoza, for example, would be able to make nothing of questions about the meaning of life or existence. For him, all understanding was to be in the form of narrow causal explanations. Every individual in nature had an individual cause and nature was seen as the cause of itself: so there would be no room for looser forms of understanding or misunderstanding. A wish for some wider account of why things are as they are would just be a kind of superstition, a relic of a view that nature is not all that exists. Again, this attitude could be seen as perverse, in that a kind of generalized bewilderment, or wonder—Plato’s thauma—or an unfocused discomfort in the world, might be seen as the basic motivation behind philosophy and many other types of inquiry. So it might seem paradoxical to define an activity out of existence by insisting that only focused misunderstandings need to be resolved.
(j) There can be parallel uncertainties about objects of moral understanding. In one way, it is obvious that great evils and great goods have been taken to be enigmatic to the point of mystery. Attempts to understand them have not been conspicuous in philosophy. This has been more the territory of religious

16

UNDERSTANDING UNDERSTANDING

thinking or of literature—or both, in the work of Dostoevsky. As with the previous area of unspecific meaning or significance, there can be the strongest temptation to simplify or reduce. On the other hand there can be a persistent suspicion that everything important can be lost by simplification. The way to get an understanding of an appalling massacre may be to catalogue its causes and contexts. This may or may not seem adequate for an understanding of a perceived evil. Any ambivalence may be an echo from some of the other areas of understanding that have been listed. It might even come from some desire for mystery, which comes out in a reluctance to accept prosaic causal explanations: to regard explaining as explaining away, or even as justification. It is not evident what (“exactly”) might be wanted in wanting to understand some monstrous or saintly character or action, or whether satisfaction is really wanted at all. A patient, methodical approach might start from ground level, reflecting on the motivation for any, small-scale moral deed and then (as it were) projecting upwards from there. So understanding morality might be a systematic matter grounded in an account of abstract notions of duty, virtue, or responsibility.
There could be two difficulties. One might be the thought that morality itself—moral character and behavior—should be sui generis, understandable only in its own terms and particularly not reducible to anything else. Another might be that moral theory is workable most of the time but could be inadequate in cases of monstrous wickedness or supererogatory virtue. This is a particularly problematic form of the notion to be discussed in chapters 5 and 6, of a boundary or limit to understanding. The subject, more than most others, leaves a concern about whether understanding or explanation is actually what is required. Suppose, as is quite possible, that some genetic modification was found in all recognized moral monsters and was absent elsewhere. Would this satisfy anyone worried by an understanding of evil? Someone who was teleologically minded might go on to ask why some people and not others had this genetic modification. Then, a further biological explanation might or might not be available; but again, how satisfying might it be? As with the understanding of other people and of ourselves, it may be a matter of temperament or history whether we really want transparency.
(k) Following a narrative is one of the central metaphors that force themselves on anyone thinking about understanding, along with seeing, reading, and grasping. Stories can make up important (and surely immediate) objects for understanding. To follow a plot seems fairly straightforward. Someone who comes out of a theater after a performance of Hamlet and claims not to have understood the play can be helped in various ways. Alternatively, two people can attend the same play and understand completely different stories in it, even without reaching for a further imagery of “levels.”
As with other central metaphors or models, the understanding of stories has led some writers—notably Paul Ricoeur—to reduce other sorts of understanding to the same pattern. Reading a text may be seen as less basic than fol-

WHAT WE UNDERSTAND

17

lowing the narrative within it. Seeing the point of a novel may be understood in narrative terms. Narrative understanding may seem particularly helpful in thinking about significance or purposes. It may sound appealing to say that other models of understanding are grounded in a capacity to grasp the sense of stories, in the same way that linguistic significance may be thought to be rooted in some wider type of symbolism. Broad terms of art such as discourse
(and the French récit) can blur the differences between stories, the language of stories, and the meanings in that language.
The understanding of stories, like the understanding of texts, may need to take some account of intention. And, also in a similar way, opinions have varied extremely. On the one side, an author’s intention may be thought to be unknowable and even irrelevant. This is a conclusion that can be reached easily enough from ancient or mythological stories. On the other side, enormous critical resources may be devoted to authorial biographies and correspondence, presumably with some hope of getting closer to an author’s mind. And again, both positions seem untenable at the extreme.
(l) Far more widely—and too widely to generalize sensibly—the arts have always been seen as an object for understanding. A piece of music or a painting seems significantly different from a sentence or a story as something to be understood. This has always seemed unpromising territory for reductionism.
There have been debates for centuries about whether, or how far, music can or should be regarded (or rather heard) as a language. But the prospects of forcing Bach’s intentions in The Art of the Fugue into the same mold as Shakespeare’s intentions in Hamlet would appear very slight. No one could imagine that the definition of an artistic object is an uncontroversial one, particularly after the efforts of twentieth-century artists to blur any boundaries between art and the rest of life. Yet, equally, no one could doubt that The Art of the Fugue presents a different challenge for understanding from a daily newspaper.18
Kant’s Critique of the Power of Judgment was a work that stemmed from his awareness that the knowledge as analyzed in his first and (he believed) second critiques was not enough. In a small table at the end of his second
Introduction he showed concisely that, for him, art was not to be an object of understanding, but of taste. “Faculty of Cognition” [Erkenntnißvermögen] was lined up with understanding [Verstand], with nature as its object. A “feeling of pleasure and displeasure” [Gefühl der Lust und Unlust] was lined up with judgment, with art as its object.19 This might have been a bureaucratically tidy way to deal with the place of understanding in the arts—there was to be none— but it may also have been too stipulative to be convincing.
(m) Mathematics as an object of understanding has sometimes seemed far too attractive. The immediacy and certainty of geometrical understanding impressed Plato greatly. Mathematics is an area that gives rise to an apparent matching of objects of cognition with styles or methods of cognition. The certainty of knowledge is matched with the certainty of what is known: also a key

18

UNDERSTANDING UNDERSTANDING

point for Plato.20 Understanding and knowledge in mathematics tend to be deceptively close. In some cases, if you understand something that is valid or true, then you also know that it is valid or true.
The obvious trouble is that not much understanding or knowledge shares any of these features. Many objects of understanding are not grasped immediately. They may not suggest any particular matching mode of understanding. Most areas of understanding have nothing much to do with an automatic perception of truth. In the seventeenth century—as maybe for
Plato—the reigning model was geometry. The further platonic tendency, to rank higher and lower forms of cognition, led to an inevitable view in which geometrical cognition came out as highest and purest, with other forms trailing behind. The matching style of understanding for geometry, naturally, was visual. Again as for Plato, the validity of theorems would be seen by the eyes of the mind.
The explosion in mathematics since that time has made it implausible to generalize beyond the vaguest terms about objects of mathematical understanding. Even when “objects” is taken in an ontologically noncommittal sense, the range of what can be understood in mathematical understanding is too wide for tidy summary. More modestly still, some theorists might wish to deny that mathematics has “objects” at all, in any sense. Then presumably, mathematical understanding or intuition would be (“no more than”) a specialized branch of psychology.
(n) Lastly, and most generally of all, nature might be held to be the object of understanding in the natural sciences, or in all human knowledge. In the sense where nature means, loosely, everything, this is barely controversial.
There are some well-trodden areas of argument. Understanding individual facts, events, or regularities within nature may be thought to be different in kind from understanding nature as a whole. The latter may be thought to be impossible or too ill-defined to be worth discussion. The bland-sounding claim that nature is all there is may imply some attitude to the supernatural and the nonnatural. Religious or moral understandings may be excluded from the natural, for example, while the natural and the factual become positively interdefined. The understanding of nature seems to raise in the most general form questions about intelligibility (which will be debated in chapter 4). Can we understand nature because it possesses some (primary) quality of intelligibility? Or is intelligibility, as Kant thought, a property that my mind projects on to the world it experiences? Are these genuine alternatives?
The view that nature can be understood has an important ideological rôle. The view that all of nature can be understood may have been crucial in the development of modern science, as a support and endorsement for (literally) endless research. The view that all of nature can be understood in one standardized way—specifically through the grasp of universal laws—has been moderated since its heyday in the seventeenth century, when it formed the

WHAT WE UNDERSTAND

19

most ambitiously reductionist understanding of understanding: “our approach to the understanding of the nature of things of every kind should be . . . one and the same; namely, through the universal laws and rules of nature.” This declaration by Spinoza ended in the most ambitious terms: “I shall, then, treat of the nature and strength of the emotions, and the mind’s power over them, by the same method as I have used in treating of God and the mind, and I shall consider human actions and emotions just as if it were an investigation into lines, planes, or bodies.”21
This list could be continued to include almost any area of knowledge that presents problems of understanding: economics, chaos, complexity, animal behavior, and many other fields could be imagined. Or a different list might be based on more general features such as structure, form, function, purpose, and so on.
Or there could be many subcategories: the understanding of motives, intentions, expressions, desires, and so on. It might beg some questions to insist that the listing here includes the main types of objects of understanding, but that is a harmless enough assumption for the sake of continuing discussion.
A more tendentious claim would be that some of these categories can be collapsed into some—or even one—of the others in a reductionist fashion. This would be most evident with the extreme example of Spinoza, just cited. For him, all objects of understanding were to be regarded as being like geometry, and were all to be understood in the same way. If that were practicable, this book could end here. Less rigid forms of essentialism may be more appealing.
The understanding of texts and of the use of language were both fashionable in the twentieth century, superseding historical and artistic understanding as hermeneutical favorites from the nineteenth century, but they should be no more convincing as universal stereotypes. It ought to be obvious how an emphasis on one sort of object of understanding can be used to exclude others.
Anyone who starts from a view that the understanding of anything else has to be like an understanding of physics (or hieroglyphics or the language of flowers) is likely to end up with unconvincingly dogmatic conclusions. A quick survey, as in this chapter, should suggest how hard it might be to generalize even about a single model of understanding. It should not be, for example, a model of a single person—or mind—trying to understand something: the “object” of understanding (“outside” the mind). To accept that would be to pitch understanding immediately into a pattern of Cartesian knowledge—essentially “my” knowledge of “the world outside me.” Maybe that is unavoidable at some point, but certainly not as an assumption from the start. When a court seeks an understanding of a statute or a precedent, the judges may, as individuals, be trying to understand, but the correct understanding of understanding would surely not be a subjectivized one. It begs too many questions to take “How is this text to be understood?” as “How am I to understand this text?” The very idea of “objects” of understanding, though harmless when taken loosely, may

20

UNDERSTANDING UNDERSTANDING

imply a pattern of subject-against-object that some might find suspicious. Heidegger, for example, claimed to find philosophical significance in the etymology of the German Gegenstand, as an object standing against a subject.22 Those who follow Hegel in seeing self-understanding as essentially constitutive of the self may have particular difficulties about a model of an understanding subject apart from an object understood.
Tastes have varied over how far it may matter to have such a disparate collection of objects for understanding. The idea that one understands disparate examples of anything by finding something in common to them is itself to take a side in one ancient philosophical dispute, against nominalism. The idea that diversity can only be helpful—Wittgenstein was fond of quoting “I’ll show you differences”—may be to take another side. For the time being, we can just accept diversity as an interesting challenge.

Chapter Two

How we understand

It may seem unnecessary to have a separate chapter on how we understand after a lengthy account of the sorts of things or objects that we understand.
Perhaps it should be unnecessary, but in reality it certainly is not. Aristotle made one of the first and most famous declarations that ways of investigating might vary with the subject matter investigated:
Our discussion will be adequate if it has as much clearness as the subjectmatter admits of; for precision is not to be sought for alike in all discussions, any more than in all the products of the crafts. . . . It is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits: it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician demonstrative proofs.1

As the context of this passage made clear, this was a jibe at Plato’s inclination to straitjacket different types of cognition into a single model: one where the object was matched in clarity and certainty with the method of knowing it.
There would be a scale (“higher” and “lower,” of course). At the top, the best and purest objects would be known in the best and purest way, and so on, downwards. Aristotle may have rejected part, but not all, of that tidy picture.
The top form of knowing was displaced in favor of different styles suitable to their subject matters. But the idea that objects of cognition and methods of cognition might be matched was retained—and amplified, in fact.2 Aristotle’s point was exactly that ethical matters might require ethical understanding, or at any rate something other than mathematical understanding.
21

22

UNDERSTANDING UNDERSTANDING

That thought, taken to an extreme, can provide support for broadly conservative positions. Most plainly, the view that religious belief and practice can only be understood through a specific form of religious understanding has a direct payoff in making religion immune to criticism. Taken literally, that has the drawback of making religion literally unintelligible to the nonreligious.
Michael Oakeshott held the view that politics was not amenable to what he called “rationalism.” By implication, a different—more humane, less harshly unsympathetic—form of understanding was appropriate.3 This made political argument somehow sui generis, dependent on the absorption of a traditional culture and, again, inaccessible to those who did not partake in the appropriate assumptions.
It is an appealing view, that each sort of object for understanding should have its own matching form or method of understanding. Mathematics could be understood mathematically, religions religiously, people personally, history historically, and so on. A further thought—certainly in Oakeshott’s mind, for example—was that trouble comes when the method fails to match the object.
And indeed, as Aristotle said, there can be obvious errors of this kind: ethics seems ill-suited to mathematical understanding (unless demonstrated ordine geometrico, in the seventeenth century). All very tidy, but there are a number of problems. First, the immunization against criticism looks much too easy.
Second, this looks like ad hoc theorizing. Third—maybe the same point—one can ask what sort of theory we assume. It sounds not far from dormitive powers in sleeping potions to assert that history is understood through a particularly historical understanding. (Some people are good at mathematics, but we should be wary about reifying that into the possession of a good faculty of mathematical understanding.) As will be seen in the next chapter, understanding—unlike knowledge—makes a bad subject for critical theorizing. By what right can anyone say that there are correct or incorrect ways of understanding anything? What support could that have? What sort of theory would work? Fourth—the theme of this chapter—there are many cases where a neat match between object and style of understanding is not plausible, and others where the appropriate style of understanding is wholly unclear. Some good physicists have said that quantum mechanics cannot be understood;4 which can only mean that quantum mechanics cannot be understood in one
(expected) way, though it can in others (maybe unclearly grasped). And in some of the most difficult areas of all—in our understanding of each other, for example—it is exactly the method, style, or form of understanding that is baffling. To be told that human beings do (or should) understand each other in a human (or humane) way is to be told very little. What is the contrast, anyway?
A Martian way?
One of the weaknesses of “hermeneutics” as a subject has been a tendency to sweep too much together too simply. All the objects listed in the previous chapter present “hermeneutical problems” and all, if you like, can form the

HOW WE UNDERSTAND

23

subject of hermeneutical methods. That is harmless enough as long as it is only seen as a matter of labeling. Whether there is really anything in common between, say, personal intuition, a grasp of sculptural form, and an understanding of number theory is not obvious. All are described as forms of understanding (in English). All may be embraced by “hermeneutics.” Then, after these naming ceremonies, the real work starts. (A parallel point could be made about an inclination to play up Verstehen and Geisteswissenschaft in contrast with presumed methods in the natural sciences. The distinctions implied by the terminology may be suggestive, but they need more than the terminology to support them.)
This chapter, following the inclusive pattern of the previous one, lists different styles, models, methods, or forms of understanding. The approach is deliberately relaxed. Some of the items listed may overlap, or might be deleted altogether, but at this stage the aim is not to exclude, simplify, or reduce.
The project for the book is to understand understanding, and so it might make a good start to reflect on what is being understood: a mental process, a technique, something that just happens, a social practice, an achievement, an aim, something that is always (or never) possible? Theories or accounts need to show some self-consciousness of circularity. If understanding is said to be like seeing, for example, then it is worth asking if that enables you to see what understanding is itself, and how far that might seem helpful. Conversely, it may be too easy to assume that to understand understanding, or to have a
“theory” about it, is to be able to articulate a definition of it (formerly: an analysis). That would take for granted that understanding something means being able to define it, which, in broad terms, is plainly not so.
(a) The simplest model for understanding, and one of the most pervasive, is visual representation. “I see what you mean” is so natural an expression that the metaphor in it is almost invisible. To understand can be to form a (“mental”) picture, and not only in overtly visual cases. I can say that I understand
The Art of the Fugue when I have followed the score and have seen the lines of counterpoint. I can say that I do not understand quantum mechanics because
I am not able to form a picture of anything that is at the same time a particle and a wave. But no one could imagine that visual imagery could provide a comprehensive account of understanding—as where, in definitional terms, A understands x can be taken in all cases as A can form a picture of x. That sort of literalism is not remotely convincing, though it is much clearer than an apparently metaphorical version, for example, from John Ziman: “When . . . we say that an individual ‘understands’ a non-cognitive entity, we imply that she has established an internal mental structure representing that entity.”5
The real difficulty with visual imagery is in getting away from it. There should be no problem in persuading anyone that it will not do as even a partial theory to “explain” understanding. Despite this, it would be almost impossible to think about understanding without reverting to visual metaphor. As

24

UNDERSTANDING UNDERSTANDING

Plato must have realized, it captures at least two features of understanding that can seem important: immediacy and subjectivity. “I see your point” can be sudden, as in: then I didn’t see it and now I do. The shift from incomprehension to understanding may be like not-seeing followed by seeing. Also, just as no one can see for me, so no one can understand for me. In some way, I have to do it for myself. In platonic terms, you can turn people toward the light, but you cannot make them see.6 Some—but far from all—understanding is irreducibly personal. In a way it has to be something that I do, or achieve. This feature may lead us to think of what Wittgenstein criticized as “mental processes.” (Popper: “the activities or processes covered by the umbrella term
‘understanding’ are subjective or personal experiences.”7) Again, there is the strongest platonic precedent: just as I see with my eyes, so I understand with my mind.
As an “explanation” of understanding, seeing does not take us far. It is all too possible to be able to see a situation and still not to understand. For anyone perplexed by perception it may not take us anywhere: just from one problematic notion to another. As with perception, understanding raises questions about the kind of theory or account that would be satisfying. Not, presumably, some account of “what happens when I understand,” any more than philosophers since the twentieth century have felt capable of giving an account of what happens when I see. (Happens where? In my brain? In my mind?)
Phenomenologically, sudden understanding can feel like suddenly seeing an aspect of something. That might be a starting-point for some kind of neurological account. How far such an account would be a genuine explanation of a feeling is part of a general problem about the philosophy of mind.
Understanding (like seeing) usually, but not always, has a (“intentional”) content. Taking an example of some complicated personal relationship, it is hard to see how “Suddenly I saw what was happening” could have a helpful physicalist explanation in terms of what goes on in my brain. Maybe what happened could be put in those terms (once more in platonic terms, “I woke up”), but the content of what was understood would remain untouched. In his Preface to The Portrait of a Lady, Henry James described a central step when his heroine sees her position:
She sits up, by her dying fire, far into the night, under the spell of recognitions on which she finds the last sharpness suddenly wait. It is a representation simply of her motionlessly seeing, and an attempt withal to make the mere still lucidity of her act as ‘interesting’ as the surprise of a caravan or the identification of a pirate.8

What she saw was a situation that it took the first forty-one chapters of the novel to describe. Any condensation would be a mutilation, not a reduction.
It is easier to see what is wrong with visual imagery than to see how understanding can be portrayed without it. This may be an area where the

HOW WE UNDERSTAND

25

lines between “theory” and metaphor are worryingly blurred. Here is Locke, deep in metaphor: “The understanding, like the eye, whilst it makes us see and perceive all other things, takes no notice of itself; and it requires art and pains to set it at a distance and make it its own object.”9
(b) The other pervasive model for understanding is one of a capacity.
“Speaking ontically,” wrote Heidegger, “we sometimes use the expression ‘to understand something’ to mean ‘being able to handle a thing,’ ‘being up to it,’
‘being able to do something.’”10 Being able to read, speak, and write French may be thought to be not just symptoms (or conditions) of understanding
French, but to constitute the understanding itself, without remainder. This is quite plausible in that it may be unclear what else really matters. Not—
Wittgenstein for one might wish to insist—some inner mental state of understanding French. In the reading of Baker and Hacker, understanding is “akin to an ability.” In linguistic terms, “the question of how it is possible for a person to understand new sentences boils down to the question of how it is possible for a person to be able to do those things which manifest understanding, namely react to, use and explain the meanings of new sentences.” More bluntly, for Ryle, “overt intelligent performances are not clues to the workings of minds; they are those workings.”11
There must be some cases where a capacity to do something would be a necessary condition for understanding. If someone claimed to understand
French but did not respond appropriately to questions in French, showed no signs of being able to speak it, and was unable to write it, there would be wellfounded doubts about the claim. Equally, there must be doubts about how far this could be stretched. It is easy to imagine cases where one might be reluctant to say that a paralyzed patient could not understand signs or language, even though lacking any capacity to respond. Anyone convinced by John
Searle’s Chinese room argument might be reluctant to equate even the most impressive linguistic capacities with an understanding of language.(An efficient translation machine does not understand the material it processes.)12
Anyone struck by visual imagery—as (a) earlier—might wonder how some cases can be forced into the form of capacities. Wittgenstein himself wrote of an understanding that consists in “seeing connections” [Zusammenhänge sehen], and of the “peace of mind” [Beruhigung] that this can bring.13 When I suddenly see a certain personal situation, of risk or betrayal, for example, my capacity to act may seem wholly irrelevant. I may act differently thereafter, or may not; but either way, my subsequent actions seem detachable from my achievement of understanding. This was the case, for example, with the passage from Henry James just mentioned. The relationship between the “mere still lucidity” of his heroine’s “motionlessly seeing” and her subsequent actions had nothing to do with her capacity or disposition to do anything. A similar example, more familiar to philosophers, was given by Iris Murdoch in The Idea of Perfection, where she wrote of freedom as “a function of the progressive

26

UNDERSTANDING UNDERSTANDING

attempt to see a particular object clearly”—to understand—in a case where what “happens” is entirely private, within someone’s mind.14
(c) Once again, a linguistic model for understanding may seem applicable well beyond an understanding of oral or written languages. An understanding of language may seem peculiar as a basis for any comparison, given that it is scarcely self-explanatory itself. Not only is it uncertain how we understand language, but it is unclear what sort of account or theory might be helpful. Theories of meaning at the end of the twentieth century tended to focus on how to make sense, rather than how sense was understood, perhaps assuming, as noted, that these were harmlessly symmetrical. The obvious immediacy and directness of a grasp of your first language can be misleading. It makes an appealing model to absorb an immediate understanding of gestures, music, emotions, and social behavior. Features in the use of language—most trickily metaphor—can be used themselves as metaphors in various types of understanding so naturally that it is almost unnoticed. The idea that religious understanding, for example, is somehow poetical relies on a double image: first, plainly, on the metaphor of poetry for religion; second, and less overtly, on the image of linguistic metaphor itself as a presumed model in understanding.
The big problem is that linguistic understanding, though not at all obscure in one sense—after all, it seems to work straightforwardly nearly all the time—is so elusive in its conditions. There may or not be necessary or sufficient conditions for making sense—this book is not about theories of meaning—but linguistic understanding is as resistant as any other forms to stipulative restrictions. Both holistic and atomistic accounts of meaning may assume that some wider or narrower knowledge is required to make sense. Whether or not that is right, any conditions for understanding sense are far more wayward. It may be possible for me to understand some expression with none of the apparently requisite background. It may be impossible for me to understand another expression despite having all the requisite background. Worse, what applies for me may not apply for you, and there is no good reason why it should. The logician W. E. Johnson remarked in 1929: “If I say that a sentence has meaning for me no one has a right to say it is senseless.” He was talking about Wittgenstein. It is not necessary to believe in private, inner meanings to agree with Johnson.15
(d) Possibly a helpfully vaguer model for understanding can be based on some general idea of interpretation. This may seem useful, in that it sounds as though understanding is being explained to some extent in terms of doing something (namely “interpreting”). It is also valuable in that much understanding does, undeniably, consist of, or contain, some degree of unambiguous interpretation. Most prosaically, I need an interpreter to get me to understand what is said in Chinese. I can be persuaded that the understanding of some texts—ancient, fragmentary ones, for example—is largely or entirely a matter

HOW WE UNDERSTAND

27

of interpretation. But this example reveals the problem. The point of an emphasis on interpretation may be that (“an act of ”) understanding is not, as it were immediate or direct, as might be suggested by a model of reading in one’s own language. Interpretation suggests some intermediary agent or stage between mere awareness and understanding, possibly in a strong form where no pure awareness is considered possible and where all experience is “interpreted” through conceptual or linguistic mediation. Interpretation may also suggest forms of understanding beyond the narrowly linguistic. The point of an extreme view that all understanding is interpretation would have to be that understanding should not be seen (i.e., understood) as analogous to direct seeing, where features of a situation are simply read off reality. Rather, the rôle of the understanding mind or person would have to be more active, itself contributing something ineliminable to what was understood.
Any view that understanding has to be conceptually mediated owes an evident debt to Kantian views of perception and cognition. The strongest view, that understanding could be nothing but interpretation, in the sense of someone’s interpretation, is associated with Nietzschean perspectivism, maybe wrongly. These views will be discussed in chapters 4 and 5.
There is an historical angle, significant in the original development of this subject by Schleiermacher. The position that some or all texts need interpretation—that they do not simply speak to us for themselves—has an important theological dimension. Literalists in all Abrahamic traditions have insisted that direct attention to the pure word will impart the true meaning to the believer’s mind, without need for priestly gloss or sophistication. (Ricoeur has written against “the positivist illusion of a textual objectivity closed in upon itself and wholly independent of the subjectivity of both author and reader.”)16 Others have thought that error is either likely or inevitable without interpretation. Again, at the extreme, there is also the position that any understanding has to be an interpretation mediated by theological or rabbinic history, and the ambition of unmediated understanding is an illusion. These debates, at least in Europe, have been tied to arguments, extending to warfare, about the place of ecclesiastical authority. There have been some curious ironies. The archetypally protestant figures of Kant and Nietzsche (and
Ricoeur) can be most easily linked with a stress on the impossibility of unmediated experience: a strange reversal from the protestant platform of the centuries after the reformation.
As with “hermeneutics” as a general label, “interpretation” may suffer from being too unclearly wide to be informative. Its main use may be as a reminder that some—or all—understanding is an active, not purely passive, process. But it does not get us much further in finding out what the activity is. The implication in the simple metaphor—understanding is like interpreting another language—can be suggestive, but can also be radically misleading.
Understanding one’s own language, to take the most plain counterexample,

28

UNDERSTANDING UNDERSTANDING

may not be at all like understanding or interpreting from another language, and so the metaphor of interpretation must be handled with care.
(e) Scientific understanding may—like several other types listed here—be considered to be separate and special, or to be the basic type to which others can or should be reduced. “Science” may be considered to be coterminous with
“knowledge,” or to be a special compartment of it. The means or styles of scientific understanding may be thought to be systematized more successfully than vaguer forms. The possession of explanations in some specific form— often lawlike—can seem central. There are helpful paradigms. The appearance of comets may appear unintelligible until a set of explanatory laws is formulated. The recurrence of a comet is then understood. A variety of freak phenomena seem to be of no interest until a set of laws of electromagnetism enables them to be understood as interrelated in important ways. In some areas, an understanding of a subject may seem to consist only of a grasp of some laws or rules: we don’t know what quarks are, but we know all the rules of their behavior; this is all there is to understanding them.
Once again, the model is powerful, but possibly misleading. A connection between understanding, explanation, and law can be crystallized into definitions. To understand is to have an explanation. An explanation will be in the form of a law. Science will be lawlike. So a comprehensive set of laws would pave the way to a full understanding (of “nature”). Such understanding—as with “interpretation”—will be mediated rather than direct. If you possess the appropriate apparatus of scientific concepts, then understanding will be available to you.
Everything in the details of this neat story will be controversial for philosophers of science. It ignores, for example, the place of metaphors and models in scientific understanding. More fundamentally, there are problems in the relationship between explanation (of any sort) and understanding. The relationship can be made automatic by stipulation (as in the quotations from
Spinoza at the end of the previous chapter: “Our approach to the understanding of the nature of things of every kind should be . . . one and the same; namely, through the universal laws and rules of nature”). To understand is just to be aware of a cause-or-reason. All causes will be interconnected. The drawback is that one’s relation to any explanation is itself as least as murky as the notion of understanding that it may be supposed to illuminate. One “has,” “is aware of,” “possesses,” or “grasps” an explanation—all of which are roundabout ways of saying that an explanation itself has to be understood. Explanation is not a terminal key to understanding (any more than “seeing”) because although explanations may lead to understanding, they still have to be understood themselves. Worse, any link between understanding and explanation can be seen to be arbitrary. There will be many cases where a situation is understood without any overt explanation—it may be just “seen” or “read”—and where any post hoc articulation seems inadequate.17 The passage quoted from

HOW WE UNDERSTAND

29

Henry James (in [a] earlier) gives an example. It would be absurd to argue that the heroine did not suddenly understand her predicament. Yet she did not understand it through arriving at an explanation except insofar as her entire predicament could be called both an explanation and what she came to understand. She just saw it.
The relationship between understanding and explanation in general must be less tidy than some writers would hope. Von Wright noted that “understanding” has a “psychological ring” which “explanation” does not; also that understanding is “connected with intentionality in a way that explanation is not.”18 A sufficient explanation in any case may be defined as one that gets someone to understand whatever is explained, but the vagueness of that formulation points to the source of the difficulty. Broadly scientific explanation may presuppose a broadly impartial, rational, suitably educated understander for the sake of making any progress at all. Accounts of explanation—including von Wright’s—tend to do this. Yet the problem is very obvious. The clarity or simplicity of some scientific laws will create an impression that an explanation for a phenomenon can be stated (“The only reason why the pressure of this gas in increasing is because its temperature is rising”). This obscures the fact that an explanation has to be related to someone’s understanding, if only in principle. It will always be easier to expand on canons of explanation than on the hazards concealed by this “in principle.”19
The dominance of scientific explanation as a model for understanding has clear historical roots. The French Preface to Descartes’ Principles of Philosophy equated wisdom with knowledge. To know would be to understand. Over the horizon was the prospect that all knowledge would be interconnected. So total understanding would be achieved. If a starting-point for wonder or bafflement is a certain type of curiosity about nature, and if the achievement of a certain type of understanding is what resolves it, then scientific explanations may be directly effective. The defects in this sunny picture may be more striking today than its merits.
On the whole we do know how to answer questions in the general form:
How does this work? It would be foolish to obscure or to complicate this obvious fact because of equally obvious uncertainties about how widely it can be applied. No one is going to agree about how far, if at all, “scientific” forms of explanation are viable in economics or in social anthropology, or what “scientific” means in those contexts. The fact that a central model may be hard to characterize unambiguously does nothing to make it less effective in many important practical cases. Whether anything other than a pragmatic, instrumental account is available is itself a problem in (and for) the philosophy of science.
(f ) Mathematical understanding would seem a strange candidate as a model for any other kind if it had not been adopted and advocated so forcefully by Plato. This is one area where there is little to add to section (m) in the

