Download the full version of the ebook at ebookfinal.

com

Underivative duty British moral philosophers from

Sidgwick to Ewing 1st Edition Thomas Hurka

https://ebookfinal.com/download/underivative-duty-british-
moral-philosophers-from-sidgwick-to-ewing-1st-edition-
thomas-hurka/

OR CLICK BUTTON

DOWNLOAD EBOOK

Download more ebook instantly today at https://ebookfinal.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Great Philosophers From Socrates to Foucault James Garvey

https://ebookfinal.com/download/great-philosophers-from-socrates-to-
foucault-james-garvey/

ebookfinal.com

Socratic Moral Psychology 1st Edition Thomas C. Brickhouse

https://ebookfinal.com/download/socratic-moral-psychology-1st-edition-
thomas-c-brickhouse/

ebookfinal.com

Stochastic Processes From Applications to Theory 1st

Edition Pierre Del Moral

https://ebookfinal.com/download/stochastic-processes-from-
applications-to-theory-1st-edition-pierre-del-moral/

ebookfinal.com

The Elements of Politics 1st Edition Henry Sidgwick

https://ebookfinal.com/download/the-elements-of-politics-1st-edition-
henry-sidgwick/

ebookfinal.com
Bioinformatics from genomes to therapies 1st Edition
Thomas Lengauer

https://ebookfinal.com/download/bioinformatics-from-genomes-to-
therapies-1st-edition-thomas-lengauer/

ebookfinal.com

Thomas Ewing Jr Frontier Lawyer and Civil War General

SHADES OF BLUE GRAY 1st Edition Ronald D. Smith

https://ebookfinal.com/download/thomas-ewing-jr-frontier-lawyer-and-
civil-war-general-shades-of-blue-gray-1st-edition-ronald-d-smith/

ebookfinal.com

Rudder From Leader to Legend 1st Edition Thomas M.

Hatfield

https://ebookfinal.com/download/rudder-from-leader-to-legend-1st-
edition-thomas-m-hatfield/

ebookfinal.com

From Plain Fare to Fusion Food British Diet from the 1890s
to the 1990s Derek J. Oddy

https://ebookfinal.com/download/from-plain-fare-to-fusion-food-
british-diet-from-the-1890s-to-the-1990s-derek-j-oddy/

ebookfinal.com

Poodles For Dummies 1st Edition Susan M. Ewing

https://ebookfinal.com/download/poodles-for-dummies-1st-edition-susan-
m-ewing/

ebookfinal.com
Underivative Duty
This page intentionally left blank
Underivative Duty
British Moral Philosophers from
Sidgwick to Ewing

EDITED BY

Thomas Hurka

C L A R E N D O N P R E S S · OX F O R D
3
Great Clarendon Street, Oxford ox2 6dp
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
# the several contributors 2011
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First published 2011
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data available
Typeset by SPI Publisher Services, Pondicherry, India
Printed in Great Britain
on acid-free paper by
MPG Books Group, Bodmin and King’s Lynn

ISBN 978-0-19-957744-6

1 3 5 7 9 10 8 6 4 2
Preface

Most of the papers in this volume derive from a conference on British Moral Philo-
sophers from Sidgwick to Ewing held at the University of Toronto’s Centre for Ethics
in April 2008. I am grateful to the Centre for its administrative support of the
conference, and to the university’s Jackman Humanities Institute and Department of
Philosophy for financial support. Thanks also to Terry Teskey for helping prepare the
papers for publication and compiling the Bibliography.
T.H.
This page intentionally left blank
Contents

List of Contributors viii

Introduction 1
Thomas Hurka
1. Common Themes from Sidgwick to Ewing 6
Thomas Hurka
2. Pleasure and Hedonism in Sidgwick 26
Roger Crisp
3. Ideal Utilitarianism: Rashdall and Moore 45
Anthony Skelton
4. McTaggart on Love 66
Dennis McKerlie
5. Has Anyone Ever Been a Non-Intuitionist? 87
Jonathan Dancy
6. Mistakes about Good: Prichard, Carritt, and Aristotle 106
T. H. Irwin
7. The Birth of Deontology 126
Robert Shaver
8. Eliminativism about Derivative Prima Facie Duties 146
Philip Stratton-Lake
9. Ross on Retributivism 166
Michael J. Zimmerman
10. A. C. Ewing’s First and Second Thoughts in Metaethics 183
Jonas Olson and Mark Timmons

Bibliography 212
Index 221
List of Contributors

Roger Crisp is Uehiro Fellow and Tutor at St Anne’s College, Oxford, and Professor
of Moral Philosophy at the University of Oxford.
Jonathan Dancy is Professor of Philosophy at the University of Reading and at the
University of Texas at Austin.
Thomas Hurka is Chancellor Henry N. R. Jackman Professor of Philosophical Studies
at the University of Toronto.
T. H. Irwin is Professor of Ancient Philosophy in the University of Oxford and a
Fellow of Keble College, Oxford. From 1975 to 2006 he taught at Cornell University.
Dennis McKerlie is Professor of Philosophy at the University of Calgary.
Jonas Olson is Docent (Reader) in Practical Philosophy at Stockholm University.
Robert Shaver is Professor of Philosophy at the University of Manitoba.
Anthony Skelton is Assistant Professor of Philosophy at the University of Western
Ontario.
Philip Stratton-Lake is Professor of Philosophy at the University of Reading.
Mark Timmons is Professor of Philosophy at the University of Arizona.
Michael J. Zimmerman is Professor of Philosophy at the University of North Carolina
at Greensboro.
 Introduction
Thomas Hurka

The essays in this volume concern a school of British moral philosophers active from
the 1870s through to the 1950s and including as principal figures Henry Sidgwick,
Hastings Rashdall, J. M. E. McTaggart, G. E. Moore, H. A. Prichard, E. F. Carritt,
W. D. Ross, C. D. Broad, and A. C. Ewing. Unfortunately, there is no satisfactory
name for this school. They are often called ‘non-naturalists’ or ‘intuitionists’, in
reference to their shared metaethical view that moral judgements can be objectively
true but are distinct from all other truths, so we know them by a distinctive moral
intuition. But these labels ignore their equally important shared views in normative
ethics, such as that all normative truths can be expressed using simple concepts such as
‘good’ and ‘ought’ and that the most basic moral duties are underivative, so there is no
further explanation why they hold. One could try calling the school ‘underivativists’,
‘primitivists’, or even ‘quietists’ (since they thought our everyday belief that we have
moral duties needs no philosophical vindication), but those names again highlight just
some of their shared beliefs and are hardly self-explanatory. Until a better name comes
along, they may have to be just the school of moral philosophers from Sidgwick to
Ewing.
This school had its peak influence between 1900 and 1939, but that faded soon
after World War II as new views emerged under the influence either of logical
positivism and its successors or of ordinary-language philosophy in the Oxford
version of J. L. Austin or the Cambridge one of Ludwig Wittgenstein. Critics on
the positivist side targeted mainly its non-naturalist metaethics (they had little interest
in normative ethics). To them the claim that moral statements such as ‘pleasure is
good’ can be objectively true violated a verificationist criterion of linguistic meaning
or, less contentiously, was inconsistent with a scientific worldview. This led many of
them to embrace non-cognitivism about moral judgement, according to which
‘pleasure is good’ expresses an emotion or issues a prescription and therefore cannot
be true.1

1
A. J. Ayer, Language, Truth, and Logic (London: Gollancz, 1936), chap. 6; Charles L. Stevenson, Ethics and
Language (New Haven: Yale University Press, 1944).
2 THOMAS HURKA

Those on the ordinary-language side also rejected non-naturalism, but partly on a

different basis. For them the assumption that moral judgements must assert truths, which
the non-naturalists tended to make, ignored the many non-assertoric ways language
can function; this was a different impetus from non-cognitivism.2 But these critics also
questioned the school’s normative views. Thus, many of them thought it a mistake to
formulate moral principles using only ‘thin’ concepts such as ‘good’ and ‘ought’. One
claimed that predicative uses of ‘good’, as in the ‘pleasure is good’ debated by Sidgwick,
Moore, and Ross, are ungrammatical; the only legitimate English use of ‘good’ is attribu-
tive, as in ‘good knife’ or ‘good human being’.3 Others found it more illuminating to
conduct ethical thought using subtler ‘thick’ concepts such as ‘generous’, ‘kind’, ‘coura-
geous’, and ‘just’.4 There were also objections to the idea that moral duties can be
underivative. One said talk of obligation or the moral law is incoherent apart from belief
in a divine legislator, another that claims about moral duty have no reason-giving force
unless connected by neo-Aristotelian arguments to ideas about how the virtues expressed
in doing one’s duty contribute to one’s leading a flourishing or good human life.5
These various arguments, plus a strong sense that it was new days in philosophy, led
many philosophers of the 1950s to dismiss the earlier school’s writings as worthless, and
for several decades they were barely read. Philosophers would of course know the first
chapter of Moore’s Principia Ethica, but often viewed it as just the best-known
statement of the non-naturalism exposing whose errors had led to the sounder views
of later years, and almost never discussed the book’s far more brilliant and original sixth
chapter. They would likewise know in general about Ross’s concept of prima facie
duty, but not that he gave different accounts of it in his two books;6 a student of
the 1960s or 1970s would not usually read his actual account of what these duties
are, let alone his discussion of what states of affairs are good. (Perfectionist views about
the good such as Moore’s and Ross’s were almost totally ignored in these years;

2
J. L. Austin, How to Do Things with Words (New York: Oxford University Press, 1962); R. M. Hare, The
Language of Morals (Oxford: Clarendon Press, 1952).
3
P. T. Geach, ‘Good and Evil’, Analysis, 17 (1956). Geach’s grammatical claim about English is refuted
by, among other things, the opening lines of the Bible, which report how, after creating each aspect of the
world, God ‘saw that it was good’; thus, after creating light God ‘saw the light, that it was good’. God was not
here feeling relief that he created good light rather than bad light (whatever that could be); he was seeing that
the existence of light is simply or predicatively good. Yet surely Geach would not accuse the author of these
lines or the millions of readers who have felt no puzzlement about them of confusion about the English
language.
4
G. E. M. Anscombe, ‘Modern Moral Philosophy’, Philosophy, 32 (1958), 15–17; for a parallel claim
about aesthetics, recommending ‘dainty’ and ‘dumpy’ over ‘beautiful’, see J. L. Austin, ‘A Plea for Excuses’,
Proceedings of the Aristotelian Society, 56 (1955–6), 9.
5
Anscombe, ‘Modern Moral Philosophy’, makes both these objections; for a more recent version of the
second, see Thomas Baldwin, G. E. Moore (London: Routledge, 1990), 118–19.
6
W. D. Ross, The Right and the Good (Oxford: Clarendon Press, 1930), 19–20, 28–9; Foundations of Ethics
(Oxford: Clarendon Press, 1939), 51–5, 79–86. John Searle’s ‘Prima Facie Obligations’, in Joseph Raz (ed.),
Practical Reasoning (Oxford: Oxford University Press, 1978), ascribes to Ross as his ‘official view’ an
understanding of prima facie duties that Ross explicitly rejects in both books, and then proposes as a new
and better understanding the one Ross clearly intends in The Right and the Good.
 I N T RO D U C T I O N 3

mentioning them in public could even elicit laughter.) And lesser writers of the school
such as McTaggart, Rashdall, Carritt, and Ewing were not read at all or even remem-
bered; they fell into a black hole. In the 1970s John Rawls stimulated interest in
Sidgwick by rightly treating him as the greatest of the classical utilitarians, and this
interest had a historical-scholarly side. But it linked Sidgwick far more to earlier
nineteenth-century figures than to the twentieth-century school that followed him.7
In mainstream moral philosophy it was as if most of the school’s works did not exist.
Philosophy keeps changing, however, and interest in the school is now beginning to
revive. The last decade has seen new editions of works by Sidgwick, Ross, Prichard,
and Moore,8 and more frequent references to their and their contemporaries’ writings;
they are slowly becoming more visible. One reason is that moral philosophy is
returning to some of their views. Though by no means the currently dominant
metaethical view, non-naturalism now has prominent defenders.9 (They sometimes
say their views are less philosophically contentious, because less metaphysically com-
mitted, than, say, Moore’s, but I see little basis for these claims.) The idea that the basic
moral duties are underivative is also gaining adherents, especially given what many see
as the failure of alternative programmes such as the neo-Aristotelian one proposed in
the 1950s. And there has been a convergence with the methodology of at least later
members of the school such as Prichard, Ross, and Broad. They saw their task as that of
theorizing an everyday morality they saw as complex, pluralistic, and subtle, or of
systematizing it in a way that retains rather than obliterates those features. As current
moral theory has become more aware of the subtleties in everyday moral thought, it
has become more sympathetic to writers who shared that awareness.
Moreover, to read the school’s works is to find anticipations of many supposedly
new contributions of the last few decades. The ‘buck-passing’ or ‘fitting-attitude’
analysis of value was defended by Sidgwick, Broad, and especially Ewing but rejected
by Moore and Ross; their debates about it have much to teach us today. The ‘Frege–
Geach’ objection to non-cognitivism was made by Ross as early as 1939,10 while
the ‘repugnant conclusion’ prominent in the recent literature on population ethics
was formulated in a single-life version by McTaggart, who even called it a ‘conclu-
sion [that] would . . . be repugnant to certain moralists’.11 The distinction between