30

UNDERSTANDING UNDERSTANDING

previous chapter on mathematics as an object for understanding, exactly because Plato felt that objects of cognition and types of cognition should match up. The attractions of mathematical cognition were its immediacy and its infallibility. The excellence of the object guaranteed, and was guaranteed by, the excellence of the intuition. Undiluted Platonism is far from dead, as can be seen from a popular work by Roger Penrose:
When mathematicians communicate, this is made possible by each one having a direct route to truth, the consciousness of each being in a position to perceive mathematical truths directly, through this process of “seeing.” . . . Communication is possible because each is in contact with the same externally existing platonic world.20

One of the appeals of a non-platonic view of mathematics is just that it might not make mathematical understanding sui generis and, by definition, superior. The implication would be that an ability to do and to understand mathematics would be an extension of an ability to do something else; not, anyway, unique. It is not easy to see that anything important is at stake here.
A psychological capacity to be a mathematician may seem unique in that it can be possessed to a high degree by those lacking many other skills, but that is hardly the point. One of the difficulties in extracting any interest is the very diversity of mathematical understanding. An extraordinary capacity for basic mental arithmetic, for example, can be present when any other ability for mathematics (or anything else) is lacking. Conversely, some good mathematicians are hopeless at simple arithmetic.
There seems little chance of retrieving the platonic purity of distinct modes of cognition, each matching distinct subject matters. Apart from any other drawbacks, that would lead to a faculty psychology of bewildering complexity. The worthwhile point could be a negative one. To force all styles of understanding into a single, simple pattern—or even a few patterns—might be to lose something important. The understanding that impressed Penrose in the earlier quotation is the apparently direct “seeing” of the validity of a proof.
Suddenly, with no perceptible external changes, you see what you did not see a moment before—perhaps a train of thought of some intricacy. Without accepting that this shows anything about the truth of platonism—as Penrose hoped—we can at least take the phenomenon at face value, and not try to turn it into something else, by nothing-but reduction.
(g) Much the same could be said of diverse types of aesthetic understanding. Taking only the visual arts, it seems undeniable that people may possess good or bad understandings of proportion, space, color, balance, and so on.
Musicians can understand pitch, rhythm, harmony, proportion, and much more. All this and more can be flattened in an eighteenth-century manner into “taste,” although the point of such an all-embracing reduction must be extremely dubious. Again, it is possible to stipulate, as Kant did, that aesthetic

HOW WE UNDERSTAND

31

discrimination has to operate through “taste” rather than “understanding” (see
(l) in the previous chapter). As a mere stipulation this cannot be questioned, though we can point out its consequences. How helpful is it to say that the spatial judgment of a sculptor or the grasp of time in a composer are to be seen as matters of taste? This seems to rely on an underlying opposition between intellect and emotion that would be hard to sustain beyond the roughest sketch. Kant himself seems to have been aesthetically tone deaf, so this may not have been a concern to him, but a denial of different senses of aesthetic discrimination must be a plain denial of much creative experience. It makes sense to say that you cannot understand how, for example, Schubert achieved some emotional effects with such apparently artlessly naïve writing. There may be nothing in the score to explain it satisfactorily. An interpretation, analysis, or explanation may help, but also it may not. To reduce this kind of understanding to sympathy or taste would seem to be to miss the whole point: that one may experience the appropriate feelings but not understand why, and one can seek to find out.
(h) Equally, moral understanding has often been seen as a prime candidate for reduction. Some thinkers who have refused to accept that moral understanding has particular objects, such as evil, have still believed that it can be separated usefully from other kinds of understanding, from a subjective or phenomenological point of view. Others have sought to show that what looks like a special kind of understanding is really a form of instinct, sensibility, emotional reaction, or culturally conditioned response.
As with mathematical understanding, there is a commonsense position that cannot be taken too far. Clearly there are people who have a less or more well-developed moral sense. To say this much is not to be committed to any general moral theory, merely to note that some people may be mean, selfish or unkind while others may be generous or selfless. To go further might be as untenable as the kind of psychology that posits a mathematical faculty for those who are good at understanding mathematics. (One current aberration is even worse: talk of a moral or mathematical gene that allegedly explains prowess in moral discrimination or in mathematics.)
But this may be an area where the difficulty of general theorizing should not persuade us into ignoring it altogether. A view that there is no such thing as moral understanding is very close to a view that any moral judgment or discrimination is (“only,” “nothing but”) a matter of sensibility. This looks like an acceptance of the challenge in the Kantian dilemma that morality must either be itself or something else.
(i) There might be thought to be a generic, nonspecific kind of understanding that can be seen as the objective of a certain type of education. The language applied to this will be one of grasping, seeing the point, making connections, consolidating.21 The relevant emphasis might be a contrast with information or factual knowledge. The view that understanding as an aim or

32

UNDERSTANDING UNDERSTANDING

process is educationally distinct from knowledge is sometimes associated with claims made by the humanities against the natural sciences. That cannot be right as it stands—except to a hardline positivist—since even the barest factual information has to be understood. And a view of any sort of learning as purely informational must be mistaken. A more plausible line could be that educational understanding can be a matter of reordering or reappraising what may be known already, in contrast with the acquisition of new knowledge. Yet again that seems defensible in commonsense terms, although the distinctions on which it rests might be hard to maintain. A central insight from Plato is that knowledge (information) can be served up to you but you can only understand for yourself.22 This may follow directly from his ruling visual metaphor: I can show you something, but you can only see it for yourself. Or there could be an association with understanding as capacity ([b] earlier). To understand trigonometry might be to be able to solve and explain trigonometrical problems.
Understanding in education may be as elusive as elsewhere, but it has been the object of more reflection than in other areas. Anyone thinking about learning or teaching is bound to reflect on how understanding can be encouraged or stimulated. This will be an important practical question. What sort of art it is—a skill or a type of knowledge, in platonic opposites—remains a matter of debate. Plato himself saw that this debate was linked closely to the place and nature of philosophy: the archetypal area of the humanities for understanding, where new or extra knowledge seems beside the point.
(j) Most unclearly and controversially of all, there is intuitive or even mystical understanding. Taken in a purely subjective sense, it would be very hard to deny persistent reports of feelings of illumination, wholeness, and integration adding up to what might be felt to be enhanced understanding, either of the world as a whole sub specie æternitatis, or of some aspect of life within it.
Such language has been used widely, as any mystical anthology will show. On a less exalted level, many might wish to claim that some form of personal insight or intuition provides a route to understanding that does not fit into any other tidier category, even one of sympathy.
Once more, visual metaphor may seem to apply; though it may be used in a disconcerting way. Visionary writers may have expressed their insights in terms of what they had seen—pictorial narratives or allegories—but in explicit proxy for some other (no doubt “deeper”) form of understanding. There may be intimations of ineffability.
This is an interesting and important area because it raises questions about the limits to our expression and understanding (to be aired in chapter 6).
Some philosophers find it provokingly vague. Few mystical writers can have presented their insights as veridical, factual reports, to be assessed in terms of their empirical truth or falsity. Anyone who thinks that Dante’s account of paradise, or Piers Plowman, was meant as a real travelogue has things badly

HOW WE UNDERSTAND

33

wrong. One of the major problems about understanding is that if someone claims to have understood, or to have understood better, in some circumstances it is hard to argue. This applies with particular force to personal illuminations.
Here is a significant difference from a claim to knowledge, where some objective criterion seems not unreasonable. As with the previous variety of understanding (i), this takes us near to the controversial topic of wisdom. Some would deny that wisdom can be discussed usefully at all. Maybe it is exhaustively reducible to specialized knowledge. Others might base it in some form of understanding that is itself not reducible to classifiable forms of knowing.
This will be the subject of chapter 7.
(k) Finally there is personal sympathy. It would be bizarre to deny its existence, or to say that we are not able to recognize it. Everything else is unclear: whether it is qualitatively different from any of the other sorts of understanding listed, how far it can be applied, whether it is a capacity shared by everybody or only by some. Many would question whether sympathy is to be understood as a variety of understanding at all. It might be seen as an emotional, not intellectual, quality. But that distinction is itself question-begging.
An opposed point of view might be that a great deal of what we consider to be intellectual understanding is really (or even “nothing but”) sympathy, that is, emotion. Historical understanding has been the plainest example.
Understanding as sympathy shares some of the features mentioned already. Most strikingly, the scope for systematic theorizing seems drastically limited. Personal intuition, insight, or sympathy is notoriously wayward: sometimes instant, sometimes absent altogether, even in the same person. It may be too capricious to be of any serious philosophical concern or, at the opposite extreme, it may be of the utmost importance. Could anyone, for example, carry out the thought experiments required by Kant’s various versions of the categorical imperative without some understanding of what it is like to have another person’s perspective? It might be possible to frame a debate where understanding-as-sympathy was argued to be irrelevant to an intellectual interest in understanding or, on the other side, to be absolutely fundamental.
As in the previous chapter, the list could go on.
It could be argued that each species of knowledge—historical, anthropological, economic, and so on indefinitely—requires its own style of understanding. It might then follow either that the subject is too fragmentary to yield any conclusions or, conversely, that some wide theory is needed, to embrace all varieties. In a prosaic sense, someone who is good at economics may be bad at anthropology and in that sense may have a good economic understanding and a bad anthropological one. Taken seriously, that would lead to an untenable psychology of faculties. Various philosophical positions have included or implied views about understanding. A holist may favor wholesale understanding, an atomist piecemeal. A nominalist may stress the place of linguistic

34

UNDERSTANDING UNDERSTANDING

understanding, a realist the place of seeing what is so. A scholastic—or a caricature of one—might stress the importance of classification and definition. A linguistic philosopher might have dwelt on the differences between understanding why, how, and that in English. A neurophilosopher might want to tell us that names, numbers, music, and personal relationships are registered in different areas of the brain.
There is a real puzzle about what sort of account of understanding would satisfy. This may be a typical philosophical puzzlement where we are not quite sure of either our exact problem or of the sort of solution that might work.
The extreme openness of the subject may be one of the reasons why it has received far less direct attention than, say, knowledge, where a straightforward definitional, Socratic account has seemed obviously useful, if elusive. In understanding understanding, it is not clear either what we want to understand or how we think we should understand it. A casual and disorderly list evades any statement of a theory—a “general theory”—of understanding. As noted earlier, a desire for such a theory can be question-begging. The implication would be would that the way to understand something—understanding—would be through formulating or grasping a theory. That path need not be wrong, or harmful, but it should be followed with care.
There are points of technique that are relevant to some extent. Anyone training to be a teacher can learn useful skills to help or speed up students’ understanding. There are tips that can be useful in the understanding of controversial texts, primitive societies, mathematical theorems, and so on. Augustine’s On
Christian Teaching, the foundational work in hermeneutics, began practically:
“There are certain rules for interpreting the scriptures which . . . can usefully be passed on to those with an appetite for such study.”23 Much of Schleiermacher’s work on hermeneutics consisted of advice (“rules”) on how to understand, starting from his ruling paradigm of ancient (and particularly biblical) texts and the
“art” of their explication. His approach was undogmatic and open-minded.
It may be that this is the best we can do. The attraction or temptation in simplification and reduction is obvious, but also misleading. Several of the styles or models of understanding listed in this chapter have struck various writers as “basic,” to which all others can or should be reduced. The understanding of language (linguistic meaning), which interested Schleiermacher so much, could be the most compelling today. The prospect of finite rules, in the sense of grammars, and general background conditions, seems more attainable than in vaguer territories. Yet Schleiermacher himself believed that the “art
[Kunst] of understanding” would depend on the special circumstances of what was understood. This amounted to a denial of a place for a single, unified theory: “Hermeneutics as the art of understanding does not yet exist in a general manner, there are instead only several forms of specific hermeneutics,” and “The prescriptions of the art of understanding are more precisely determinable if they are related to a specific given, from which the special hermeneutics arise.”24

HOW WE UNDERSTAND

35

Psychological facts might have some place, but an “art”—presumably nonreducible—would remain. This must have been correct in that any account of understanding supposed to tell us “what happens” when we understand would be liable to the subversion found most persuasively in Wittgenstein’s
Philosophical Investigations. Understanding as a “process” to be described or
“analyzed” has serious drawbacks: most plainly, that it has to be postulated to
“explain” understanding while then itself becoming problematic:
We are trying to get hold of the mental process of understanding which seems to be hidden behind those coarser and therefore more readily visible accompaniments. But we do not succeed; or, rather, it does not get as far as a real attempt. For even supposing I had found something that happened in all those cases of understanding,—why should it be the understanding? And how can the process of understanding have been hidden, when I said “Now
I understand” because I understood?! And if I say it is hidden—then how do
I know what I have to look for? I am in a muddle.25

One possible response can be to revert to mystery. Kant identified one crucial point in the section on the schematism of pure concepts of the understanding in the first Critique. He was thinking of perceptual recognition rather than understanding, where his problem was one of how general concepts can be made to apply to particular experience, but this is just as relevant for any case of understanding. We read that “The concept of a dog signifies a rule in accordance with which my imagination can specify the shape of a four-footed animal in general, without being restricted to any single particular shape that experience offers me or any possible image that I can exhibit in concreto.” And then still more strangely: “This schematism of our understanding with regard to appearances and their mere form is a hidden art in the depths of the human soul [eine verborgene Kunst in den Tiefen der menschlichen Seele], whose true operations we can divine from nature and lay unveiled before our eyes only with difficulty.”26 The unsatisfactoriness of this passage might well lead us to think that any attempt to characterize a process of recognition or understanding is likely to end badly.
More positively, an account of understanding in terms of imagery or metaphor (rather than process) might seem more productive, especially when combined with some sense of history. Eighteenth-century writers, for example, were particularly drawn to images of visual or pictorial representation. To think, to imagine, and to understand were all to be viewed in terms of seeing.
(Foucault elaborated on this in Les mots et les choses.) Later, pragmatically or instrumentally inclined writers were more struck by imagery of grasping and capacities. A two-part theory about understanding can be glimpsed. First: understanding is to be understood through metaphor. We cannot do without some imagery, whether it is seeing, grasping, connecting, or gathering. Second: such imagery is inescapably historical. The images that appeal to one era

36

UNDERSTANDING UNDERSTANDING

may seem specious to another. Hence, the universality of a mathematicalmechanical model that seemed so convincing to optimistic natural philosophers in the seventeenth century might have come to seem arrogantly narrow to pessimistic post-Darwinians and post-Freudians in the twentieth century.
More extremely: there can be no right images, only persuasive ones at particular times.
Some of this is justifiable. It is remarkably hard to say anything about understanding without resort to metaphor. Visual imagery in particular seems almost unavoidable in trying to pin down the immediacy and suddenness of some forms of understanding, to the extent that it is doubtful whether
“I see what you mean” can be decoded sensibly into any allegedly “literal” version. One problem, evidently, is that no imagery could be pressed far enough to cover everything at any time in history. The mechanical imagery that
Descartes applied so enthusiastically to the whole physical world was explicitly ruled out as a way of understanding the rational soul, since it was “specially created.”27 There is also the difficulty in any form of historical relativism. The claim that understanding can only be understood historically is itself either historical or not. If not, then it is self-contradictory. If so, then why accept it now?
Analytically inclined philosophers might aspire to reach useful results by reflecting on conditions for understanding, or for successful understanding.
Such an approach might avoid the perils of psychologism inherent in any view of understanding as a mental process. The general form of a theory of conditions for understanding might be: A understands “s” if A . . . and If A . . . A understands “s.” The trouble is that any sort of binding or general conditions can be so elusive. It is hard to think of any nontrivial cases where necessary or sufficient conditions would guarantee that understanding is achieved. In linguistic cases, which seem the most helpful, the conditions necessary or sufficient for understanding, say, a Chinese sentence, may be minimal, they may be wholly arbitrary, or they may amount to a complete grasp of Chinese. The prospects for saying anything sensible look slim.28
And yet there is an asymmetry between understanding and misunderstandings. It may be difficult or impossible to make someone understand something. It may be difficult or impossible to determine with finality that something has been understood as it is intended to be. But it can be possible to prevent understanding or (far more significantly) to determine when something has not been understood. There may not be right ways to understand, but there can be wrong ones. If you think that personal sympathy is the best route to an understanding of quantum mechanics, or visual imagination the best route to a grasp of Islam, you are making definite mistakes. Reductive theorists might like to strengthen this tentative point: to say that whole styles of understanding are misplaced. The most well-trodden route is that any specifically personal or intuitive (or even humane) way of understanding other

HOW WE UNDERSTAND

37

people has to be excluded in favor of something allegedly “scientific.” Specifically aesthetic understanding can be denied in favor of “taste” or (mere) emotional response. Historical understanding can be denied in favor of sympathy or empathy (and vice versa). The generalized prospect is of some critical account, with which we can at least rule that some types of understanding are inappropriate. This might bring the philosophy of understanding into some kinship with the theory of knowledge, where critical theorists were concerned not to determine what you can know, but to rule out what you cannot (the classic text: “I had to deny knowledge to make room for faith”).29 An attractive thought, but a hard one to press very far. If, for example, some musicians claim—as they certainly do—Olivier Messiaen was the best-known case—to understand harmony almost entirely in terms of colors (literal colors, not musical tone-colors: synaesthesia), how, if at all, can we begin to argue or disagree? The facts seem beyond dispute. One (strongly visual) form of understanding is being used in a (strongly auditory) context that seems entirely inappropriate. What theory of understanding, positive or negative, could deal with this? Or with mathematicians—Ramanujan was famous for this—who had strong and apparently instinctive feelings for certain numbers? Was this a mistake? On the one hand, it looks as though a theory of knowledge is needed to exclude inappropriate forms of understanding. On the other, it is hard to see how this could work, or what force it might have.
The next chapter looks at the relationship between understanding and knowledge, and between theories of understanding and theories of knowledge. A preliminary view from the surveys in these first two chapters is that understanding is vastly more elusive.

This page intentionally left blank.

Chapter Three

Understanding and knowledge

It is easy enough to set up a debate over the priority of knowledge and understanding. On the one side, what you know must come first. What you understand can be seen as part of what you know. More radically, you can always ask: What do I need to know in order to understand? On the other side, you have to understand knowledge before considering what you know. So understanding trumps knowledge.
Jumping to a linguistic level: on one side, meaning must take precedence before truth. Until you understand what is meant you cannot know whether it is true. But then again, maybe some knowledge is needed to understand a language—knowledge of some assumed theory of meaning, for example; or, more simply, knowledge of the language itself.
Michael Dummett’s assessment of Frege’s place in the history of philosophy gave precedence to meaning over knowledge:
Before Descartes, it can hardly be said that any one part of philosophy was recognized as being . . . fundamental to all the rest: the Cartesian revolution consisted in giving this role to the theory of knowledge. . . .
Frege’s basic achievement lay in the fact that he totally ignored the
Cartesian tradition, and was able, posthumously, to impose his different perspective on other philosophers of the analytical tradition. . . .
For Frege the first task, in any philosophical enquiry, is the analysis of meanings. . . .

But then, it emerged, an understanding of meanings requires a tacit knowledge of a theory of meaning. So, since “meaning is correlative to understanding,” knowledge required understanding required knowledge.1
39

40

UNDERSTANDING UNDERSTANDING

There have been other versions of such debates. In the seventeenth century, the insistence of Descartes that subjective knowledge should be the first item on the metaphysical agenda was answered by Spinoza’s case that human knowledge had to be understood in a context of a prior grasp of human nature and, before that, of nature as a whole. In the twentieth century, assertions of the universality of hermeneutics were answered by reminders that historical understanding requires contexts of presumed knowledge. There were also abstract methodological disputes about the precedence between “scientific” knowledge and hermeneutical understanding. A political context hung in the background. Epistemology could pose as radical, critical, or subversive, in contrast with historically conservative hermeneutics. Questions about the priority of knowledge against understanding could be recast as questions about the priority—and hence, by implication, the superiority—of philosophical theories of knowledge or of understanding.
All this matters because the theory of knowledge has been a huge endeavor in Western philosophy. Thought about understanding may seem, in comparison, haphazard and less productive. But if understanding has a good claim to the philosophical high ground, history may need to be revisited.
Epistemology was the subject of some skepticism in the last decades of the twentieth century.2 To some, hermeneutics offered a more promising route.
How genuine were the alternatives?
An argument of this chapter is that questions about the priority of understanding only arise, and only make much sense, in contrast with past claims made for knowledge. There would be no need to care about priority if it were not for the established Cartesian claim to first philosophy. A need— or apparent need—for a theory or account of understanding has to be seen against the background of past desires for a theory of knowledge. Expectations for a theory of understanding may be shaped by past hopes for a theory of knowledge, both in terms of its intentions and its form. But understanding is not like knowledge. Whether or not a viable theory of knowledge is, or ever has been, possible, an account of understanding modeled on similar ambitions would be misguided.
In the most general way: What was the point of trying to think about knowledge? The first of Descartes’ Rules for the Direction of the Mind—his earliest philosophical work—said that “The aim of our studies should be to direct the mind with a view to forming true and sound judgements about whatever comes before it.” The basic question for Descartes was: Which beliefs should
I accept? They were seen propositionally, in terms of their explicit content.
Among the possible beliefs on offer, some were acceptable, others not. His hope was to maximize his acceptable beliefs by the use of discriminatory rules or methods. Tests would be available to sort between acceptable and unacceptable beliefs. The upshot would be knowledge: the possession of truth. The

UNDERSTANDING AND KNOWLEDGE

41

sciences, which were “nothing other than human wisdom,” would “consist wholly in knowledge acquired by the mind.”3
In the background to the search for knowledge or truth was Christianity, a religion defined in terms of explicit propositions that its adherents were meant to accept. The distinction between acceptable and unacceptable propositions was of the utmost significance. It mattered a great deal to the individual soul to subscribe to the correct propositions, as can be seen from the Tridentine Profession of 1564, commended in the bull Injunctum nobis of Pius IV:
I accept and profess, without doubting, the traditions, definitions and declarations of the sacred Canons and Oecumenical Councils and especially those of the holy Council of Trent; and at the same time I condemn, reject and anathematize all things contrary thereto, and all heresies condemned, rejected and anathematized by the Church. This true Catholic Faith (without which no one can be in a state of salvation), which at this time I of my own will profess and truly hold, I, . . . , vow and swear, God helping me, most constantly to keep and confess entire and undefiled to my life’s last breath.

Here are some common features shared by this large religious context and by theorizing about belief and knowledge in the footsteps of Descartes:
(a) Belief—and hence knowledge—was viewed in terms of content: propositions believed. The Council of Trent produced what must have been the most ambitious attempt to nail down exactly what believers should—and should not—believe. The idea was that if only the correct propositions could be listed and accepted, then all would be well. Descartes took a coldly reductionist view of wisdom and knowledge. Wisdom was nothing but knowledge.
Knowledge was nothing but the assembly and accumulation of truths.
(b) Beliefs mattered. For the Catholic believer (and also the Protestant, in a way which is not relevantly different here), assent to the correct propositions was vital to salvation. The definition of the faith was propositional, in the sense of what was believed. Getting the definition right was overwhelmingly important. Defending it could be a matter of life or death, to the point of martyrdom and war. In philosophical terms, beliefs might be held to have intrinsic connections with action and with knowledge. Your beliefs affected and would be affected by your actions. Part of the point of believing correctly would be to enhance the possibility of acting correctly.
(c) Beliefs had relevant support. A difference between a belief and a taste would be that it would always be possible to ask why a belief was held. The reason would not necessarily be a matter of proof or even evidence; but there would be a crucial distinction between the relevant support and the cause of a belief. In Christian terms, relevant support was meant to be noncausal. Even in the case of the most forceful direct evidence—an immediate, personal vision—it would be the assent or acceptance of the believer that counted.
Because belief had to be to some (extremely controversial) extent voluntary (“I

42

UNDERSTANDING UNDERSTANDING

of my own will profess and truly hold,” earlier), it had to be a matter of choice, distinct from causality. Hume wrote that “A wise man . . . proportions his belief to the evidence.”4 Even when writing against Christianity, he retained an attachment to appropriate support. Although he thought that many beliefs were a matter of animal instinct, the proportioning of belief to evidence by the wise man was surely voluntary. Otherwise, where was his wisdom?
(d) The support for beliefs provided them with legitimation. Appropriate proof, support, or evidence led to appropriate beliefs—or, in some circumstances, knowledge. The means of legitimation were wildly varied. The shortest route might be a crude appeal to authority, but by 1600 this impressed only bigots. The search was afoot for something better, and it went on through the correct method (for Descartes himself ), the appropriate origin (for empiricists), or the right legalistic “birth certificate” to show a “pure and lawful pedigree”(for Kant).5 Legitimation could be genetic (in terms of origins), foundationalist, systematic, pragmatic, or some cocktail of any of these. Because beliefs mattered, why they were held was a question of great importance. In
Christian terms, there was a long quest for evidence or proof through philosophical theology and, latterly, the philosophy of religion. Western Christianity developed a uniquely elaborate body of defensive and apologetic argument.
In philosophical epistemology, the search was for relevant justification that might legitimize true beliefs as knowledge.
(e) The formulation of creeds was an intrinsically critical activity. The whole point, historically, was to draw lines between orthodoxy and heresy. The intention, from earliest times, had been to define what the believer should, and could not, believe. The Tridentine Profession was only the most aggressive manifestation. Positive and negative beliefs were interdefined to the point of inseparability. In parallel philosophical terms, it is hard to see what could have been the purpose of epistemology if it was not meant to be critical. An account or definition of knowledge might have some intrinsic interest to some, but its only point or value could come from its use in sifting between genuine and spurious beliefs or knowledge. Just as the difference between true religion and superstition might be said to lie in the authority of witness or evidence, so the difference between astrology and astronomy might lie in their sources in illegitimate or legitimate methods of inquiry, or in the force of their predictive powers.
(f ) “I of my own will profess and truly hold, I, . . . , vow and swear, God helping me”: this kind of belief was necessarily first-person, subjective. What mattered was what each person chose to believe (with God’s help: a touchy point theologically, but not relevant now). Belief was not communal, or, if it was, it would not help the individual nonbeliever. Your beliefs were essentially yours, and this was important. Equally, and famously, Cartesian epistemology was egocentric. It started from me, as the searcher for truth. The whole point of the method was to enable me to ask, of any belief: Why should I accept

UNDERSTANDING AND KNOWLEDGE

43

that? The opening sections of Part I of Descartes’ Principles of Philosophy were cast, uneasily, in the first-person plural: “Since we began life as infants . . . the only way of freeing ourselves from these opinions . . . once we have doubted these things” and so on. But very soon there had to be a lurch into the singular to make the opening arguments at all convincing: “it is possible for me to think I am seeing or walking, though my eyes are closed . . . it may perhaps be the case that I judge I am touching the earth even though the earth does not exist at all” and so on.6 The emphasis was never on what we, as a community, should find acceptable, but on what I, the searcher for truth, should accept.
The reasons for this were themselves egocentric: the community, like the earth, could be imagined away by the determined doubter. With rare exceptions, this continued to be the tenor of modern thought about knowledge.
Historically, modern epistemology did in fact originate from overwhelmingly
Christian imperatives. The dedicatory letter to Descartes’ Meditations explained the overtly theological intentions of the book. In the work of Locke there was a need to find a path between the external authority of the Roman
Church and the inner light of enthusiasm. In Hume, an entirely Christian framework of attitudes toward belief and its importance was directed against the content of Christian beliefs.
It would make a tall order to show that postreformation Christianity engendered modern epistemology. It would be rash (and is not necessary) to present structural similarities as positive causal links. But it is easier to show how an absence of interest in Christianity can go along with something that looks quite different from most modern epistemology. It can be illuminating to see how a philosopher wholly untroubled by either Christian or anti-Christian concerns could take an entirely different view on the place of knowledge and the theory of knowledge. This can be more telling than generalized historical argument about the relationship between religion and philosophy, or than Weberian speculation on the contexts for an intellectual position.
Spinoza offers a unique example of an important modern philosopher who had no interest in Christian priorities. For him, religious belief was a matter of practice or observance—what he called “obedience”—not propositional content. Natural theology, in the sense of a system of views about the origin, nature, and workings of the universe, was identified entirely with natural science (“philosophy”) as an object of certain knowledge. Religion was entirely different, a matter of “history and language.” It was not defined propositionally, by what was believed, but by a way of life. Its legitimation was to be not philosophical but moral. Radically, for a long list of theological tenets: “on these questions it matters not what beliefs a man holds,” since
“faith demands piety rather than truth.” The only constraint on beliefs was craftily negative:

44

UNDERSTANDING UNDERSTANDING

faith requires not so much true dogmas as pious dogmas, that is, such as move the heart to obedience; and this is so even if many of those beliefs contain not a shadow of truth, provided that he who adheres to them knows not that they are false. If he knew that they were false, he would necessarily be a rebel, for how could it be that one who seeks to love justice and obey God should worship as divine what he knows to be alien to the divine nature?7

That is, you can believe any propositions you like so long as you do not know that they are untrue. Further, a distinction between a reason and a cause for any beliefs was repudiated. In religious terms this might have had some connection with Spinoza’s Jewish background. (To be Jewish—at least on some interpretations—a causal qualification can be adequate: having a Jewish mother, or being brought up in a community observing Jewish law or custom.)
A propositional account is not needed either for definition or for defense. A creed is not required. In philosophical terms, Spinoza’s determinism would have kept him from a workable distinction between reasons and causes for any beliefs. What I feel to be a freely chosen reason with which I support one of my convictions will have had some cause, possibly unknown to me. Apologetics, still less missionary conversion, will not be an issue.
Spinoza’s understanding of knowledge was not egocentric. He repudiated the starting-point of Descartes in the isolated self, probably regarding the machinery of the method of doubt as unsound. Knowledge was not the first issue. Knowing was an activity of human beings, who were part of nature.
Until nature and the place of human beings in it had been investigated, nothing could be assumed about knowledge. The question could never be interestingly “What do I know?” It might be, on any subject, “What is known?”
(Wittgenstein was to note in On Certainty, “It would be correct to say: ‘I believe . . .’ has subjective truth; but ‘I know . . .’ not”; later, that “There is something universal here; not just something personal.”)8
It would be an exaggeration to think that Spinoza simply relegated the question “What do I know?” to a subordinate status, below an understanding of a wider nature. The theory of knowledge was not demoted as much as relocated. It was not needed in the place that it had occupied in Christian philosophy, as part of the apparatus used to legitimize a definitional set of propositions. A combative, credally-based religion needs a theory of knowledge that can offer endorsement or legitimation to the choice of favored propositions.
An opponent of such a religion may want to subvert that endorsement. A thinker from a quite different tradition may not care one way or the other.
A theory of knowledge was supposed to be universal—it was not just for me as an individual, or a society, or for a particular discipline. The Cartesian
“I” was not René Descartes in person, but anyone (critics have suggested any male mind) who was prepared to seek the truth. Christianity was supposed to be a religion capable of saving any true believer, universally. Spinoza’s view of religions was not like that. They were specific to the societies in which they