7
J. B. Schneewind, Sidgwick’s Ethics and Victorian Moral Philosophy (Oxford: Clarendon Press, 1977).
8
Henry Sidgwick, Essays on Ethics and Method, ed. Marcus G. Singer (Oxford: Clarendon Press, 2000);
W. D. Ross, The Right and the Good, ed. Philip Stratton-Lake (Oxford: Clarendon Press, 2002); H. A. Prichard,
Moral Writings, ed. Jim MacAdam (Oxford: Clarendon Press, 2002); G. E. Moore, Ethics, ed. William H. Shaw
(Oxford: Clarendon Press, 2005).
9
Thomas Nagel, The View From Nowhere (New York: Oxford University Press, 1986), chap. 8;
T. M. Scanlon, What We Owe to Each Other (Cambridge, Mass.: Harvard University Press, 1998), chap. 1;
Russ Shafer-Landau, Moral Realism: A Defence (Oxford: Clarendon Press, 2003).
10
Ross, Foundations of Ethics, 33–4.
11
J. M. E. McTaggart, The Nature of Existence, 2 vols. (Cambridge: Cambridge University Press, 1927), ii.
452–3.
4 THOMAS HURKA

agent-neutral and agent-relative ‘oughts’, the former assigning the same moral goal
to all agents while the latter assigns different goals, was described in just those terms
by Sidgwick and Broad.12 Moore’s principle of organic unities is the first clear
statement of ‘moral contextualism’, the view that the moral difference a given factor
makes can vary from context to context, depending on what other factors it is
combined with. (Moore applied this principle only to claims about the good, but
extending it to right-making factors is a small step.13) The idea that we have stronger
duties to people close to us, such as our family, friends, and co-nationals, was
discussed by Broad under the heading of ‘self-referential altruism’,14 while Rashdall’s
idea that loving something good is another intrinsic good, which was central to his
account of virtue as well as to those of Moore, Ross, and others, has been resurrected
in recent writing about both value and virtue.15
The increased philosophical interest in the school has also led to an increase in
historical interest, as philosophers do not just cite an isolated text from Ross or Broad
while developing some view of their own, but try to understand that philosopher’s
position as a whole, including both the relations between its parts and its place in the
larger context of work of its time. And there is fascinating work to be done here. Not
only is the intellectual quality of the subject works high, but, given their long neglect,
they have not been much studied, which leaves considerable scope for original research.
A scholar of this school can raise new interpretive questions about its members rather
than just add to debates that have run for decades; she can make new discoveries about
Prichard or Ewing rather than contribute yet another book on the ethics of Aristotle
or Kant.
In response to these developments, a conference on ‘British Moral Philosophers
from Sidgwick to Ewing’ was held at the University of Toronto’s Centre for Ethics in
April 2008. Eight papers in this volume derive from that conference, with one an
amalgam of two given on Ewing’s metaethics; two other papers were added later. After
an opening survey of the school as a whole, the papers discuss specific issues that arise in
the work of one or a few of its members. And these tend to be new issues rather than
ones that have already been extensively discussed: they include McTaggart on the value
of love, Prichard and Carritt as critics of Aristotle’s ethics, Ross on retributivism and the
criminal law, and Ewing’s proposal of a ‘middle way’ between non-naturalism and
non-cognitivism. The papers are works of historical scholarship but also philosophically
engaged, assessing their subject philosophers’ ideas and relating them to contemporary

12
Henry Sidgwick, The Methods of Ethics, 7th edn. (London: Macmillan, 1907), 420–1, 497–8, and
‘Mr. Barratt on “The Suppression of Egoism”’, Mind, os 2 (1887), 411; C. D. Broad, ‘Self and Others’, in
Broad’s Critical Essays in Moral Philosophy, ed. David Cheney (London: George Allen & Unwin, 1971), 267.
13
For this step see Shelly Kagan, ‘The Additive Fallacy’, Ethics, 99 (1988); and F. M. Kamm, Morality,
Mortality, vol. 1: Rights, Duties, and Status (New York: Oxford University Press, 1996), chap. 2.
14
Broad, ‘Self and Others’, 279–80.
15
Robert Nozick, Philosophical Explanations (Cambridge, Mass.: Harvard University Press, 1981), 428–36;
Thomas Hurka, Virtue, Vice, and Value (New York: Oxford University Press, 2001); Robert Merrihew Adams,
A Theory of Virtue: Excellence in Being for the Good (Oxford: Clarendon Press, 2006).
 I N T RO D U C T I O N 5

debates; they want to know what is persuasive in them and what is not. And while
some authors are critical of their subjects, all treat their works as ones that are important
and that can contribute to present-day understanding.
The desire to raise new topics explains a feature of the volume that some may find
surprising: that it contains no essay primarily on Moore. This is due partly to the fact
that authors were invited to write on a philosopher of their choice and no one chose
Moore. But behind that is the fact that of all members of the school, Moore has been by
far the most discussed, especially recently, when the centenary of Principia Ethica in
2003 prompted several celebratory conferences and volumes on Moore’s ethics, to
which some of this volume’s authors contributed.16 Moore appears in many of this
book’s papers, as a parallel with or contrast to its main subject, but his open-question
argument and even his principle of organic unities are no longer topics of the under-
discussed kind this volume tries to address. (There is also no paper primarily about
Broad, but that is more an accident.)
There are any number of topics worth exploring in the rich body of moral-
philosophical writing published in Britain between Sidgwick’s Methods of Ethics
(1st edition 1874) and Ewing’s Second Thoughts in Moral Philosophy (1959). This
volume’s hope is that by discussing some of these it can stimulate interest in others
and in the school that produced them.

16
‘Centenary Symposium on G. E. Moore’s Principia Ethica’, Ethics, 113 (2003); Terence Horgan and
Mark Timmons (eds.), Metaethics After Moore (Oxford: Oxford University Press, 2006). See also Susana
Nuccetelli and Gary Seay (eds.), Themes From G. E. Moore: New Essays in Epistemology and Ethics (Oxford:
Oxford University Press, 2007).
 1
Common Themes from Sidgwick
to Ewing
Thomas Hurka

The philosophers discussed in this volume—Henry Sidgwick, Hastings Rashdall, J. M. E.

McTaggart, G. E. Moore, H. A. Prichard, E. F. Carritt, W. D. Ross, C. D. Broad,
and A. C. Ewing—form a unified and distinctive school in the history of ethics.
They all taught at Oxford or Cambridge in the late nineteenth or early twentieth
century, and their principal writings run from the first edition of Sidgwick’s Methods
of Ethics in 1874 through to Ewing’s Second Thoughts in Moral Philosophy of 1959.
But their period of greatest influence was in the first four decades of the twentieth
century. Before 1900 (and perhaps for some time after) British moral philosophy
was dominated by the Idealism of T. H. Green, F. H. Bradley, and Bernard
Bosanquet; after World War II logical positivism and ordinary-language philosophy
spawned new approaches to the subject that rejected the earlier school’s ideas. But
in the first third of the century theirs were the dominant voices in the field.
The school’s members differed on many important topics. Sidgwick, Rashdall,
McTaggart, and Moore were consequentialists, while Prichard, Carritt, Ross, and
Broad defended versions of deontology. Sidgwick thought there is only one intrinsic
good, pleasure; the others endorsed a plurality of goods, including perfectionist goods
such as knowledge and virtue. Sidgwick and Moore thought the most reliable moral
judgements are about abstract principles, Carritt and Ewing that they concern particular
cases. But underneath these disagreements was a set of shared assumptions about what
morality is and how moral theory should be pursued. These allowed their disputes to
arise on a common platform, so they were always substantive rather than tied up in
methodology. They also distinguished the school both from earlier figures in the history
of ethics such as Aristotle and Kant and from much present-day ethics. This chapter will
try to identify these common assumptions, show how they were shared, and comment
on their merits.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 7

Conceptual Minimalism
The first common assumption is what I will call ‘conceptual minimalism’, the view that
all normative judgements can be expressed using just a few basic concepts. Some
members of the school recognized only one irreducible normative concept: for Sidg-
wick, what one ‘ought’ or has ‘reason’ to do; for Rashdall and the Moore of Principia
Ethica, what is ‘intrinsically good’; and for Broad and Ewing in certain writings, what is
‘fitting’. Others, such as Prichard, the later Moore, and Ross, thought there were two
basic concepts, often ‘ought’ and ‘good’. But the number of underivative normative
concepts they employed was always very small.
This marks a contrast with much present-day ethics, which employs a larger number
of distinct normative concepts and spends considerable time debating their relations.
The Sidgwick-to-Ewing school were certainly aware of these concepts, but took either
of two lines about them. One was to argue that they can be reductively analysed using
the basic normative concepts and some more or less determinate description; the other
was to deny that they are normative. Either way they denied, of many familiar concepts,
that they are both irreducible and normative.
Thus their basic normative concepts were all ‘thin’ concepts such as ‘ought’ and
‘good’ rather than ‘thick’ ones such as the virtue-concepts ‘courageous’, ‘generous’, and
‘malicious’. (A thin concept says nothing about what non-normative properties an item
must have to fall under it, but a thick concept does. A courageous act, for example, must
somehow involve facing danger or accepting the risk of harm.) Some present-day
philosophers hold that thick concepts are irreducible to thin ones,1 but our school
held that they can all be reductively analysed. To the view that the virtue-concepts do
‘not admit of being stated in definite formulae’, Sidgwick replied, ‘our notions of special
virtues do not really become more independent by becoming more indefinite: they still
contain, though perhaps more latently, the same reference to “Good” or “Well-being”
as an ultimate standard’. And only if they do can we explain the difference between the
virtues and their cognate vices, for example, between courage and foolhardiness or
between generosity and profusion.2 Others gave reductive analyses of the concept of
virtue in general, which Ross, for example, took to involve a desire either for acts that
are independently right or for states of affairs that are independently good.3 And the
general point that the school treated only the thin concepts as basic is surely unconten-
tious: the title of Ross’s great book is not The Courageous, Generous, Kindly, Dumpy, and
Dainty; it is The Right and the Good.

1
See e.g. John McDowell, ‘Non-Cognitivism and Rule-Following’, in S. H. Holtzman and C. M. Leich
(eds.), Wittgenstein: To Follow a Rule (London: Routledge & Kegan Paul, 1981), 144. For a reply to this
argument, see Daniel Y. Elstein and Thomas Hurka, ‘From Thick to Thin: Two Moral Reduction Plans’,
Canadian Journal of Philosophy, 39 (2009).
2
Henry Sidgwick, The Methods of Ethics, 7th edn. (London: Macmillan, 1907), 392. Unless otherwise
specified, all references will be to the 7th edition of this work.
3
W. D. Ross, The Right and the Good (Oxford: Clarendon Press, 1930), 134.
8 THOMAS HURKA

In addition, the school did not draw irreducible distinctions between uses of their thin
concepts. They often distinguished between moral and non-moral goodness, for example,
but moral goodness was just the same property of intrinsic goodness when had by a
particular kind of object, say, an attitude to something independently right or good. (For
Ross, therefore, moral goodness is just intrinsic goodness when had by virtue.4) Nor did
they use the present-day concept of ‘welfare’, ‘well-being’, or what is ‘good for’ a person
as distinct from what is simply ‘good’. Rashdall and Moore defined ‘my good’ as that part
of what is intrinsically good that is located in me, as did Sidgwick. For him, ‘my good’ is
what I ought to desire—his general definition of goodness—‘assuming my own exis-
tence alone to be considered’, that is, considering only states of myself.5 Sidgwick did
accept a ‘good for’ concept that Moore did not, and used it to argue, as Moore would
later deny, that egoism is internally consistent. But this was the concept of agent-relative
goodness, or of what is good from a particular person’s point of view, so he and perhaps
only he ought to desire it. This is again not the present-day concept of ‘welfare’; it is
another that is reductively analysed using Sidgwick’s one basic concept ‘ought’.6
If these are reductive analyses of normative concepts, Prichard took the alternative
line with the concept of what is a ‘good to’ a person, which he said is not a normative
or even evaluative concept, but makes just the descriptive claim that something will
satisfy a person’s desires or, more accurately, give her pleasure. Ross had a similar view
about attributive uses of ‘good’, as in ‘good knife’ and ‘good liar’; these too make only
the descriptive claim that something will be an effective means to some purpose, so
there is again no genuine value-concept independent of intrinsic goodness.7
The school likewise made no irreducible distinctions between uses of ‘ought’, for
example, between moral ‘oughts’ on the one side and prudential or rational ones on
the other. Since for them all genuine ‘oughts’ were categorical and, following Kant, all
categorical ‘oughts’ were moral, the only genuine ‘oughts’ were moral. This view was
reflected in Sidgwick’s classing egoism as a ‘method of ethics’. The conflict between
it and utilitarianism is not, as many present-day philosophers would say, between
morality and prudence or rationality; it is between two ethical theories making
competing claims using the same concept of ‘ought’. Or consider Prichard’s and
Ross’s claim that there is no moral duty to pursue one’s own pleasure.8 They did not

4
Ibid. 155. See also Sidgwick, Methods of Ethics, 1st edn. (London: Macmillan, 1874), 93 n.; Hastings
Rashdall, The Theory of Good and Evil (London: Oxford University Press, 1907), i. 138, 174–5; and A. C. Ewing,
‘A Suggested Non-Naturalistic Definition of Good’, Mind, 48 (1939), 5. Note that much of Rashdall’s book
incorporates material from articles published as early as 1885; though I will cite this material from the book, it will
often have been written earlier.
5
Rashdall, Theory of Good and Evil, ii. 98 n.; G. E. Moore, Principia Ethica (Cambridge: Cambridge
University Press, 1903), 98–9; Sidgwick, Methods of Ethics, 112.
6
Sidgwick, Methods of Ethics, 420, 497–8. For a fuller defence of this reading of Sidgwick on ‘good for’, see
my ‘Moore in the Middle’, Ethics, 113 (2003), 611–12.
7
H. A. Prichard, Moral Writings, ed. Jim MacAdam (Oxford: Clarendon Press, 2002), 172–6; Ross, Right
and the Good, 65–7, and Foundations of Ethics (Oxford: Clarendon Press, 1939), 255–7.
8
Prichard, Moral Writings, 10 n., 135, 171, 204; Ross, Right and the Good, 21, 24–6, 151, and Foundations of
Ethics, 72–5, 129–30, 272–4, 284.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 9

add that of course there is a prudential or rational duty to pursue one’s pleasure,
assuming instead that where there is no moral duty there is no duty at all.
The non-moral ‘ought’ is often represented by Kant’s hypothetical imperative,
about which the school again took either of two lines. Sidgwick at one point suggested
the ‘wide-scope’ reading of this imperative, on which it is not a hypothetical with an
imperative consequent but a command to make a hypothetical true, namely ‘make
it the case that: if you have some end, you take (what you believe are) effective means
to it’. On this view the hypothetical imperative is really a categorical imperative with a
distinctive content, enjoining a kind of coherence between one’s ends, beliefs, and
acts.9 But Prichard and Ross again took the alternative line, holding that the hypo-
thetical imperative is not normative and therefore not an imperative at all, making only
the descriptive claim that a certain act is necessary for achieving some end.10
The view that there are no ‘oughts’ other than moral ones11 helped ground the
school’s belief that it is a mistake to ask ‘Why be moral?’ or ‘Why ought I to do what
I morally ought?’: if the only ‘oughts’ are moral ones, these questions cannot arise.
Though most associated with Prichard, this belief was in fact widely shared. Moore said
the question ‘Why should I do my duty?’ is ‘puzzling’, since it reduces to ‘“Why is duty
duty?” or “Why is good good?”’, while Carritt wrote: ‘If anyone ask us, “Why ought I
to do these acts you call my duty?” the only answer is, “Because they are your duty.”’12
Sidgwick may seem to fall outside this consensus, since he said we ask ‘Why should I do
what I see to be right?’ whereas we do not ask ‘Why should I believe what I see to be
true?’ But his explanation was that in the first case we are torn between different
substantive views about what is right and express our uncertainty by asking the
question; this implies that if we had no doubts about what is right, there would be
nothing to ask. And there can be nothing if the only ‘oughts’ are moral.13
Of course this view about ‘ought’ does not show that any particular moral claims,
such as Sidgwick’s consequentialist or Prichard’s deontological ones, are true. But it
does help defend these claims against sceptical attacks. Consider the instrumentalist

9
Sidgwick, Methods of Ethics, 37. For more recent defences of the wide-scope reading, see Patricia
S. Greenspan, ‘Conditional Oughts and Hypothetical Imperatives’, Journal of Philosophy, 72 (1975); and John
Broome, ‘Normative Requirements’, in Normativity, ed. Jonathan Dancy (Oxford: Blackwell, 2000).
10
Prichard, Moral Writings, 54–5, 126–8, 135, 143–4, 166; Ross, Foundations of Ethics, 48.
11
What about epistemic or logical ‘oughts’ that say we ought to believe a conclusion given certain
evidence or premises? The school did not discuss these but could have taken them to involve hypothetical
imperatives, of the form: ‘If you want to believe the truth, believe this conclusion.’ For Prichard and Ross this
would mean the epistemic ‘ought’ is not really normative but says only that believing a given conclusion is an
effective means to believing what is true.
12
G. E. Moore, The Elements of Ethics, ed. Tom Regan (Philadelphia: Temple University Press, 1991), 17–18;
E. F. Carritt, The Theory of Morals (London: Oxford University Press, 1928), 29.
13
Sidgwick, Methods of Ethics, 5–6. For the same reason, the school would have rejected as empty a
question some present-day philosophers discuss: whether an act’s being morally right gives us reason to
perform it. For them talk of ‘reasons’ was just another way of talking about what we ought to do, and if the
only ‘oughts’ are moral, the question reduces to whether an act’s being one we morally ought to do makes it
one we morally ought to do. The answer is no, but on entirely trivial grounds.
10 THOMAS HURKA

view that what we ought rationally to do is always whatever will best satisfy our current
desires, which sceptics say fulfilling narrowly ‘moral’ duties does not always do. Given
the school’s conceptual minimalism, the instrumental principle must be restated as a
moral one, using the same ‘ought’ as consequentialism and deontology and saying that
what we ought morally to do is satisfy our desires. But then the principle just seems
false. If someone desires above all to get rich and knows the most effective way of doing
so is to kill a relative and inherit her wealth, is it true that he ought simply to kill his
relative or that he ought simply not to? Surely it is the latter.14
The school’s shared minimalism did not mean they had no disagreements about the
normative concepts; on the contrary, there were lively disputes about, for example, the
concept ‘good’. Moore and Ross thought ‘good’ a simple, unanalysable concept,15
whereas Sidgwick, Broad, and Ewing analysed it reductively as what one ‘ought’ to
desire or what it is ‘fitting’ to desire.16 The ensuing debate was vigorous, but it took
place against the backdrop of the shared view that, whatever exactly they are, the basic
normative concepts are few in number.
This shared minimalism has several merits. First, it allows illuminating moral ex-
planations that non-minimalist views cannot give. For example, if we analyse courage
as involving, roughly, accepting the risk of harm to oneself for the sake of sufficiently
great intrinsic goods, such as preserving one’s nation, we can explain why the sacrifice
of the Spartans at Thermopylae was courageous but refusing a robber’s demand for
‘A penny or your life’ is foolhardy—something we could not do if courage and
foolhardiness were irreducibly thick concepts applied only by quasi-perceptual intuition.
Second, minimalism makes all normative questions substantive rather than conceptual.
Imagine that we ask which life would be better for a child we are raising: a life with more
virtue but less pleasure or one with more pleasure and less virtue. If this question can even
arise, the pleasure and virtue must at some level have the same kind of value, and we are
asking which in this case has more of that value. If we like, we can call the value of the virtue
‘moral’ and connect that of the pleasure to ‘welfare’, but this does not change the core issue,
which is the substantive one of which of two states has more of a common value. And the
labelling only invites confusion, by suggesting that the issue turns somehow on conceptual
questions about what ‘moral goodness’ and ‘well-being’ in the abstract consist in.
Or consider the conflict Sidgwick wrestled with, between a principle saying we
ought to promote the pleasure of everyone and one saying we ought to promote only
our own pleasure. This conflict again presupposes that the two principles use the same
concept, now the same ‘ought’, so the issue is which of two claims using that concept

14
Prichard, Moral Writings, 7–9, 18–20, 23–30.
15
Moore, Principia Ethica, 6–7, 9–10, 17, 21, 37, and ‘A Reply to My Critics’, in The Philosophy of
G. E. Moore, ed. P. A. Schilpp (Chicago: Northwestern University Press, 1942), 554–81; Ross, Right and
the Good, 75–8, 91–4, 131–2, and Foundations of Ethics, 278–83.
16
Sidgwick, Methods of Ethics, 92 n., 112, 381, 388; C. D. Broad, Five Types of Ethical Theory (London:
Routledge & Kegan Paul, 1930), 277–8; Ewing, ‘A Suggested Non-Naturalistic Definition’, and The
Definition of Good (London: Routledge & Kegan Paul, 1947), chap. 5.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 11

has greater weight. We could call the first principle ‘moral’, given its content, and the
second ‘prudential’, but this would again suggest, misleadingly, that the issue turns on
conceptual questions about ‘morality’ and ‘prudence’. Minimalism again has the merit
of treating a substantive issue as just substantive.
So the Sidgwick-to-Ewing school built their moral theories using a small roster of
basic concepts, but what resulted often had a rich and varied content. Moore combined
his one concept of intrinsic goodness with a principle of organic unities and recursive
principles about, for example, the goodness of loving what is good to construct a highly
elaborate account of the things that are intrinsically valuable; Ross had a similarly
complex theory of what is right. Their work illustrates what can be called the
‘vinaigrette’ approach to moral theory. It is well known that the key to a good
vinaigrette is to be a spendthrift with the oil and a miser with the vinegar; in moral
theory, it is to be a miser with moral concepts and a spendthrift with moral claims.

Non-Naturalism
The Sidgwick-to-Ewing school are best known for their non-naturalist metaethics,
which combines the realist thesis that some moral judgements are objectively true with
the autonomy-of-ethics thesis that they are neither reducible to nor derivable from
non-moral judgements, such as those of empirical science, metaphysics, or religion. On
the one hand there is moral truth; on the other hand it is a distinctive or sui generis truth.
Let us consider these theses in turn.
Early in the school’s history the realist side of non-naturalism seems to have been
simply assumed. In the preface to the first edition of The Methods of Ethics Sidgwick
remarked casually that he would assume we can know ethical truths. The first sentence
of Moore’s Principia Ethica said, without feeling any need for supporting argument, that
philosophical ethics is concerned with which ethical judgements are true, while Ross
affirmed ‘a system of moral truth, as objective as all truth must be’.17
The school were doubtless influenced here by the grammar of moral judgements,
which parallels that of uncontroversially truth-apt judgements such as those of science.
But they also seem not to have been aware of serious alternatives to realism, in
particular non-cognitivist ones. One might think they had a non-cognitivist theory
to hand in Hume, but they all read him as a subjective naturalist, for whom the claim
that an act is right merely reports the psychological fact that the speaker or some group
approves of the act. Having refuted that naturalist view, often at considerable length,18
they took themselves to have refuted subjectivism more generally.

17
Sidgwick, Methods of Ethics, p. vii; Moore, Principia Ethica, 1; Ross, Right and the Good, 15 (see also
pp. 29–30).
18
See e.g. Sidgwick, Methods of Ethics, 25–8, 31; Moore, Ethics (London: Oxford University Press, 1912),
chaps. 3–4; Ross, Right and the Good, 11, 80–104, and Foundations of Ethics, 22–6; and Broad, Five Types, 84–6,
259–64.
12 THOMAS HURKA

When non-cognitivist theories began to appear in the 1930s, the school’s members
were initially hostile, accusing writers like A. J. Ayer of applying a general thesis about
language to ethics without seriously considering whether it fitted that case19 and raising
particular objections that were sometimes weak but sometimes, as in Ross’s anticipa-
tion of the Frege–Geach objection, more telling.20 But as writers like C. L. Stevenson
and R. M. Hare began to ground their non-cognitivism in specifically ethical claims
about the motivating power of moral judgements, some in the school became
more sympathetic. Moore famously flirted with non-cognitivism in his 1942 ‘Reply
to My Critics’,21 Broad eventually found himself inclining toward it,22 and Ewing
incorporated non-cognitivist elements in the compromise metaethics of Second
Thoughts.23
It is probably true that none of the school fully and permanently abandoned realism,
but the fact that some considered doing so suggests that their realism, however much
associated with them by later critics, was not vital to their overall ethical view. Much
more crucial, I would argue, especially for their normative theorizing, was their belief
in the autonomy of ethics, which the non-cognitivists shared.
If the school recognized no ‘oughts’ other than moral ones, their autonomy-of-
ethics thesis was equivalent to a broader thesis about the autonomy of the normative
from the non-normative: values in general are distinct from non-evaluative facts, and
no ‘ought’ can be derived from an ‘is’. Their principal argument for the claim was the
‘open-question’ argument, commonly attributed to Moore but, as many have noted,
used earlier by Sidgwick, Rashdall, and others.24 Moore’s version of the argument
was lengthier than the earlier writers’, which helped give it more influence on
later philosophy. It was also more combative in tone. Sidgwick had said that since
Bentham’s equating of ‘good’ with ‘pleasant’ would turn the hedonist ‘pleasure is
good’ into a tautology, Bentham ought to be read in some more charitable way;
Moore pounded away at the error relentlessly.25 Sidgwick’s attitude was, ‘That would

19
Carritt, ‘Moral Positivism and Moral Aestheticism’, Philosophy, 13 (1938), 132–3, 140; Broad, ‘Some
Reflections on Moral Sense Theories in Ethics’, in Broad’s Critical Essays in Moral Philosophy, ed. David
Cheney (London: George Allen & Unwin, 1971), 190, and ‘Critical Notice of H. A. Prichard, Moral
Obligation’, Mind, 59 (1950), 566.
20
Carritt, ‘Moral Positivism and Moral Aestheticism’, 133–4; Ross, Foundations of Ethics, 33–4.
21
Moore, ‘Reply to My Critics’, 544–5. The flirtation was, to be sure, temporary. After Moore’s death
Ewing wrote, ‘I think I ought to mention that Moore completely retracted this statement [that he was
‘strongly inclined’ to accept a non-cognitivist view] in the later years of his life (and here Blanshard would
confirm what I say). Moore told me orally that he still held to his old view, and further that he could not
imagine whatever in the world had induced him to say that he was almost equally inclined to hold the other
view’ (Ewing, ‘G. E. Moore’, Mind, 70 (1962), 251).
22
Broad, ‘A Reply to My Critics’, in The Philosophy of C. D. Broad, ed. P. A. Schilpp (New York: Tudor,
1959), 817.
23
Ewing, Second Thoughts in Moral Philosophy (London: Routledge & Kegan Paul, 1959), chap. 2.
24
Moore, Principia Ethica, 10–17; Sidgwick, Methods of Ethics, 26 n., 109, and Lectures on the Ethics of
T. H. Green, Mr. Herbert Spencer, and J. Martineau (London: Macmillan, 1902), 145; Rashdall, Theory of Good
and Evil, i. 47–8.
25
Sidgwick, Methods of Ethics, 26 n.; Moore, Principia Ethica, 17–20.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 13

be stupid, therefore Bentham didn’t really mean it’; Moore’s was, ‘Bentham meant it,
therefore he was really stupid’.
The non-cognitivists grounded the open-question argument in the claim that
normative judgements are intrinsically motivating whereas descriptive judgements
are not—that is why the former cannot be derived from the latter. But this was not a
major issue for the earlier school. Moore remarked casually that when we believe
something is good we are usually motivated to pursue it, but he did not think it
important to specify that claim further, nor did he connect it to his open-question
argument.26 More generally, the issue of moral motivation, or of internalism vs.
externalism about moral judgement, was not one that much concerned the school or
on which their views are easy to discern. (Though many of them seem to have been
externalists, they often expressed themselves in what could be internalist language.27)
For them, the sui generis character of normative judgements was something one just sees
or is led by argument to see, independently of claims about action-guidingness.
A common present-day response to the open-question argument is that it ignores the
possibility of non-analytic property-identities. Just as water is identical to H2O even
though ‘water’ does not mean the same as ‘H2O’, critics say, so goodness could be non-
analytically identical to a natural property such as pleasantness even though ‘good’ does
not mean ‘pleasant’.28 But Sidgwick, Moore, and the others could respond that the
identity of water and H2O depends on specific features of natural-kind properties that are
not present in normative properties like goodness. The property of being water is the
property of having that underlying structure, whatever it is, that explains the behaviour
of the stuff we find in lakes, rivers, and so on, but the property of goodness has no such
inner complexity. It could be identical to a natural property only if it were analytically
identical to that property, and the open-question argument shows it is not.29
Whatever its basis, the autonomy-of-ethics thesis was vitally important to the
Sidgwick-to-Ewing school, and they spent a surprising number of pages combating
naturalistic programmes grounding ethics in evolutionary biology, empirical
psychology—from associationism in the nineteenth century to Freudianism in the
twentieth—Idealist metaphysics, and religion.30 For them, it was crucial that ethical
truths are a category apart, knowable only by distinctively ethical means.