UNDERSTANDING AND KNOWLEDGE

45

had developed. Religious beliefs were not absolute truths but were to be understood as part of a social fabric. He caused great offense by portraying
Judaism in this way.
A theory of knowledge in the Cartesian mold provides a poor model for any account of understanding, to the extent that any comparison is difficult. In classic analytical form, the hope might be to find conditions that will fit into the format A understands “p” [or that p] if and only if . . . just as epistemologists tried to fill the gap in A knows [that] p if and only if. . . . The assumptions would be that understanding, like knowledge, will be personal—it belongs to the individual, A; and that it will be propositional—of proposition p. For the epistemologist, following the hints in the Theætetus, part of the gap in A knows
[that] p if and only if . . . may be filled by the relevant reasons that A has for a true belief in p. These reasons will provide A’s justification for a claim to know
[that] p, and also the legitimation for p in a set of beliefs. They will be used as a critical tool to discriminate between legitimate and illegitimate beliefs. (For a conventional empiricist, a proposition grounded in possible experience will seem legitimized. For Descartes, at least as he avowed piously in the last proposition of the Principles, legitimacy was conferred by “the authority of the
Catholic Church and the judgement of those wiser than myself.”) There were good religious and philosophical reasons why theories of knowledge should look like this. The need at the most general level, after all, was to help me to decide which propositions I should accept against a context where this should be of great importance to me.
But none of this either applies or works with understanding, just as it did not even work with knowledge for Spinoza (or for many philosophers in the ancient world). There is no reason why questions about understanding should start from the individual self. Some of the examples in the two previous chapters show this. When a court seeks an understanding of a statute, it could only be a distortion to say that what really happens is that the judges on the bench are each, individually, trying to understand. The understanding is necessarily not subjective. Where a linguist is seeking to understand some undeciphered symbols, what is sought is not a personal, subjective grasp, but an understanding, for all future readers. There should be no reason to insist that such cases should be reduced to first-person versions.
In the same way, there is no reason to insist that understanding has to be propositional, either primarily or essentially: although of course propositions or meanings are undeniable objects of some varieties of understanding. Seeing a situation—an almost paradigmatic image for understanding—can only be forced into a propositional straitjacket with great implausibility.
More significantly, legitimation need not be an issue. Conditions for understanding, as argued already, can only have the most doubtful status. The notion of a relevant justification seems completely beside the point. Understanding,

46

UNDERSTANDING UNDERSTANDING

after all, is known for its waywardness. Someone can fail to understand a predicament despite apparently overwhelming evidence or justification and then suddenly understand everything in an instant. There is no arguing with this. There are no reasons to insist that understanding should be sufficiently grounded to qualify. If someone shows a surprising understanding of Japanese, of counterpoint or of prime number theory despite an apparently sketchy education, this may be curious but is not altogether remarkable. For several centuries it seemed plausible to ask where an item of knowledge originated, as though that might help with legitimation or endorsement. (Both textbook empiricism and textbook rationalism fitted that pattern.) For understanding that may be pertinent with some cases but with others it will not. The exact source of a person’s sympathetic intuition for the feelings of others is of no relevance to its value or validity. There are ways of denying this, most plainly by narrowing down the types of understanding that one is willing to accept. A dogmatic behaviorist, for example, who maintains that understanding is to be understood solely in terms of overt capacities, may wish to insist that someone cannot understand something unless certain observable tests are passed. But why swallow that?
It follows that a theory or account of understanding, unlike a theory of knowledge, could not be used critically. This must be historically important in that some thinkers—notably Kant and the early Wittgenstein—felt that the function of philosophy, in whole or in part, was to be critical. From the seventeenth century, the religious function of epistemology was to find a location for Christian propositional beliefs alongside, above, or outside other accepted beliefs. The inherited secular function of epistemology must have been to identify types of support, reasons, or justification that can be used to discriminate between reliable and untrustworthy beliefs. In many cases
(though not all) if I say I understand, no further legitimation is either possible or necessary. There are limited, stipulative, circumstances in which conditions for understanding may apply (“A candidate shall not be considered as having an adequate understanding of electrical engineering unless . . .”); but these will be convincing mainly where knowledge and understanding are almost interchangeable. (It is also possible to imagine: “I passed my examination in electrical engineering although I never felt that I really understood the subject at all.”)
The universality of classical modern epistemology may have been ambitious, even on its own favorite territories. The aim was to nail down knowledge as such, not local forms of knowledge in physics, history, or economics.
In contrast, for Spinoza, knowledge would be universal; beliefs would be local, and the connection between the two would be less definite than in the work of Christian philosophers. He relied on a sharp break between knowledge—for him, natural science—and beliefs, both religious and secular. It is not evident that thought about understanding could be universalist, even if

UNDERSTANDING AND KNOWLEDGE

47

that were desirable. A set of rules or conditions by or through which anyone could understand anything would make a puzzling ambition. What could be the point?
“How do I know . . . ?” both embodied the historical, Cartesian claim to priority for knowledge and was itself a central element in the Cartesian theory of knowledge. There could be no gain in trying to outbid that question by asking “How do I understand . . . ?,” for the reasons that should have been plain from this chapter.
There is no reason for a veto on the question “How do I understand?,” but there are good reasons to believe that attempts to answer it will not lead to a valuable “theory” of understanding. A model that proved productive for knowledge is likely to be less productive for understanding; though in fact one difficulty is created in both areas. “How do I . . . ?” leaves an ambiguity over the kind of answer required. Twentieth-century writers tended to evade the hazards of psychologism by turning to “conditions”: What were the conditions for the use of “A knows that . . . ?,” although that left entirely unclear whose conditions these might be, or where they came from, or why they should matter. Asking: How do I understand? produces the same ambiguity. What do I expect as a reply? Neurology? Conditions for the use of a concept or of an
English word? What could be done with a reply anyway?
The epistemologist starts from “How do I know?” because Cartesian doubt placed the self first. There are no parallel considerations to promote
“How do I understand?” either as the first question to be asked about understanding or as the first question on a wider philosophical agenda. Cartesian doubt does not work, and will not apply, for understanding. (Maybe rather the reverse. Wittgenstein’s private language arguments could be seen as an attempt to show that I cannot even understand my own discourse with myself unless it is grounded appropriately in a discourse shared with others.) This is a more radical point than it might seem at first sight. How, or whether, I understand, literally as an individual, can be a question in some contexts— educational, for example: everyone else in the class understands this formula, why don’t I?—but there can be no reason to imagine why these should be typical, or even interesting. If you bought a book called Understanding Relativity, you would be very surprised to find that it was meant to explain relativity to any literate reader, but not, as it happened, to you. The double entendre in the
Cartesian “I” might work for knowledge, but it is ridiculously implausible for understanding. Descartes sitting by his stove worrying about what he understood would be worse than unpersuasive.
God’s understanding, like God’s knowledge, may have been assumed to be universal and absolute. Yet the universalization that seems plausible for human knowledge sounds absurd for understanding. Achieving an impartial, objective
“absolute” understanding can be vital in some contexts—in the natural sciences,

48

UNDERSTANDING UNDERSTANDING

obviously—but there are other areas where it could only be a delusion. This would apply not only in the disputed areas of Geisteswissenschaft in the social sciences, but in plain cases of personal understanding. A nineteenth-century realist writer might create a God’s-eye portrayal of a marital argument, but in a real-life situation the idea that an objective understanding can be attained is an illusion that could be accepted only by someone who had spent too long in counseling or analysis. But on the other hand, it is possible to overplay this point. In Being and
Time, Heidegger made heavy weather of self-understanding, apparently believing that it had to be prior to other types of understanding—“Da-sein is a being that does not simply occur among other beings. Rather it is ontically distinguished by the fact that in its being this being is concerned about its very being.
Thus it is constitutive of the being of Da-sein to have, in its very being, a relation of being to this being”—and so on. He made some telling points against
Descartes. It should be necessary to understand sum before understanding cogito, for example.9 But he said nothing to demonstrate that self-understanding has to take special precedence. The idea that I have to understand myself before I can understand anything else is nonsense. If Heidegger thought that (which seems unlikely, in those bald terms) he could only have been trying to put understanding into the place given by Descartes to knowledge.
The upshot of all this is that generalized arguments about priority between understanding and knowledge, or between the theory of understanding and the theory of knowledge, would be misplaced. That applies as well to the specific case of such arguments that held some appeal in the twentieth century, on the priority between truth and meaning. You cannot say whether a sentence is true until you understand what it means. So understanding looks prior to knowledge. But then to understand what a sentence means you require some knowledge (minimally, of the language in which it is expressed).
So knowledge precedes understanding (a condensed caricature of the view from Dummett at the beginning of this chapter). To be able to understand an expression, tacit knowledge of a theory of meaning was required. But there is no reason to model understanding in that way upon knowledge; and in particular no reason to suppose that understanding needs “conditions” of a kind that would mirror those in a theory of knowledge.
So far the case in this chapter has been largely negative. Genuine contrasts in priority between understanding and knowledge will not get far. Priority is a concept derived from, and dependent on, a theory of knowledge, unsuited to thought about understanding. More speculatively, the very idea of what looks like a philosophical theory may derive from the pattern of the theory of knowledge. Whatever a theory of understanding might be, it would not be like that.
So why raise the subject of priority at all? One answer is that the negative point is worth making. Study of understanding can be ignored or deferred

UNDERSTANDING AND KNOWLEDGE

49

on the grounds that knowledge has to be handled first. The postponement can be indefinite. This has to be seen as a mistake.
A more positive view comes from looking at the range of subjects that may be covered by a treatment of understanding. Taking only a selection of questions from the two previous chapters: How thoroughly do I understand myself? Is it possible to understand people in the past? Are there barriers of understanding between societies or religions? Can understanding be timeless or must it be specific to a time or place? Can there be an impartial understanding of an author’s meanings? Is mathematical understanding special, sui generis? and if so, what follows? How far is scientific understanding to be taken as typical? Of these questions, only the last can be reworked plausibly into a question about knowledge. The remainder are not so much prior to questions about knowledge as unrelated to them.
And yet there are some areas where issues about understanding do seem to occupy the philosophical high ground. That goes for the extreme cases to be examined in the next chapters, where an appeal to unintelligibility blocks any further discussion. If the social practices of your tribe are unintelligible to mine, then a debate between us on what our practices presuppose about our beliefs will be impossible. If I cannot understand your religious convictions, then I will not understand how to question your assent to them. Simply, if I cannot understand what you are saying then I cannot know whether it is true.
This may be taken in a plain sense—if you speak Tibetan I will not understand you—or in some specialized way, where we speak the same language but where I insist that I cannot understand what you say because of theoretical reservations about meaningfulness.
The scope of the subject is so wide that anything like a “general theory of understanding” would surely be a fantasy: another respect in which understanding and knowledge are not helpfully commensurable. An emphasis on only one style or form of understanding—linguistic or textual understanding would be obvious candidates—might create an impression of generality, but this should hardly be convincing. Yet a pessimism about general theorizing need neither open the way for claims for piecemeal, uncritical ad hoc understanding nor exclude specific cases of plain unintelligibility. (It does not follow that anything goes just because it is hard to construct a theory of anything.) A more positive conclusion might be that liberation from a model of understanding based on epistemology might be beneficial. Instead of looking for a fundamental, reductionist theory of understanding, we could simply try to understand it, in its manifold forms.

This page intentionally left blank.

Chapter Four

Intelligibility

The platonic pedigree of intelligibility seems both unavoidable and unfortunate. Behind a notion of intelligibility was the intellect, as one of the parts or faculties of the human mind. As the eyes saw, so the intellect understood. A distinction between “sensibles,” accessible by perception, and “intelligibles,” accessible through the intellect, cannot be defended.
The first point of interest is the misleadingness of the analogy between understanding and vision. It makes apparent sense to ask about visibility “in principle.” It might even make sense to think about perfect or ideal visibility, or an ideal spectator. Alluring parallels with understanding can be created.
Could something be intelligible, or unintelligible, in principle? Could there be a perfect (absolute, objective, divine) understanding? Such parallels are less impressive if you start from sound or smell rather than sight. A sound that is audible in principle can be more than anomalous. There may be a best seat in a concert hall to hear a musical performance, but that will still not be the place to hear the perfect or ideal sound. Objective listening is less plausible than objective seeing. Why should understanding be like seeing rather than hearing?
It might seem tempting to skip over intelligibility altogether, as a relic of defunct philosophical distinctions and terminology. That would be premature.
It is possible to ignore the platonic faculty-psychology of the understanding intellect and to ask simply what we, as people, can and cannot understand, and why. The next chapter will look at possible failures in understanding, ranging from skepticism to relativism. Then the following chapter will consider the

51

52

UNDERSTANDING UNDERSTANDING

view that anything important can be beyond understanding—a view central not merely to some theological positions. Both those discussions will raise questions about what can be understood, and what that means.
An initial question—much more puzzling than it sounds—is how much we want to understand. Two assumptions—maybe, again, going back to Plato— are that more understanding must be better than less, and that perfect transparency is an ideal. Such views might have played an ideological rôle in the development of the sciences. An obvious exception might be in understanding ourselves and others. Many people would balk at the thought that others could achieve even a competent understanding of them (however that might be taken). Some people would reject the thought that they understand themselves adequately, and might regard that Socratic ambition as unattainable.
The most doctrinaire assertion of the uniform intelligibility of everything, in the same terms, was, once more, to be found in Spinoza, in the extraordinary passage from his Ethics already quoted at the end of chapter 1:
Our approach to the understanding of the nature of things of every kind should be . . . one and the same; namely, through the universal laws and rules of nature. . . . I shall, then, treat of the nature and strength of the emotions, and the mind’s power over them, by the same method as I have used in treating of God and the mind, and I shall consider human actions and emotions just as if it were an investigation into lines, planes, or bodies.1

Spinoza thought that universal, uniform knowledge or understanding was to be pursued, not as an end but because human beings possessed some natural drive to seek it: “for the man who is guided by reason, the final goal, that is, the highest desire whereby he strives to control all the others, is that by which he is brought to an adequate conception of himself and of all things that can fall within the scope of his understanding.”2
All of this has to be more than questionable. Chapters 1 and 2 reviewed objects and styles of understanding. There need be absolutely no presumption of uniformity for either. Even if an ideal end for science is not assumed, questioning has to go on and answers will be available. An extremely productive model of scientific inquiry was grounded on a principle that research must always be pursued. The boundaries to that model are not constrained, and may be expanded energetically. Spinoza, for one, plainly believed that there were no boundaries. To understand something is to know its cause. Any individual thing or event has a cause because nothing can be uncaused. So (in brief ) anything can be understood, and in the same manner. The trouble with this is not only its deep implausibility but that there is no reason to accept it. Taking only a few of the examples from chapter 1, why should anyone want to assume that history, language, law, and music can or should be understood in the same way or to a similar degree? One reason might be an all-embracing use of “nature”

INTELLIGIBILITY

53

to incorporate, literally, everything; but that would beg the question. There is no reason to take for granted that human actions can be forced into the same causal model as anything else in nature. This might be a useful starting-point, and a productive methodological axiom, but it is hard to support without circularity. A simple platonic view might propose a universal human understanding that works in a single general way; but what sort of claim would that be, and where is the evidence for it?
A desire for universal, uniform understanding can only be seen as historically and geographically peculiar. It was essential to seventeenth-century
European science, as a counteremphasis to an older view that there was, or should be, a multiplicity of styles of knowledge, just as there was a multiplicity of seemingly unrelated powers and forces in nature—a view damagingly associated with Aristotle. There can be little doubt that a certain type of bafflement or wonder must be a concomitant of certain types of understanding.
It is a truism among ethnographers that societies untouched by the modern west may not even begin to see the point of anthropologists’ inquiries: what is interesting, puzzling, in need of explanation, or what sort of explanation might seem appropriate. To ask: Why do we want understanding, and how much? must provoke the further questions: Who are we, and when?
A doctrinaire view can be satisfying at a cost of being self-supporting: we want as much understanding as possible, and believe it to be available in the form of causal or lawlike explanations about the workings of nature (i.e., everything). The obvious problem is that we may not want this, and may not believe it. A view that human beings may be understood in important ways—physiologically or anatomically, as mammals—like pigs or dolphins was enormously important when first accepted, and was enormously productive. It might even be productive to ask how far—behaviorally, for instance—human beings can be understood as pigs or dolphins are understood. But nothing could license an assertion that human beings must be understood in all ways as pigs or dolphins are understood (whatever that could mean). A uniformity of understanding can be pragmatically or regulatively useful while being indefensible absolutely. (For some, such as Nancy
Cartwright, such uniformity may not even seem useful.)3 A contrary, Aristotelian view, that each subject needs to be understood in its own appropriate way, to its own appropriate degree, sounds agreeably relaxed but is entirely unhelpful. To see “what a human life is, has been, and can be,” as
Bernard Williams has put it, “This requires a proper understanding of the human sciences, and that requires us to take seriously humanity, in both senses of the term.”4 But then what is the appropriate way to understand humanity? Do we seek to find this out? Or decide it?
In the most provincial, twenty-first century, anglophone terms, it is far from plain what “we” want from intelligibility. The romantic era left a powerful resistance to the hope that people—ourselves or others—can or should be

54

UNDERSTANDING UNDERSTANDING

transparent to understanding. The idea of being “completely understood” may be disconcerting or repellent. There are areas—most patently, creativity and various forms of moral extremity—where thorough understanding not only seems elusive but where we have to be unsure what it is we want, or might find satisfying. This must apply even to the most confident enthusiasts for genetic or evolutionary explanation. Mozart or Stalin may have had something in their inheritance or their upbringing that made them as they were. In both cases it could have been something so exceptional that any alleged explanation might turn out to be just as baffling as the capacity that was supposed to be explained.
Modernists in the first half of the last century played with intelligibility.
This was very common in the visual arts. Finnegans Wake was the most provocative case in a written work. No one—not even the most exhaustive commentator—could aspire to a full understanding of every word, nor even imagine how such an understanding would contribute to an understanding of the whole. Part of Joyce’s aim must have been to confront a simple model of reading a clear meaning from a text. A single, uniform style of understanding was evidently inadequate. There could be no clear reply to the question:
What do you want from this book? The example could be (and has been) extrapolated to any creative writing, and then to any writing at all; but there is no need to go so far to see that perfect understanding can be a mirage. One reaction might be to point to the objective, neutral, literal prose of a scientific report, where a perfect fit between an author’s intention and a reader’s understanding may seem attainable. Yet despite many years of theorizing in the twentieth century, attempts to establish such literal meanings as foundational came to nothing. Language that, as it were, cannot be misunderstood may not be impossible, but it can only function within a context that itself has to be understood and accepted. It can be a special case, not a basis for other meanings.
One conclusion that might seem to follow is that a desire for understanding is, or should be, wholly relativized. What needs to be understood will depend entirely on the assumptions within a society at a particular time. As concepts develop, what we want to understand will change. So do not expect any timeless thoughts about desires for understanding or intelligibility.
Some of this must be sensible. Someone time-transported from the Middle Ages could not readily understand many of the artifacts of the modern world, obviously because the necessary concepts—electricity, instant communication, and so on—would be missing. But does it follow by analogy that someone transported from Manhattan to the jungles of New Guinea could not understand some of the practices to be found there because the necessary concepts would be missing? To an extent that would be right but uninformative. The missing concepts might add up to a whole way of life, not a set of transferable apparatus or terminology. More trickily, does it follow by analogy that something thought to be unintelligible today might become intelligible

INTELLIGIBILITY

55

in the future when the right concepts are discovered or developed? An interesting example might be the relationship between mind and body. Some writers have believed that connections between consciousness and physical processes cannot be understood now, and that the explanatory laws or theories which would elicit such connections cannot be imagined, but that such laws or theories might be devised some day.5 There might be more than one assumption about intelligibility behind this belief. There could be the optimistic view that nothing is intrinsically unintelligible: more bluntly, that science will succeed eventually. Or there might be a view that unintelligibility is only a matter of defective conceptual, interpretative apparatus: with better vision we could see better. Neither assumption can be taken for granted, although both may be necessary to provide motivation for further inquiry.
There are other cases where forms of intelligibility have been assumed in the past that now seem inappropriate or completely irrelevant. This applies to teleological or narrative understandings of coincidences or disasters. Why assume that the question of the relation between mind and body is more like the question “What virus caused my illness?” than like the question “What sins caused my illness?”
It might be thought straightforwardly false that sins cause illness and true that viruses do cause illness. So any attempt to understand the cause of illness with a concept of sin would be not merely out of place or time but simply wrong. More will be said on this later, in chapter 6. The immediate point is that a desire for understanding may seem entirely relative to a time or place.
And that is hardly controversial in outline. In the loosest terms, the framework of modern science supports expectations for unlimited understanding. It is easy to imagine that without that framework the desire for understanding might be quite differently directed. (Augustine lashed excessive scientific inquisitiveness, as he saw it, as a kind of greed comparable to the lusts of the flesh.)6 The trouble, as with any form of historical relativism, is in seeing the normative implications for the present. A mother may have the strongest desire to understand why her child, rather than any other, has been stricken with a fatal disease. What she may think she wants is a teleological account in terms of providence or fate. Is it possible to suppress a desire for that form of understanding only by classing it as an anachronism? Or does one—limply— just say that it would constitute “another form” of understanding, neither wrong nor inappropriate in contrast with an account from orthodox modern pathology? By analogy with critical theories of knowledge, a critical theory of understanding might show that one should not want to understand inappropriately (just as a convinced Kantian might cease to want to prove the existence of God from metaphysical premises). Whether or not that is possible, surely it would take more than historicised relativism to make it convincing.
The answer to “Why should I not seek an understanding like this?” would have to be a lot more than “Because we do not think like that here and now.”

56

UNDERSTANDING UNDERSTANDING

It would have to be more like “Because it is not correct to think like that,” where “correct” would be more than local.
Nevertheless, it is still possible to comment on current tastes in intelligibility. The reasons why we want to understand, in general, can be traced either to native curiosity or to the ideology of modern scientific method: the value of more knowledge, the inevitability of its growth, and so on. Just as vaguely, it may well be true that confidence in universal intelligibility has decreased, at least in the parts of the world where such things are considered at all. A relaxed acceptance of differing styles of understanding (discourses, récits) may go along with an explicit repudiation of a single (privileged, “Enlightenment”) channel. If this is right, it may be connected more to mood or fashion than to any theoretical basis. There are some strange ironies. A decline in faith in the transparency of human rationality has often been blamed on Marx and
Freud—“masters of suspicion”—both of whom were hardline determinists
(Freud: “There is nothing arbitrary or undetermined in the psychic life”).7 A decline in faith in comprehensively systematic understanding has sometimes been blamed on Gödel, who was a hardline platonist.
How far we want understanding to extend is a puzzle for which no systematic answer may be possible. Some people may just feel comfortable with areas of mystery, others very uncomfortable. Perhaps this only matters where clear limits are suggested, beyond which understanding must fail, or where the limits of what can be understood are much narrower than might be expected. These points will be the subjects of the following chapters. It is worth pointing out again that any difficulty in formulating a single acceptable theory or account of understanding does not imply in itself that a diversity in styles of understanding is acceptable, or that I can want to understand in any way I choose. I may long to see the world with the alleged simplicity of my peasant ancestors, and may study long and hard to achieve this. The reasons why it will not work are hardly obscure. Desires for understanding, though perhaps wayward, must be constrained in some ways, by knowledge and by practical realities. It is difficult to want to be uncurious in a deeply curious age. To not want to find out what exists and how things work may be beyond anyone with a scientific mentality. In fact this may show what is characteristic of a scientific mentality. Here is another reason why intelligibility remains an important subject.
Something is intelligible if someone can understand it. If someone can understand something, then it is intelligible. So is intelligibility a sort of property?
If so, is it a property of whatever is understood, or a property that depends to some degree on a relation to a person who is doing the understanding? In different terms: on the one side, some writers have thought that if there were no order, lawlikeness, or even rationality in the world then nature would be unintelligible.8 (There may follow a transcendental argument: since we can under-

INTELLIGIBILITY

57

stand nature, it must have certain features which may then be unfolded.) On the other side, it seems obvious that intelligibility must bear some relation to human powers.
This debate, and these alternatives, could seem artificial in that a commonsense compromise seems ready at hand. If a book is on sale in a store we expect it to be intelligible to whatever readership the author and publisher had in mind. If it is not, there are good grounds for complaint. The book (so to speak) possesses intelligibility in relation to its intended audience. So intelligibility can be taken as analogous to a secondary quality, like color, taste, or smell, rather than to a primary quality, like shape or extension. It is somehow grounded in things, but also somehow relative to people.
This easy compromise cannot be maintained with any generality. There may be mathematical cases where it is possible to say that no one could understand something in the strong sense that there is something that is objectively unintelligible. No one could understand how there could be an integer between 3 and 4 or a prime between 13 and 17. This does not mean that most people find this difficult, or even that God would find it difficult, but that it cannot be done. Less controversially, if what looks like an ancient text turns out to be merely decoration or random scribbling, it seems reasonable to believe that no one could understand it as a text and that its unintelligibility as a text has nothing to do with any relation to the readers trying to understand it: it is really, objectively unintelligible. Trying to determine whether a text is intelligible is not trying to find out anything about people (or possible people), but about the text (“itself ”).
In philosophical terms, a distinction between primary (objective) and secondary (partly subjective) qualities proved historically unstable. The extent to which qualities were grounded in things while being relative to people’s experience proved elusive. Worse, a commonsensical approach does not get far with the very cases where intelligibility can prove interesting. Should one say that nature is intelligible? Some parts of it are, but do we have to think in general only about nature-for-us, such as we can understand it, or is some kind of lawlike order really there? There could be some connection with rationality, in that irrational behavior and unintelligible behavior may seem to be interlinked almost by definition. There is certainly a connection with modality—possibility—in that something is intelligible if someone can understand it: the sense of “can” may be slippery. Does it mean “can” in practice, in theory, in principle? On the philosophical rough ground of questions about objectivity, rationality and modality the prospect of simple, commonsense answers gets more remote.
The main source for thought about the location of intelligibility must be in the work of Kant, although he did not address the subject in those terms. He reserved the term intelligibel for a negative use, meaning “nonsensory,” while

58

UNDERSTANDING UNDERSTANDING

denying that there must be intelligible objects for the understanding as there were appearances as objects for the sensibility.9
To begin, we do understand nature. “We are in possession of a pure natural science,” he wrote. The problem for him was how this was “possible.”10 In the first edition of the Critique of Pure Reason, understanding and nature were intimately connected: “The understanding is . . . not merely a faculty for making rules through the comparison of the appearances; it is itself the legislation for nature, i.e., without understanding there would not be any nature at all.”11
The connection was through the typically Kantian knot of experience, law, and universal necessity. Nature is subject to laws. It is “the existence of things, insofar as that existence is determined according to universal laws.”12 The
“highest” of these laws “are not borrowed from experience, but rather must provide the appearances with their lawfulness and by that very means make experience possible.”13 So if no law, then no experience and no nature. Our understanding makes experience possible because without the order provided by its “legislation” we could make no objective judgments. Because laws had to be universally compelling they could not be drawn from experience. They were “truly universal laws of nature, that exist fully a priori.”14
So much was straightforward enough. More uncertain was the part played by the “unity” of nature in the later parts of the first Critique. Kantian science presupposed “systematic unity,” which implied uniformity, consistency, and universality. More problematically, it also suggested purposiveness. This unity could not be read from nature or be found in nature, but had to be applied to it. The idea of systematic unity “should only serve as a regulative principle for seeking this unity in accordance with universal laws of nature.”15
Here “regulative” meant not constitutive (sometimes “hypostatic”: substantive). We were supposed to think as if nature were a systematic unity:
For the regulative law of systematic unity would have us study nature as if systematic and purposive unity . . . were to be encountered everywhere to infinity. For although we may light on or reach only a little of this perfection of the world, yet it belongs to the legislation of our reason to seek for it and presume it everywhere.16

Later, in the second Introduction to the Critique of the Power of Judgment, he went further, in suggesting darkly that “particular empirical laws . . . must be considered in terms of the sort of unity they would have if an understanding
(even if not ours) had likewise given them for the sake of our faculty of cognition, in order to make possible a system of experience in accordance with particular laws of nature.”17 He kept worrying in this area until the end of his life, as can be seen from the many notes under headings such as “What is physics?” and “How is physics possible?” in his notes published as the Opus
Postumum. His continuing preoccupation was that nature was intelligible
(through physics) not because we can discern an order or unity in it, but

INTELLIGIBILITY

59

because our understanding18 assumes order or unity. To ascribe unity to nature itself—to imply that we understand it because of its properties of unity or orderliness—was a particularly reprehensible error, frequently characterized as Schwärmerei—enthusiasm (often associated by Kant with the name of Spinoza). “To take the regulative principle of the systematic unity of nature for a constitutive one, and to presuppose hypostatically, as a cause, what is only in the idea as a ground for the harmonious use of reason, is only to confuse reason.”19
There is no need to go into the details of Kant’s thinking—still less its development or its ornate terminology—to see the central points. The two notions of transcendental possibility and the regulative principle remain immensely persuasive: “we will . . . be able to study a priori the nature of things in no other way than by investigating the conditions, and the universal
(though subjective) laws, under which alone such a cognition is possible as experience.”20 Here, the “conditions” that make cognition “possible” amount to the apparatus of concepts and assumptions without which we cannot understand nature. Because that apparatus is as it is, nature for us must be as it is, and there is no point in thinking that it could be otherwise. We can understand nature only because of a certain orderliness. Without it, our world would be chaotic and judgments could not be formed. This orderliness is not found by us in nature but taken for granted by us as a “condition” for our experience, judgments, and thought. The guarantee that we do understand—or that we do not radically misunderstand—lies in the very fact that the order in nature comes from us, not from nature (“in itself ”). “If nature meant the existence of things in themselves, we would never be able to cognize it, either a priori or a posteriori.”21
A regulative approach to order took intelligibility as a methodological assumption. We cannot know whether things are systematically ordered, but we must think as if they are, maybe because thought is not possible otherwise.
“Ideas” of the systematic unity of nature cannot read be from it, but provide regulative guidance, without which understanding, again, would not be “possible.” Such ideas “are not created by nature, rather we question nature” according to them.22
The simplifying metaphor that comes to mind is one of projection. The orderliness or lawfulness of nature is projected into it, so the fit between understanding and nature is automatic. A metaphor of projection locates understanding between seeing and interpreting. What we understand is there, but only as we have to interpret it. Using unKantian language, rationality or intelligibility becomes transferred from things to us. We can understand things not because they are intelligible, but because our conceptual apparatus interprets things for us in the only way in which we can understand them. “For truth and illusion are not in the object, insofar as it is intuited, but in the judgement about it insofar as it is thought.”23