26
Moore, Principia Ethica, 131.
27
For a careful analysis of Sidgwick on this issue, see Robert Shaver, ‘Sidgwick on Moral Motivation’,
Philosopher’s Imprint, 6 (2006).
28
See e.g. David O. Brink, Moral Realism and the Foundations of Ethics (Cambridge: Cambridge University
Press, 1989), chap. 6.
29
I am indebted here to unpublished writing by Derek Parfit.
30
See e.g. Sidgwick, Methods of Ethics, 31, 505, and Lectures on Green, Spencer, and Martineau, 1–14, 60–79,
143–53; Rashdall, Theory of Good and Evil, ii. 189–413; Moore, Principia Ethica, chaps. 2–4; Ross, Right and the
Good, 12–15, and Foundations of Ethics, 12–17; Broad, ‘Critical Notice of Julian Huxley’s Evolutionary Ethics’,
in Broad’s Critical Essays in Moral Philosophy, and ‘Symposium on the Relations Between Science and Ethics’,
Proceedings of the Aristotelian Society, 42 (1941–2); Ewing, Definition of Good, 23–5, 73–4, 106–10, 134, and
‘Symposium on the Relations Between Science and Ethics’, Proceedings of the Aristotelian Society, 42 (1941–2).
14 THOMAS HURKA

These means were, of course, intuitive, involving the direct apprehension of moral
truth, often as self-evident. The school’s members regularly said, when some normative
issue reached its crux, that the decision must be made by each person examining his
own moral consciousness, or giving his own intuitive verdict on the question.31 They
did not say this with pleasure; they would have been delighted if there were some more
reliable route to moral knowledge, or one less prone to unresolvable disagreements.
But they thought there was no such route; all proposals for one failed. At the end,
moral judgement had to involve an immediate apprehension of sui generis truth.
That said, there were differences among them about which intuitive judgements are
most trustworthy. Sidgwick placed most credence in judgements about highly abstract
moral principles, such as the axioms of prudence and benevolence discussed in the
‘Philosophical Intuitionism’ chapter of his Methods of Ethics. This was also Moore’s
official view, as stated in the preface to Principia Ethica and reflected in his bald claim, in
Ethics, that it is self-evident that right acts always maximize the good. But it is hard to
confine oneself to that approach, and another part of Moore’s official doctrine did not.
This was his claim that, to decide whether some generic state of affairs X is good, we
must imagine a possible world containing only X and judge whether that world is
good. Here a judgement about the value of X in the abstract is reached via a judgement
about a particular possible world.32
The opposite view was taken by Carritt and Ewing, who thought the most reliable
intuitive judgements concern particular cases, with general principles mere abstractions
from them.33 But Prichard, Ross, and Broad took a middle line. Unlike Sidgwick and
Moore, they did not think we can know abstract principles by reflecting on them just
as abstract principles; intuitions are elicited only in particular situations. But what we
intuit in a particular situation is that an act’s having some non-moral property tends
to make it right or wrong, for example, that its being the keeping of a promise tends to
make it right.34 Though prompted by a particular situation, the intuition is implicitly
general, since it implies that any act of keeping a promise is, other things being equal,
right. And this implication, though not equivalent to the principle that we ought, other
things being equal, to keep our promises, is sufficiently close to it that our grasp of the
principle follows by a small step.

31
See e.g. Sidgwick, Methods of Ethics, 400–1; Rashdall, Theory of Good and Evil, i. 59, 69, 70–1, 75, 78;
Moore, Principia Ethica, pp. viii, 59, 75–7, 92, 143–4, 197, and Ethics, 102; Prichard, Moral Writings, 2; Carritt,
Theory of Morals, 28, 72; Ross, Right and the Good, 39–40; Broad, Five Types, 131, 233; Ewing, The Morality of
Punishment (London: Kegan Paul, Trench, Trubner, 1929), 6–7, 14, 17–18, 185–7, and Definition of Good,
15–18.
32
Sidgwick, Methods of Ethics, 379–86; Moore, Principia Ethica, p. viii, and Ethics, 76–7; Moore, Principia
Ethica, 83–4, 91, 93–5, 187–8.
33
Carritt, Theory of Morals, 30–1, 70–1, 84–5, 114–15, 138–9; Ewing, Morality of Punishment, 2, 119 n.,
160–1, 174–5, 179, 187–8, 202.
34
Prichard, Moral Writings, 4–5, 13, 77; Ross, Right and the Good, 32–3 and Foundations of Ethics, 84, 168–71,
184; Broad, Five Types, 145–6, 177–8, 271–2.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 15

A common charge against the school is that their reliance on intuitive judgements
made them dogmatic, announcing personal prejudices as universal moral truths.35 A
more subtle charge is that while Sidgwick recognized the fallibility of moral intuition,
later writers such as Moore, Prichard, and Ross tended to dogmatism.36 The latter
charge may be fair against Moore in some of his moods, as when he simply announced
that consequentialism is self-evidently true, but applied more broadly it is not.
Sidgwick laid down four conditions for genuine intuitions of self-evidence: aside
from two requiring that the propositions intuited be clear and mutually consistent, the
principal ones required us to test by introspection that what we have really is an
apprehension of a proposition as apparently self-evident rather than, say, a reflex
echoing of common opinion, and to check that other people share our intuition, so
there is consensus about it.37
But these conditions were absolutely shared by the later writers. Moore too
emphasized the fallibility of moral intuition, saying ‘in every way in which it is possible
to cognise a true proposition, it is also possible to cognise a false one’, and noting that
others’ dissent from a proposition we believe should lessen our confidence in it.38
Rashdall, Ewing, and others insisted just as much as Sidgwick on distinguishing
genuine apprehensions of apparent self-evidence from superficially similar psychological
states, while Prichard said, ‘I don’t think the apprehension of the self-evident easy to
reach’.39 Moreover, the reliance of many in the school on everyday moral judgements—
on what Ross called the ‘existing body of moral convictions of the best people’, which is
‘the cumulative product of the moral reflection of many generations’40—reflects their
acceptance of something like Sidgwick’s consensus test. For them, moral judgement was
not so much a matter of what ‘I’ think as of what ‘we’ think.
In fact, this reliance on common-sense moral opinions generates another objection
to the school that is harder to answer but that is in some tension with the charge of
dogmatism: that the school were morally conservative, offering philosophical defences
of the everyday moral views of their time but rarely proposing significant reforms to
them. This charge is not entirely fair. A theorist who works within common-sense
morality can note inconsistencies in its current application and propose reforms to
remove those, as Sidgwick did when he urged extending our concern with human
pleasure to include that of animals; Ross allowed similar innovations.41 But it remains

35
See e.g. Alasdair MacIntyre, After Virtue (London: Duckworth, 1981), 14–19.
36
See e.g. Bart Schultz, ‘Introduction: Henry Sidgwick Today’, in Essays on Henry Sidgwick, ed. Bart
Schultz (Cambridge: Cambridge University Press, 1992), 28–9, 59 n. 84; and Robert Shaver, ‘Sidgwick’s
Minimal Metaethics’, Utilitas, 12 (2000), 263–6.
37
Sidgwick, Methods of Ethics, 338–42.
38
Moore, Principia Ethica, pp. x, 75–6, 143–4, and Elements of Ethics, 163, 167–8.
39
Moore, Elements of Ethics, 162–3; Rashdall, Theory of Good and Evil, i. 211–13; Ewing, Morality of
Punishment, 85–7, 191; Prichard, ‘Letter to John Laird of July 30, 1938’, Bodleian Library, University of
Oxford, Ms. Eng. Lett. C. 131 fols. 18–29, fol. 27.
40
Ross, Right and the Good, 41.
41
Sidgwick, Methods of Ethics, 414; Ross, Right and the Good, 39, and Foundations of Ethics, 189–90.
16 THOMAS HURKA

true that neither Sidgwick, Rashdall, Moore, nor Ross was a moral revolutionary, and
their work might well have been more profound if they had considered more radical
revisions to everyday moral beliefs. That they did not stemmed largely from their
fallibilism, however, or their anti-dogmatic view that one’s own intuitions need to be
tested against other people’s, since one’s own may so easily be wrong.

Moderate Moral Theory

Though the school often started their theorizing with common-sense morality, they
did not stop there, or think common-sense views are already in the best condition
possible. On the contrary, they thought the task of moral philosophy is to theorize
common sense, or reduce its many and conflicting judgements to a consistent and
systematic whole, with a small number of fundamental principles explaining the rest.
There were differences about how far this systematization should go. Early in the
school’s history Sidgwick made very strong demands on theory. His requirements of
clarity and consistency, at least as he interpreted them, amounted in effect to a demand
for completeness, so an acceptable moral theory must yield a determinate verdict about
what is right and wrong in every conceivable circumstance. Since he thought com-
pleteness impossible if there are competing principles or values, he concluded that the
only acceptable theory is monistic both about the right, accepting just the one principle
to maximize the good, and about value, with just one intrinsic good, pleasure.
Sidgwick’s immediate successors, the ideal consequentialists Rashdall, McTaggart,
and Moore, retained his monism about the right but rejected his monism about value,
accepting a number of other goods in addition to pleasure, including perfectionist ones
such as knowledge and virtue. It is not the proper business of philosophy, Moore
wrote, to ‘search for “unity” and “system”, at the expense of truth’,42 and the truth is
that there is more than one good. Even later, Prichard, Carritt, Ross, and Broad applied
the same argument to the consequentialist principle: it is false to moral experience to
restrict the basic duties to a single one. As Ross said, ‘it is more important that our
theory fit the facts than that it be simple’,43 and the facts now included a plurality of
grounds of duty, so an acceptable theory must be pluralistic not only about the good
but also about the right.
Even these later writers, though, were not radical pluralists. They did not recognize
327 fundamental goods or eighty-four underivative duties. Ross, for example, thought
there were only four intrinsic goods and, though more tentatively, only six basic
duties—even fewer if some are subsumed under a more general duty to promote the
good. Though a pluralist, he was a moderate one, committed like Sidgwick to making
our moral judgements more systematic by connecting them to a few more general
ones. Thus aesthetic appreciation and personal love were for him not distinct intrinsic

42
Moore, Principia Ethica, 222.
43
Ross, Right and the Good, 19 (see also p. 23); Broad, Five Types, 283–4.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 17

goods but were reducible to the more fundamental goods of pleasure, knowledge, and
virtue, while the duty not to lie was just an instance of the duty to keep promises.44
And his central concept of prima facie duty served to show, as common sense itself
cannot, how recognizing plural duties need not lead, as Sidgwick thought, to logical
inconsistency. Though Ross did not follow the systematizing path as far as Sidgwick
did, he was still very much on it.
What the later writers represent, therefore, is a moderate pluralism, one in between a
strong demand for systematicity and precision like Sidgwick’s, which allows only fully
monistic theories, and the anti-theory view of writers like F. H. Bradley and his fellow
Idealists in the nineteenth century and some neo-Wittgensteinians today, according to
which our moral judgements cannot be systematized at all, because they involve just
particularized perceptions of right and wrong in particular cases.45 And it avoids these
extremes because it rejects an assumption common to them both: that morality can be
theorized successfully only if it can be theorized completely, so its principles always
yield determinate verdicts. (Sidgwick thought this goal can be achieved, the anti-
theorists that it cannot.) Rejecting this assumption as unwarranted, Moore, Ross, and
the others proceeded to partly theorize a moral reality they thought was partly (though
not wholly) theorizable, or only partly capable of being made determinate. Ross, for
example, thought we can have certain knowledge of the principles of prima facie duty,
such as that we ought other things equal to keep promises, but can never be certain
what our duty all things considered in a particular situation is, because we can never be
certain how those principles weigh against each other.46
There is a question about where these philosophers thought the unavoidable
indeterminacies lie. Are they in the moral truth itself or only in our ability to know
it, that is, are the indeterminacies metaphysical or merely epistemic? Ross seems to have
thought that there is always a determinate truth about how good something is and
about how its goodness compares with that of other things.47 If he extended this view
to right—and it is hard to see how he could not—he would hold that there is always a
determinate truth about what our duty proper in a given situation is. It is just that
neither we nor our best moral theory can say definitively what it is.
This moderation about moral theory is illustrated in the group’s approach to the
measurement of values. The Idealists were sceptical about such measurement, citing
(surprisingly for present-day readers) pleasure as the paradigm of a good that cannot be
quantified. The writers after Sidgwick agreed that we cannot assign precise cardinal
measures to pleasure and other goods, but insisted that we can nonetheless assign rough
cardinal ones. McTaggart thought we can never say that two pleasures are exactly

44
Ross, Right and the Good, 21, 23, 24–6, 140–1.
45
See e.g. F. H. Bradley, Ethical Studies, 2nd edn. (Oxford: Clarendon Press, 1927), 157, 193–9, and
Principles of Logic, 2nd edn. (London: Oxford University Press, 1922), 269–70; also John McDowell, ‘Virtue
and Reason’, The Monist, 62 (1979).
46
Ross, Right and the Good, 23, 29–32, 41–2.
47
Ibid. 143.
18 THOMAS HURKA

equally intense, but was certain that he got more than twice as much pleasure from a
plate of turtle soup as from a plate of pea soup. Ross too thought we can say that one
pleasure is at least twice as intense as another, while Rashdall and Moore made similarly
rough cardinal claims, for example, that aesthetic contemplation and personal love are
‘by far’ the greatest goods.48 The fact that we cannot measure precisely is no bar to our
measuring as far as we can.

Moral Explanation
A scientific or ethical theory can serve two functions. One is to discover new truths,
which in ethics means using the theory’s principles to arrive at moral judgements about
particular cases that we could not otherwise make. The other is to explain truths we do
know; here the principles tell us why particular moral judgements we already confi-
dently make are true.
Sidgwick seems to have been most interested in the first function, about correct
particular judgements. This is reflected in his emphasis on methods of ethics rather than
principles, where a single method of determining right and wrong can be combined
with different explanatory principles.49 It also shows in his attitude to his ‘dualism of
the practical reason’, the conflict he saw between the egoist claim that we should
pursue just our own pleasure and the utilitarian claim that we should pursue the
pleasure of all. Sidgwick thought this dualism would be resolved if there were a God
who rewarded right conduct in an afterlife,50 but Moore and Broad disagreed. Egoism,
they claimed, says the one right-making characteristic is maximizing the agent’s
pleasure, utilitarianism that it is maximizing the general pleasure, and no merely
extensional equivalence between the two can resolve this conflict about explanation.51
But then Sidgwick’s belief that the conflict can be resolved suggests that he did not see it
as about explanation. He read the two claims extensionally, as saying that an act is right
if and only if it maximizes our own or the general pleasure, and a conflict between
extensional principles is resolved if their verdicts coincide.
Other philosophers of the period emphasized the second, explanatory function. The
Idealists often accepted the consequentialist structure of hedonistic utilitarianism and
allowed that utilitarianism yields broadly correct verdicts about particular cases. But
they thought the vital philosophical question is what explains those verdicts, and the
explanation, they insisted, is perfectionist. As Bradley put it: ‘What we hold to against
every possible modification of Hedonism is that the standard and test is in higher and

48
J. M. E. McTaggart, Studies in Hegelian Cosmology (Cambridge: Cambridge University Press, 1901), 117;
Ross, Right and the Good, 143, and Foundations of Ethics, 183; Rashdall, Theory of Good and Evil, ii. 49–50;
Moore, Principia Ethica, 188.
49
Sidgwick, Methods of Ethics, 8–9, 83.
50
Ibid. 503–9.
51
Moore, Principia Ethica, 102–4; Broad, Five Types, 255–6.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 19

lower function, not in more or less pleasure.’52 Many members of our school had a
similar view. Moore thought that most of the time we should follow a set of rules
obeying which will preserve society and so promote overall good, and that these rules
will be the same given any plausible theory about what is good.53 But he nonetheless
thought the central philosophical question is ‘What is intrinsically good?’, because the
answer to that question explains why the rules are correct.54 And his critique of
hedonism argued that even if that view can capture our convictions about which pleasures
are best, by considering their effects on future pleasures, it does not capture our convic-
tions about why they are best, which concerns what they are pleasure in now.55
A similar emphasis appears in Prichard, Carritt, and Ross. They sometimes criticized
consequentialism for yielding the wrong results about particular cases, as when Ross
said it tells us to break a promise if that will promote 1,001 units of good rather than the
1,000 that would follow from keeping it.56 But their more common objection was that
consequentialism gives the wrong explanations for its results even when those are right.
Thus they asked whether the reason we ought to keep a promise is that this will help
maintain an institution that will benefit people in the future or simply that we made a
promise. Their insistence that the second is the right explanation was their main
objection to consequentialism, and Prichard in particular gave it a striking form. He
said that by deriving the duty to keep promises from a more general one to promote
the good, consequentialism turns the duty to keep promises into a quite different duty
to promote the good—if that is what explains the first duty, that is what it at bottom is.
But this distorts the moral phenomena, turning what we recognize as one duty into
something it is not; in trying to explain the duty to keep promises, consequentialism
destroys it.57
This emphasis on the explanatory function of moral theory fits several other aspects
of Prichard’s and Ross’s view. One was their faith in common-sense morality. If most
people already usually know what is right, they do not need moral theory to tell them
what is right; philosophy’s task is instead to explain why it is right. Another was their
moral epistemology. If the primary moral intuition is that an act’s having a certain
property, such as that of keeping a promise, tends to make it right, then the primary
intuition is itself explanatory, and in testing moral theories against beliefs about
explanation we are testing them against the most secure moral knowledge we have.

52
Bradley, ‘Mr. Sidgwick’s Hedonism’, in Collected Essays, vol. 1 (Oxford: Clarendon Press, 1935), 97.
On the frequent (though not exceptionless) extensional equivalence of Idealist ethics and utilitarianism see
Bradley, Ethical Studies, 138–41, and T. H. Green, Prolegomena to Ethics, ed. A. C. Bradley (Oxford:
Clarendon Press, 1883), secs. 332, 356.
53
Moore, Principia Ethica, 155–64.
54
Ibid. 27, 77, 90–1, 138, 140, 158, 184, 189, 222.
55
Ibid. 94–5.
56
Ross, Right and the Good, 34. See also Prichard, Moral Writings, 2, and Carritt, Theory of Morals, 39–40,
108–9.
57
Prichard, Moral Writings, 10, 29–30; Carritt, Theory of Morals, 69–70; Ross, Right and the Good, 17, 19,
24, 36–9, and Foundations of Ethics, 65–9, 113, 187.
20 THOMAS HURKA

Inherent Explanations
Though the school pursued moral explanations, they also had a distinctive view about
how those should proceed, proposing mostly what I will call conceptually ‘inherent’
rather than ‘external’ explanations.58
Inherent explanations explain common-sense moral judgements by connecting them
to principles that are more abstract but use similar concepts, so they are continuous with
the common-sense judgements and concern the same general subject. Because they are
more abstract, the principles have independent appeal and can therefore both explain
the judgements and increase our warrant for believing them. But they operate within
the same circle of concepts rather than concerning some other, allegedly more funda-
mental topic.59
Thus Sidgwick grounded utilitarianism in the principles that one should not prefer a
lesser good at one time to a greater good at another, or a lesser good for one person to a
greater good for another.60 These are principles of impartiality and unify many
particular moral claims, but they are only more abstract versions of ideas about caring
for the future and for others that are already present in everyday moral thought. While
stating those ideas more explicitly and precisely, they do not replace them with
something different.
The same holds for the account of virtue given first by Rashdall and then by Moore,
Ross, and others.61 It treats virtue as a higher-level intrinsic good involving morally
appropriate attitudes to items with other, previously given moral properties. Thus one
form of virtue is having a positive attitude, such as desiring, pursuing, or taking pleasure
in, something else that is good, such as another person’s pleasure. Another is having the
negative attitude of trying to prevent or being pained by something evil, such as
another’s pain. Just as the positive attitude fits a positive value and therefore is virtuous,
so a negative attitude fits a negative one. This account of virtue is more abstract than
anything in common-sense moral thought, but it also resonates with that thought,
which can recognize in its less formal conceptions of benevolence and compassion the
ideas of positive-to-positive and negative-to-negative concern or, as the children’s
book Madeline puts it, of ‘smil[ing] at the good and frown[ing] at the bad’.
Or consider Moore’s formulation of retributivism using his principle of organic
unities. It says that while vice is intrinsically evil, as is pain, the combination of vice and
pain in the same person’s life is good as a combination, and sufficiently good that

58
In earlier writings I have contrasted these as ‘structural’ vs. ‘foundational’ explanations, but the second
of these terms in particular may be misleading. See my ‘Moore in the Middle’, 627–8, and ‘Normative Ethics:
Back to the Future’, in The Future for Philosophy, ed. Brian Leiter (Oxford: Clarendon Press, 2004), 253–61.
59
It is not part of inherent explanation that the explanatory principles can be known only by reflecting on
common sense. They can be self-evident in themselves, as Sidgwick and Moore held.
60
Sidgwick, Methods of Ethics, 381–2.
61
Rashdall, Theory of Good and Evil, i. 59, 63–5, 76, 137, 174–5, 214; ii. 41–2; Moore, Principia Ethica,
203–4, 214–22; Ross, The Right and the Good, 134–5, 163.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 21

adding the pain to the vice makes the overall situation better.62 This analysis illuminates
the structure of retributivist claims and has important implications, for example, that
while deserved pain is good as deserved it is also evil as pain, so the morally appropriate
response to it mixes satisfaction that justice is being done with pain at the infliction of
pain. But it does not ground everyday retributivism in some other, less contentious
claim; it merely reformulates it more abstractly.
Likewise for Broad’s treatment of what he called ‘self-referential altruism’.63 It holds,
against utilitarianism, that our duty concerning others is not to treat them impartially
but to care more for those who are in various ways closer to us, such as our family and
friends. Broad’s analysis unifies a variety of common-sense claims about loyalty or
partiality and invites further inquiry about exactly which relations make for closeness of
the relevant kind. But it does not justify partiality in other terms. On the contrary, it
assumes partiality, saying each person should care more about his family and friends
because he should care more about those who stand in special relations to him.
The contrasting external approach arises from dissatisfaction with merely inherent
explanations, whose principles, it complains, are too close to the judgements they are
meant to yield to explain rather than just restate them. A genuine moral explanation,
externalists say, must connect an everyday moral claim to one that uses different
concepts and concerns some other, more fundamental, topic.
The last half-century has been dominated by external (in this sense) projects in
ethics. Some have tried to ground moral judgements, either as a whole or individually,
in something outside morality itself, such as the language of morals (Hare), self-interest
non-morally construed (Gauthier), the metaphysics of the person (Rawls, Scheffler,
Parfit, Brink), or the demands of practical consistency (Gewirth, Korsgaard). Others
have tried to ground particular moral claims in ones that, while still moral, concern
some different topic. Examples include the justification of retributive punishment
in terms of distributive fairness (Morris) or of our duty to benefit others in terms of
our own good or flourishing morally conceived, as in neo-Aristotelian virtue ethics
(Anscombe, Foot, Hursthouse). But in every case the explanatory principles use
different concepts from the judgements being explained and concern some other,
allegedly more fundamental, topic.
The grand exemplar of the external approach is Rawls. Denying that our everyday
judgements about equality and rights are properly justified if taken on their own, he set
out to show that they would be chosen by rational contractors in a specified ‘original
position’, using what he argued are the correct principles of rational choice. Since the
specification of Rawls’s original position depends on moral judgements, his justification
is not extra-moral, like ones citing the language of morals or non-moral self-interest.
But it is external in my sense, since ideas about rational contracting are far removed
from everyday talk of equality and rights—they are nothing like what people have

62
Moore, Principia Ethica, 215–16.
63
Broad, ‘Self and Others’, in Broad’s Critical Essays, 279–82.
22 THOMAS HURKA

in mind when engaged in that talk—and in Rawls’s view can explain it precisely
because they are far away.
The distinction between inherent and external explanation is one of degree rather
than kind, and there can be in-between cases. Moreover the Sidgwick-to-Ewing
school did sometimes give non-inherent explanations, as when Sidgwick rested the
rationality of egoism on the claim that individuals are metaphysically distinct or Ross
subsumed the duty not to lie under the duty to keep promises.64 But their general
tendency was to prefer inherent explanations, and this tendency had several grounds.
One was their faith in common-sense moral judgements. Externalism is often
motivated by distrust of everyday intuition: if that is unreliable, then a persuasive
justification for a moral claim can only be found elsewhere. But if one thinks common
sense contains mostly true judgements, it is natural to think it also contains, even if
implicitly, the materials for sound explanations of them. Those materials may need to
be extracted and made more precise, but they are there.
Another was their belief that the more ambitious external approach does not succeed,
for two familiar reasons. Its moral explanations do not in fact yield the results they are
intended to, or do not unless they tacitly assume what they are meant to prove. And
even if they did yield those results, they would give the wrong explanation for them.
Prichard took both these lines about instrumental justifications of morality that
say we should promote others’ happiness and keep promises because doing so will
maximize our own happiness. He argued, first, that it is not always true that fulfilling
these duties will maximize our happiness,65 and second, that even when it is true, this
is not the reason why we ought to fulfil them. Instrumental views turn the duties to
benefit others and keep promises into duties to promote our happiness, which is not
what they intuitively are. Given his particular conceptual views, Prichard gave this last
argument a radical form. Holding that the instrumental or hypothetical imperative
is only descriptive, he said instrumental justifications of morality resolve the moral
‘ought’ into something that is not an ‘ought’ at all and so deny normativity altogether.66
But even if the instrumental ‘ought’ is a real ‘ought’, he could say that instrumentalism
turns other-regarding duties into self-regarding ones and so distorts the moral phenomena.
Many in the school had a similar view about Aristotle’s more high-minded egoism,
which says we ought to promote others’ happiness or keep promises because, if done
with the right motives, this will manifest virtue on our part, which is an essential part of
the eudaimonia or flourishing that is our one ultimate goal. Sidgwick attacked the
conceptual underpinnings of this view, saying it fails to distinguish the questions ‘What
ought I all things considered to do?’ and ‘What will make my life go best?’, or assumes
without argument that the answers to these two questions must always be the same.67

64
Sidgwick, Methods of Ethics, 404, 498; Ross, Right and the Good, 21.
65
Prichard, Moral Writings, 26, 32, 180.
66
Ibid. 9, 29–30, 43, 116, 122–3, 143, 144–5, 150, 169, 183, 188–93, 236–40, 241.
67
Sidgwick, Methods of Ethics, 404–5.
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 23

Ewing thought the egoistic framework forced Aristotle into implausible claims, for
example, that if it is all things considered right for a person to sacrifice his life, the
excellence this courageous act will add to his life must be greater than the excellence of
the thousands of other virtuous acts he could perform if he continued to live, so his act
involves no real sacrifice.68 But even apart from these difficulties, Aristotle’s view gave
the wrong explanation of our other-regarding duties, making them at bottom self-
regarding. Prichard tied this charge to an implausible hedonistic reading of Aristotle, on
which by ‘good’ Aristotle always meant ‘conducive to the agent’s pleasure’.69 But his
criticism can be separated from this interpretation, since whatever it is, my eudaimonia
must be a state of me, and any view that grounds all ‘oughts’ in a state of me makes
all ‘oughts’ ultimately self-regarding. And the school found more specific egoistic
elements in Aristotle’s ethics. Sidgwick said Aristotle lacked the modern concept of
benevolence, since the virtue closest to it on his list, liberality, is shown just as much in
tasteful expenditure on a fine house for oneself as in spending on other people.70
Others were repelled by Aristotle’s description of the megalopsychos or ‘proud’ man,
who takes pleasure in being more virtuous than other people and finds it beneath his
dignity to give others small benefits; only when great things are at stake will he deign to
act. Rashdall commented on ‘Aristotle’s revolting picture of the high-souled man
(megalopsychos)’,71 while Ross said the description of the megalopsychos ‘betrays some-
what nakedly the self-absorption which is the bad side of Aristotle’s ethics’.72
The school had a similar view of Kant’s attempt to ground the moral duties in the
first formulation of his categorical imperative. This attempt is again largely external,
since the question whether the universalization of a given maxim can be coherently
thought or willed is remote from everyday thought about promoting others’ happiness
or keeping promises. But it again is open to two objections. One is that it yields the
wrong results: it is not true, for example, that a world in which everyone makes lying
promises cannot be conceived.73 And even if it did yield the right results, it would give
the wrong explanation for them. Prichard wrote: ‘No one could suppose that the reason
why an act ought to be done consists in the fact that everyone could do it. Even Kant
could not have supposed this. The difficulty escaped him because it didn’t occur to him
that his criterion of moral rules must express what, on his view, is their reason.’74