60

UNDERSTANDING UNDERSTANDING

Still in simple terms, the power of this model is that is seems to deal with the deep-looking question: How is it that we can understand nature?—as though this were some remarkable feat or coincidence. It appears to tilt questions about intelligibility firmly toward questions about our understanding, which seems to make sense. The very idea of a “rational nature” or a “rationally ordered nature” sounds archaic, after all.
There are two large, interconnected difficulties.
First: what kind of story was this? Kant himself had no problem with powers of the mind. Reason is “the faculty of the unity of the rules of understanding under principles.” Yet “Our reason itself (subjectively) is a system.” It is a capacity
(Vernunftvermögen), not reified but still nevertheless part of the mind’s activity. It
“orders” concepts.24 So it serves a double purpose: as an activity of the mind and as a standard or set of principles to which thinking has to conform. The laws of reason were not meant to be psychological laws; rather the reverse, they could not be psychological laws to work as they did. Some commentators have presented
Kant’s ambivalence positively. Christine Korsgaard writes that he tried to bring two conceptions of reason together: “to explain the normative force of the principles by showing that they are constitutive of mental activity itself.”25 Less generously, it is possible to ask how this could work. If it is a psychological theory, about the real workings of the “minds” of real people, where is the evidence? This might seem a gross inquiry in the face of Kant’s subtlety, but it must be pursued. If the mind is constructed in certain ways, so that it must understand nature in certain ways, whose mind is this? Attempts have been made to depsychologize Kant,26 but the main alternative seems to be an optimistic reliance on “logic,” as though that improved any explanatory clarity. The “conditions” under which “a cognition is possible,” for example, in the earlier quotation from the Prolegomenon: would they be psychological? Kant’s reference to “universal (though subjective) laws” would suggest so. Logical conditions could only be construed as propositions implicitly assumed, in some logical relation of presupposition with judgments. And would the force of such logic then be purely subjective?—surely not—or objective?—in which case, what would be the contribution of the individual?
A second difficulty derives from Kant’s idealism. To the blunt question:
Is intelligibility a characteristic of things (nature, “the world”) or does it depend on us (our minds)? Kant would seem to have to answer: the latter. The world is a “sum of appearances.” Nature is the sum total of objects of possible experience.27 So do we understand things or do we not? Or is our understanding constrained (in a nontrivial way) by the fact that we use the apparatus that we use? With a projective theory, the rôle of the object perceived or understood becomes cryptic. This is scarcely news, in that the shadowy nature of the thing-in-itself has always been a delicate point. Even taking account of
Kant’s warily negative characterization,28 it is still possible to be puzzled about how an understanding connects with something understood. The clear answer should be that this is a meaningless question. Nothing can be known about

INTELLIGIBILITY

61

such a connection. Yet there are uneasy passages in the section of the first Critique “On the regulative use of the ideas of pure reason.” Principles or laws (“of the homogeneity, specification and continuity of forms”) were supposed to be assumed by scientific inquiry. These laws judge the parsimony of fundamental causes, the manifoldness of effects, and the consequent affinity of the members of nature in themselves reasonably and in conformity with nature [. . . eine daherrührende Verwandtschaft der
Glieder der Natur an sich selbst für vernunftmässig und der Natur angemessen urteilen], and these principles therefore carry their recommendation directly in themselves, and not merely as methodological devices.29

No one could call this clear, but one point is particularly difficult. How could anything be judged in conformity (or be commensurate) with nature if nature in itself is unknown and any relation with nature is unknowable? If nature is, as Kant insisted elsewhere, nature as experienced, then his principles might well be “methodological devices,” but how could they operate “directly in themselves” rather than pragmatically? More straightforwardly (for example), a “law of homogeneity” may be useful, as Kant said, and we can accept that its lawlikeness must be brought to possible experience not drawn from it. We can accept that it can be used in a heuristic way, “only asymptotically, as it were,
i.e., merely by approximation.”30 Yet we may still want to know whether it is right or not: whether nature is indeed homogeneous in the way that the “law” suggests. Maybe this is not answerable, but then Kant’s projective approach to understanding may imply more idealism, or subjectivism, than might be apparent at first. That may be historical reinforcement for the view that a balanced compromise between a purely subjective interpretation of intelligibility and an objective interpretation is not so easy.
There is a parallel between Kant’s understanding of intelligibility and his short treatment of secondary and primary qualities in the Critique of Pure Reason: “Colors are not objective qualities of the bodies to the intuition of which they are attached, but are . . . only modifications of the sense of sight, which is affected by light in a certain way. Space, on the contrary, as a condition of outer objects, necessarily belongs to their appearance or intuition.”
Kant gave an example a few pages later. A rainbow is “a mere appearance in a sun-shower.” What about the raindrops in it? “not only these drops are mere appearances, but even their round form, indeed even the space through which they fall are nothing in themselves, but only mere modifications or foundations of our sensible intuition; the transcendental object, however, remains unknown to us.”31
To ask whether intelligibility is a property or quality of objects might produce an equally reserved reply. The characteristics of nature that make it intelligible—“unity” and so on—are “conditions.” The understanding of nature in itself will remain beyond us.

62

UNDERSTANDING UNDERSTANDING

A notion of intelligibility “in principle” might seem to entail an objective siting for intelligibility. No one person may be able to follow every detail of a large computer proof, but it can still be intelligible in principle (the Appel-Haken solution of the four-color problem is an example). This need not imply that the proof possesses some objective quality of intelligibility—only that no one person has enough time or concentration to understand it. More generally, where anything is held to be intelligible in principle, one thought could just be that somebody (or a group of people) could understand it with enough effort.
The subject of intelligibility-in-principle need not be so humdrum. The suggestion that something could be understood in principle is ambiguous. It might be meant in a merely factual sense. (No one has yet deciphered an ancient inscription, but more examples might turn up and more might be learned; so the inscription is now unintelligible in practice but intelligible in principle.) Or it might be meant in some stronger sense where no practical circumstances can be represented in which something could be understood, but where it is still thought to be intelligible in principle, or in some absolute way. Then there is a connection with so-called logical possibility. Something may be intelligible in principle if it is logically possible for it to be understood, or for someone to understand it. Such a formulation is not without difficulties, even allowing for the problematic nature of logical possibility. Does it require a logically possible someone who does understand, or just an actual anyone who possibly understands? The latter seems empty, taken in either a de re or a de dicto sense. An actual person does not or cannot understand something, but he or she possibly understands. Or: it is possible that someone might understand. Both readings sound paradoxical, the former less so. No actual person does or can understand something but some possible person could understand it.
The someone who understands-in-principle has been taken to be God: in particular the God of seventeenth- and eighteenth-century philosophers.
God’s understanding was supposed to be infinite. What would not be intelligible to mortals would be intelligible to God. That would provide a context of meaning for absolute intelligibility. This was something like a modernized version of divine providence or judgment—impenetrable to humans, and making sense only to God. God was meant to understand how Descartes might be in error even when he could not understand it at all himself. God became a mathematical intellect instead of a judge. One hope could be that such language might be translated into the neutral terms of a purely objective, absolute perspective, although the translation would miss an essential point. A
(literally visual) view or perspective may not have to be somebody’s, but it does have to be from somewhere. A purely objective view may be not only difficult in practice but impossible. This is a point at which any analogy between understanding and vision has to be given up. The visual metaphor of a point of view provides no guidance, one way or the other, on objective (or still less

INTELLIGIBILITY

63

“absolute”) understanding: a point rather surprisingly missed by some writers who rely on seemingly neutral “representations.”32
Unintelligibility-in-principle may cause less trouble. It seems reasonable to say that something can be absolutely unintelligible, or unintelligible in principle, and to mean by this that there are no conceivable circumstances in which it could be understood by anyone, even, if you like, by God. This is not wholly incontrovertible, in that any collection of data can be covered by an infinite number of theories; so any “unintelligible” information is open to endless arbitrary interpretations. But leaving that aside, it might be thought that an assertion that 3 + 2 = 6 cannot be understood at all, except in the trivial sense of a misuse of signs. The point is hardly, as Descartes put it, that “our nature is incapable” of conceiving as possible “things which God could have made possible, but which he has nevertheless wished to make impossible.”33
There is absolutely no sense to be understood.
A lack of symmetry is plausible. Something “absolutely” intelligible “in principle” positively requires a possible someone who could, in some circumstances, understand it. There has to be someone, because intelligibility has to imply someone’s ability to understand. If you don’t like God or an absolute perspective, try Martians. Something unintelligible in principle does not call for anyone who does not understand. It may suggest merely that there is no context in which understanding can be represented. This can be factual—as of now, no Martians—but it is nonetheless as “absolute” as it needs to be. The next chapter will look at a number of less decisive barriers to understanding— examples in which whole classes of discourse or knowledge are said to be unintelligible in one way or another.
These abstract points come out with more force in the specific case of linguistic understanding. A text can be unintelligible in the simple sense that no one can or could make anything of it. (In this case there cannot even be an appeal to endless arbitrary interpretations. If there were such interpretations, they would have to be somebody’s, which ex hypothesi they are not.) Such unintelligibility is as strong as any context requires. But intelligibility in principle does call for the strongest possible context: for the positive supposition of a possible someone who actually could understand, even when nobody actually does understand. A text has a meaning. That can only be taken as saying: someone could understand it. To which it is not irreverent to ask: Who?
There must always be someone, or a plausibly possible someone, maybe even in the past. Absolutely objective, or purely impersonal meaningfulness can only be an aim, as in the ideal of a scientific report. A theory of meaning might be pragmatic—how to make yourself understood—but it could never be simply objective—how to create meaning.
So can anything be said in general about the intelligibility of nature? It was Einstein who remarked that the most incomprehensible thing about nature is its

64

UNDERSTANDING UNDERSTANDING

comprehensibility, so that thought cannot be dismissed as simply naïve.34 A further step into the strong “anthropic” position—that a match between our understandings and physical laws is highly noteworthy—may be less pardonable.
Surprise that we understand anything at all has its roots in a feeling that we could have it all wrong; which in turn could mean that there might be some wholly different perspective that might be “right,” or more right than ours. A refusal to admit this might seem like a dogmatic denial of the possibility of mystery. Certainly, it has a religious dimension, in that the most familiar version of a wholly different perspective was the view that God, or the gods, will see what we cannot see. This goes back at least to Heraclitus: human nature does not have right understanding, but divine nature does. Many modern writers have not been content to leave it at that. Descartes, typically, was quite ready to admit that God (and angels) may take a different view of things, but
(with a pragmatic shrug of the shoulders, no doubt) he thought that we will never know what this is, so we had better manage with what we have: “Why should this alleged “absolute falsity” bother us, since we neither believe in it nor have even the smallest suspicion of it?”35
It is not clear whether Kant’s elaborations added anything. Our understanding of nature is not a direct vision but an interpretation mediated through our concepts, marshaled by our reason. Nature in itself is unknowable. The fact that we understand cannot be surprising because, in short, our understanding makes for us the only nature that we can understand. One of the aims of the critical philosophy was to enable us to realize that any puzzlement about nature outside our understanding might be humanly unavoidable, but was also misplaced. The advance on the pragmatism of Descartes was in the suggestion that we all have a disposition to roam beyond the boundaries of our reason. That is doubtful enough, taken empirically; but the real difficulty is in the view that our minds have to work as they do work. For the sake of argument we may accept that our world would not be as it is for us if we did not have the framework of concepts and ideas that we do have. This framework is a Kantian condition for experience and thought. But the alternative need not be unconditioned, direct experience (of things-in-themselves).
It might as well be a changed or developed framework of concepts, giving us a different world. Even if you cannot see without spectacles, that is hardly a constraint if an open-ended number of pairs of spectacles will be available. If concepts of space, time, and causality have changed since 1800 in ways that
Kant thought inconceivable, what justification can there be to regard any particular set of concepts as limiting?
The rigidly uniform approach of Spinoza mentioned earlier in this chapter (“Our approach to the understanding of the nature of things of every kind should be . . . one and the same; namely, through the universal laws and rules of nature”) may seem anachronistically unpromising. But if we are thinking about the intelligibility of nature, one of Spinoza’s theories does remain of

INTELLIGIBILITY

65

some interest: the arcane-sounding doctrine of the infinite attributes of God or nature. The same object or event could be conceived in two entirely different ways, as extension and as thought, with no causal relationship between them. This was the basis of his anti-Cartesian account of the relation between body and mind. He also held that there were an infinite number of other ways in which nature could be conceived, but that we do not know what these are
(or we cannot know what they are). In plain terms, this was a view that we do understand nature as it is, but we have to accept that it may be understood in many other ways which we may not (or cannot) understand. There was no need for the external, nonhuman perspective traditionally adopted by the eye or mind of God. People understand in various ways, so to them nature can be fully intelligible. There may be other ways in which nature can be understood in which it would also be fully intelligible.36
None of this is without problems. Commentators have been divided on whether this was meant subjectively (i.e., to do with the conceiving) or objectively (i.e., to do with aspects of things). It may be attractive to say that a human action may be conceived as bodily (physiologically) or as mental
(intentionally or psychologically). This produces a position very like twentieth-century neutral monism. The serious obstacle lies in characterizing distinct ways of understanding. How are inanimate objects to be conceived in different ways? Everything is “animate,” although to differing degrees.37 So do rocks think, to some small degree?
The Cartesian view must have been that God can always do better. Kant’s view was that our understanding is constrained significantly by our use of concepts that bear only on possible experience. Despite its difficulties, the helpful element in Spinoza’s theory might be that understanding is possible, and may be accurate, but it can never be complete. There is no point in thinking of nature as unintelligible; but there may be ways in which it is intelligible that we do not understand (yet, if you are an optimist).

This page intentionally left blank.

Chapter Five

Failures of understanding

The thought that human understanding is imperfect has not been uncommon. What is the standard of perfection? Where does it come from? Why should it matter, given that human understanding is, after all, the only understanding that we understand?
The contrast in the past—one of total transparency—must have been
God’s understanding, or understanding in heaven. In paradise, Beatrice told
Dante, what is held as faith will be “known through itself,” not by demonstration. Strong support was given by a visual image of understanding. As we see dimly and distantly on earth, so in heaven we shall see clearly and directly.
Proofs, the “eyes of the mind,” will no longer be needed, as truth is apprehended directly.1
If this is no longer credible, it might seem that there is not much more to say. We are left with human understanding, which has its own standards, and nothing else. For that matter, the picture of an absolute, perfect perspective is less convincing than it might have seemed in the past. The eye that sees literally everything literally immediately would appear to have the same shortcoming as a map at 100% scale (as explored in a terse fable, “On Exactitude in Science,” by Borges).2 A concept of vision without attention, focus, and selection may be too attenuated to make useful sense. Yet again, this can be brought out by thinking about the auditory alternative. Perfect hearing makes no sense if it means hearing everything as it is, or as it sounds. People could hear better than they do in some respects. It is not pointless to say that in some ways people do not hear as well as dogs (though what they might do with dogs’ hearing is a curious question), but it is surely entirely pointless to
67

68

UNDERSTANDING UNDERSTANDING

conclude that human hearing is imperfect. The parallel with understanding looks straightforward. I could understand many things in many ways much better than I do. My understanding is imperfect by human standards, but what other standards could apply?
So does it follow that we understand everything just as well as we need, that any thought of general failures in understanding is simply mistaken? One answer is that in specific cases we cannot always be certain what sort of understanding is appropriate, so we cannot be certain whether we are failing or not.
If my model for understanding people is the way in which I hope to understand myself, then maybe I do not understand other people very well. Or: if
(as Wittgenstein asserted in the Tractatus) “to understand a proposition means to know what is the case if it is true,”3 then maybe we are in error in believing that certain elements of discourse consist of propositions that we understand.
Or: if I am impressed by the elegance of some mathematical understanding, then maybe I can be persuaded that all other forms of understanding should aspire to its clarity and immediacy. Which takes us nearly back to the original line of thinking, that maybe all human understanding falls short of some ideal.
The Cartesian paradigm, of perfect mathematical understanding by an intellectual God, may carry little conviction today, but there are many other models of understanding whose application may be less universal but may also be far less decisive. The most obvious will be understanding between people, at many levels. It is possible to assert quite persuasively that in some important ways men and women cannot understand each other, or that people from differing cultures cannot understand each other, or even that I can never hope to understand myself. Quite persuasively: but how validly? The general assertion— that all human understanding, in all ways, falls short of divine perfection— leaves little to say. The next chapter will look at the archetypally vague notion that anything—or rather anything important—can be beyond understanding.
(That seems to imply either that something is unintelligible in principle or that our understanding does have some determinate or inscribed general limitations.) This chapter will catalogue some views on specific types of alleged failures in understanding, and will ask what they imply.
Two preliminary points come out from earlier chapters.
First, bearing in mind some of the reservations in chapter 3: there are some ways in which a treatment of understanding does undercut any conclusions on knowledge. The linguistic case is the clearest. If I cannot understand what you say, then I cannot know whether it is true or false. The religious case is the most often heard. If you cannot understand my religious beliefs and practices, then how can you know whether they are justified? More explicitly, a claim not to understand must be more generally subversive than a claim not to know.
Historically, one way of advancing beyond the Cartesian obsession with knowledge was to cultivate an interest in meanings. Another, more radical

FAILURES OF UNDERSTANDING

69

move, might have been to question the intelligibility of the first steps in
Descartes’s mise-en-scène. My knowledge of the world outside my mind became fundamental to first philosophy because I was supposed to be able to suspend my belief in that world, if only for the sake of argument. But if such a supposition could not be portrayed coherently, then maybe its portrayal would not be properly intelligible and the whole priority for knowledge would be undermined.
Some philosophers have hoped to give a form to understanding in areas where they think they cannot possess knowledge. Kant felt able to say a good deal about his thing-in-itself (“x,” the “transcendental object,” noumena, variously) while also saying that he knew nothing about this elusive territory, or while saying that his knowledge of it was merely negative. Critics have always been able to respond that his whole apparatus was not merely unknowable but unintelligible: that he had failed to construct a story that held together enough to make sense. And this, if sustainable, would be devastating.
A pose of not even being able to understand a rival’s theories was a cliché of twentieth-century philosophy. It reached a level of caricature in Ayer’s Language, Truth and Logic, where whole areas of previous debate were ruled out as so unintelligible that further discussion of their truth would be a waste of ink.
That general project may have been bankrupt before it was even finalized, but the strategy within it remains potentially applicable for specific cases. There are important areas where incapacity or unwillingness to understand can forestall any further discussion. There are areas where an inability to decide the appropriate form of understanding really can prevent any discussion at all. In both respects, the Balkan wars of the 1990s provide examples. How far could the warring factions ever understand each others’ histories, hatreds, and ambitions? How far could outsiders, however well-intentioned and well-informed, ever understand what was going on? What sort of understanding would have been helpful anyway?
Second, from the end of the previous chapter: limits on understanding within specific contexts should hardly be unusual. The real difficulty comes from positive assertions of intelligibility or meaningfulness in principle, not from barriers or restrictions. No set of rules or conditions can guarantee that anything will be understood, in general or in any specified circumstances. But there are many conditions that may stand in the way of understanding.
Schleiermacher could have had this in mind when he noted that general hermeneutics was a study of the conditions in which misunderstanding becomes possible. It “rests on the fact of the non-understanding of discourse.”4 Some conditions will be obviously practical. Others may be barriers of principle.
The former, practical category is large and, on the whole, not interesting in itself. There are very many practical obstacles to understanding of every sort. People speak different languages, come from different backgrounds, and

70

UNDERSTANDING UNDERSTANDING

hold differing opinions. Some are clever, some are stupid. Some have specific talents—for music, mathematics, personal relationships, languages, and so on—that widen their horizons of understanding. Others have disabilities or shortcomings that bar them from types of understanding. Imaginations may be limited. Prejudices may be ineradicable. There can also be innumerable objective-looking barriers. Texts may be hard to decipher. Intricate calculations may be beyond the concentration of most people. Events may resist simplification. Histories and memories may be partly lost. People die.
Many failures of understanding will come into this large practical category. The ways to diminish or remove them will also be practical: education, hermeneutical skill and patience, hard work, tolerance, sympathy, insight. All these can be extremely important. The main point of caution for a philosopher is that generalizations will be fruitless. One of the reasons that can make
Schleiermacher a frustrating figure to philosophers was his insistence on ad hoc, piecemeal hermeneutical rules. But this should not be surprising. The impediments to the understanding of (say) a biblical text will not, on the whole, be matters of general principle (although some may be). They will be matters of language, historical knowledge, religious presuppositions, textual corruption, cultural assumptions, and sectarian prejudice, as well as the perceptiveness and intelligence of the reader. Some of these factors can be regulated methodically, others not, but there will be no “general method.” Generalizations about how and why people actually do or do not understand may be unreliable. Theorizing based on assumptions about human nature will be worse than risky. Theories assuming what the (average) rational person should not or cannot understand in principle will rest on shaky ground. The distinction between understanding in practice and “in principle” is not a clear one.
Nor need it be tied directly to a suspect opposition between empirical observation and a priori speculation.
This needs to be kept in mind when thinking about the other category of barriers to understanding, as presented by alleged matters of principle rather than practicality. That category is of far more interest to philosophers, and has shown itself in a variety of forms:
(a) The originating thought behind modern skepticism was that the mind may not possess the capacity to grasp reality accurately or adequately.
In the history traced by Richard Popkin, the originating theological thought was that God has not made me so that my mind is up to the task of grasping reality as it is.5 Typically, in his dialogue, Things Above Reason of 1681,
Robert Boyle wrote:
I am induc’d to think that God, who is a most free Agent, having been pleas’d to make Intelligent Beings, may perhaps have made them of differing Ranks, or Orders, whereof Men may not be of the Principal; and that whether there be such Orders or no, he hath at least made us Men, of a limited nature (in general) and of a bounded Capacity.6

FAILURES OF UNDERSTANDING

71

Either theologically or nontheologically, there was an assumption about standard human capacities. The underlying point of modern (but not ancient) skepticism—that there is some impediment in principle between the mind
(my mind) and the world outside it—rested on factual assumptions about the normal constitution of a mind. The Cartesian man (and he was certainly a man) who went through the recommended process of intellectual demolition and rebuilding started with various abilities that qualified him to stand proxy for the reader. He understood doubt, certainty, knowledge, evidence, and other essential epistemological tools. He dreamt, or understood dreaming. He understood elaborate philosophical metaphors: demolition and reconstruction. (Dressing the mind in the scanty conceptual garb of the “pure enquirer” does not disguise all this.7) This subject is normally presented in terms of knowledge, and not understanding. Descartes himself must have seen the point of ancient skepticism about standards of truth; but his method of doubt was more far-reaching. The evil demon at the end of the first Meditation subverted not only his capacity to determine the truth but his reason, his ability to understand anything coherently.
A great deal of attention has been paid to the arguments for and against (often history-free, context-free) versions of skepticism, almost none to what its truth would actually mean. The theological basis for early modern versions, if anything, made them more not less persuasive than subsequent restatements. At least some explanatory context was given, however improbable. In the hands of Pascal, for example, there was a wellworked-out story of how the human mind was too damaged by original sin to achieve understanding on earth. “Without this most incomprehensible of all mysteries we are incomprehensible to ourselves.” The contrast was provided by God and heaven: “Let us allow to the Pyrrhonists what they have so often claimed, that truth is neither within our grasp nor is it our target. It does not reside on earth but belongs in heaven, in God’s bosom, and we can know it only as much as he is pleased to reveal.”8 That backdrop may not be to later tastes, but without it skepticism loses the context that gave it a point. If I am a natural organism evolved and living in nature, why should my understanding be inadequate to the task for which it is so evidently fitted? If there is no contrast beyond human understanding, how can a radical deficiency be characterized? What nontheological context could exist in which our minds or senses might have something intrinsically and radically wrong with them?
Philosophical skepticism was an extension of the commonplace view that there are many things that, in practical terms, I do not or cannot understand.
Just as I am not smart enough for superstring theory and too lazy to learn
Finnish, so my mind may not be not fitted to achieve certainty. The barrier to understanding lies not in the makeup of my mind personally, but in the makeup of the human mind. Boyle again, making florid use of visual imagery:

72

UNDERSTANDING UNDERSTANDING

Eugenius: As when the Eye looks into a deep Sea, though it may pierce a little way into it, yet when it would look deeper, it discovers nothing but somewhat which is dark and indistinct, which affects the sensory so differingly from what other more genuine objects are wont to do, that by it we easily discern, that our sight fails us in the way before it arrives at the bottom, and consequently that there may be many things conceal’d there, that our sight is unable to reach.9

(b) Kant believed he was writing not only about his own mind, or the minds of eighteenth-century scholarly gentlemen, but about any rational being. Despite the endless subtlety of his thought, there was a starting assumption that all human beings had the same inclinations: for example, to extend their thinking beyond the proper bounds of reason. The opening words of the Preface to the first edition of the Critique of Pure Reason announced that
“Human reason has the peculiar fate . . . that it is burdened with questions that it cannot dismiss, since they are given to it as problems by the nature of reason itself, but which it also cannot answer, since they transcend every capacity [Vermögen] of human reason.” Insofar as his theme was understanding as much as knowledge (or reason, pure and practical, or the understanding—Verstand—which he used as a term to cover wide cognitive functions), his presupposition was that we all must share the same powers and the same limitations. Insofar as that presupposition was argued rather than taken for granted, the form of his arguments must have been implicitly transcendental: unless human cognition was essentially uniform, some people would be able to understand what others could not. That would be obviously true in a straightforwardly empirical sense; but the limits of cognition were not meant to be a matter of native capacity or upbringing. In Kant’s case this was a matter of protestant conviction—that even the plainest mind could grasp the essentials: in what concerns all human beings without exception nature is not to be blamed for any partiality in the distribution of its gifts, and in regard to the essential ends of human nature even the highest philosophy cannot advance further than the guidance that nature has also conferred on the most common understanding.10

His aim was to work out what the mind could do to its limit [Schranke], as it were from within, in line with his invaluable distinction:
Boundaries [Grenzen] (in extended things) always presuppose a space that is found outside a certain fixed location, and that encloses that location; limits
[Schranken] require nothing of the kind, but are mere negations that affect a magnitude insofar as it does not possess absolute completeness.11

The trouble is that “negations that affect a magnitude” of understanding cannot be generalized except in a circular way, where people cannot understand

FAILURES OF UNDERSTANDING

73

what they cannot understand. As argued in the previous chapter, the extent to which this is genuinely limiting is uninteresting without a noncircular way of characterizing what and how people do understand.
Some readers have found Kant’s Analytic impressive in its general theses but unsatisfactory on their detailed support, while on the other hand the
Dialectic offers a treasury of stimulating detailed argument in the context of a wholly incredible general pathology of illusion. Maybe that is because limits to understanding can be more convincing when not generalized to the whole of humanity. The problem with subjective or personal barriers to understanding is that they may be entirely sensible in practical terms, but they become indefensible when projected into matters of principle. If taken in a psychological sense, they require assertions about the necessary composition or working of the human mind that can always founder on history or on counterexamples. If they are taken “conceptually,” a problem arises of finding a nonpsychological account of concepts. But even if this were available, further difficulties would arise. No doubt the contents of our conceptual toolbox may limit our comprehension; but, once again, it does not follow that new tools cannot be acquired, or old ones adapted. As Hume had noted: “Every event, before experience, is equally difficult and incomprehensible; and every event, after experience, is equally easy and intelligible.”12
(c) An alternative is that there really are things that cannot be understood in some sense that is not subjective or psychological: in terms of a shaky dichotomy, to do with things, not us. (Even if you agree with Kant that rationality is located in us, not in nature, it by no means follows that irrationality may not be located in nature.) So then there would be objective limits or boundaries to understanding. Leaving aside theological cases until the next chapter, the most promisingly decisive candidates come from logic and mathematics. In logic, contradictions and various kinds of paradoxes may be thought to be beyond comprehension. One consideration behind mathematical finitism may have been that actual infinities should play no part in mathematical thinking, perhaps because they are in some way unintelligible. Complexity and chaos may, as a matter of principle, imply more information than can be comprehended by anyone. Elizabeth Anscombe took the view—of
Parmenides in particular—that “the ancients never argued from constraints on what could be thought to restrictions on what could be, but only the other way round.”13 In anachronistically modern terms, unthinkability would be objective, not conceptual or subjective.
You can reach such conclusions readily enough by narrowing the styles of understanding that will be acceptable to you. If you insist—implausibly—that all understanding has to be pictorial—that you cannot understand something unless you can form a mental picture of it—then by that standard large parts of mathematics (and much besides) will become classed as unintelligible. A little less vacuously, genuine difficulties in understanding may follow from the

74

UNDERSTANDING UNDERSTANDING

adoption of unsuitable models. The most common example is wave-particle duality, where visual imagination is a definite impediment to understanding, to the extent that the best route to understanding is to put it aside.
The right way to understand quantum phenomena is through an understanding of quantum theory. It need not follow in general, though, that anything can be understood because an explanatory theory can be constructed to cover it. This might seem tempting because consistent theories can be formulated where imagination plays no part in their understanding (or a very limited part, as with string or brane theories). So it might seem that understanding can be understood exhaustively in terms of acceptance of an explanatory theory. This sounds attractive, for example, with infinities. The more exotic areas of pure mathematics will give many other examples, where the “subject” understood appears to exist solely in the theorems in which it is expressed. A further extension might be to so-called paraconsistent logics, where expressions that would be excluded contradictions in standard logics may be accepted as axiomatic bases for variant systems, and where explanation or understanding can be suggested in terms of formal modeling.14 Then it would follow, in effect, that nothing need be unintelligible in principle.
That line of thinking is grounded in the distinction drawn in its classical form by Descartes at the opening of the sixth Meditation, between imagination and “clear and distinct conception” or intellection. The basic thought was simple and alluring. Imagination relies on the fickle powers of the body. Some people can imagine well, others hardly at all. Most people can imagine a triangle, very few can imagine a thousand-sided figure, or chiliagon. Nobody can imagine the difference between a chiliagon and a 999-sided figure, yet nobody can deny that the difference is genuine. So: it is conceived by the mind, rather than imagined. Conception may then be extended into a form of objective or theoretical representation: plane geometry, for example. The affinity with understanding is also alluring. First we understand what we can grasp with the imagination. Then we can understand what is grasped by the imagination as extended by conception: complicated theorems in geometry. Then we can understand what is formulated solely in terms of conception without imagination: non-Euclidean geometry. So, ultimately, we can accept that understanding can be associated with pure representation rather than imagination.
Just as understanding can be narrowed arbitrarily to cover pictorial imagination, so—apparently—it can be widened to include any form of representation. And since there are no formal limits to representation there are no objective limits to intelligibility.
This train of reasoning was not followed through by Descartes himself, though the slippery distinction between imagination and conception can be seen in his work as well as anywhere. Imagination was meant to be purely personal. (Purely psychological would be incorrect for Descartes in particular because of his allocation of functions between mind and body. The important