68
Ewing, Ethics (London: English Universities Press, 1953), 28–9.
69
Prichard, ‘The Meaning of agathon in the Ethics of Aristotle’, in Moral Writings.
70
Sidgwick, Outlines of the History of Ethics, 5th edn. (London: Macmillan, 1902), 62, 122.
71
Rashdall, Theory of Good and Evil, i. 205. See also Carritt’s remark about ‘the egoistic self-righteousness of
Aristotle’s philautos’, in ‘An Ambiguity of the Word “Good”’, Proceedings of the British Academy, 23 (1937), 69.
72
Ross, Aristotle (London: Methuen, 1923), 208.
73
Prichard, Moral Writings, 60; Carritt, Theory of Morals, 79–82; Broad, Five Types of Ethical Theory, 130–1;
Ross, Kant’s Ethical Theory (Oxford: Clarendon Press, 1954), 29–33, 45–7.
74
Prichard, Moral Writings, 59. Thomas E. Hill, Jr. has recently made the same criticism, saying that if we
ask why slavery is wrong, the claim that a world in which everyone acts on the maxim of a slave-owner and
none on that of a slave is logically impossible is not the right answer; see his ‘Kantian Normative Ethics’, in
The Oxford Handbook of Ethical Theory, ed. David Copp (New York: Oxford University Press, 2006), 488.
24 THOMAS HURKA

External moral explanations cannot be dismissed en bloc; each must be examined on

its own merits. But the experience of the last half-century is hardly encouraging, as
time and again external projects have been found to fail. This leaves the conceptually
inherent approach of the Sidgwick-to-Ewing school, which, while not as exciting as
externalism would be if successful, can nonetheless be deeply illuminating. Sidgwick
on impartiality, Moore on organic unities, Rashdall, Ross, and the others on virtue—
these and analyses like them unify particular moral judgements under an independently
plausible head while illuminating their internal structure. Everyday morality is subtle
and complex, with more internal coherence than may initially appear. By working
within that morality, inherent explanations can uncover that subtlety and coherence
as externalists, in their eagerness for more grandiose justifications, too rarely do. While
less ambitious than the alternative, inherent explanation is more likely to yield
substantial results and in my view has a better record of doing so.
It is not that the Sidgwick-to-Ewing school gave every inherent explanation
possible; there were aspects of common-sense morality they seem to have been blind
to. Thus, most of them did not see how common sense permits us to care somewhat
more about our own happiness than about other people’s, and distinguished on that
basis between acts of benefiting others that are strictly required and ones that are
supererogatory. (Ross thought that when doing so will not violate any deontological
constraint, our duty is always to maximize the good impartially.75) So they did not
theorize that permission or the many fascinating issues it gives rise to. Nor did they
always take inherent explanation as far as they could. Prichard and Ross thought the
duty to keep promises cannot be analysed further, but it can be if we see the act of
promising as transferring a moral right, understood as a package of permissions for
oneself and duties for others, to another person, by the exercise of a higher-level moral
power.76
So the school’s importance does not rest on their having given every possible
inherent moral explanation. It is rather that, by largely restricting themselves to that
style of moral theorizing, they demonstrated its considerable merits.

Underivative Duty
If we ask for a single, central belief of the Sidgwick-to-Ewing school, it is that some
moral duties, the fundamental ones, are underivative. If we ask why these duties hold,
there is no other answer than that they do. They cannot have a moral justification,
because they are morally basic, nor can they have an extra-moral one, because none of
those succeed. But the claim that duty is underivative can be made at three different
levels.

75
Ross, Right and the Good, 39.
76
See H. L. A. Hart, ‘Are There Any Natural Rights?’, Philosophical Review, 64 (1955).
 C O M M O N T H E M E S F RO M S I D G W I C K T O E W I N G 25

The first concerns the normative realm as a whole. Here the idea is that normative
judgements are sui generis, neither reducible to nor derivable from non-normative
judgements such as those of science; the ‘ought’ is in general distinct from the ‘is’. The
school usually combined this claim with normative realism, yielding their metaethical
non-naturalism. But I have argued that what mattered most to their theorizing was the
sui generis claim on its own.
The second level concerns moral judgements more narrowly. Here the school held
that judgements about how one ought morally to act are likewise underivative, not
only from non-normative judgements but also from any other normative judgements;
there are no non-moral ‘oughts’ from which moral ‘oughts’ derive. They based this
claim partly on the minimalist view that there are no other ‘oughts’ than the moral one,
but also on the substantive view that no ought-statements other than those commonly
called ‘moral’ are true.
Finally, later writers of the school, such as Prichard, Ross, and Broad, applied the
idea of underivativeness to deontological duties such as to keep promises or make
reparation. These duties, they held, do not derive from a more general one to promote
good consequences; the main reason we ought to keep promises or compensate those
we have harmed is just that we ought to.
This last application was not shared by consequentialists such as Sidgwick, Rashdall,
and Moore, but they did share the first two. If asked why we ought to promote other
people’s happiness or knowledge, they would have said there is no answer other than
that we ought to. (The claim that happiness or knowledge is good either needs to be
supplemented by the further claim that we ought to promote whatever is good—
which claim is then underivative—or already includes an ‘ought’ in its content.) No
less than the deontological duties of Prichard, Ross, and Broad, the supreme conse-
quentialist duty holds just because it does.
Though in one sense modest, this view contrasts with vast stretches of western moral
philosophy, which have aimed precisely at explaining why we ought to fulfil basic
moral duties. The Sidgwick-to-Ewing school rejected these attempts, holding that
they fail to yield their intended conclusions and, even if they did, would give the
wrong explanations for them. The only approach that avoids these pitfalls—that
generates the right verdicts for the right reasons—treats some moral duties as under-
ivative. This may not be an entirely distinctive view in the history of ethics, but no
other group expressed it so clearly or used it as the springboard for such searching
analyses of the moral judgements we actually make.
 2
Pleasure and Hedonism in Sidgwick
Roger Crisp*

Hedonism
The classical utilitarian tradition reached its apogee in the work of Henry Sidgwick, in
particular in his Methods of Ethics.1 In various ways, including his explicit philosophical
intuitionism and the weight he attached to the egoistic point of view in his ‘dualism of
practical reason’, Sidgwick broke with that tradition. But his commitment to hedonism
remained unshakeable throughout his life.
What kind of hedonism did Sidgwick avow? Unlike certain of his predecessors,
Sidgwick clearly distinguished between the descriptive and the prescriptive or normative:
‘Utilitarianism . . . is an ethical, and not a psychological doctrine: a theory not of what
is, but of what ought to be. Therefore, more particularly, it does not include . . . the
proposition that in human action, universally or normally, each agent seeks his own
individual happiness or pleasure’ (U 3). Sidgwick denied psychological hedonism. He
believed that human beings have ultimate desires for objects other than pleasure and
the relief of pain (ME 1.4.2.3; see LE 102, 117), and may be led to act in ways that they
know will be, overall, worse for them in hedonistic terms: ‘“Video meliora proboque,

* I am grateful for comments and discussion to the participants of the conference on ‘British Moral
Philosophy’ convened by Tom Hurka at the University of Toronto in April 2008. My debt to Hurka, as I
hope the chapter makes clear, is especially great. An earlier version of the paper appeared as ‘Sidgwick’s
Hedonism’ (and in Italian as ‘L’edonism di Sidgwick’), in P. Bucolo, R. Crisp, and B. Schultz (eds.), Henry
Sidgwick: Happiness and Religion (Catania: Dipartimento di Scienze Umane dell’Università degli Studi di
Catania, 2007), 104–57. That paper itself had benefited from comments and discussion at a conference on
Sidgwick convened by Placido Bucolo in Catania in November 2006.
1
The Methods of Ethics, 7th edn. (London: Macmillan, 1907). Hereafter referred to as ME; references are to
book, chapter, section, and paragraph, respectively. Other works are referred to by the following abbrevia-
tions: FPM: ‘Fowler’s Progressive Morality’ (1885); GE: ‘Green’s Ethics’ (1884); GU I and II: ‘Grote on
Utilitarianism I and II’ (1871); HUG: ‘Hedonism and Ultimate Good’ (1877); LE: Lectures on the Ethics of
T. H. Green, H. Spencer, and J. Martineau (1902); MSES: ‘Mr Spencer’s Ethical System’ (1880); PD: ‘Pleasure
and Desire’ (1872); PE: The Principles of Political Economy (1883); U: ‘Utilitarianism’ (1873). All essays are
reprinted in Sidgwick, Essays on Ethics and Method, ed. M. G. Singer (Oxford: Clarendon Press, 2000).
References are to paragraph unless otherwise stated.
 PLEASURE AND HEDONISM IN SIDGWICK 27

deteriora sequor” is as applicable to the Epicurean as to anyone else’ (PD 7; see PE, p. 63
n. 1). His hedonism, then, is evaluative (a theory of the good).2
Again, Sidgwick was unusually clear about the distinction between the good for an
individual, on the one hand, and overall or general good, on the other, and saw in this
distinction an important difference between ancient and modern ethics. Modern
thinkers, he believed, tended to think in terms of general good and universal ends,
while for a Greek ‘the primary question as naturally and inevitably took an egoistic
form.3 The Good which he studied was “good for himself ” ’ (HUG 2).4 I am not here
adverting to the distinction between:

Egoistic hedonism: Each agent has strongest reason to promote the greatest balance of
pleasure over pain in her own life;
and

Impartial hedonism: Each agent has strongest reason to promote the greatest balance
of pleasure over pain overall, her own pleasures and pains counting equally with
those of others.

This is a distinction not between theories of the good but between ‘methods of ethics’
or theories concerning how we should act, and is a topic for another paper. The
distinction I have in mind is as follows:

Welfare hedonism: What is good for any individual is the greatest balance of pleasure
over pain;

and

Global hedonism: The only good or value is the greatest balance of pleasure over pain.

Welfare hedonism, as Sidgwick understood it, is a theory about ‘happiness’ (U 5;

MSES, p. 221 n. 2). It is quite consistent with welfare hedonism to admit non-hedonistic
values, as long as they are not values which constitute welfare or well-being. I might

2
Sidgwick links the normative (i.e. reasons for action) with the evaluative at e.g. LE 331.
3
Sidgwick criticizes T. H. Green in particular for seeking to collapse the distinction between one’s own
good and that of others (GE 252–3).
4
Tom Hurka has suggested to me that Sidgwick did not use the concept of ‘good for’ as equivalent to
‘well-being’. A search of the Past Masters database appears to provide many counter-examples to Hurka’s
suggestion. On Green, for example, consider the following from the frontmatter to LE: ‘Green is not able to
maintain an idea of a True Good which is non-competitive, which does not admit of the distinction between
“good for” self and “good for” others.’ Sidgwick does recognize that the notion of the ‘Ultimate Good for
man’ can be used in a ‘non-egoistic’ way, to refer to what it is ultimately reasonable for a person to aim at (ME
3.14.1.1). But he is contrasting that with the ‘egoistic’ use, in which it is equivalent to ‘well-being’. So, for
example, he appears to be using the egoistic sense a few pages later in ME, when he notes: ‘We may make the
distinction [between Virtue as a means to human well-being and an end] clearer by considering whether
Virtuous life would remain on the whole good for the virtuous agent, if we suppose it combined with
extreme pain’ (3.14.2.3).
28 RO G E R C R I S P

deny global hedonism, claiming, for example, that beauty, in itself, adds to the value of
the universe, while insisting that the only thing that adds to the value of a life for the
person living it is pleasure. Sidgwick himself accepted both welfare and global forms of
hedonism (see e.g. ME 3.14.3.2). In this chapter I shall concentrate on welfare hedonism,
though it is worth noting that the arguments for and against welfare hedonism often
carry across directly to global hedonism.
Sidgwick again stood out from some of his utilitarian predecessors in seeing clearly
that the word ‘good’ does not mean ‘pleasant’. First, although attributions of goodness
are frequently related to judgements of the pleasantness of the object in question, the
goodness often corresponds to a particular kind of pleasantness. So even if a whole-
some wine, for example, were productive of a higher-than-average balance of
pleasure over pain in the longer term, we would not call it good on that account
(ME 1.9.2.1). Further, although we think an individual is the final arbiter of the
pleasantness of his or her own experience, in cases of goodness we often—as, for
instance, in aesthetics—allow for the possibility of good taste or judgement (ME
1.9.2.2). Finally, Sidgwick anticipates G. E. Moore’s ‘open question’ argument,
noting that equating the two terms would be to reduce hedonism to a tautology
(ME 1.9.2.4; LE 145).5
So how are we to understand Sidgwick’s conception of a person’s welfare or ‘good
on the whole’? After some discussion, he defines the notion as follows:
[A] man’s future good on the whole is what he would now desire and seek on the whole if all the
consequences of all the different lines of conduct open to him were accurately foreseen and
adequately realised in imagination at the present point of time. (ME 1.9.3.4)

Sidgwick notes that this definition involves only certain facts, and involves no
judgements of value or ‘dictates of reason’. We might also object to it on the ground
that it takes no account of the phenomenon of weakness of will, the existence of
which led Sidgwick to reject psychological hedonism and indeed psychological
egoism more generally. Sidgwick suggests it is more in line with common sense to
grant some authority to my desire for my good on the whole, and revises his definition of
‘ultimate good on the whole for me’ accordingly as ‘what I should practically desire if my
desires were in harmony with reason, assuming my own existence alone to be consid-
ered’ (ME 1.9.3.6).
This view is a version of what Rabinowicz and Rønnow-Rasmussen call a ‘fitting-
attitude’ (FA-)analysis of goodness, and it is in danger of running into what they call
‘the wrong kind of reasons objection’.6 Consider a case in which an evil demon
threatens to inflict great agony on me if I do not desire something that appears to be

5
G. E. Moore, Principia Ethica (Cambridge: Cambridge University Press, 1903), 9–17. Moore credits
Sidgwick with the idea at p. 17.
6
W. Rabinowicz and T. Rønnow-Rasmussen, ‘The Strike of the Demon: On Fitting Pro-attitudes and
Value’, Ethics, 114 (2004), 393.
 PLEASURE AND HEDONISM IN SIDGWICK 29

irrelevant to what is ultimately good for me, such as the possession of a saucer of mud.7
I now have a strong reason to desire the mud, but it is hard to see why we should
accept its possession as a constituent of my welfare or well-being. This objection is
still controversial, and several lines of response have been offered to it on behalf of
FA-analyses. But what lies behind the objection is the dependency of desire upon
goodness, which any account of goodness in terms of desire is likely to reverse. We
desire things, such as pleasure, because they are good, often because they are good for us.
‘Good for’, then, is better seen as a primitive notion, one that should not be elucidated
in terms of fitting attitudes.
It is important to note that Sidgwick’s welfare hedonism concerns primarily not how
well-off a person is at any point in time, but how well their life goes for them as a
whole. Pleasure and pain can be traded off against one another, so that a person’s
ultimate good or greatest possible happiness is to be understood as the ‘greatest
attainable surplus of pleasure over pain’ (ME 2.1.1.4). Further, ultimate good should
be understood in a temporally neutral way, so that the value of some future pleasure to
me does not depend, in itself, on when in my life it occurs: ‘this equal and impartial
concern for all parts of one’s conscious life is perhaps the most prominent element in
the common notion of the rational—as opposed to the merely impulsive—pursuit of
pleasure’ (ME 2.2.1.1 n. 1). This is not, of course, to deny that distance in time can
affect the probability of my obtaining some pleasure.

Arguments for Hedonism

Sidgwick’s arguments for hedonism need to be understood in the context of his overall
epistemological position. One major component of that position is his philosophical
intuitionism,8 according to which certain propositions are ‘self-evident’ and a person
who properly and reflectively grasps them can be justified in believing them on the
basis of that grasp. Like many hedonists, Sidgwick tends not to begin with positive
arguments for the hedonistic position before moving on to consider alternative posi-
tions and objections.9 Rather, he outlines his arguments in response to these positions
and objections. His first argument ‘appeals to the immediate intuition of reflective
persons’ (HUG 12; see ME 3.14.5.1; LE 107, 127–8). It is important to note that the

7
See R. Crisp, ‘Review of Kupperman, Value . . . and What Follows’, Philosophy, 75 (2000).
8
For further discussion see R. Crisp, ‘Sidgwick and the Boundaries of Intuitionism’, in P. Stratton-Lake
(ed.), Ethical Intuitionism (Oxford: Clarendon Press, 2002).
9
U, p. 5 may be interpreted as a conceptual argument for hedonism, according to which when pleasure is
understood sufficiently broadly it ‘cannot be distinguished from Happiness, except that Happiness is rather
used to denote a sum or series of those transitory feelings each of which we call a Pleasure’. But such an
interpretation ignores Sidgwick’s strictures elsewhere against turning substantive ethical judgements into
tautologies through stipulating senses. His claim here is rather the substantive one that happiness is pleasure,
broadly understood—or rather, as Sidgwick himself characteristically notes immediately—the greatest
balance of pleasure over pain.
30 RO G E R C R I S P

self-evident truth of welfare hedonism or its supporting principles,10 according to Sidg-

wick, does not consist in their being obvious. And by ‘immediate’ here, Sidgwick does not
mean to imply that acceptance of the proposition will occur as soon as any reflective
person considers it. Indeed, that goes against the very idea of reflection. His view is rather
that welfare hedonism will be accepted by anyone who reflects in the right way upon it.
He is here addressing the individual reader, and is quite aware that such a reader may
disagree with him.11 There is at this point something of a difficulty in Sidgwick’s
ethical epistemology.12 In ME 3.11.2 Sidgwick asks himself which conditions an
apparently self-evident proposition must meet if it is to be established ‘in the highest
degree of certainty attainable’. The first three conditions are relatively unproblematic:
the terms of the proposition must be clear and precise, the self-evidence of the
proposition must be ascertained by careful reflection, and the proposition must be
consistent with other beliefs held by the same subject. The fourth condition—which
we might call the consensus condition—is more troublesome:
[I]f I find any of my judgments, intuitive or inferential, in direct conflict with a judgment of some
other mind, there must be error somewhere: and if I have no more reason to suspect error in the
other mind than in my own, reflective comparison between the two judgments necessarily
reduces me temporarily to a state of neutrality.

Though Sidgwick is fully aware that ‘several cultivated persons do habitually judge
that certain ideal goods are ends independently of the pleasure derived from them’
(HUG 12), he tends—as we shall see shortly—to deal with this fact by providing
arguments against the non-hedonistic position rather than evidence that we should
suspect error in the minds of non-hedonists. These arguments might themselves
provide material for an evidence-claim, if it turned out that the anti-hedonists’ position
could be changed to hedonism through reflection upon them or if it could be shown
that the anti-hedonists had failed to consider the arguments in the first place. Attempt-
ing to show this would be especially appropriate since some of the arguments are
intended to explain how the relation between ‘ideal’ goods and pleasure might have
led the anti-hedonists mistakenly to attribute value to non-hedonistic goods. Never-
theless, it remains a brute fact that, if anything, a greater proportion than in Sidgwick’s
day of ‘cultivated persons’ are non-hedonists about welfare, and that many such
persons have heard the arguments of Sidgwick and other hedonists and remain

10
E.g. the views that nothing is good out of relation at least to some consciousness or feeling (ME 1.9.4.1),
that it could not be rational to aim at beauty apart from contemplation by human beings (1.9.4.2; see 1.9.4.4;
3.14.4.1), or that virtue is desirable only in so far as it promotes desirable conscious life (3.14.2.3).
11
Here we may note a similarity between Mill’s ‘proof ’ of the utility principle and Sidgwick’s position,
the difference consisting primarily in that, while Mill appeals to the reader to consider his desires, Sidgwick
asks for a straightforward judgement. (See J. S. Mill, Utilitarianism, ed. R. Crisp (Oxford: Oxford University
Press, 1998), chap. 4, para. 10.)
12
See R. Crisp, ‘Intuitionism and Disagreement’, in M. Timmons, J. Greco, and A. Mele (eds.),
Rationality and the Good: Critical Essays on the Ethics and Epistemology of Robert Audi (New York: Oxford
University Press, 2007).
 PLEASURE AND HEDONISM IN SIDGWICK 31

unpersuaded. In these circumstances, Sidgwick’s own consensus principle appears to

require him to suspend judgement on the question of whether welfare hedonism
provides the best account of well-being.
This is not to say that, if the relevant parties suspend judgement, debate on such
contested issues can no longer proceed. What will change is the conception of his or
her own view held by each party. Rather than thinking that she has seen the truth, and
that her opponent has not, she will accept that she has no justification for thinking that
she is closer to the truth than her opponent, but continue to insist, with the aid of
whichever arguments she wishes, that this is how things appear to her.13
Sidgwick’s second argument for hedonism consists in ‘the results of a comprehensive
comparison of the ordinary judgments of mankind’ (HUG 12; see ME 3.14.5.1;
LE 127). Here we see, rather than Sidgwick’s philosophical intuitionism, his Aristote-
lian commitment to the role of dialectic in ethics.14 Consider, for example, Aristotle’s
own discussion of happiness. In a central chapter of his Nicomachean Ethics (1.7) Aristotle
offers his famous ‘function’ argument to the conclusion that the human good consists
in virtuous rational activity. In the following chapter he attempts to shore up that
conclusion in part by showing how his conclusion is consistent with, indeed supported
by, many of the ‘ordinary judgments of mankind’.
Against the non-hedonists, Sidgwick notes that the ideal goods do produce pleasure
in several ways, and that common sense approves of them roughly in proportion to the
degree of such productiveness (HUG 12; ME 3.14.5.2). He claims that this is obvious
in the cases of beauty and freedom. As noted above, he is rather too sanguine about
this. For a clear denial of Sidgwick’s position on beauty, for example, consider Moore’s
case of the beautiful universe which Moore believes has value even if it is never seen by
anyone.15 Sidgwick accepts that knowledge is harder for the hedonist to deal with, but
notes first that common sense is especially impressed by knowledge that bears fruit.
Even when it is the result of ‘blue skies research’, it may receive some approbation on
the grounds that such apparently pure research often has unforeseen useful results, that
it provides the enquirer with the innocent pleasures of satisfying curiosity, and that the
intellectual disposition towards the acquisition of such knowledge is likely to be
valuable as a whole. Further, just as common sense tends to set limits on the acquisition
of knowledge according to its fruitfulness, so the same is true of the question of how far
another important alleged ideal good, virtue, is to be pursued in independence from its
valuable hedonistic consequences.
I find Sidgwick’s arguments powerful and in large part persuasive. But because of the
prevalence of serious and reflective disagreement, I suggest that Sidgwick’s own
consensus condition requires ‘epistemic humility’ of all of us as far as the correct

13
See Sextus Empiricus, Outlines of Scepticism, ed. and trans. J. Annas and J. Barnes (Cambridge: Cam-
bridge University Press, 2000), secs. [15], [31], [151], [196], etc.
14
See e.g. Aristotle, Nicomachean Ethics, ed. and trans. R. Crisp (Cambridge: Cambridge University Press,
2000), 7.1, 1145b2–7.
15
Moore, Principia, 83–4.
32 RO G E R C R I S P

account of well-being is concerned.16 This has implications for contemporary ethics, in

which welfare hedonism is usually quickly dismissed, often with an appeal to Robert
Nozick’s famous experience machine objection (see below). Hedonism remains as serious
a contender as any other for the title of correct theory of welfare, and in that respect the
arguments of Sidgwick and other hedonists deserve significantly more sympathy and
attention than they currently receive.

Pleasure
Sidgwick’s view, then, is that what is ultimately good for me is pleasurable experience,
and that a life becomes better for me the greater the balance of pleasure over pain
in that life. Sidgwick interprets the notions of pleasure and pain broadly, to include
‘respectively all kinds of agreeable and disagreeable feelings’ (ME 2.1.1.4; see U 5). This
breadth enables him immediately to sidestep any simple charge of sensualism.
Recall welfare hedonism as stated above: What is good for any individual is the
greatest balance of pleasure over pain. A welfare hedonist can be understood as
answering only the first, or both, of the following questions:17
1. The substantive question: Which goods and bads constitute well-being?
2. The explanatory question: What makes these goods and bads, respectively, good
and bad for the subject?
One may be a welfare hedonist only at the substantive level, claiming, for example, that
though pleasure is the only good, it is good because it perfects human nature. Here the
answer to the explanatory question is perfectionist, not hedonist. I take it that a ‘true’
hedonist will provide hedonistic answers to both questions, claiming that what makes
pleasurable experience good is its pleasurableness or pleasantness, and nothing else.
Sidgwick does not draw this distinction, but given his attack on the notion of ideal or
non-hedonistic goods and his claim against Mill that the ‘higher’ pleasures may be
preferred by the hedonist only on grounds of their pleasantness (see below), we can
safely assume that he is a true hedonist.
What, then, according to Sidgwick, is pleasantness, pleasurableness, or pleasure?
(Most) theories of pleasure can usefully be divided into two types—what Wayne
Sumner has called internalism and externalism.18 According to internalism, pleasurable
experience has a special, introspectible ‘feeling-tone’, and it is their possessing this
feeling-tone that different pleasurable experiences have in common and explains our
calling them all ‘pleasures’. It is this kind of view which Sidgwick has in mind in his
discussion of Alexander Bain in ME 2.2.2.3, when he asks whether pleasure is to be

16
See S. McGrath, ‘Moral Disagreement’, in R. Shafer-Landau (ed.), Oxford Studies in Metaethics (Oxford:
Oxford University Press, 2008), vol. 3, esp. sec. 3.
17
See R. Crisp, Reasons and the Good (Oxford: Clarendon Press, 2006), 102–3.
18
L. W. Sumner, Welfare, Happiness, and Ethics (Oxford: Clarendon Press, 1996), 87–91.
 PLEASURE AND HEDONISM IN SIDGWICK 33

understood as ‘a measurable quality of feeling . . . independent of its relation to volition,

and strictly undefinable from its simplicity?—like the quality of feeling expressed by
“sweet”, of which also we are conscious in varying degrees of intensity’.
Sidgwick is considering this internalist view of pleasure as an alternative to a version
of preference hedonism—the view of Spencer and Bain which he himself had previously
held (U 5) that pleasure is to be understood as a feeling which prompts the will to
maintain it when present or to produce it when absent.19 This view is a kind of
externalism, in that it characterizes pleasure by reference to some mental state inde-
pendent of pleasure as a special kind of introspectible feeling-tone. In his discussion
of this view, Sidgwick notes first (ME 2.2.1.2) that any such stimulation must often
be assumed to be latent, as in the case of relaxation after exertion or of someone
who has repressed his desires for certain things which would give him pleasure. Further
(ME 2.2.2.3), exciting pleasures have a volitional strength out of proportion to their
intensity as pleasures, and some experiences which create a strong stimulus towards
avoidance are either not painful or only slightly painful (Sidgwick’s characteristically
apt example is that of being tickled).
One option here would be to expand the scope of the definition beyond desire to
other attitudes—perhaps any kind of favourable attitude. According to attitudinal hedo-
nism, then, pleasure consists in any experience, or feeling, towards which the subject has a
favourable attitude. Of course, the nature of the attitudes in question would need to be
elucidated, but it would enable its proponent to avoid the problems of latency and
volitional strength faced by the Spencer–Bain form of preference hedonism.
Sidgwick does not consider this option. His preferred definition of pleasure is offered
as an alternative to internalism and Spencer–Bain preference hedonism. It follows a
clear rejection of internalism:
for my own part, when I reflect on the notion of pleasure,—using the term in the comprehensive
sense which I have adopted, to include the most refined and subtle intellectual and emotional
gratifications, no less than the coarser and more definite sensual enjoyments,—the only common
quality that I can find in the feelings so designated seems to be that relation to desire and volition
expressed by the general term ‘desirable,’ in the sense previously explained. I propose therefore
to define Pleasure—when we are considering its ‘strict value’ for purposes of quantitative
comparison—as a feeling which, when experienced by intelligent beings, is at least implicitly
apprehended as desirable or—in cases of comparison—preferable. (ME 2.2.2.3)

The first part of this passage is Sidgwick’s version of the ‘heterogeneity objection’,
which has led so many modern writers to reject internalist conceptions of pleasure.20