FAILURES OF UNDERSTANDING

75

point is that imagination was a matter of the powers of the imagining individual subject; it was not related to the properties of the imagined object.)
Conception contained a helpful ambiguity. It was a function of the mind as a mind in general—mental substance—in the case of Descartes, either sharing in or aspiring to the mind of God, where its perspective would be ideal and its judgment perfect.15 But it also retained aspects of human vision. These were partly metaphorical. As noted already, it is remarkably difficult to say anything about understanding or conception without reverting to seeing, perceiving, perspectives, focus, and so on. In the simplest statement, what was conceived was seen by the mind. Descartes drew some explicit comparisons in Rule Nine of Rules for the Direction of the Mind, where “We must concentrate our mind’s eye . . . to acquire the habit of intuiting the truth distinctly and clearly.” He went on to say that “We can best learn how mental intuition is to be employed by comparing it with ordinary vision.”16 This looks just like an early stylistic slip that could be tidied up easily, but the problem for Descartes was more serious than that. In a late letter, even while trying to distinguish intuition from sensation, he wrote that the mind “sees, feels and handles.”17 Conception had to be enough like seeing or imagining for the analogy to work, but not like it in other important ways. Imagination needs someone to do the imagining—either a real someone or a really possible someone (i.e., not a logicallypossible-someone or a someone-in-principle). Understanding can be narrowed, implausibly but not altogether absurdly, to what can be imagined because the affinity between understanding and visual imagery is so strong.
Widening it to what can be “conceived,” or somehow represented, is far less persuasive just because the personal element has to be missing. If “conceived” means that somebody—or even God—literally can understand, then it is uninteresting. If it means that some understanding is said to be possible in some other sense, then it may be not merely arbitrary but circular. For example, a contradiction may be thought to be intelligible because there may be some (“paraconsistent”) logical system in which it can appear. No one can imagine how this can be so, but it can be represented within some formal structure. (A modern version of the unorthodox opinion, sometimes ascribed to Peter Damian, that God could choose even to make contradictions true.)18
Understanding in principle can then be taken to cover exactly this, which seems completely paradoxical: it would become possible to understand what no one can understand. That does not mean merely that it would be possible to conceive what no one can imagine (which may be fine), or that it would be possible to conceive or represent anything (which may be true) but that it would be possible for someone to conceive what no one can conceive (which looks like nonsense).
So is there some “objective” limit to what can be understood? It is possible to dilute understanding to a point where the answer to this question is no. Anything—apparently complete nonsense, paradoxes, contradictions,

76

UNDERSTANDING UNDERSTANDING

inconsistencies, incorrect arithmetic—can be understood not just as complete nonsense (and so on), but in a sense where some arbitrary language or system can always be constructed in which apparent nonsense may fit. So in that context it can always be understood and even “explained.” The catch is that understanding and explanation have to be understanding by, or explanation to, somebody. An explanation that no one understands and no actual or possible person could understand is not just a bad explanation; it fails to explain at all; it fails to be an explanation at all. There is no content in an understanding-in-principle by a logically-possible-somebody. It would be no different from thinking that an incorrect sum might be right because a logicallypossible-somebody might find it satisfactory.
If a narrowing and a widening of understanding both lead to difficulties, are there any convincing conditions in relation to what cannot be understood?
This question is framed in vague and negative terms because precision is so elusive and a positive version has been excluded. Nevertheless, it is a question that must be of great interest to philosophers, who may feel unsympathetic toward a general view that there is anything resistant to understanding. Once more, we are not looking for “objective conditions” that would ensure intelligibility (“in principle”) but conditions that would prevent it. Once more, the obstacle is one of ambitious generality. Of course there are endless ways of being unintelligible: speaking nonsense, acting bizarrely, writing chaotic music, lapsing into ineffable mysticism. The problem in each case—and still more, generally—lies in finding or defining any limit beyond which understanding cannot be achieved.
Music, or the arts more widely, will illustrate the point. If it is reasonable to think of understanding music at all—and why should it not be?—could there be any general conditions that would stand in the way of intelligibility?
Before 1900, the answer might have been in absences of tonality, rhythm, melody, and perhaps harmony. Sounds without a key structure, stable rhythm, singable melody, and recognizable harmony might have been seen (or rather heard) as unintelligible, or not as music at all. The early critics of Debussy,
Schoenberg, and Stravinsky were not slow to point this out. But during the twentieth century, and even more since 1950, none of it was remotely plausible. That might be a symptom of a general onslaught on convention and regulation in twentieth-century art, together with a broader understanding of non-European musical styles. There can be rules, but no one could regard them as anything but conventional.
The same could be said for other objects of understanding in chapter 1.
It may be possible in many cases to draw up ad hoc rules or conditions that are said to make understanding possible. George Schlesinger’s study, The
Intelligibility of Nature, was a sustained attempt at this. Unless “nature” exhibits certain features of consistent orderliness—that can be specified in detail—“we” cannot understand it—where to “understand” means to formu-

FAILURES OF UNDERSTANDING

77

late explanatory laws or principles. There may be many cases where to understand is to grasp an explanation, and there are some cases where it is to grasp an explanation of an extremely specific kind. Until a function is discovered to generate prime numbers, it could be said that the ordering of primes is not understood by anyone and cannot be understood by anyone. Here, there are very specific agreed conditions on what counts as understanding, with extremely specific criteria of success. But this tells us nothing about whether anything in nature is in itself beyond our understanding. It may be so now, in that no theory or function is available, but that says nothing about the possible future, and certainly does not limit it.
(d) During the twentieth century, the limit on understanding that was discussed most extensively was in language. Wittgenstein’s Tractatus must have been the most dramatic attempt to set a limit to “the expression of thoughts.”19 There were some elements or aspects to language that we think we understand but where we are really in error, or where our understanding is not of the sort we think. Most widely of all, “the limits of language (of that language which alone I understand) mean the limits of my world.”20
As seen in chapters 1 and 2, linguistic understanding can be presented as central, or even basic. Understanding between people, cultures, and religions can all be presented as an understanding of each others’ sentences. Barriers to understanding can be presented as linguistic. This has been claimed of philosophical understanding, too. If only linguistic fog could be cleared, we could see. Its presence has been thought to be either incidental and removable (by
Berkeley)21 or inevitable and endemic (by the later Wittgenstein). A fundamental placing for the understanding of language, and the importance of understanding in philosophy, together have the effect of making language a basic philosophical theme. Gadamer, once again: “Man’s relation to the world is absolutely and fundamentally verbal in nature, and hence intelligible.”22
There can also be skepticism about how completely language, or what it conveys, is really understood. There may be thought to be a gap between what is meant and what is understood. It is obvious enough that some kinds of meanings are irretrievable in practice: we cannot ask Shakespeare what he meant in writing Hamlet, or Heraclitus what he meant by some of his more obscure remarks. There can be far stronger views: that what we mean to each other all the time, in speech or writing, is only imperfectly conveyed, and that some residue of meaning may remain not understood. There is also the extremist position that a speaker’s or writer’s meaning is never retrievable in itself: there can only be different understandings or interpretations—what
Donald Davidson called a “disenchantment with meanings.”23 Further, this could be another point where what we want may be uncertain. Some feel strongly that sense must be communicable intact, as it were: that there could be no meaning unless there were some fully successful meanings, or unless all meaning were fully successful. Others feel that there may always be some

78

UNDERSTANDING UNDERSTANDING

element of indeterminacy or even mystery: that defined or determinate uses of language are special, not fundamental.
In the Preface to the Tractatus, Wittgenstein pointed out that a limit to the expression of thoughts—to what can be said—can only be set from within.
“It will . . . only be in language that the limit can be set.” His idea was that in showing the basic conditions for the use of any possible language—what makes possible any expression of thought—he would be showing what language could do and hence (by implication) what it could not do. Philosophy would “set limits to what cannot be thought by working outwards through what can be thought. It will signify what cannot be said, by presenting clearly what can be said.”24 In showing how sense can be made, he thought he could show that it could not be made otherwise. You could not make sense—say anything—unless you said something, and that would have to be something definite or determinate. At the elemental level, “The requirement that simple signs be possible is the requirement that sense be determinate.” Plainly, all everyday language could not be clear (even though “its propositions, just as they stand, are in perfect logical order”), but unless some sense could be made clearly there could be no sense at all, and hence no language. “Everything that can be put into words can be put clearly.”25
This is a book not about meaning but about understanding. It is the connection made by Wittgenstein between understanding and meaning that is at issue now. Just as to make sense you must use language in a certain way, so to understand that sense I must understand it in the same way. Specifically: “A proposition is a picture of reality: for if I understand a proposition, I know the situation that it represents.” And “To understand a proposition means to know what is the case if it is true.” Even though “The tacit conventions on which the understanding of everyday language depends are enormously complicated,” a proposition has to be understood in the same way in which it is meant.26 This works if communication is assumed to work fully and successfully. If it can be assumed that I do understand what you mean—or if we can agree on what we both mean—then it may seem to follow that meaning and understanding are like two operations of the same machinery. (In terms of a view repudiated by Davidson, “communication by speech requires that speaker and interpreter have learned or somehow acquired a common method or theory of interpretation.”) In the Tractatus, if “A proposition has one and only one complete analysis,” you assemble a proposition to make sense and then I dismantle it in the only possible way to understand it.27 This was an engineer’s view of understanding. Later, Wittgenstein as a writer might have remembered the obvious and common illusion that a piece of writing that is easy to read and understand must also have been easy to write. There need be no correlation at all between the making and the understanding of sense.
Now it may or may not be true that making sense depends ultimately on making clear sense, that this depends on conveying the possibility of states of

FAILURES OF UNDERSTANDING

79

affairs and so on. It may even have been quite valid to refine the meaning of terms so that a thought is a proposition with sense, a proposition is a picture of reality and so on.28 Conditions for meaning or making sense might well be persuasive, but conditions for understanding unfortunately are not. You can insist that something is not said at all unless it is said in a certain way, but how can you insist that anything cannot be understood at all unless it is understood in a certain way? And what is a way of understanding in any event? It is puzzling how Wittgenstein could have asserted with confidence that “the understanding of general propositions palpably depends on the understanding of elementary propositions.”29 If, most improbably, this was a claim about the process of understanding—how it operates psychologically—then it was just incorrect. If it was about conditions under which understanding could be possible, then it was arbitrary.
It might be true that we could not understand each other unless we were able to talk about the same things. Frege relied on this argument in “The
Thought: A Logical Inquiry,” and an assumption that we do understand each other when talking about his example, of geometry, does not sound too ambitious.30 The paradigm is not mysterious. I go through a proof with you, explaining each move step by step, and then it seems obvious that when we get to the conclusion, you must understand it in exactly the same way as I have explained it. So meaning (making sense) and understanding (grasping sense) fit together perfectly. There are three difficulties. In less clear-cut areas than geometry the assumption that we do understand each other—that communication is successfully achieved—may be held less confidently. Then, even in a very clear case (like geometry), it remains mysterious why meaning and understanding have to work along the same lines. What if—as is actually not uncommon—you see the truth of a conclusion before going through all the premises? What if you have reached the conclusion by a different proof? Have you misunderstood, or understood differently? How is what “happens” relevant anyway? Third, why assume that such an artificial case is fundamental or paradigmatic rather than extremely particular, dependent on a great many external and artificial assumptions? (Frege’s argument evokes the absurd thought of an argumentative couple who keep a dictionary on the kitchen table to make sure that they mean the same things in marital disputes: a bad state of affairs, hardly the norm.) An assertion about meaning or making sense sounds at least plausible: unless some clear sense could be made, the communication of sense would not be possible. But for understanding that just sounds baffling: unless perfect understanding could be achieved, no understanding would be possible. Why not?
After 1929, Wittgenstein came to the view that meaning could be conveyed in many ways, that no single way could be essential, that there was no essence to meaning. If you chose, you could talk about propositions as bearers of truth, but this was a matter of your decision, not some insight into reality.31

80

UNDERSTANDING UNDERSTANDING

Understanding, like meaning, could be achieved in many ways. His later views on it were largely negative or critical, aimed against any notion that it must be a “mental process.” The emphasis came to be on how it could not work—on erroneous theories—not on how it could. If, for example, to understand a meaning was to have access to some inner, “private” object of reference, then we could not understand much, if any, of normal communication.32 This was not some restriction on the use of understanding as much as a useful reminder of a truism that the incommunicable cannot be communicated. Equally, the rest of these much-discussed passages in the Philosophical Investigations did not seek to place limits on what can be understood—I cannot “understand” how your toothache “feels to you,” for example—as much as to show how previous philosophical theories would make understanding, and hence successful communication, impossible.
Wittgenstein’s later critical position was clear enough. What remains very uncertain was his positive one: how far he thought that successful communication was possible. I do not understand an English word like “grief ” because
I have private access to some inner sample of grief-experience, which the use of the word sparks off in my mind. I understand it fully when I am able to use the word. If someone tells me that I do not really understand what “grief ” means because I have had no direct experience of grief, then this is just a mistake. That seems right. The error would be (at least) the basic one of identifying the meaning of a word with its reference. But does it follow that an emotion (such as grief ) must be exhaustively intelligible: that I need no direct experience of grief to understand grief (i.e., not the word “grief ”) fully? Or that I do not understand grief (not “grief ”) better if I have experienced it? If
Wittgenstein showed that there are no hidden linguistic corners where linguistic understanding cannot reach, a further step would still be needed to show that all understanding is like, or is based on, linguistic understanding.
This is not to suggest that some (or all) experiences are ineffable. It is to say that an account of how words are understood, or rather a negative account of how they cannot be understood, may leave more to be said about understanding. Wittgenstein could have accepted this.
(e) A more elusive barrier, or set of barriers, to understanding may seem to be presented by the idea of understanding as interpretation. In the most general terms, if all understanding has to be understood as some kind of interpretation, then it might seem to follow that no understanding can be final, complete, or even correct. Different interpretations may always be available or possible. In more narrowly linguistic, literary, or textual terms, it might seem to follow that there can be no final, complete, or even correct understanding of an expression in speech or writing. Hence, it might seem, real or true meanings are inaccessible.
Sketched in these extremely general terms, such views sound like a rehash either of an extreme skepticism or a corollary of the Kantian position already

FAILURES OF UNDERSTANDING

81

discussed in this chapter ([b] earlier). A textual or literary version can sound plausible, as noted in chapter 2 (d). Take a few obscure words from Heraclitus. There have been innumerable interpretations in more than two thousand years. No doubt there will be more in the future. Simply to conclude that no single interpretation can be right would be hasty (although that thought will have occurred to many desperate students). Undeniably, what Heraclitus intended by his use of words is beyond retrieval. What he has been taken to mean could now be taken as the meaning of his text. Undeniably, no understanding can be final or complete. So does it not follow that there can be genuine limits to understanding? We can only understand (as it were) past and present understandings of the text, never the author’s meaning in itself. So understanding must be intrinsically historical. What is more, it cannot avoid taking into account our own position as interpreters, and that must be understood historically as well.
Here there may be some risk of dwelling on views that no one wants to hold. It might be hard to find anyone who has signed up to all the conclusions in the previous paragraph. There were some rash words from Nietzsche in his unpublished notes: “The same text allows for innumerable interpretations: there is no “right” interpretation,” or “Against positivism . . . ‘there are only facts,’ I would say: no, there are no straight facts, only interpretations. We can establish no fact ‘in itself ’: perhaps it is nonsense to want something like that.”
The perspectivism of his published work sounds less conclusively dogmatic.33
Foucault adopted somewhat similar views:
If interpretation can never be completed, this is quite simply because there is nothing to interpret. There is nothing absolutely primary to interpret, for after all everything is already interpretation, each sign is in itself not the thing that offers itself to interpretation but an interpretation of other signs.

(Derrida, often associated with such a position, has actually been much more circumspect, for example, in his remarks in Spurs on Nietzsche’s fragment “I have forgotten my umbrella.”)34
Two points can be made. First: a literary or textual model for understanding may be attractive for literature or for the reading of texts. There, it has unsurprising merits over, for example, models of understanding as direct vision or as explanation through natural laws. But how far can it be generalized beyond its grounding context? Is there any value, for example, in taking understanding through the formulation of physical laws as a kind of interpretation, in the manner of reading a text? There are ways of doing this, where theories are seen as guiding metaphors, or as forms of discourse, and where apparent differences between them are seen as incommensurabilities. But something more than the inductive extrapolation of a model of understanding would be needed to sustain such conclusions. A model that can be illuminating in one context acquires no particular legitimacy for any other, especially

82

UNDERSTANDING UNDERSTANDING

when other models may make more sense. In areas where the appropriate style of understanding itself is truly uncertain, a model of textual interpretation is not immediately appealing. This applies obviously to the understanding of ourselves, of other people, or of other cultures. “To read someone like a book” is a familiar enough figure of speech; although in English it actually suggests direct transparency rather than mediated interpretation. It may be that people’s lives are understood interestingly on the model of literary narratives, although some argument would surely be needed to take this beyond a stimulating suggestion. On the other hand, biographical narratives can hardly be altogether autonomous, without some regard for natural facts. They would never make sense outside the obvious biological constraints of human lives.
Second, even with a narrow context of literary or textual interpretation it may be hard to extract significant conclusions on limits to understanding.
From the fact that innumerable interpretations can exist it does not follow that one may be not correct, nor that many be demonstrably incorrect. From the fact that no understanding of anything may be final it does not follow that any particular understanding is importantly constrained. The intentions of an author may be irretrievable (not merely in practice but in principle in many cases, not only in the distant past). Again, it does not follow that many possible understandings may not be incorrect. Nor, outside a limited range of special cases, does it follow that some “real meaning” is beyond understanding. It may be reasonable to say of a few lines of an ancient inscription: this must have a meaning, but we shall never know what it is. It would be perverse to say that of a few pages from a nineteenth-century newspaper.
That conclusion may seem wrong if meaning is understood as an author’s intention. Indeed, it is true that many authors cannot describe their own intentions adequately at the time of writing, beyond the use of the words they chose to write (Did Shakespeare “mean” something else by “Shall I compare thee to a summer’s day?”). It is also true that many authors (and not only bad ones) have a feeling that they cannot match their words to their intentions. Since the end of the nineteenth century a good deal of literary experimentation has tried to show how the sense of a text may be independent of an author’s conscious or explicit intentions, and how any understanding of a text may not be limited to a single, “correct” reading. The battle lines between classicism and romanticism were drawn out fruitfully in an exchange of literary articles in 1896. From Proust, there was a wish for clarity: nature “enables every man, during his passage on earth, to explain the most profound mysteries of life and death clearly.” (The narrator of À la recherche du temps perdu at one point realized that his duty was de tâcher de voir plus clair dans mon ravissement, “to try to see more clearly in my rapture.”) From Mallarmé, “the dreary level that suffices for intelligibility” in writing, had to be distinguished from something hidden, at the base of everything.35 The reader would be liberated to construct understandings of the text, but, in any event, the poet’s

FAILURES OF UNDERSTANDING

83

intentions in writing might remain elusive. That was hardly a new thought in itself, as can be seen from Plato’s Phædrus, where the source of poetry was given plainly as mania.36
(e) Failures in understanding take on practical importance in politics, society, or culture. No one can doubt that they can lead to intolerance, persecution, and war. If we are interested in understanding, we need to know whether such barriers are removable, or whether we have to accept that there are aspects of other peoples’ beliefs and practices that really are inaccessible.
(The addition of “in principle” here would be problematic. What would it mean to say that a member of one tribe or sect in an ancestral feud “could” understand the position of the other side, but only in principle?) And, once more, it is problematic whether we can generalize over differing fields of understanding. A person might feel completely unable to understand a particular style of painting, despite many efforts and much persuasion. Is this the same kind of failure as where someone just cannot understand how anyone could accept the life of a contemplative monk? The differences might be as significant as any similarities.
You can understand why the members of a society feel obliged to eat their dead relatives (in the sense that you have read all the relevant anthropological explanations and can expound them competently), but at the same time you can feel that you do not understand this at all (in the sense of some kind of empathy or insight). If the understanding of other cultures or societies is reduced or assimilated into linguistic understanding—we understand what they say, or how their practices are described—then failures or barriers can be nailed far more readily than with vaguer models: than with sympathy or empathy, for example. But there can be a temptation to be too tidy: to make a distinction between a set of beliefs, which can be described, explained, and therefore understood, and the holding of those beliefs, which may be thought to be somehow subjective and therefore more resistant to empathy or understanding. That temptation arises in particular with comprehensively formulated dogmatic tenets. You can write books on the theory of transubstantiation while feeling unable to understand how anyone could ever really accept it. This is a case where what is believed, in the form of a set of propositions, has been identified, codified, and refined to a point where it is supposed to be beyond ambiguity. So its acceptance or non-acceptance may seem equally clear: yes or no. There may be some beliefs like this but they must surely be few and exceptional. This theological example is itself ironic, given the fractiousness of the subject. Successive church councils aimed to define exactly what the faithful should and should not believe, in extraordinary detail. Yet who could say whether the beliefs about life after death among the mourners at a single funeral (even in our own culture) were all the same? How could one be sure? By using an ethnographic questionnaire?
Who would frame the questions?

84

UNDERSTANDING UNDERSTANDING

It is possible to take an optimistic, rationalizing view that understanding must be achievable. Where priority is given to linguistic understanding this attitude may be inevitable. As just argued in (d), propositions that are unintelligible in principle would be hard to defend. This must also apply where concepts, frameworks, or schemes of belief are taken in primarily linguistic ways (as they were by Donald Davidson in “On the Very Idea of a Conceptual Scheme”).37 Unfortunately, this gets us almost nowhere in the real world.
It may be reasonable enough to believe that we can always, in principle, understand each others’ language or concepts, but it seems unduly hopeful to conclude that we can always understand each other. Alternatively, appeals to incommensurable “mentalities” or rationalities seem merely question-begging.
So the discussion that might be expected at this point on “relativism” would be an unnecessary detour. If it were possible to set up mutually incomprehensible, isolated blocks of discourse, each with its own independent legitimation, then this might indeed present a problem; but the challenge is surely to show that this possibility has any content. It may have no resemblance to real ethnographic cases.38 For the purpose of thought-experiment in the days of Cold
War politics, or in theology before the Second Vatican Council, it might have sounded challenging to envisage mutually unintelligible dogmatic systems, but that project was never realistic in any literal sense.
Equally, it is not hard to construct pessimistic scenarios where understanding seems unachievable. Again, the depths of hatred in the Balkan Wars of the 1990s, where close neighbors raped and murdered each other, can seem genuinely unintelligible, even when all the appropriate explanatory literature, history, and commentary has been properly absorbed. The kind of understanding that you want may be accessible through imagination or vivid personal testimony, or it may seem elusive itself. Not only may you not be able to imagine what it would be like to want to kill your friendly neighbor, but you may not even be able to imagine how you could imagine that. This of course could be your personal failing, which might not be shared by a competent writer who might seek to recreate and convey some of the sentiments you may be too unimaginative to grasp. Or—more gloomily—there could be some genuine barrier, beyond which explanation is of no help. (But then the notion that “tribal warfare” is somehow unintelligible may be not without its own political assumptions.)
A barrier is of course an image. It may be helpful to speculate on how far assertions about irremovable failures of understanding between social or cultural groups depend for their force not on argument as much as on imagery that is predominantly spatial. From where you stand, the position of another society or group may seem unintelligible. You may seem wholly outside it, or your horizons may only partly overlap. (For Gadamer, the horizon is “the range of vision that includes everything that can be seen from a particular vantage point. Applying this to the thinking mind, we speak of narrowness of horizon,

FAILURES OF UNDERSTANDING

85

of the possible expansion of horizon, of the opening up of new horizons, and so forth.” For John Donne [in 1627], “We limit, and determine our consideration with that Horizon, with which the Holy Ghost hath limited us.”)39
Understanding may only seem possible from within one group or another, and you cannot be in more than one place at the same time. Inability to understand in practice may be explained by lack of imagination or ignorance, but there may also be a barrier at the frontier between two frameworks of beliefs. Such language may be natural enough (it has its place), but it is likely to mislead.
Spatial imagery obscures our ability to pick, mix, and criticize, and it ignores the results of reflective understanding. There is no room in such language to locate the understanding gained from appreciating where you may be. To understand in that way, by finding where you are, is neither to be located inside nor outside a position: the relativism suggested in the imagery of areas is wholly illusory. The thought that you may be isolated, or even nowhere, if not within one or another set of beliefs should have evident failings. Historically, maybe it is true that understanding starts at home. You must be somewhere is true literally, but it translates only misleadingly into metaphor.
It is interesting, though, that time can be relevant, where spatial imagery may be misleading. If you do come to understand, in whatever way, several different cultures, societies, or ways of life, then your understanding of one—your own—may be altered irreversibly. (Descartes, who was a great traveler in his youth, remarked that if you travel too much, you become a stranger in your own country.)40 Once you do see how a proof works it may be impossible not to see it, or even to see how you could not have seen it. That may only be a psychological curiosity, in that it is all too easily possible to forget proofs or to lose a grasp of them. It is quite possible to stop understanding what you understand now. Nevertheless, not understanding what you do understand now—what you have come to understand—may be impossible. So may understanding how you could not understand it (“how could I ever have not understood that he was a liar?”). Proust’s narrator, in coming to understand for himself the reality of the deaths of his grandmother (in the crucial passage entitled Les intermittences du cœur) and of Albertine, both some time after the event, passes into a new phase of existence, unimaginable before. He finds himself to be a different person. His frequent apprehension about the future often centers on a fear of becoming a different person, who will lacks the feelings he has at that time.
The asymmetry of time gives a more convincing explanation for apparent barriers to understanding than a language of space, based in positions, horizons, inside or outside, and so on. If you go beyond a position of provincial narrow-mindedness to a generous cosmopolitanism, the barrier between your two conditions is of course temporal, not spatial. The reason you cannot go back, at least at any one moment, is not that you cannot travel to there, but because you cannot travel to then. More strongly, there is a great deal about the past

86

UNDERSTANDING UNDERSTANDING

that no one can ever understand in any sense, simply because it has gone irretrievably, but there is no reason to dramatize this into a general historical skepticism. Much about the past can be understood, but not everything.
(f ) Levinas wrote that Western philosophy was “a reduction of the Other to the Same.” Philosophy has been like the story of Ulysses who “through all his wanderings only returns to his native island,” and: “To the myth of
Ulysses returning to Ithaca, we would like to oppose the story of Abraham leaving his country forever to go to a still unknown land and forbidding his servant to take even his son back to the point of departure.”41 This can be given a linguistic, Wittgensteinian interpretation: unless I could take part in discourse with others I could not even talk to myself. Recast in terms of understanding: until, or unless, I understand other societies and other people, I do not (or cannot) understand myself. That makes a suggestive thought, but one hard to assess. In banal terms, an ethnographer who studied only one (home) society would not learn much. A psychologist who relied only on introspection would have a limited view. A more productive reading might be in terms of standards and priorities in understanding. Why assume that an understanding of others, or of other societies, is less good than your understanding of yourself? Why start with yourself? Why not just as well the reverse? Still more pointedly: Why assume that an understanding of yourself, or of your society, is to be taken as the standard to which other understandings have to be compared or, perhaps, reduced? Why not—and this notion certainly is in Levinas—accept the irreducible otherness of other people and societies, instead of trying to measure them against a template of your own making? The urge to compare or reduce to my standards leads to an inevitably skeptical problem about understanding—I will always understand others less well than I understand myself.
This might add up to a diagnosis of the source of some supposed barriers to understanding. It might be asserted, for example, that a difference in gender constitutes one irremovable barrier in understanding—that there are at least some ways in which women and men can never understand each other, and where some common, “neutral” understanding will remain elusive.
The same might be proposed of cultural, religious, or ethnic differences.
Whatever is made of such claims—and however hard they are to substantiate convincingly—they are all undercut by a view that otherness or difference is not to be overcome, as a problem, but is to be accepted and welcomed. At the very least, this approach does make us ask what kinds of understanding we want, and why. A warning against understanding by excessive reduction or simplistic overclassification should hardly be necessary (“Why can’t a woman be more like a man?”). It need not follow, on the contrary, that there have to be distinct, isolated styles of understanding, only that some types of understanding suit some subject matters better than others, and that a fixed order of priority between them is not helpful. The intuitions of a number

FAILURES OF UNDERSTANDING

87

theorist and the intuitions of someone gifted with warm personal insight are evidently different, with different scopes, although neither can be distilled into a clear definition that would provide help to a relativist in separating them into independent fields.
The order of priority that Levinas wanted to resist was, of course, Cartesian. Like many twentieth-century thinkers, he tried to argue that a startingpoint in the isolated self and in subjective, first-person knowledge was itself dependent on some kind of third-person objectivity. In his terms, the self could not precede the other. “In Descartes, the cogito depends on the Other who is God and who had put the idea of the infinite into the soul.”42 Whether or not such arguments were effective in general, it was argued in Chapter 3 that a priority for subjective, first-person knowledge need not imply any priority for subjective, first-person understanding—insofar as that notion has any sense at all. The Cartesian myth of suspending as much of my knowledge as possible by asking what I cannot not-know cannot be mirrored so convincingly with understanding. What would I have to pretend to myself that I do not understand? My own language?
A survey of supposed failures in understanding cannot pretend to be comprehensive, any more than the surveys of objects and types of understanding in chapters 1 and 2, but it should include the main varieties. It would miss the point to extract general conclusions from such diversity. There has been some confirmation of the elusiveness of understanding as a target for philosophical theorizing. It is possible to set up standards or norms for understanding—the intellect of God, the rational mind—and to measure how far the power of human understanding falls short. But this is only as convincing as the norms or standards taken as the starting points. It is possible to align the unintelligible with whatever cannot be explained in some favored way—through scientific theorizing, for example. It is possible to insist that all understanding must be primarily linguistic, and then to go on and try to delimit the boundaries of linguistic expression. But the arbitrariness should be obvious.
On the other hand, it should not be surprising that failures of understanding should be a subject of continuing philosophical attention, and that similar-looking accounts should emerge in varying costumes. Models of understanding based on mathematics (in the seventeenth century), on a priori psychology (in the eighteenth century), or on language (in the twentieth century) all led to assertions about what cannot be understood. A looser model based on artistic understanding does not lead in that direction. So maybe it was no accident that hermeneutics became a subject of great interest in the romantic era. If any consensus was ever reached in the history of modern epistemology, it was around the acceptability of scientific and mathematical knowledge. It was not necessary to be a positivist to accept that at least the deliverances of, say, experimental physics were to count as knowledge. With

88

UNDERSTANDING UNDERSTANDING

understanding, not even such a minimal consensus could ever have been likely.
A suggestion that understanding might be modeled (typically) on our understanding of the recurrence of comets or of Euclidean geometry was never likely to impress deeply, or for long. It does not seem too rash to speculate that the appeal in aesthetic understanding in the nineteenth century was a reaction against the arbitrary boundaries implied by too narrow an admiration for the physical sciences.
The assertion that there is nothing that cannot be understood—that there are no limits to understanding—sounds like either hubris, or else idealism, where the human mind is fitted exactly to the world it has to understand.
The apparent alternative—that something may be beyond understanding— sounds even worse.

Chapter Six

Beyond understanding

In a letter of 1649, Descartes wrote to his friend Claude Clerselier: it is sufficient for me to understand the fact that God is not grasped by me in order to understand God in very truth and as he is, provided I judge also that there are in him all perfections which I clearly understand, and also many more which I cannot grasp.1

The translator catches the distinction between what Descartes said he could understand (intelligere) and what he could not “grasp” (comprehendere). The implication must have been that he could understand some of what he could not grasp. That distinction was no doubt meant to sound satisfyingly tidy, matching parallel distinctions between the light of natural reason and the light of faith, or between a knowledge that God exists and a knowledge of what God is. Nevertheless, the impression falls far short of Descartes’s usual confident clarity. What he was actually saying was that God was not properly grasped—was beyond comprehension—but was understood to some extent. A blunter Descartes might have concluded that understanding must be an inferior, inadequate form of grasping, only delivering inadequate information; but of course that would not have been the proper emphasis at all. Beyond mortal, human understanding lies not confusion but the infinitely superior understanding of God and angels.
The archetypal limit on human understanding may seem to be mortality itself. As Boyle put it:
It seems not . . . unreasonable to think, both that God has made our faculties so limited, that in our present mortal condition there should be some
89

90

UNDERSTANDING UNDERSTANDING

Objects beyond the comprehension of our Intellects (that is) that some of his creatures should not be able perfectly to understand some others, & yet that he has given us light enough to perceive that we cannot attain to a clear and full knowledge of them.2

From a different—superior—perspective, things will look different. A reluctance to accept anything beyond understanding may be connected with an unwillingness to contemplate any nonhuman perspective. The ineffable, the mystical, and the transcendent may all seem to be threats to the power of reason or the effectiveness of language. The adequacy of human understanding or “reason” of course has been a central strand in Western thinking about religion. Newman listed as one of the errors of liberalism the proposition that
“No one can believe what he does not understand.”3 It may be just that sort of idea that makes philosophers uneasy about a view of anything beyond understanding. Thomas Nagel has written of a fear of religion; by which he means not a fear of religions or religious institutions, but an apprehension that religious claims might be justified. As he puts it, “I don’t want the universe to be like that.”4 Maybe it matters to us that nothing is in principle beyond understanding because we feel that we need to be able to understand how things are.
Or there may be contrary reasons for an aversion to unintelligibility, as expressed by Gillian Rose:
To argue for silence, prayer, the banishment equally of poetry and knowledge, in short, the witness of “ineffability,” that is, non-representability, is to mystify something we dare not understand, because we fear that it may be all too understandable, all too continuous with what we are—human, all too human.5

There could be a further, anti-esoteric wish, that understanding should be openly and freely available (again “in principle”), rather than accessible only to the initiated or privileged. In many societies for much of history, understanding would not have been available on level terms to everyone. It would have be restricted by caste, gender, or religious initiation. The thought that anyone “could” “in principle” understand a ritual formula used by a priest or shaman might be just unconstructible. “Higher” forms of illumination might only be available to the initiated. Less colorfully, in the mainstream
Western tradition, understanding might only be available to believers or participants. The ambition that understanding can be for anyone may be seen as a declaration or stipulation, though there could be some embarrassment in pursuing it too far. A suggestion that “in principle” nothing is to be hidden could turn out to be arbitrary.
Religion seems to be where to start in thinking about what could be beyond understanding, partly because it is where the boldest claims can be made. This may also apply negatively, in that those disinclined to think at all about religion may feel like that just because of a suspicion of the ineffable or

BEYOND UNDERSTANDING

91

the allegedly unintelligible. But, apart from religion, there are other areas of equal importance. Great evil and great goodness have both been felt to be beyond understanding in a sense that interpretations or reductions may seem banal, trivial, or inadequate. The same may apply to great misfortune or injustice. The standard text for debate on these themes was for many centuries the
Book of Job. As the unfortunate Job said at the end of his ordeals, “I spoke without understanding of things beyond me, which I did not know.” His story, one commentator suggests, “irrespective of its historical veracity, is the biblical warning not only to the pious but also to the philosophers about a perpetual need for safeguards against intellectual hubris.”6 In neutralized, nonreligious terms, the point might be that not only will no type of understanding fit the narrative satisfactorily, but we have no idea of the type of understanding that might be satisfactory. A prosaic response might then be that there is just nothing relevant or interesting to understand (as with “Why is rain falling on me now?”). Less dismissively, there might be only unsatisfying forms of understanding (as with “Why did my child have to die?”).
Other styles and objects of understanding from the first two chapters can give rise to similar lines of thinking. There is no agreement on how human beings or “the mind” should be understood, or on whether any style of understanding could be comprehensive. One scientific ambition could be that a single style—physiological or neurological—might predominate, rendering other styles—such as understanding through humanistic, geisteswissenschaftlich narratives—redundant, as folk psychology. At the opposite extreme would be an equally unconvincing relativism, where an open variety of styles of understanding might be accepted for different purposes, none of them exclusively right or wrong. In between lies hope, as for Colin McGinn:
“surely there is something about brains that makes them conscious, whether we can know and understand it or not. We should persist in the hope that some day philosophy (or perhaps science) will find the answer.”7 This falls short of the dogma that everything must be intelligible while relying on its methodological, regulative force.
Even so, more generally, as can be seen from some of the examples in the last chapter, it is not hard to argue that nothing is, in principle, unintelligible.
A case need not amount to some version of a principle of sufficient reason.
Much more weakly, it could be simply a refusal to accept that some explanation cannot be devised to cover any data—or, in different terms, once more, that an appropriate form of understanding cannot be found for any circumstances. This line gains strength from the implausibility of many of the candidates for mystery. It is true, continuing with the same example, that there is no explanatory theory that links the physiology of the brain with conscious experience. It is also true that no one has any idea what such a theory would look like, what it might do if successful, or how it might satisfy any sense of non-understanding that we now feel. But it does not follow that we might

92

UNDERSTANDING UNDERSTANDING

want to exclude the future formulation of a working theory. This must be the most common, promissory technique for driving back mystery. We have no idea what to say, or what might be said, but someone may say it some day.
Such optimism has an allure even in technical contexts where wishful thinking might seem out of place. In the conclusion to their elementary text on
Gödel’s theorems, for example, Ernest Nagel and James Newman took a brisk approach: Gödel’s proof “should not be construed as an invitation to despair or as an excuse for mystery-mongering. . . . It does not mean . . . that there are
‘ineluctable limits to human reason’. It does mean that the resources of the human intellect have not been, and cannot be, fully formalized, and that new principles of demonstration forever await invention and discovery.” But these
“new principles” waiting forever to be invented and discovered went well beyond Gödel’s conclusions (or his own, rather gloomy view of the world).
The moral to be drawn was breezily upbeat:
Nor do the inherent limitations of calculating machines imply that we cannot hope to explain living matter and human reason in physical and chemical terms. The possibility of such explanations is neither precluded nor affirmed by Gödel’s incompleteness theorem. The theorem does indicate that the structure and power of the human mind are far more complex and subtle than any non-living machine yet envisaged. Gödel’s own work is a remarkable example of such complexity and subtlety. It is an occasion, not for dejection, but for a renewed appreciation of the powers of creative reason.8

No doubt the authors were right that keep trying is a more constructive methodological maxim than give up, but the supporting argument contains more bluster than content. If one form of understanding does not work, let us hope that another will. We have no reason to expect this, but neither do we have any reason to exclude it.
In the twentieth century, when linguistic understanding came to seem so central, a notion of being beyond understanding might seem to have been almost excluded. A strange case was the early Wittgenstein. In 1929, he spoke of “our ethical and religious expressions”: “I see now that these nonsensical expressions were not nonsensical because I had not yet found the correct expressions, but that their nonsensicality was their very essence. For all I wanted to do with them was just to go beyond the world and that is to say beyond significant language.” For him, what really mattered lay beyond significant language—beyond saying. “It is clear that ethics cannot be put into words,” just as “God does not reveal himself in the world.” It was necessary to overcome, transcend, or go beyond (überwinden) the propositions of the Tractatus to “see the world aright.”9 These remarks illustrate the point made in the previous chapter on the lack of fit between meaning (or making sense) and understanding. The early Wittgenstein believed that sense cannot be made— nothing can be said—about God and ethics. Apparent sense is really Unsinn:

BEYOND UNDERSTANDING

93

nonsense. And yet it is evident that a good deal can be understood somehow.
At any rate, whatever was beyond sense was plainly not beyond understanding
(or, presumably, we would not be able to understand was he was trying to tell us). This anomaly was removed in his later work by an insistence that “nothing is hidden.” That remark was applied to the way in which sentences represent, about which nothing could be “said” in the Tractatus, but it can be taken more widely (as it has been, in a book by Norman Malcolm).10 By the time of the
Philosophical Investigations, any rigid link between the way in which sense was made and the way in which it was understood had been severed. Just as there was no essential way of making sense, so there was no essential process of understanding. The way in which sense was understood (dismantled) did not need to match the way in which sense had been made (constructed). The key points about understanding were negative or critical: nothing had to happen
(particularly not as a process in the mind), and there was nothing hidden behind, above, beyond, or within an utterance that was the sense to be understood. What Wittgenstein wanted to deny was that any act of meaning needed to be grasped behind the meaning of what was spoken or written. Meaning was not hidden in the consciousness of a speaker or writer. Whereas in his earlier work everything important was beyond saying, in the later work nothing was beyond saying—“everything lies open to view” or, elsewhere, “the meaning of a word is what the explanation of its meaning explains.” This was not a subversion of his view in the Tractatus, that “everything that can be put into words can be put clearly,”11 as much as an extension of it to all discourse: anything that could be understood was as clear as it needed to be. Then, as Elizabeth
Anscombe said, the outcome of such a view could be that “Mystery would be an illusion—either the thought expressing something mysterious could be clarified, and then no mystery, or the impossibility of clearing it up would show it was really a non-thought.” But she added, “The trouble is, there doesn’t seem to be any ground for holding this position. It is a sort of prejudice.”12 This was a shrewd diagnosis of a philosophical hubris. Taking it further, the belief might be that a suitable form of understanding can always—in principle—be found.
It is easy to dilute the strength of that belief. It is not that we can persuade ourselves that we can always understand in different senses, rather as with Descartes’ understanding without grasping. (Newman tried “apprehending” and “understanding” along similar lines, in A Grammar of Assent.13) That path leads to an untenable relativism, with different forms or degrees of understanding, each vindicated in its own terms. (In the background looms an untenable distinction between literal and nonliteral senses, each understood differently.) It is not that some statements may always be accepted on trust, without being fully understood, in the hope that fuller understanding will develop later. (That would be like signing a contract in a foreign language because a friend told you that it was all right.) These would be painless but unconvincing ways to abolish mystery.

94

UNDERSTANDING UNDERSTANDING

The real reductive bite in the dilemma sketched by Anscombe comes from the assumption that an expression must be intelligible somehow. If it is not intelligible, it is not an expression—a unit of sense—but something else— a noise or a set of marks on paper. This is scarcely interesting if taken as the tautology that a meaningful expression must be meaningful. Rather, the thought must be that meaning is conveyed and understood; that communication does succeed. Maybe it is right that if language is to be a means of communication there must be agreement in judgments, as Wittgenstein said.14 But then what if communication fails, or is less than fully successful, or is impossible? That, after all, could be part of the point in any insistence that there may be something beyond understanding. Wittgenstein’s conviction may have been that I could not even communicate successfully with myself—record my own thoughts and experiences—unless I used a working common language.
Quite so; but what if I do not understand myself at all well, or if I am unable to identify and articulate my thoughts and feelings even to myself, even to my own satisfaction? One possibility, as in one side of Anscombe’s dilemma, would be that this would be a case of “really a nonthought.” Another, in plain terms, might be that some of my thoughts and feelings were beyond my understanding. There would be no point in asking what they were. The most striking examples will be found in psychiatric cases of severely confused identities. Someone who has been brought up in a family in which the close relationships are obscured can genuinely fail to understand what is going on, to the extent of even not understanding what is being said.15 The possibility of successful communication may indeed not exist. The transcendental argument that misunderstanding is made possible by understanding hits the buffers where misunderstanding actually prevails as the norm.
To identify understanding only with articulation can be one route toward ensuring that nothing is beyond understanding. Nothing can be said about whatever cannot be said, so any inability to understand it seems uninteresting.
Using a metaphor from the familiar visual repertory, transparency can be provided by language: the correct but limited truth in the insight that “everything that can be put into words can be put clearly.”
Far more elusive is the thought that something which can be put into words might still be beyond (or “above”)16 understanding. In the most naïve way, you might be able to understand the words but not what they meant. This has been classic territory for debates over religious language: not only over socalled negative theology or the reports of mystical experiences, but over simple-looking assertions such as “God became man.” Yet again, the difficult point would have to be not merely that such assertions may be understood in different senses—literally, analogically, symbolically, metaphorically, poetically, systematically, spiritually—but that genuine paradoxes can be created.
There can be something, as it were, usefully informative about a piece of language that really cannot be understood. This seems an unlikely and unattrac-

BEYOND UNDERSTANDING

95

tive prospect, as must be evident from the search for different senses or modes of expression that can avoid outright contradiction. Unless you suppose that words possess fundamental, literal senses whose understanding must precede the understanding of other presumed senses, there seems to be no reason at all to believe that you cannot (“really”) understand anything that makes sense for you. Nor is it necessary to be able to say, or even to know, what that sense might be. Just as countless writers in the late twentieth century argued against meanings behind the use of language, so it is possible to argue against meanings behind or beyond understanding.
Concentration on linguistic examples, or a reduction of other forms of understanding to linguistic versions, is likely to reinforce a reductive or critical view against anything of interest beyond understanding. Even if you are convinced that meaning must always be indeterminate, it should remain true that you understand only what you do understand, and not something else.
The meanings in Finnegans Wake are inexhaustible. It must be correct that they are all beyond the understanding of the most indefatigable commentator.
But it does not follow that there is any meaning that is beyond understanding. The words as they are will be understood less, or more. The fact that their understanding will be incomplete means that there is always more to understand, and not at all that the understanding has to come to an end, falling short of some full meaning.
But, as chapters 1 and 2 should have shown, there is no reason to concentrate only on linguistic understanding. Only an arbitrary stipulation could entirely rule out wordless understanding: a stipulation that you cannot be said to understand something unless you can express it articulately to yourself.
There is no need to go to mysticism, for examples. A painter’s understanding of color or proportion may be expressible in words, but equally it may not be.
Perhaps it might not even be expressible (behaviorally) in terms of competent practice. An operagoer’s understanding—of what is happening on stage—can hardly be reduced to an understanding of the words in the libretto. The illusion that reality lies hidden behind a veil of words has been unmasked so many times that its appeal must surely be exhausted.
Not only among writers in analytical traditions has there been some inclination to treat activities, behavior, narratives, or rituals in propositional terms.
The tendency to do this with ethnographic and religious understanding is a familiar source of confusion. Practices or rituals in a group or society are taken as implying the acceptance of beliefs. In turn, those beliefs are treated as though they consist of propositions believed. Given the weird distortions such treatment can create, it is hardly surprising that one reaction has been to protest that the reality of some beliefs or practices lies beyond language, and hence, perhaps, beyond understanding. Thomas Nagel has argued for “the existence of facts beyond the reach of human concepts.” His acceptance of
“facts that do not consist in the truth of propositions expressible in a human

96

UNDERSTANDING UNDERSTANDING

language” makes it possible for him to believe that there are facts “which humans never will possess the requisite concepts to represent or comprehend.”
Yet even if there could be such extralinguistic facts, it would only follow that we could never understand them if it was assumed that all understanding had to be linguistic.17
An identification of beliefs with their “contents”—with propositions allegedly believed—may well have been a consequence of debates in particularly Christian history, which distilled matters of belief down to lists of credal assertions to be accepted or denied. Even as orthodox a figure as St Thomas
Aquinas glimpsed the point of difficulty. He remarked that the Apostles’
Creed does not say that God is almighty but “I believe in God almighty.”18 He thought, at least, that the belief represented more than the acceptance of some proposition. He may have thought that the “more” was beyond or outside language, either as an object of belief (de re) or an act of believing as a disposition or practice of trust. A different viewpoint, taken, for example, by Kant,19 might be that the whole enterprise of trying to nail down what people thought in any set of propositional beliefs was hopeless. It was the very articulation and presentation of belief in linguistic form that was misleading. This had gained plausibility in a Christian context just because key tenets could be isolated for acceptance or rejection. Beliefs were supposed to have some noncausal support, encouraging a desire for their clarification and codification. It might be true that the members of a group would all hold the same beliefs if they all assented to the same list of propositions and no others. But that paradigm of uniformity would be absurd.
In the simplest, individual case you would be wrong to think that you can understand other people, their way of life, or their beliefs, just because you understand what they say—what they tell you—about themselves, their way of life, or their beliefs. It would be still worse to think that you can understand them because you can understand what you say yourself about those people, their way of life, or their beliefs: an ethnographers’ fallacy.
Only a narrowing of understanding to linguistic understanding leads to these traps.
Lafew: They say miracles are past, and we have our philosophical persons, to make modern and familiar, things supernatural and causeless. Hence is it that we make trifles of terrors, ensconcing ourselves into seeming knowledge, when we should submit ourselves to an unknown fear.
Parolles: Why, ’tis the rarest argument of wonder that hath shot out in our latter times.20

It is not too difficult to be persuaded that nothing of any interest lies beyond understanding. The openness and flexibility of understanding means that it can be extended or adapted without systematic restriction. If something can-

BEYOND UNDERSTANDING

97

not be understood in one way, then another can be found. That does not mean that everything must be intelligible, only that there are no reasons to assume anything significant is not.
The trouble with this kind of critical philosophizing is that two questions can recur: Does it work? and: What are you worried about?
The second question has ready answers. As well as the prospect of unintelligible divine providence or judgment, mentioned already, there is the possibility of unfathomable evil, irrationality, the unreachable otherness of other people, the depths of one’s own thoughts, feelings, and actions. With all these, understanding can feel all too like rationalization in the most negative sense— feel because this can be an area where argument can be unconvincing. On the other hand some may feel that the whole value in philosophy is in shedding the light of reasonable argument on to such darkly unreasonable areas.
It is possible to generalize critical arguments from obvious cases. How to understand, for example, that a city full of people can be wiped out by an earthquake, plague, or massacre? The rhetoric of bewilderment and non-understanding can be shown to be misplaced. The geophysical, epidemiological, or political causes can be exhibited. A narrative of purpose or meaning can be excluded. Understanding can then be suited to the available explanations.
Whatever may be left unexplained will be a matter of ignorance, not mystery.
This was the strategy originated by Spinoza, starting from a trivial example— why did a stone fall off a roof onto a man’s head? Only two chains of causality were allowed: to explain why the man was walking that way, and to explain how the wind happened to blow the stone off the roof.21 The implication can be that more interesting examples will differ only in scale, not principle. How to understand the killing of six million people in European death camps between
1942 and 1945? Study European history and there are the causes.
Any sense of dissatisfaction need not derive from the crudity of the explanations. It might be possible that in some cases, no explanation will satisfy.
You could say, taking Spinoza’s example (though certainly not his point of view), that the falling of the stone on the man’s head was the providential will of God. The man was being punished for his sins, even though these may not have been evident to anyone. Or, more seriously, religious narratives can be
(and have been) devised to encompass the killing of six million people. So explanations may be produced. What is to be the standard for understanding?
Will it be left as merely a removal or resolution of puzzlement? One scholar writes, “for all our knowledge, the world of the camps continues to offer an ever-receding horizon that seeks to escape our understanding every time we seek to close in on it.”22
Some people may feel happy with bluntly causal explanations. Others may require narratives that confer personal significance. Others may refuse to be satisfied at all, and may regard any attempts at comprehensive understanding as inadequate. After all, the belief that everything cannot be understood

98

UNDERSTANDING UNDERSTANDING

could be as well or as little justified as the belief that everything can be understood. Why prefer one belief—ideology—superstition—to the other? Are any standards for a satisfactory explanation to be so lax that they can be wholly subjective? It would be disturbing if such important questions could depend on matters of taste or temperament.
In many cases, feeling as if you understand may be wholly irrelevant. It is amusingly possible to feel that you do not understand a subject at all while giving competent lectures on it, or while passing examinations in it. Here, the appropriate standard for understanding may be behavioral (if you can do it, then you understand it). Other cases may be less clear. I may feel unable to understand some disaster, despite a thorough acquaintance with all the facts and despite a wide choice of explanatory theorizing, some, perhaps, religious.
Am I just mistaken?
One answer might be that there are constraints of consistency. It is possible to imagine someone who wanted to understand everything that happened in terms of astrological significance. Births, marriages, illnesses, and deaths would all be “understood” in line with astrological theories. Nothing else would feel satisfying. The explanatory power of astrological theory might be subjective: it feels better to know that a friend has died because the planets are in certain positions, and so on. That state of mind could only be maintained along with a state of unusual ignorance. Sincere beliefs about astrological causality may be held, but they may not be held without inconsistency by anyone with the most minimal knowledge of physics, or perhaps with an ordinary confidence in day-to-day causality.23 There may also be some doubt whether it would be practicable to live as a genuinely consistent fatalist, astrological or otherwise.
Unfortunately, more convincing examples are likely to be far less decisive. Replacing astrological understanding by theology, for instance, leads to no clear result. Those who believe that God can be a causal agent in a physical, historical sense, literally making events happen, will not be able to avoid some problems of inconsistency with their own accepted physical beliefs. But more subtle understandings of divine providence may not fall into this trap (and, after all, theologians have had many centuries to work out a way around it). There are also constraints of consistency within systems of religious understanding. Theological “problems” about evil, most plainly, are created by a need for explanatory narratives that take into account both universal divine concern or causality and some apparently disagreeable events or history. Difficulty in understanding may be exactly—no more than—a difficulty in assembling a consistent (or at least plausible) narrative. A further point about theological explanations may be that they still leave a gap that cannot be covered by understanding. Solace may be provided by the thought that some disaster, seen from a wider perspective, is all part of God’s plan; but that plan itself may remain inscrutable. Mean-

BEYOND UNDERSTANDING

99

ing may have been conferred, but puzzlement is just moved on by one step—divine mystery supersedes a mysterious event.
Resistance to a critical view—a continuing desire for something beyond understanding—may be no more than a feeling, and one that is hard to defend; but it is also hard to eradicate. One may take the view, for example, that comprehensive understanding of human motivations and actions—one’s own or others’—can never be attained. In itself that might be no more unacceptable that an insistence that everything can be understood. The problem might come with the account that might be needed to explain how human beings were exceptions in a world where systematic, consistent understanding was sought. That could be difficult but not impossible. It could be argued that human beings possess some qualities—free will, intuition, or original creativity, for example—that set them aside from (or above) the normal course of nature, and therefore outside any generalized form of understanding. Kant’s approach was not too different from this, in his initial advocacy of the “freedom of reason.” In fact he actually used the terminology of not getting beyond the “intelligible cause” of free actions.24 But his works illustrate how elaborate a case may be needed to stand any chance of convincing us that human actions stand apart from a form of understanding that is accepted as normal for the rest of nature. It could hardly be enough just to assert that people are different. Noncircular reasons will be needed. A traditional theological rationale will do the job for those who can accept it. Pascal was one example: human understanding is damaged by original sin, unable to grasp even its own nature properly—“If man had never been corrupted, he would enjoy in his innocent state both truth and happiness with confidence”—but this will not be to everyone’s taste. Behind this may have been an even more radical augustinian view that “language belongs to man in his fallen state and can only operate within the context of his corruption.”25 Kant, for his part, needed to go as far as an entirely separate order of understanding, the whole paraphernalia of practical reason, the Moral World, noumena, and so on.
There may be two ways to explain why we may want to go on searching for understanding when critical argument should be discouraging. First, there can be a Weberian story about disenchantment. A context of significance, meaningfulness, or enchantment has been lost. So expectations for understanding should have been pared back accordingly. No modern person should expect to find meaning in the appearance of a comet or the eruption of a volcano. In the same way, to try and understand “why” a child has been killed by a horrible illness is simply a relic of medieval superstition. If we were properly rational, any quest for understanding beyond the scope of attainable knowledge would be considered similarly.
Much of the detail in any such historical thesis will be debatable: the starting date for modernity, the direction of cause and effect between social and intellectual trends, the sweeping generalization about what “we” are supposed

100

UNDERSTANDING UNDERSTANDING

to think. But the central idea is more than plausible. Any critical theory in philosophy needs some historical or anthropological underpinning to explain why we want to do something that the theory forbids: In Kant’s case, why we want to extend our reason beyond the bounds of possible experience; in the early
Wittgenstein’s, why we want to say what can only be shown. This may be a presupposition about human nature—A. W. Moore writes of “our aspiration to infinitude”26—people just are like that—leaving room for a factual objection that some people are not. Weber offered a hypothesis. Magic, and then religion declined, leaving large areas of experience lacking explanations which they had possessed in the past. It seems not unlikely that a desire for understanding can persist when the scope for it has diminished. No claim need be made about humanity—or modern, Western humanity—as a whole, only about the effects of education or enlightenment. As you become better informed, your unfocused sense of wonder will narrow into orderly research. Your search for understanding will systematize into an accumulation of knowledge. Miracles are past. The same process of rationalization is said to account for the flattening of esoteric understanding. The hidden insights of magi and priests lose their charm as knowledge become democratized. Understanding becomes, in principle, open to anyone: modern and familiar.
As an explanatory story, this may work well with the development of the natural sciences. Lightning ceases to be a sign of anger from the gods and becomes a measurable phenomenon of electromagnetism. To understand lightning, you only need to know about electromagnetism. Extrapolation from the past into the present and the future must be more speculative. We may well be resistant to a thoroughgoing electroneurological account of human thought because our ancestors believed human beings to be other than elaborate animals. There may be some areas where a sense of bafflement—an inability to understand—is really no more than a lack of necessarily detailed knowledge. It may be that any feeling that some depths of human nature may be beyond understanding is just a superstitious relic of past (mainly religious) beliefs. Perhaps so; but such feelings could also be right, or justified. It is certainly possible that one model of knowledge that has worked in the past may not apply universally in the future. There may be a systematic human science in the future. Understanding people—ourselves and others—may become a matter of teachable and learnable techniques—a matter of codified knowledge. Mozart’s creative facility seems beyond understanding, but that only means we do not know what was so special about his brain. It may be more than routine skepticism to feel cautious about that. (The long-running Soviet research project on Lenin’s brain never came up with anything.)
More generally, Weberian historical and sociological theorizing may be plausible, but it cannot assume any critical force. Human beings are evolved animals that appear to exist in line with identifiable causes in a large universe, and so on. Present beliefs in purpose or providence may be as redun-

BEYOND UNDERSTANDING

101

dant as past beliefs in astrology and magic; but that inference is purely inductive, and there may be some relevant differences. If someone feels strongly inclined to understand human life within a narrative of sin and redemption, against a wider narrative of divine providence, there may not be too much that can be said.
Historical or sociological accounts may not be meant to have any present or future critical application, so this may not be a shortcoming. A second kind of reason for a search for understanding is equally speculative but even more abstractly theoretical. A desire for understanding derives from puzzlement.
Puzzlement must always exist from some perspective of questions unanswered. Yet it can persist when questions are answered or—more importantly—the ground for them has shifted. This can be an historical point—as where it is obviously possible to worry about questions that ceased to matter many centuries ago. A stronger historical point—associated with Collingwood—might be that philosophical questions matter only within the historical context within which they arose. A still stronger point might be that wonder or bewilderment will be the consequences only of a particular point of view, from a particular perspective. When the point of view is changed, the desire for understanding diminishes, or should diminish.
Any feelings either that there is—or there is not—something of significance beyond understanding could derive from several misplaced startingpoints. An analogy between understanding and vision, for example, might lead one to think that the unintelligible can be as well grounded as the invisible. An acquaintance with the writings of critical philosophers might lead one to think that the mind is inclined to range beyond understanding just as it has been supposed to range beyond the bounds of possible experience or expression. To realize that those are false parallels should be to be cured of thinking in these ways. More tellingly, assumptions about the context for understanding may determine what seems problematic. Most plainly, if I start subjectively, from questions about what I can understand, then some areas—such as the perceived quality of your experiences—may be beyond my understanding.
If I start by assuming that the understanding of myself and the attitudes of my society is straightforward, then the understanding of others, and of other societies, may become problematic, maybe to a point where some elements seem to end up beyond (my) understanding. As seen in the previous chapter, such assumptions can mislead. It is scarcely surprising that a great deal is beyond literally my understanding. To become interesting, that subjective sense has to be generalized in a way which robs it of most of its force: to beyond the understanding of anyone, any rational person or anyone in a particular group. There is the further thought that a first-person, subjectivized perspective cannot be taken for granted as a starting-point. Why assume that your perspective has to be brought into relation with mine? Why not the opposite? A desire to understand can be a desire to reduce or assimilate. A respect for otherness and

102

UNDERSTANDING UNDERSTANDING

difference might alter that perspective, also altering what seems to need understanding, or to seem beyond understanding.
Again, it seems natural to slide toward an essentially critical line of argument—to begin by asking why we might want to feel that anything of importance might be beyond understanding, and to end up concluding that no likely reasons are good enough. They might be an historical hangover from a more enchanted past or they might be a symptom of starting in the wrong place, or from the wrong direction. Asking why anyone might be worried about the prospect of something beyond understanding itself becomes a therapy to cure the worries: they can be forgotten, or should never have arisen.
Does this work, though? For whatever reason, if you want to feel puzzled—if you like mystery—if you feel that there must always be something that you cannot understand, or that cannot be understood—maybe nothing can stop you. There is no reason to hold that our current modes of understanding are comprehensive, or that other modes of understanding may not be beyond our current understanding.
The prospect that anything beyond your—or anyone’s—current understanding might be interesting or, more strongly, particularly significant, is an entirely different matter. Both mystery and paradox have been thought to be significant in some long-lasting and widespread religious and mystical traditions. Some allowance must be made for exaggeration. Many mystical writers have managed to say a good deal about what cannot be said (as Russell wrote sharply of Wittgenstein).27 Ineffability has to be taken with a pinch of salt.
There was a wave of antirationalizing sentiment in the nineteenth century, insisting romantically that the most essential elements in human life were exactly those beyond mechanical understanding. Important choices might be just those beyond the reach of reason, where the dark force of the will came into play. “Where the understanding despairs,” wrote Kierkegaard, typically,
“faith is already present in order to make the despair properly decisive, in order that the movement of faith may not become a mere exchange within the bargaining sphere of the understanding.”28
Nineteenth-century writing was full of characters—Dostoevsky’s Devils contains an entire zoo of them—whose motivations and intentions were beyond their own understanding and, in some cases, may seem to test the understanding of the reader.
A philosopher’s critical response might be to remark that styles of understanding need not be limited to the mechanical or the pseudomathematical, or to straightforward visual imagination. Simply because something seems beyond one form of understanding, it need not be beyond another, or still less beyond understanding altogether. There could be no good reason to rule our the sympathetic understanding of feelings, though no one could think that this could be reducible to a mechanistic model. Equally, there is no reason to

BEYOND UNDERSTANDING

103

rule out the understanding of mathematical theories whose content cannot possibly be pictured by the most powerful imagination. If the activities of a
Raskolnikov or a Stavrogin really made no sense at all, then Dostoevsky’s plots would be unintelligible, which they are not.
But just as an attachment to mystery can be seen as a relic of romanticism, a repudiation of it might be (an even more dated) rationalism. Why not simply accept the otherness of the point of view of a different consciousness, or of a different society, without seeking to encompass it in one’s own way of thinking? Does it not give a narrow view of creativity to rule out unimagined possibilities? On the other hand, to place restrictions on the extension of understanding hardly seems liberal. There may be some political currents under the surface. On the one side, toleration seems to require the recognition of separate points of view that should not be assimilated into each other. From there, understanding can be presented as a kind of intellectual appropriation or colonialism. On the other side, mystery can be seen as a blockage to free inquiry. Liberty seems to require an absence of barriers to understanding as a prerequisite for a free trade in ideas. These are false opposites, based in an untenable relativism. There is no reason to adopt fields, frameworks, or horizons of understanding that either do or do not suggest their own limits. If understanding is denied a metaphor of location, then the question of getting beyond it or not getting beyond it should hardly arise.

This page intentionally left blank.

Chapter Seven

Wisdom

The usual suspect is Descartes. The first page of his first philosophical work made a bland identification of wisdom with knowledge: “The sciences as a whole are nothing other than human wisdom.” In the French Preface to his
Principles of Philosophy, about twenty years later, he started more promisingly:
“the word ‘philosophy’ means the study of wisdom, and by ‘wisdom’ is meant not only prudence in our everyday affairs” and then continued in what he clearly found to be a more comfortable vein: “but also a perfect knowledge of all things that mankind is capable of knowing, both for the conduct of life and for the preservation of health and the discovery of all manner of skills.
In order for this kind of knowledge to be perfect it must be deduced from first causes.” A page later he had settled back into his real theme: “In truth it is only God who is perfectly wise, that is to say, who possesses complete knowledge of the truth of all things; but men can be said to possess more or less wisdom depending on how much knowledge they possess of the most important truths.”1
Wisdom, in other words, was knowledge: more knowledge, more wisdom. That attitude, of course, has been decried both before and after
Descartes. It has not been difficult for critics to point out that knowledgeable people—and latterly scientists—can be conspicuously lacking in wisdom. (At this point, really unkind critics may bring up Descartes’ own fatally unwise career move to Sweden.) The apparent reduction of wisdom to knowledge has been blamed for the separation of values from the pursuit of facts and for widespread scientific irresponsibility, both allegedly endemic in modern Western thought and culture. One full-length critique opens the charge-sheet with
105

106

UNDERSTANDING UNDERSTANDING

“killing, torture, enslavement, poverty, suffering, peril and death.”2 A less polemical diagnosis, from Stephen Toulmin, has sought to retrieve some notion of humane, practical rationality from the wreckage of mechanistic theorizing blamed, again, largely on Descartes. In the background was supposed to lie a contrast between pure platonic knowledge (episteme) and Aristotelian practical wisdom (phronesis).3
Many philosophers in the twentieth century were cautious about wisdom, relying on a contrast between literary, consoling Philosophies and impersonally logical or scientific philosophizing. This contrast may have originated from Russell (who also exemplified in his own life a sharp distinction between philosophical knowledge and practical unwisdom) or from William James’s opposition between the tough and the tender-minded.4 It is easy enough to sketch out a table of dichotomies, if only as a target for critical deconstruction, in rough order of increasing improbability: knowledge analysis critique atomism sciences Naturwissenschaft facts theory episteme objective researcher information tough west political left

wisdom synthesis exposition holism humanities
Geisteswissenschaft
values practice phronesis personal sage education tender east political right

Before this trails off into a reductio ad absurdum one could ask where philosophy might be supposed to fit: maybe “philosophy” or “philosophizing” should be on one side and a “Philosophy” on the other. Kant himself stressed that philosophy cannot be learned, that “we can at best only learn to philosophize.” To him “the path to wisdom, if it is to be assured and not impassible or misleading, must for us human beings unavoidably pass through science.”
And again, “Science (critically sought and methodically directed) is the narrow gate that leads to the doctrine of wisdom.”5 A notable problem for Russell, at one extreme, was that he was so keen on the rhetoric of the left hand column of this table that he allowed only a residual, diminishing function for his own philosophizing:

WISDOM

107

I believe the only difference between science and philosophy is, that science is what you more or less know and philosophy is what you do not know. Philosophy is that part of science which at present people choose to have opinions about, but which they have no knowledge about. Therefore every advance in knowledge robs philosophy of some problems which it formerly had.6

Positivist assumptions about knowledge are likely to lead to similar conclusions, if usually in less directly blunt forms. It is worth remembering that the reductionist impulse against Philosophies that advertise wisdom comes exactly from the obvious partisanship or banality of many prescriptions on offer. One origin of German-language positivism in the 1930s was disenchantment with the dark wisdom of blood and soil. From the heyday of Philosophies, during the hellenstic period, the standard therapies for life’s problems—or rather the worries in the lives of comfortable, educated men7—can be easy enough to satirize. With any such recipe for practical wisdom, the temptation to ask simply: What does this come down to? and: Does it actually work? must be strong. As Hume put it sharply in his essay on The Sceptic: “The empire of philosophy extends over a few; and with regard to these, too, her authority is very weak and limited.”8
But, even in these hostile and prejudiced terms, where to put understanding? To understand is not to know, and any reduction of understanding to knowledge could only impress in the most restrictive of contexts. Again, to understand at least some necessary truths is to know that they are true. This could be one characterization of a priori knowledge, but it could hardly be extended to any wider application of understanding. Attempts to do that can appear arbitrarily unconvincing. Descartes, for example, wrote with crass insensitivity to the ailing
Princess Elisabeth in 1645 of how knowledge would lead not merely to virtue but to happiness: “it seems to me, Seneca should have taught us all the principal truths whose knowledge is necessary to facilitate the practice of virtue and to regulate our desires and passions, and thus to enjoy natural happiness.”9
Understanding and wisdom share some characteristics. Both are vaguer and apparently wider than knowledge. They have both attracted less attention than knowledge from philosophers in the modern West. There are also clear differences. By no means all understanding is practical. On the contrary, understanding may be contemplative, passive, or visionary. Wisdom may also exist, as it were, only on paper, but maybe its practical dimension is more significant. (It would be comical to be wise only in theory.) One argument of this chapter is that the connection between understanding and the nature of philosophy is a close one. The etymological root of philosophy as a love of wisdom may seem embarrassing for various reasons, starting with the impracticality of philosophers, notorious since Thales fell down his well, and ridiculed even in Plato.10 Nevertheless, if philosophy is not supposed to generate knowledge, then maybe it generates something else.

108

UNDERSTANDING UNDERSTANDING

This book could not open with a straightforward reply to the simple question:
What is understanding anyway? because nothing can be taken for granted about the kind of answer to be given. Anyone in pursuit of understanding can be diverted from the trail by two related methodological questions. First, what is the aim of the search? (A concept, a process, a set of conditions, the use of a word?) Second, what sort of understanding of understanding is sought? (A glimpse of a concept? A grasp of some linguistic usage or of some other technique? An experience of some process, mental or otherwise? A theory or narrative? Illumination?) Most obviously of all, any presumption that a definition makes a good starting point is disastrously flawed. What could make us assume that definitions produce understanding, or that a definition tells us what something is?
These ramifying questions cannot be brushed aside. They bring out the connections between the nature of understanding and the nature of philosophical inquiry. A simple platonic-realist theory can see both understanding and philosophical inquiry in strongly visual terms, with objects to be discerned and studied (though Plato himself was not so doctrinaire). A nominalist may have a grasp of understanding as a capacity to use words, where the field of inquiry may be the uses of words in a language. Those who want to stress the importance of interpretation in understanding may also interpret philosophy as a historical series of reinterpretations. None of these readings need imply harmful circularities (as Heidegger saw, in dismissing other forms of
“hermeneutical circle”).11 On the contrary, it ought to be important to avoid an inconsistent (philosophical) understanding of understanding. That sounds elementary, but may not be so easy. An observation that might be drawn from chapters 1 and 2 is that any single understanding of understanding is likely to be partial. Naturally, it is possible to insist on some limiting definition, metaphor, or account, although the point in doing that is hard to see. A uniform rationalizer may seek to persuade us that all understanding has to be through the perception of proximate reasons-or-causes (later: through explanatory laws). A uniform representationalist may believe that all understanding has to be portrayed in terms of the viewing of mental images. We may also decline to accept such narrowing accounts, and it then becomes hard to see where the argument could go from there.
No assumption need be made that we do understand understanding correctly—for example, that the surveys in chapters 1 and 2 need be taken as accurate accounts of what understanding really is, against which any theories have to be measured. A more modest view is that we should not hold or presuppose accounts of understanding which are inconsistent with philosophical or hermeneutical practices which we want to maintain. This is where the assertion of uniform theories, analyses, or definitions of understanding may prove difficult. It may be possible to offer an account as prescriptive or stipulative rather than inclusive or descriptive and then to repudiate some under-

WISDOM

109

standings of understanding that do not fit the preferred pattern—though, again, the point of doing that might be uncertain. One could, for example, exclude intuitive personal sympathy from one’s view of understanding, on the grounds that any kind of intellectualized theory of it would seem far too elusive to be valuable. The upshot might be a conclusion that I can understand what you say—your words—but that it is mistaken to suggest that I understand you. The only test for such a ruling could be one of reasonable persuasiveness, where it would surely fail straightaway.
Wittgenstein provides a curious puzzle case. In his later work he stressed the importance of attaining an Übersicht—an overview, synoptic view, or surview—as an aim in philosophy. In the notes published as Zettel, he wrote
(of what he saw as a misunderstanding on mathematical infinity) that “the aim is a synoptic comparative account . . . [Das Ziel ist eine übersichtliche, vergleichende Darstellung].” Not to understand was not to have an overview [Du verstehst, d.h. übersiehst . . . nicht]. That important model of understanding may have been a relic of the strongly visual image used at the end of the Tractatus: to see the world aright [sieht er die Welt richtig],12 but it would seem awkward to bring into line with a linguistic, capacity-based account of understanding that has been attributed to him elsewhere, in his later writings. Of course, there may be different kinds of uses of “understanding” that make up the concept.13 To understand may be both to see or realize (passively), and to possess a linguistic capacity (actively), though how far both interpretations could be maintained, liberally, at the same time may be debatable. The external, objective, third-person standards applicable to understanding-as-a-capacity have no relevance to subjective understanding-as-Übersicht. To achieve philosophical insight may not be to be able to do anything differently: things may just look different, subjectively (“I have changed his way of looking at things”).14 One possibility is that here we see a specimen of Wittgenstein’s antitheoretical or nontheoretical approach—inclusive description rather than explanation— where two very different styles of understanding are only described. In any event, a single consistent account would not be easy to extract.
One conclusion of chapter 3 was that expectations for a theory or account of understanding might derive from a Cartesian model for a theory of knowledge. So one aim might be presumed to be some theory or definition that would enable us—or rather me—to distinguish between real and mistaken understandings, as a basis for developing a fuller or better understanding.
Then, such a theory might vie for priority with a theory or definition of knowledge. But that whole picture would be wrong. There is no reason to give any special emphasis or priority to the question: How do I understand? along the lines of the significance given to the Cartesian question: How do I know?
There need be no analogy in the understanding of understanding for the legitimation of subjective experience or consciousness in the Cartesian theory of knowledge. So there is no need for a philosophical “theory” to explain how my

110

UNDERSTANDING UNDERSTANDING

understanding is supposed to work, in analogy to the a priori psychologizing about knowledge that filled the works of both rationalist and empiricist epistemologists for two centuries after Descartes. An appropriate source for knowledge has no parallel in an appropriate source for understanding. It matters to the epistemologist how I come to know something because this may provide endorsement or legitimation for my knowledge. It need not matter at all how I come to understand something. A “method” need not be relevant.
Theorizing about knowledge evolved from speculative psychology into a search for the conditions for the application of concepts—from asking how I get to know that s into asking about the conditions for the use of I know that
s. But, once more, there is still no analogy with understanding. It may seem interesting to inquire into the conditions for the use of I understand that s, perhaps in the hope of understanding understanding. Plenty of specific conditions might be assembled but the prospects for any general account would seem minimal.
There is certainly no need to insist dogmatically that philosophy is supposed to generate understanding rather than knowledge. That would be to assume some clear distinction, and order of priority, between understanding and knowledge, as well as an unnecessarily doctrinaire view of philosophy. It would be a formidable challenge to show that philosophy delivers nothing but knowledge (or, conversely, delivers no knowledge at all). A less restricting view should require less supporting argument. The range of imagery associated with understanding—appreciating connections, realizing, grasping, fitting together, seeing perspectives, sensing proportions—might suit the diversity of philosophical activity better than a single-minded search for truth. Both passive understanding (seeing, realizing) and active understanding (grasping, acquiring a capacity) have analogues in philosophy. Coming to understand is presumably a more accurate portrayal of a philosophical goal than the gaining of extra knowledge. Items of knowledge—facts, or whatever—have a certain objective plausibility—more facts, more knowledge—whereas items of understanding make no sense at all. Heuristically or educationally, one kinship of philosophy is with mathematics, where an acquaintance with bits of information (formulas, algorithms) is insignificant in comparison with an understanding of them: which means both to see their point and to be able to manipulate them. Teaching can be portrayed—no doubt unfairly—as the transmission of knowledge. Understanding, famously, can be taught only erratically, if at all. It is something, maybe with some help, that you have to do for yourself.
Even the most aridly technical characterization of philosophy lacks a point without some reference to understanding. A linguistic philosopher from the 1950s, convinced that the only rôle for philosophy was the cleaning and polishing of intellectual equipment (“conceptual analysis”), would still need to believe that such a procedure had to be accepted—understood—to

WISDOM

111

be adopted. The point of the activity might be portrayed behaviorally—to make intellectual equipment more useful—but that, too, would be in the context of a view of understanding, where to understand might include being able to do things with words.
If understanding is so important in philosophy, it seems reasonable to ask why it has not been the subject of more direct philosophical attention. There is of course the hermeneutical tradition, often concentrating on styles or techniques of understanding in specific fields, such as law, textual analysis, or biblical criticism. But where is there a tradition of argument about the nature and rôle of understanding itself, and of understanding in philosophy? One speculative reply might be that thinking about knowledge—and latterly, meaning— has seemed more productive or (more directly) easier. Theorizing has seemed possible about what can or cannot be known, and about how sense can or cannot be made. Such theorizing can be used critically, to discriminate between acceptable and rejected items of knowledge or sense. It can have a wider social function, in drawing lines around acceptable disciplines: “science,” history, economics, theology. Definitions of knowledge—or analyses of the concept of knowledge—have seemed within reach. Projects to determine theories of meaning have been described at length.
Understanding, in contrast, has been neglected. This could be because the subject is so intangible. Descartes set the scene clearly for theories of knowledge: I want to know about the world outside me. The scene for theorizing about meaning followed on. How do you and I make sense to each other?
None of this comparative clarity applies with understanding. There can be no presumption of a first-person, subjective starting-point. (Again: a court’s attempt to understand a statute, or a scholar’s attempt to decode a text, will not be reducible the attempts of individuals to form subjectivized understandings.) In some fields—personal or communal relations—it is quite possible that understanding cannot be attained. There need be no presumption of success. The value of a critical theory of understanding seems elusive. One aim might be to determine that something or other was unintelligible (“in principle”), along the lines of critical accounts of knowability or meaningfulness.
Yet—as seen in chapters 4, 5, and 6—that aim is remarkably hard to pin down in any usefully general way. An account of how sense is made—a theory of meaning—might be imaginable, whereas an account of how sense is understood seems merely baffling.
One conclusion might be that if understanding is so slippery, it does not merit much concern: no wonder the attention has mostly gone elsewhere. Yet that would be ironical, given the place of understanding in philosophy. No stance on positivism (for or against) is needed to accept that, whatever philosophy is, it cannot be a mere accumulation and arrangement of items of knowledge. (Even A Philosophy as a systematic body of would-be Objective
Truths has to be understood to have any interest or effect. Nobody could

112

UNDERSTANDING UNDERSTANDING

think that a philosophy is simply devised to be there to be admired.) If philosophers want understanding, the understanding that they want can be as protean as any other kind: personal insight, Übersicht, theoretical narratives, explanations, agreed readings, and so on. An obvious constraint (as just mentioned) is not to propound an explicit theory about understanding that conflicts with one’s own practice in seeking understanding through philosophy. That can be less straightforward than it sounds: more than a few philosophers have found themselves writing works that ought to be unintelligible by their own standards. More interestingly, the fugitive quality of understanding might be reflected by some openness in the characterization of philosophy. In Plato’s Phædrus, writing was said to be dead and unambiguous, in contrast with living speech. In a written dialogue this assertion stood as an open contradiction between form and content. Early in the Phædrus an overt caution was given on interpretation. Nevertheless, there have been innumerable interpretations of the dialogue, to the extent of provoking the thought that the author may have been trying to signal how understanding cannot be controlled.15
So what about wisdom?
Judge: I have heard your case, Mr. Smith, and I am no wiser now than when
I started.
F. E. Smith: Possibly not, my lord, but far better informed.

An unimpressive line of thinking might be caricatured like this: Understanding seems broader and vaguer than knowledge; wisdom seems broader and vaguer than knowledge; so wisdom has some kinship with understanding; so if philosophy seeks understanding, then philosophy seeks wisdom.
It might be more productive to start from another direction. Philosophical reticence about wisdom might stem from a healthy skepticism about the practical efficacy of philosophy, or from a wariness about understanding. To be wise you need understanding as well as information, or knowledge; but understanding is so much harder to pin down. A single story about the origin, nature, or limits of knowledge has seemed possible to many philosophers. A single account of understanding has not, and with good reason. There is no need to constrict methods, styles, or forms of understanding. That should be liberating rather than negative. Philosophy does not need to have a single aim, but it does have some aims, including clear insight and a capacity to think through problems. Working toward such aims may be enjoyable, but the destination must surely justify the journey. Understanding may be passive, intuitive, or contemplative. It may also be active—changing the world—as an outcome of practical philosophizing. Together, both forms of understanding might even add up to a kind of wisdom.

WISDOM

113

Proust, mentioned several times in these pages, had an aversion to explicit intellectual theorizing (disparaged as grande indélicatesse). His narrator finally understood that the point of his lifetime’s (apparently pointless) experiences was to be understood and narrated in his book, and that the narration might constitute the understanding.16 This was far more than a trivially entertaining paradox. It is always possible to complain that you do not understand understanding without some theory, or explanation, or analysis, and to insist that this is still lacking. Here could be where an understanding of the variety and elusiveness of understanding rightly mirrors the diversity and elusiveness of philosophy. It is in seeing how understanding can be seen that we can understand how variously philosophy can be understood.

This page intentionally left blank.

Notes

INTRODUCTION
1. Theætetus, 155d.
2. Job, 42:3 ( JPS).

CHAPTER 1
1. See papers in Knowing Our Own Minds, eds. Crispin Wright, Barry C. Smith, and Cynthia Macdonald, Oxford: Oxford University Press, 1998.
2. Treatise of Human Nature, I, 4, 6.
3. Second Meditation, in Philosophical Writings, trs. John Cottingham, Robert
Stoothoff, Dugald Murdoch, and Anthony Kenny, Cambridge: Cambridge University
Press, 1985/91, vol. II, pp. 22–23.
4. See Neil Cooper, “Understanding People,” Philosophy, 75, no. 293, July 2000.
5. Sources of the Self, Cambridge: Cambridge University Press, 1989, Chs. 1 and 2.
6. Pierre Clastres, Chronicle of the Guyaki Indians, trs. Paul Auster, London:
Faber, 1998, p. 2.
7. Truth and Method, trs. Joel Weinsheimer and Donald G. Marshall, London:
Sheed and Ward, 2nd. ed., 1989, pp. 327–8.
8. Philosophical Investigations, trs. G. E. M. Anscombe, Oxford: Blackwell, 1953,
§§156–78.
9. Truth and Method, p. 164.
10. Quentin Skinner, “Meaning and Understanding in the History of Ideas”
(1969), in Visions of Politics, Cambridge: Cambridge University Press, 2002, Vol. I,
p. 86.
115

116

NOTES TO CHAPTER 2

11. Truth and Method, pp. 475–6.
12. Michael Dummett, The Interpretation of Frege’s Philosophy, London: Duckworth, 1981, p. 77; Paul Ziff, Understanding Understanding, Ithaca: Cornell University
Press, 1972.
13. Michael Dummett, “What Is a Theory of Meaning ?(I),” in The Seas of Language, Oxford: Clarendon Press, 1993, p. 4.
14. Andrew Bowie, “The Meaning of the Hermeneutic Tradition in Contemporary Philosophy,” in Anthony O’Hear, ed., Verstehen and Humane Understanding, Cambridge: Cambridge University Press, 1997, p. 136.
15. Wittgenstein, Philosophical Grammar, ed. Rush Rhees, trs. Anthony Kenny,
Oxford: Blackwell, 1974, I, §24; Philosophical Investigations, I, §597.
16. E.g., Charles Taylor, “Theories of Meaning,” in Philosophical Papers I, Cambridge: Cambridge University Press, 1985.
17. Michael Dummett, Frege, Philosophy of Language, London: Duckworth, 1973,
p. 2.
18. See, e.g., Peter Kivy, New Essays on Musical Understanding, Oxford: Oxford
University Press, 2001.
19. Kant, Critique of the Power of Judgment (1790), Paul Guyer and Eric
Matthews, trs., Cambridge: Cambridge University Press, 2000, p. 83.
20. Republic, V, 478a.
21. Spinoza, Ethics, trs. Samuel Shirley, Indianapolis: Hackett, 1992, Preface to
Part III.
22. Heidegger, What Is a Thing?, trs. W. B. Barton and V. Deutsch, South Bend:
Regnery/Gateway, 1967, pp. 54, 137.

CHAPTER 2
1. Nicomachean Ethics, I, 3, W. D. Ross and J. O. Urmson trs., revised J. Barnes,
Oxford: Clarendon Press, 1984.
2. Nicomachean Ethics, VI, 3.
3. Michael Oakeshott, “Rationalism in Politics,” Cambridge Journal, I, 1947.
4. Richard P. Feynman, QED, The Strange Theory of Light and Matter, London:
Penguin, 1990, p. 9f.
5. John Ziman, Real Science: What It Is and What It Means, Cambridge: Cambridge University Press, 2000, p. 290.
6. Republic, VII, 518c.
7. Karl Popper, Objective Knowledge, Oxford: Clarendon Press, 1972, p. 162; or
Ziff, “understanding what is said is essentially a matter of data processing of some sort.” Ziff, Understanding Understanding, p. 17.

NOTES TO CHAPTER 2

117

8. Henry James, The Portrait of a Lady, Harmondsworth: Penguin, 1966, p. xvii.
9. Locke, An Essay Concerning Human Understanding, Introduction, 1.
10. Being and Time, trs. Joan Stambaugh, Albany: SUNY Press, 1996, §31, p. 134.
11. G.P. Baker and P. M. S. Hacker, Language, Sense and Nonsense, Oxford: Blackwell, 1984, pp. 351, 350; Gilbert Ryle, The Concept of Mind, Harmondsworth: Penguin,
1963, p. 57.
12. E.g., in John Searle, Minds, Brains and Science, London: BBC, 1984, Ch. 2.
13. Philosophical Investigations, I, §122; “Philosophy” (1933), in Philosophical Occasions 1912–1951, eds. James Klagge and Alfred Nordmann, Indianapolis: Hackett,
1993, p. 175.
14. Iris Murdoch, “The Idea of Perfection” (1962), in ed. Peter Conradi, Existentialists and Mystics, London: Chatto & Windus, 1997, p. 317.
15. M. O’C. Drury, in R. Rhees, ed., Ludwig Wittgenstein, Personal Recollections,
Oxford: Blackwell, 1981, p. 118.
16. “On interpretation,” in Philosophy in France Today, trs. K. McLaughlin, ed. A.
Montefiore, Cambridge: Cambridge University Press, 1983, p. 195.
17. See David-Hillel Ruben, Explaining Explanation, London: Routledge, 1990,
p. 14.
18. G. H. von Wright, Explanation and Understanding, London: Routledge &
Kegan Paul, 1971, p. 6.
19. See Nancy Cartwright, The Dappled World, Cambridge, Cambridge University Press, 1999, p. 33.
20. Roger Penrose, The Emperor’s New Mind, London: Vintage, 1990, p. 554.
21. E.g., Paul H. Hirst, “Curriculum Integration,” in Knowledge and the Curriculum, London: Routledge & Kegan Paul, 1974, pp. 150–151 (and many other points in that volume).
22. See M. F. Burnyeat, “Wittgenstein and Augustine De Magistro,” Aristotelian
Society Supplementary Volume LXI, 1987, pp. 19–22.
23. Augustine, On Christian Teaching, trs. R. P. H. Green, Oxford: Oxford University Press, 1997.
24. F. D. E. Schleiermacher, “Hermeneutics and Criticism,” p. 5 and “General
Hermeneutics,” p. 267, in Hermeneutics and Criticism and Other Writings, ed. Andrew
Bowie, Cambridge: Cambridge University Press, 1998.
25. Philosophical Investigations, I, §153.
26. Critique of Pure Reason, trs. Paul Guyer and Allen W. Wood, Cambridge:
Cambridge University Press, 1998, A141/B180.
27. Discourse on the Method, V, in Philosophical Writings, vol. I, p.141.
28. See Jürgen Habermas “On Hermeneutics’ Claim to Universality,” in The
Hermeneutics Reader, ed. K. Mueller-Vollmer, Oxford: Blackwell, 1986, p. 301.
29. Critique of Pure Reason, Bxxx.

118

NOTES TO CHAPTER 4

CHAPTER 3
1. Dummett, Frege: Philosophy of Language, pp. 666–7; also see Ch. 1 (h), note
12.
2. See, e.g., Charles Taylor, “Overcoming Epistemology,” in Philosophical Arguments, Cambridge: Harvard University Press, 1995; Richard Rorty, Philosophy and the
Mirror of Nature, Princeton: Princeton University Press, 1980, Ch. III.
3. Philosophical Writings, vol. I, p. 9.
4. In Enquiry Concerning Human Understanding, X.
5. Iritique of Pure Reason, A86–87/B119–20: Geburtsbrief . . . reinen und gesetzmäßigen Abkunft.
6. In Philosophical Writings, vol. I, Part I, §§1, 2, 9, 11.
7. Theological-Political Treatise, S. Shirley, trs., Indianapolis: Hackett, 1998, pp.
169, 168, 166.
8. On Certainty, D. Paul and G. E. M. Anscombe trs., Oxford: Blackwell, 1969,
§§179; 440.
9. Being and Time, pp. 10, 21, 43.

CHAPTER 4
1. Ethics, Preface to Part III.
2. Ethics, Part IV, Appendix, 4.
3. Cartwright, The Dappled World, pp. 13–19.
4. Bernard Williams, Making Sense of Humanity, Cambridge: Cambridge University Press, 1995, p. 88.
5. David Chalmers, The Conscious Mind: In Search of a Fundamental Theory, New
York: Oxford University Press, 1996, pp. 276–7.
6. Confessions, X, xxxv (54).
7. Freud, Psychopathology of Everyday Life (1914), trs. A. A. Brill, Harmondsworth: Penguin, 1938, p. 177.
8. At length: George Schlesinger, The Intelligibility of Nature, Aberdeen:
Aberdeen University Press, 1985.
9. Critique of Pure Reason, A538=B566, A256/B311.
10. Prolegomena to any future metaphysics, trs. Gary Hatfield, Cambridge: Cambridge University Press, 1997. References are to sections and marginal page numbers.
§15, 4:295.
11. A126: Es ist . . . der Verstand nicht bloß ein Vermögen, durch Vergleichung der
Erscheinungen sich Regeln zu machen: er ist selbst die Gesetzgebung für die Natur, d.i. ohne
Verstand würde es überall nicht Natur . . .

NOTES TO CHAPTER 4

119

12. Prolegomena, §14, 4:294.
13. Critique of Pure Reason, ibid.
14. Prolegomena, §15, 4:295.
15. Critique of Pure Reason, A692=B720.
16. Critique of Pure Reason, A700=B728.
17. Critique of the Power of Judgment, Introduction § IV, pp. 67–8.
18. Strictly, “reason” or “power of reason” [Vernunftvermögen], not “understanding” in Kantian terminology: Critique of Pure Reason, A302/B359, A307/B363. The distinction can be crucial, but not here (see Susan Neiman, The Unity of Reason: Rereading Kant, New York: Oxford University Press, 1994, pp. 48–62).
19. Critique of Pure Reason, A693=B721. For Kant on Spinoza see the author’s
“Intelligibility: The Basic Premise?,” Iyyun, The Jerusalem Philosophical Quarterly, 50,
July 2001.
20. Prolegomena, §17, 4:297.
21. Prolegomena, §14, 4:294.
22. Critique of Pure Reason, A645=B673.
23. Critique of Pure Reason, A293/B350: Wahrheit oder Schein sind nicht im Gegenstande, sofern er angeschaut wird, sondern im Urteile über denselben, sofern er gedacht wird.
24. Critique of Pure Reason, A302/B359; A738=B766; A643=B671.
25. The Sources of Normativity, Cambridge: Cambridge University Press, 1996, p.
236.
26. Notably by P. F. Strawson, The Bounds of Sense, London: Methuen, 1966.
27. Critique of Pure Reason, A696=B724: die Welt ist eine Summe von Erscheinungen;
Prolegomena, §17, 4:296.
28. E.g., Critique of Pure Reason, A235/B294–A289/B346.
29. Critique of Pure Reason, A658=B686; A661=B689.
30. Critique of Pure Reason, A663=B691.
31. A228/B44; A46/B63. A different reading: Rae Langton, Kantian Humility,
Oxford: Clarendon Press, 1998, Ch. 8.
32. See A. W. Moore, Points of View, Oxford: Clarendon Press, 1997, p. 7.
33. Letter to Mesland, 2 May 1644, Philosophical Writings, vol. III, p. 235.
34. Quoted by Schlesinger, The Intelligibility of Nature, p. xiii.
35. Heraclitus, Fr 78, trs. T. M. Robinson, Toronto: University of Toronto Press,
1987; Second Replies, Philosophical Writings, vol. II, p. 103.
36. See E. M. Curley, Spinoza’s Metaphysics: An Essay in Interpretation, Cambridge: Harvard University Press, 1969, pp. 144–53.
37. Ethics, Part II, 13, Scholium.

120

NOTES TO CHAPTER 5

CHAPTER 5
1. Paradiso, II, 43–44: Lí si vedrà ciò che tenem per fede, non dimostrato, ma fia per sé noto
Also Convivio, III, 15; after Summa Theologiæ,1a. 2, 1.
2. Jorge Luis Borges, in The Aleph, Andrew Hurley, trs., London: Penguin, 2000.
3. Tractatus Logico-Philosophicus, D. F. Pears and B. F. McGuinness, trs., London: Routledge, 1966, §4.024.
4. “General Hermeneutics,” in Hermeneutics and Criticism and Other Writings, p.
227.
5. Richard Popkin, The History of Skepticism from Erasmus to Spinoza (revised from 1960), Berkeley: University of California Press, 1979.
6. Robert Boyle, A Discourse of Things Above Reason Inquiring Whether a Philosopher should admit there are any such, London: Jonathan Robinson, 1681, pp. 83–4.
7. Bernard Williams, Descartes, Harmondsworth: Penguin, 1978, p. 47 ff.
8. Pascal, Pensées, trs. Honor Levi, Oxford: Oxford University Press, 1995, §164
(Sellier numbering).
9. Things Above Reason, p. 49.
10. Critique of Pure Reason, A831=B859 .
11. Prolegomena to any Future Metaphysics, §57, 4:352; also Critique of Pure Reason, A761=B789, A767=B795.
12. Dialogues Concerning Natural Religion, VIII.
13. Collected Philosophical Papers, Vol. I, Oxford: Blackwell, 1981, Introduction, p. viii. 14. Graham Priest, Beyond the Limits of Thought, Cambridge: Cambridge University Press, 1995, pp. 188–94 with references there to his other writings.
15. See Jean-Luc Marion, Sur la théologie blanche de Descartes, Paris: Presses Universitaires de France, 1981.
16. Philosophical Works, I, p. 33.
17. Philosophical Works, III, p. 331 (letter to Silhon, 1648): [cette connaissance] . . . votre esprit la voit, la sent et la manie.
18. See Irven Michael Resnic, Divine Power and Possibility in St Peter Damian’s De
Divine Omnipotentia, Leiden: Brill, 1992, e.g., pp. 99–102.
19. Preface, p. 3.
20. Tractatus, 5.62..
21. G. Berkeley, Principles of Human Knowledge, Introduction, §§22–25.
22. Truth and Method, p. 475.
23. “Radical Interpretation” (1973), in Inquiries into Truth and Interpretation,
Oxford: Clarendon Press, 1984, p. 126.

NOTES TO CHAPTER 5

121

24. 4.114, 4.115.
25. 3.23; 5.5563; 4.116.
26. 4.021: Der Satz ist ein Bild der Wirklichkeit: Denn ich kenne die von ihm dargestellte Sachlage, wenn ich den Satz verstehe; 4.024; 4.002.
27. Donald Davidson, “A Nice Derangement of Epitaphs,” in Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, Ernest LePore, ed., Oxford:
Blackwell, 1986, p. 446: Tractatus, 3.25.
28. 4; 4.01.
29. 4.411: das Verständnis der allgemeinen Sätze hängt fühlbar von dem der Elementarsätze ab.
30. A. M. Quinton, trs., reprinted in Philosophical Logic, ed. P. F. Strawson,
Oxford: Oxford University Press, 1967, pp. 28–9.
31. Philosophical Investigations, I, §136.
32. Philosophical Investigations, I, §243.
33. Nietzsche, Nachlass 1885–1887: Kritische Studienaugabe, Giogio Colli and
Mazzino Montinari, eds.,Berlin: de Gruyter, 1980, vol. 12, pp. 39, 315; On the Genealogy of Morality, III, §12.
34. Michel Foucault, “Nietzsche, Freud, Marx,” Cahiers de Royaumont, Paris:
Minuit, 1967, Jon Anderson and Gary Hentzi, trs., in Foucault, Essential Works, Vol.
2, Harmondsworth: Allen Lane, 1998, p. 275; Jacques Derrida, Spurs: Nietzsche’s Styles,
Chicago: University of Chicago Press, 1978, pp. 122–34.
35. Marcel Proust, “Contre l’Obscurité” and Stéphane Mallarmé, “Le Mystère, dans les Lettres,” La revue blanche, XI, 1896, pp. 72 and 214–15, quoted from JeanYves Tadié, Marcel Proust, Euan Cameron, trs., London: Viking, 2000, pp. 249–50; À la recherche du temps perdu, Jean-Yves Tadié et al. eds. Paris: NRF Pléiade, 1987–1989, vol. I, p. 153.
36. 245a.
37. Inquiries into Truth and Interpretation, pp. 184–185; and Ziff, Understanding
Understanding, ch. VIII.
38. See Dan Sperber, “Apparently Irrational Beliefs,” in Rationality and Relativism, Martin Hollis and Steven Lukes, eds., Oxford: Blackwell, 1982.
39. John Donne, Funeral Sermon on Magdalen Herbert, Lady Danvers, 1 July 1627;
Truth and Method, p. 302.
40. Discourse on the Method, I, in Philosophical Writings, I, p. 114.
41. Emmanuel Levinas, Totalité et infini (1971), Paris: Livres de poche, 1994, p.
33; “La trace de l’autre” (1963), in En découvrant l’existence avec Husserl et Heidegger,
Paris: Vrin, 1994, p. 191.
42. Totalité et infini, p. 85.

122

NOTES TO CHAPTER 6

CHAPTER 6
1. Philosophical Writings, Vol. III, p. 378.
2. Things Above Reason, pp. 20–21. See also Jan W. Wojcik, Robert Boyle and the
Limits of Reason, Cambridge: Cambridge University Press, 1997.
3. Apologia Pro Vita Sua (1864), ed. M. J. Svaglic, Oxford: Oxford University
Press, 1967, Note A, p. 260.
4. The Last Word, Oxford: Oxford University Press, 1997, p. 130.
5. Gillian Rose, Mourning Becomes the Law, Cambridge: Cambridge University
Press, 1996, p. 43.
6. Job, 42:3 ( Jewish Publication Society trs.); Idit Dobbs-Weinstein, Maimonides and St. Thomas on the Limits of Reason, Albany: SUNY Press, 1995, p. 51.
7. The Character of Mind, Oxford: Oxford University Press, 2nd. ed., 1997, p. 39.
8. Ernest Nagel and James R. Newman, Gödel’s Proof, New York: New York University Press, 1958, pp. 101–102.
9. “Lecture on Ethics,” Cambridge, November 17, 1929, in Philosophical Occasions, p. 44; Tractatus, 6.421, 6.432, 6.54.
10. Philosophical Investigations, I, §435; Tractatus, 4.121, 4.1212; Norman Malcolm, Nothing Is Hidden: Wittgenstein’s Criticism of his Early Thought, Oxford: Blackwell, 1986.
11. Wittgenstein, Philosophical Investigations, I, §126; Philosophical Grammar,
Rush Rhees, ed., Anthony Kenny, trs., Oxford: Blackwell, 1974, II, §23, p. 59; Tractatus, 4.116.
12. “Parmenides, Mystery and Contradiction” (1969), in Collected Philosophical
Papers, vol. I, p. 8.
13. John Henry Newman, An Essay in Aid of a Grammar of Assent (1870), Notre
Dame: University of Notre Dame Press, 1979, Ch. 5; also Gadamer, Truth and Method,
Part III, 2(B).
14. As in Philosophical Investigations, I, §242: Zur Verständigung durch die Sprache gehört . . . eine Übereinstimmung in den Urteilen.
15. E.g., Ronald D. Laing and Aaron Esterson, Sanity, Madness and the Family,
Harmondsworth: Penguin, 1970, particularly Ch. 5.
16. The Greek hyper, as in the most frequently cited text, Pseudo-Dionysius Areopagite, The Divine Names and Mystical Theology, John D. Jones, trs., Milwaukee: Marquette University Press, 1980.
17. Nagel, Thomas, “What is it like to be a bat?,” in Mortal Questions, Cambridge:
Cambridge University Press, 1979, p. 171.
18. Summa Theologiæ, IIaIIæ. 1, 2.
19. Religion within the Boundaries of Mere Reason (1793), in Allen W. Wood and
George Di Giovanni trs. and eds., Religion and Rational Theology, Cambridge: Cambridge University Press, 1996, pp. 186, 205–206.

NOTES TO CHAPTER 7

123

20. All’s Well That Ends Well, Act II, Scene iii, opening.
21. Ethics, Part I, Appendix.
22. Robert Jan van Pelt, The Science of Holocaust Research and the Art of Holocaust
Denial, University of Waterloo Research Lecture, 1999, p. 53.
23. Cf. Bernard Williams, “Deciding to Believe,” in Problems of the Self, Cambridge: Cambridge University Press, 1973, pp. 149–51; Jonathan E. Adler, Belief ’s Own
Ethics, Cambridge: MIT Press, 2002.
24. Critique of Pure Reason, A557–B585: nur bis an die intelligible Ursache, aber nicht über dieselbe hinaus kommen.
25. Pensées, §164; Nicholas Hammond, Playing with Truth: Language and the
Human Condition in Pascal’s Pensées, Oxford: Clarendon Press, 1994, pp. 19–20.
26. Points of View, Ch. 9, §4.
27. Tractatus, 1922, Introduction, p. xxi.
28. Concluding Unscientific Postscript, David F. Swenson and Walter Lowrie, trs.,
Princeton: Princeton University Press, 1968, p. 209.

CHAPTER 7
1. Rules for the Direction of the Mind (c1628) and French Preface to the Principles of Philosophy (1647), Philosophical Writings, vol. I, pp. 9 and 179–80.
2. Nicholas Maxwell, From Knowledge to Wisdom, Oxford: Blackwell, 1984, p. 1; or see also David Conway, The Rediscovery of Wisdom, Basingstoke: Palgrave, 2000.
3. Stephen Toulmin, Cosmopolis, Chicago: University of Chicago Press, 1990. p.
192.
4. William James, Pragmatism (1907), Cambridge: Harvard University Press,
1975, Lecture I.
5. Critique of Pure Reason, A837=B865 (Kant explained the distinction fully as early as his “Announcement of the programme of his lectures for the winter semester
1765–66,” in Theoretical Philosophy 1755–1770, David Walford and Ralf Meerbote, trs. and eds., Cambridge: Cambridge University Press, 1992, pp. 291–4); Critique of Practical Reason (1788), in Practical Philosophy, Mary J. Gregor, trs. and ed., Cambridge:
Cambridge University Press, 1996, pp. 253–4, 270–71.
6. The Philosophy of Logical Atomism (1918), in Logic and Knowledge, ed. R. C.
Marsh, London: Routledge, 1992, p. 281.
7. An opposed view: Martha Nussbaum¸ The Therapy of Desire, Princeton:
Princeton University Press, 1994, pp. 53–6.
8. Essays, Moral, Political and Literary (1742) Oxford: Oxford University Press,
1963, p. 171.
9. Letter of 4 August 1645, Philosophical Writings, vol. III, p. 258. See Andrea
Nye, The Princess and the Philosopher, Lanham: Rowman and Littlefield, 1999, Ch. 9.

124

NOTES TO CHAPTER 7

10. E.g., Republic, V, 474a; VI, 487be.
11. Being and Time, §32, pp. 142–4; §63, pp. 290–91.
12. Zettel, trs. G. E. M. Anscombe, Oxford: Blackwell, 1967, §273; Tractatus,
6.54; see also Philosophical Investigations, I, §§5, 122.
13. Ich will lieber sagen, diese Gebrauchsartsen von “verstehen” bilden seine Bedeutung, meinen Begriff des Verstehens, Philosophical Investigations, I, §532.
14. Philosophical Investigations, I, §144: Ich habe seine Anschauungsweise geändert.
15. Phædrus, 275b–277a; 229c–230a.
16. À la recherche du temps perdu, Vol. IV, pp. 461, 478.

Bibliography

Unattributed translations are by the author.
No references are given for standard texts where any edition may be used.
Adler, Jonathan E. Belief ’s Own Ethics. Cambridge: MIT Press, 2002.
Anscombe, G. E. M. Collected Philosophical Papers, Vol I. Oxford: Blackwell, 1981.
Aristotle. Nicomachean Ethics. W. D. Ross and J. O. Urmson trs., revised J. Barnes.
Oxford: Clarendon Press, 1984.
Augustine. On Christian Teaching. R. P. H. Green, trs. Oxford: Oxford University
Press, 1997.
Baker, G. P. and Hacker, P. M. S. Language, Sense and Nonsense. Oxford: Blackwell,
1984.
Bowie, Andrew. “The Meaning of the Hermeneutic Tradition in Contemporary Philosophy.” In Verstehen and Humane Understanding. Anthony O’Hear, ed. Cambridge: Cambridge University Press, 1997.
Boyle, Robert. A Discourse of Things Above Reason Inquiring Whether a Philosopher should admit there are any such. London: Jonathan Robinson, 1681.
Burnyeat, M. F. “Wittgenstein and Augustine De Magistro,” Aristotelian Society Supplementary Volume, LXI, 1987.
Cartwright, Nancy. The Dappled World. Cambridge, Cambridge University Press,
1999.
Chalmers, David. The Conscious Mind: In Search of a Fundamental Theory. New York:
Oxford University Press, 1996.
Clastres, Pierre, Chronicle of the Guyaki Indians. Paul Auster, trs. London: Faber, 1998.
Conway, David. The Rediscovery of Wisdom. Basingstoke: Palgrave, 2000.
Cooper, Neil. “Understanding People.” Philosophy, 75, no. 293, July 2000.
Curley, E. M. Spinoza’s Metaphysics: An Essay in Interpretation. Cambridge: Harvard
University Press, 1969.
125

126

BIBLIOGRAPHY

Davidson, Donald. Inquiries into Truth and Interpretation. Oxford: Clarendon Press,
1984.
——— . “A Nice Derangement of Epitaphs.” In Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson. Ernest LePore, ed. Oxford: Blackwell, 1986.
Derrida, Jacques. Spurs: Nietzsche’s Styles. Chicago: University of Chicago Press, 1978.
Descartes, René. Philosophical Writings. John Cottingham, Robert Stoothoff, Dugald
Murdoch and Anthony Kenny, trs. Cambridge: Cambridge University Press,
1985/91.
Dobbs-Weinstein, Idit. Maimonides and St. Thomas on the Limits of Reason. Albany:
SUNY Press, 1995.
Donne, John. Funeral Sermon on Magdalen Herbert, Lady Danvers. London, 1627.
Drury, M. O’C. “Conversations with Wittgenstein.” In Ludwig Wittgenstein, Personal
Recollections. Rush Rhees, ed. Oxford: Blackwell, 1981.
Dummett, Michael. Frege: Philosophy of Language. London: Duckworth, 1973.
——— . The Interpretation of Frege’s Philosophy. London: Duckworth, 1981.
——— . “What Is a Theory of Meaning ?(I).” In The Seas of Language. Oxford: Clarendon Press, 1993.
Feynman, Richard P. QED, The Strange Theory of Light and Matter. London: Penguin,
1990.
Foucault, Michel. “Nietzsche, Freud, Marx.” trs. Jon Anderson and Gary Hentzi, trs.
Cahiers de Royaumont, Paris: Minuit, 1967. In Foucault, Essential Works, Vol. 2.
Harmondsworth: Allen Lane, 1998.
Frege, Gottlob. “The Thought: A Logical Inquiry” (1892). A. M. Quinton, trs. In
Philosophical Logic, ed. P. F. Strawson, Oxford: Oxford University Press, 1967.
Freud, Sigmund. Psychopathology of Everyday Life (1914). A. A. Brill, trs. Harmondsworth: Penguin, 1938.
Gadamer, Hans-Georg. Truth and Method (1960/75). Joel Weinsheimer and Donald
G. Marshall, trs. London: Sheed and Ward, 1989.
Habermas, Jürgen. “On Hermeneutics’ Claim to Universality.” In The Hermeneutics
Reader. K. Mueller-Vollmer, ed. Oxford: Blackwell, 1986.
Hammond, Nicholas. Playing with Truth: Language and the Human Condition in Pascal’s Pensées. Oxford: Clarendon Press, 1994.
Heidegger, Martin. Being and Time (1927). Joan Stambaugh, trs. Albany: SUNY Press,
1996.
——— . What Is a Thing? W. B. Barton and V. Deutsch, trs. South Bend:
Regnery/Gateway, 1967.
Heraclitus. Fragments. T. M. Robinson, trs. Toronto: University of Toronto Press,
1987.
Hirst, Paul H. “Curriculum Integration.” In Knowledge and the Curriculum, London:
Routlege & Kegan Paul, 1974.

BIBLIOGRAPHY

127

Hume, David. Essays, Moral, Political and Literary (1741/42), Oxford: Oxford University Press, 1963.
James, William. Pragmatism (1907). Cambridge: Harvard University Press, 1975.
Kant, Immanuel. “Announcement of the programme of his lectures for the winter semester 1765–66.” In Theoretical Philosophy 1755–1770. David Walford and Ralf
Meerbote, trs. and ed. Cambridge: Cambridge University Press, 1992.
——— . Critique of Pure Reason (A1781/B1787). Paul Guyer and Allen W. Wood, trs.
Cambridge: Cambridge University Press, 1998.
——— . Prolegomena to any future metaphysics (1783). Gary Hatfield, trs. Cambridge:
Cambridge University Press, 1997.
——— . Critique of Practical Reason (1788). In Practical Philosophy. Mary J. Gregor, trs. and ed. Cambridge: Cambridge University Press, 1996.
——— . Critique of the Power of Judgement (1790/1793). Paul Guyer and Eric
Matthews trs. Cambridge: Cambridge University Press, 2000.
——— . Religion within the Boundaries of Mere Reason (1793). In Allen W. Wood and
George Di Giovanni trs. and eds. Religion and Rational Theology. Cambridge:
Cambridge University Press, 1996.
Kierkegaard, Søren. Concluding Unscientific Postscript. David F. Swenson and Walter
Lowrie, trs. Princeton: Princeton University Press, 1968.
Kivy, Peter. New Essays on Musical Understanding. Oxford: Oxford University Press,
2001.
Korsgaard, Christine. The Sources of Normativity. Cambridge: Cambridge University
Press, 1996.
Laing, Ronald D. and Esterson, Aaron. Sanity, Madness and the Family. Harmondsworth: Penguin, 1970.
Langton, Rae. Kantian Humility. Oxford: Clarendon Press, 1998.
Levinas, Emmanuel. Totalité et infini (1971). Paris: Livres de poche, 1994.
——— . “La trace de l’autre” (1963). In En découvrant l’existence avec Husserl et Heidegger. Paris: Vrin, 1994.
Malcolm, Norman. Nothing Is Hidden: Wittgenstein’s Criticism of his Early Thought.
Oxford: Blackwell, 1986.
Marion, Jean-Luc. Sur la théologie blanche de Descartes. Paris: Presses Universitaires de
France, 1981.
Mason, Richard. “Intelligibility: The Basic Premise?” Iyyun, The Jerusalem Philosophical
Quarterly, 50, July 2001.
Maxwell, Nicholas. From Knowledge to Wisdom. Oxford: Blackwell, 1984.
McGinn, Colin. The Character of Mind. Oxford: Oxford University Press, 2nd. ed.,
1997.
Moore, A. W. Points of View. Oxford: Clarendon Press, 1997.

128

BIBLIOGRAPHY

Murdoch, Iris. “The Idea of Perfection” (1962). In ed. Peter Conradi, Existentialists and Mystics. London: Chatto & Windus, 1997.
Nagel, Ernest and Newman, James R. Gödel’s Proof. New York: New York University
Press, 1958.
Nagel, Thomas. “What is it like to be a bat?” In Mortal Questions. Cambridge: Cambridge University Press, 1979.
——— . The Last Word. Oxford: Oxford University Press, 1997.
Neiman, Susan. The Unity of Reason: Rereading Kant. New York: Oxford University
Press, 1994.
Newman, John Henry. Apologia Pro Vita Sua (1864), M. J. Svaglic, ed. Oxford: Oxford
University Press, 1967.
——— . An Essay in Aid of a Grammar of Assent (1870). Notre Dame: University of
Notre Dame Press, 1979.
Nietzsche, Friedrich. Nachlass 1885–1887: Kritische Studienausgabe, Giogio Colli and
Mazzino Montinari, eds.,Berlin: de Gruyter, 1980, Vol. 12.
Nussbaum¸ Martha C. The Therapy of Desire. Princeton: Princeton University Press,
1994.
Nye, Andrea. The Princess and the Philosopher. Lanham: Rowman and Littlefield, 1999.
Oakeshott, Michael. “Rationalism in Politics.” Cambridge Journal, I, 1947.
Pascal, Blaise. Pensées, Honor Levi, trs. Oxford: Oxford University Press, 1995
Penrose, Roger. The Emperor’s New Mind. London: Vintage, 1990.
Popkin, Richard. The History of Skepticism from Erasmus to Spinoza (revised from 1960).
Berkeley: University of California Press, 1979.
Popper, Karl. Objective Knowledge. Oxford: Clarendon Press, 1972.
Priest, Graham. Beyond the limits of thought. Cambridge: Cambridge University Press,
1995.
Proust, Marcel. À la recherche du temps perdu. Jean-Yves Tadié et al. eds. Paris: NRF
Pléiade, 1987–1989.
Pseudo-Dionysius Areopagite. The Divine Names and Mystical Theology. John D.
Jones, trs. Milwaukee: Marquette University Press, 1980.
Resnick, Irven Michael. Divine Power and Possibility in St Peter Damian’s De Divine
Omnipotentia. Leiden: Brill, 1992.
Ricoeur, Paul. “On interpretation.” In Philosophy in France Today. K. McLaughlin, trs.
A. Montefiore, ed. Cambridge: Cambridge University Press, 1983.
Rorty, Richard. Philosophy and the Mirror of Nature. Princeton: Princeton University
Press, 1980.
Rose, Gillian. Mourning Becomes the Law. Cambridge: Cambridge University Press,
1996.
Ruben, David-Hillel. Explaining Explanation. London: Routledge, 1990.

BIBLIOGRAPHY

129

Russell, Bertrand. The Philosophy of Logical Atomism (1918). In Logic and Knowledge. R.
C. Marsh, ed. London: Routledge, 1988.
Ryle, Gilbert. The Concept of Mind. Harmondsworth: Penguin, 1963.
Schleiermacher, F. D. E. Hermeneutics and Criticism and Other Writings. Andrew
Bowie, ed. Cambridge: Cambridge University Press, 1998.
Schlesinger, George N. The Intelligibility of Nature. Aberdeen: Aberdeen University
Press, 1985.
Searle, John. Minds, Brains and Science. London: BBC, 1984.
Skinner, Quentin. “Meaning and Understanding in the History of Ideas.” In Visions of
Politics, Cambridge: Cambridge University Press, 2002, Vol. I.
Sperber, Dan. “Apparently Irrational Beliefs.” In Rationality and Relativism. Martin
Hollis and Steven Lukes, eds. Oxford: Blackwell, 1982.
Spinoza, Benedictus. Ethics (1677), Samuel Shirley, trs. Indianapolis: Hackett, 1992.
——— . Theological-Political Treatise (1670), Samuel Shirley, trs. Indianapolis: Hackett, 1998.
Strawson, P. F. The Bounds of Sense. London: Methuen, 1966.
Taylor, Charles. “Theories of Meaning.” In Philosophical Papers, Vol. I. Cambridge:
Cambridge University Press, 1985.
——— . Sources of the Self. Cambridge: Cambridge University Press, 1989.
——— . “Overcoming Epistemology.” In Philosophical Arguments. Cambridge: Harvard University Press, 1995.
Toulmin, Stephen. Cosmopolis. Chicago: University of Chicago Press, 1990. van Pelt, Robert Jan. The Science of Holocaust Research and the Art of Holocaust Denial.
University of Waterloo Research Lecture, 1999. von Wright, G. H. Explanation and Understanding. London: Routledge & Kegan Paul,
1971.
Williams, Bernard. Descartes. Harmondsworth: Penguin, 1978.
——— . “Deciding to Believe.” In Problems of the Self. Cambridge: Cambridge University Press, 1973.
——— . Making Sense of Humanity. Cambridge: Cambridge University Press, 1995
Wittgenstein, L. Philosophical Occasions 1912–1951, James Klagge and Alfred Nordmann, eds. Indianapolis: Hackett, 1993.
——— . Tractatus Logico-Philosophicus. D. F. Pears and B. F. McGuinness, trs. London:
Routledge, 1966 (with Introduction by Bertrand Russell).
——— . Philosophical Grammar, Rush Rhees, ed. Anthony Kenny, trs. Oxford: Blackwell, 1974.
——— . Philosophical Investigations. G. E. M. Anscombe, trs. Oxford: Blackwell, 1953
&c.

130

BIBLIOGRAPHY

——— . Zettel. G. E. M. Anscombe, trs. Oxford: Blackwell, 1967.
——— . On Certainty. D. Paul and G. E. M. Anscombe, trs. Oxford: Blackwell, 1969.
Wright, Crispin, Barry C. Smith, and Cynthia Macdonald, eds. Knowing Our Own
Minds. Oxford: Oxford University Press, 1998.
Wojcik, Jan W. Robert Boyle and the Limits of Reason. Cambridge: Cambridge University Press, 1997.
Ziff, Paul. Understanding Understanding. Ithaca: Cornell University Press, 1972.
Ziman, John. Real Science: What It Is and What It Means. Cambridge: Cambridge University Press, 2000.

Index of Names

Einstein, A., 63

Adler, J. E., 98n23
Anscombe, G. E. M., 73, 93–94
Appel, K. (and Haken, W.), 62
Aristotle, 21–22, 53
Augustine, 12, 34, 55
Ayer, A. J., 69

Feynman, R., 22n4
Foucault, M., 36, 81
Frege, G., 2, 3 9, 79
Freud, S., 8, 55
Gadamer, H.-G., 2, 12–13, 77, 85
Gödel, K., 56, 92

Baker, G. P. (and Hacker, P. M. S.), 25
Berkeley, G., 77
Borges, J.-L., 67
Bowie, A., 14
Boyle, R., 70–72, 89–90
Burnyeat, M., 32n22

Habermas, J., 36n28
Hammond, N., 99n25
Hegel, G. W. F., 20
Heidegger, M., 20, 25, 48, 108
Heraclitus, 64, 81
Hirst, P., 31n21
Hume, D., 7, 42–43, 73, 107

Cartwright, N., 29n19, 53
Chalmers, D., 55n5
Clastres, P., 10
Collingwood, R. G., 101
Conway, D., 106n2
Cooper, N., 8n4
Curley, E., 65n36

James, H., 24–25, 29
James, W., 106
Job, 5, 91
Johnson, W. E., 26
Joyce, J., 54, 95

Dante, 32, 67
Davidson, D., 77–78, 84
Derrida, J., 81
Descartes, R., 2–4, 61 8, 29, 36, Ch. 3 passim, 62–65, 68–69, 71, 74–75, 85,
87, 89, 93, 105–107, 109–111
Dobbs-Weinstein, I., 91n6
Donne, J., 85
Dostoyevsky, F., 16, 102–103
Dummett, M., 13–14, 15n17, 39–40, 48

Kant, I., 17–18, 27, 30–31, 33, 35,
37n29, 42, 46, 55, 57–65, 69, 72–73,
96, 99–100, 106
Kierkegaard, S., 102
Kivy, P., 17n18
Korsgaard, C., 60
Laing, R. (and Esterson, A.), 94n15
Langton, R., 62n31
131

132

INDEX OF NAMES

Levinas, E., 9, 86–87
Locke, J., 25, 43
Malcolm, N., 93
Mallarmé, S., 83
Marion, J.-L., 75n15
Marx, K., 56
Mason, R., 59n19
Maxwell, N., 106n2
McGinn, C., 91
Messiaen, O., 37
Moore, A. W., 63n32, 100
Murdoch, I., 25–26
Nagel, E. (and Newman, J. R.), 92
Nagel, T., 90, 95
Neiman, S., 59n18
Newman, J. H., 90, 93
Nietzsche, F., 27, 81
Nussbaum, M. C., 107n7
Nye, A., 107n9
Oakeshott, M., 22
Parmenides, 73
Pascal, B., 71, 99
Penrose, R., 30
Peter Darnian, 75
Pius IV, 41
Plato, 1–3, 5, 15, 17–18, 21, 24, 30, 32,
45, 51–53, 83, 107–108, 112
Popkin, R., 70
Popper, K., 24
Priest, G., 74n14
Proust, M., 82–83, 85, 113
Pseudo-Dionysius, 94n16

Ramanujan, S., 37
Ricoeur, P., 16, 27
Rorty, R., 40n2
Rose, G., 90
Ruben, D.-H., 28n17
Russell, B., 102, 106
Ryle, G., 25
Schleiermacher, F., 27, 34, 69
Schlesinger, G., 56n8, 76
Searle, J., 25
Seneca, 107
Shakespeare, W., 82, 96–97
Skinner, Q., 13
Socrates, 8, 34
Sperber, D., 84n38
Spinoza, B., 15, 19, 28, 43–46, 52, 59,
65, 97
Strawson, P., 60n26
Taylor, C., 9, 15n16, 40n2
Thales, 107
Thomas Aquinas, 67n1, 96
Toulmin, S., 106 van Pelt, R. J., 98n22 von Wright, G., 29
Weber, M., 99–101
Williams, B., 53, 71n7, 98n23
Wittgenstein, L., 5, 13–14, 20, 24–26,
35, 44, 46–47, 68, 77–80, 86, 92–94,
100, 102, 109
Wojcik, J. W., 90n2
Wright, C. et al., 7n1
Ziff, P., 14n12, 24n7, 84n37
Ziman, J., 23