19
The view is stated in ME 1.4.2.1, but only tentatively for the purposes of the argument at that point,
with a reference forward to 2.2.2 in n. 1. This note implies that Sidgwick is, in that passage in book 2, merely
qualifying and limiting the view. I believe he is best understood as rejecting it—see below in the main text.
20
E.g. James Griffin: ‘The trouble with thinking of utility as one kind of mental state is that we cannot find
any one state in all that we regard as having utility—eating, reading, working, creating, helping. What one
mental state runs through them all in virtue of which we rank them as we do?’ (Well-Being: Its Meaning,
Measurement, and Moral Importance (Oxford: Clarendon Press, 1986), 8).
34 RO G E R C R I S P

Pleasurable experiences are so various that introspection is unable to identify any single
feeling-tone common to all of them. And in place of the notion of the stimulus to the
will, Sidgwick inserts the idea that pleasure—that is, what pleasurable experiences have
in common and which makes them pleasures—is their being, ‘when experienced by
intelligent beings . . . at least implicitly apprehended as desirable’. Let me call this the
apprehension account of pleasure.
There are two ways in which the role of apprehension in this account may be
understood. According to the first—the actual apprehension account—pleasure occurs
only when some feeling is in fact apprehended as desirable. Given that such apprehen-
sion is possible only for the ‘intelligent beings’ Sidgwick mentions, this would imply
that ‘lower’ animals, as well as certain human infants and certain mentally defective
humans, are unable to experience pleasure. Further, some of our most pleasant
experiences seem to be those in which we are engrossed in what we are doing,
where the question of the desirability of our experience at that time is far from our
minds.21 To avoid these difficulties, Sidgwick may be read as offering a hypothetical
apprehension account, according to which a feeling is pleasurable if its subject would
apprehend it as desirable if that subject were intellectually capable of such apprehension
and were to contemplate her own experience in the appropriate way.
As Sidgwick himself recognizes (ME 2.2.2.4), we sometimes think certain experi-
ences preferable to others for non-hedonistic reasons. Sometimes these experiences
need not even be pleasurable. Of such cases, Sidgwick says:
[I]t is not really the feeling itself that is preferred, but something in the mental or physical conditions
or relations under which it arises, regarded as cognisable objects of our common thought. For
certainly if I in thought distinguish any feeling from all its conditions and concomitants—and also
from all its effects on the subsequent feelings of the same individual or of others—and contemplate
it merely as the transient feeling of a single subject; it seems to be impossible to find in it any other
preferable quality than that which we call its pleasantness, the degree of which is only cognisable
directly by the sentient individual. (ME 2.2.2.5; see LE 124–6)

But consider a case in which I judge sensations of blueness to be preferable to

sensations of redness.22 I do not have any non-hedonistic reason for this judgement;
nor, however, do I have any hedonistic reason. It may even be that I do not enjoy
sensations of blueness any more than sensations of redness. On the apprehension
account, however, it seems that these sensations must be counted, by definition, as
pleasures.
Now Sidgwick begins ME 2.3.1.1 in the first edition as follows:
Let, then, pleasure be defined as feeling that is preferable or desirable, considered merely as
feeling, and therefore from a point of view from which the judgment of the sentient individual is
final: and not considered in respect of its causes, or of the relations of the sentient individual to

21
See e.g. M. Csikszentmihalyi, Flow: The Psychology of Optimal Experience (New York: Harper & Row, 1990).
22
I owe this example to Hurka.
 PLEASURE AND HEDONISM IN SIDGWICK 35

other persons or things, or of any other facts that come directly within the cognizance and
judgment of others beside the sentient individual.

At LE 130, responding to a charge of tautology by Green, he claims: ‘The statement

that Ultimate Good is a feeling of a certain quality, the quality being estimated by the
judgment of value implicitly passed on it by the sentient being at the time of feeling
it,—this proposition is certainly not a tautology.’ By the seventh edition the relevant
sentence in the ME reads:
Let, then, pleasure be defined as feeling which the sentient individual at the time of feeling it
implicitly or explicitly apprehends to be desirable;—desirable, that is, when considered merely as
feeling, and not in respect of its objective conditions or consequences, or of any facts that come
directly within the cognisance and judgment of others besides the sentient individual.

As Hurka has pointed out to me, there is a question why the judgement of the sentient
individual should be taken as final, since value judgements can be false. This may be why
Sidgwick made the change in later editions. The ‘therefore’ in the original sentence
explains what Sidgwick means by the claim that the individual’s judgement is final: only
someone experiencing a feeling is qualified to judge that feeling merely as feeling. This
allows that an individual may be mistaken as to the value of a feeling if she is considering
it, for example, in the light of its ‘objective conditions’. What Sidgwick would need to
show is that my non-hedonic judgement about blue sensations is somehow based on
‘objective conditions’. For example, you may note that I am judging my experience of
the sky on a sunny day as preferable to my experience of a pillar-box, and infer that I am
basing my judgement on colour. But, again, this need not be the case. My sight could be
defective in such a way that I see blue things as red, and vice versa. My access to the
colour of my sensations is as privileged as that to ordinarily pleasurable states.
The implication of this position, then, is that pleasure consists in a feeling such that,
were an intelligent being to experience that feeling, she would judge it, considered
purely as a feeling, to be desirable. This might be thought to involve an objectionable
kind of self-verification: if I judge F to be good, then it is good.23 But note that my
judging does not make F good, in the explanatory sense. F is made good by its being the
kind of feeling that is judged to be good purely as a feeling. Mere ‘judging good’ is not
a good-making property.
At times Sidgwick drops the reference to apprehension, and speaks merely of
‘desirable consciousness’ (e.g. ME 3.14.3.2; 3.14.5.5; 3.14.5.7). Such passages suggest
that the reference to apprehension may in fact not have been essential to his apparently
primary definition, and have been an attempt to capture the separate epistemological
point about the privileged access of an individual to the quality of her own feelings
considered merely as feelings. At 3.14.4.1 Sidgwick defines ‘Ultimate Good’ as ‘Desir-
able Consciousness’, and goes on:

23
This is another point I owe to Hurka.
36 RO G E R C R I S P

According to the view taken in a previous chapter [2.2], in affirming Ultimate Good to be
Happiness or Pleasure, we imply (1) that nothing is desirable except desirable feelings, and
(2) that the desirability of each feeling is only directly cognisable by the sentient individual at the
time of feeling it, and that therefore this particular judgment of the sentient individual must be
taken as final on the question how far each element of feeling has the quality of Ultimate Good.

Let me call this the desirable consciousness account, captured in statement ‘(1)’ in the
above quotation. Hurka has suggested to me that one problem with this account is that
it makes the claim ‘pleasure is good’ tautologous (see the discussion of Green’s criticism
at LE 129–31) and that this may lead Sidgwick into an objection based on the open-
question argument. But Sidgwick’s claim is not about the meaning of the word
‘pleasure’. His claim is that the thing that we call ‘pleasure’ consists in desirable
consciousness. Of course, once one has accepted the view that pleasure is so con-
stituted, it is no longer an open question whether it is good or desirable, but that in
itself is not a problem: all such definitions attempt to close such questions. But I think
Hurka’s point does direct our attention in the right direction. For the projects of
explaining what pleasure is and of evaluating it seem separate. In other words, the
desirability of pleasure cannot be one of its defining characteristics. We should be able
to agree on what pleasure is, and then move on to evaluating it. But stripped of the
notion of desirability, the desirable consciousness account is left with nothing but the
claim that pleasure is consciousness of some kind or other. That may be true, but it is
close to vacuous.
Here is a further, related point. The desirable consciousnesss account will of course
incorporate the same restriction on ‘desirability characterizations’ as the apprehension
account—this must be consciousness or feeling which is desirable considered purely as
such—but will still run into the ‘blue sensations’ problem above. Since the words
‘valuable’ or ‘good’ could be substituted for ‘desirable’ in this account, it seems hard to
characterize as either internalist or externalist. No reference need be made to external
attitudes; nor, however, is there any commitment to pleasure or pleasantness as an
introspectible and homogeneous feeling or sensation.
One way to avoid the blue sensations problem would be to allow pleasantness itself
as the only relevant desirability characterization. So it will be only when I find blue
sensations preferable to red on grounds of their pleasantness that my experiencing such
sensations will count as pleasures. This way, one might think, is blocked for Sidgwick,
since—unless some further externalist account of pleasure is offered—it consists in the
reintroduction of internalism about pleasure. But in fact I believe that Sidgwick is best
understood as such an internalist.24 Sidgwick speaks of pleasure, in the core of his
definition, as ‘a feeling’ (ME 2.2.2.3) and as a ‘kind of feeling’ (ME 2.2.2.4). Further, he
goes on to speak of pleasantness as itself a quality cognizable by the subject (ME 2.2.2.5)
and as a ‘ground’ for a judgement of preferability (ME 3.14.4.1). If all we had in play

24
See n. 35 below.
 PLEASURE AND HEDONISM IN SIDGWICK 37

were feelings, non-hedonistically construed, which were then apprehended as desir-

able, it is hard to see how pleasure could itself be cognizable or a ground for a
judgement. Rather, the non-hedonistically construed feeling would be immediately
apprehended as desirable, and pleasure would be merely this combination of the feeling
and its apprehension.
I suggest, then, that Sidgwick is at heart an internalist about pleasure, who is
misled—as have been so many other philosophers—by the heterogeneity argument
into offering an externalist view which is open to serious objections. Pleasurable
experiences are indeed very different from one another. But what they have in
common is their pleasantness, and this can be taken to be a primitive aspect of our
experience which requires no elucidation through reference to the will, desires, or any
other independent mental capacity or state.

Objections and Replies

Since Sidgwick believes welfare hedonism to be highly plausible in itself in the light of
impartial reflection, he often, as I have mentioned, develops lines of argument in
support of hedonism in the form of responses to objections. Several of these objections,
emanating from writers such as T. H. Green and F. H. Bradley, are of their time, and
Sidgwick often provides a convincing rebuttal. Consider, for example, the impossibility
of conception objection—Green’s claim that ‘pleasure as feeling, in distinction from its
conditions that are not feelings, cannot be conceived’. As Sidgwick notes, this contra-
dicts not only common sense and the views of empirical psychologists, but an assump-
tion which underlies many of Green’s arguments elsewhere (ME 2.3.1.4). The view
also has counter-intuitive implications. If it were correct, then we could not compare
geometrical angles of objects without comparing their sides, since angles cannot be
conceived of apart from sides (HUG 13).
Another example is a non-additivity argument found in both Bradley and Green, that
pleasures cannot be added since they occur in series and not at the same time. Again
Sidgwick notes the counter-intuitiveness of the argument as applied elsewhere (HUG
14; ME 2.3.1.5). If Bradley and Green are right, then we could not add periods of time,
since they occur in series. Nor is it clear why the good must be possessed all at once.25
A hedonist will of course not believe that it can be. All that hedonism needs is that
some quantitative judgements are possible of pleasure and pain, since this implies
pleasures can be summed in thought (LE 9–10). Nor should we accept that the mere
fact that pleasures are transient is a cause of pain either at the time or subsequently. It is

25
I cannot resist quoting the following dry remark by Sidgwick, a rare case of his well-attested sense of
humour entering his philosophical writing: ‘The man who has philosophised himself into so serious a quarrel
with the conditions of human existence that he cannot be satisfied with the prospect of never-ending bliss,
because its parts have to be enjoyed successively, and under the condition of being successively desired—such
a man, I venture to think, is not a typical çæØ [man of practical wisdom]’ (GE 249; see also LE 108, 119).
38 RO G E R C R I S P

true that we often gain much pleasure from enduring sources, such as friendships. But
this is a mere exemplification of the ‘paradox of hedonism’ that we often gain more
pleasure by pursuing objects other than pleasure itself (ME 2.3.2.1–2).
One common modern objection to welfare hedonism rests on the notion of evil
pleasures.26 The welfare hedonist, it is objected, must accept that the pleasures of a sadist
are good, whereas it is clear that they are not only bad, but evil. This objection is
especially problematic for a global hedonist who claims that pleasure is the only value.
A welfare hedonist can draw the distinction between ‘good for’ and ‘good’, and might
even accept that, while sadistic pleasures are good for the sadist, they are indeed morally
evil. But this response is of course unavailable to Sidgwick. His answer—though he
nowhere explicitly considers the issue—would be that the objection rests on common-
sense morality, and that any plausibility that form of morality has must rest upon a
foundation of universal hedonism. So it cannot provide a solid basis for an objection to
universal hedonism itself.
Another objection to which Sidgwick’s welfare hedonism is open is what we might
call the philosophy of swine objection.27 Sidgwick does indeed offer us a broad conception
of pleasure, encompassing not only sensual or bodily enjoyment but the intellectual or
‘higher’ pleasures of, say, music or literature. But, the objection goes, he is committed to
the view that, say, enjoying full appreciation of a late Beethoven quartet is valuable for
exactly the same reason as a drink of beer on a warm day—its pleasantness. But such
aesthetic activity is of an entirely different order from the lower, animal pleasures of
satisfying bodily desires. Hedonism, in other words, is unacceptably reductionist.
John Stuart Mill famously attempted to deal with the philosophy of swine objection
within a hedonist framework by drawing a distinction between quantity and quality of
pleasure, and arguing that some pleasures, because of their quality, are ‘higher’ than
others:28
It is quite compatible with the principle of utility to recognise the fact, that some kinds of pleasure
are more desirable and more valuable than others. It would be absurd that while, in estimating all
other things, quality is considered as well as quantity, the estimation of pleasures should be
supposed to depend on quantity alone.
If I am asked, what I mean by difference of quality in pleasures, or what makes one pleasure
more valuable than another, merely as a pleasure, except its being greater in amount, there is but
one possible answer. Of two pleasures, if there be one to which all or almost all who have
experience of both give a decided preference, irrespective of any feeling of moral obligation to
prefer it, that is the more desirable pleasure. If one of the two is, by those who are competently
acquainted with both, placed so far above the other that they prefer it, even though knowing it
to be attended with a greater amount of discontent, and would not resign it for any quantity of

26
See e.g. J. Harsanyi, ‘Morality and the Theory of Rational Behaviour’, in A. Sen and B. Williams (eds.),
Utilitarianism and Beyond (Cambridge: Cambridge University Press, 1982), 56.
27
Mill, Utilitarianism, 2.4.
28
Ibid. 2.4–8.
 PLEASURE AND HEDONISM IN SIDGWICK 39

the other pleasure which their nature is capable of, we are justified in ascribing to the preferred
enjoyment a superiority in quality, so far outweighing quantity as to render it, in comparison, of
small account.29

Sidgwick has several objections to Mill’s account. The first (U 6; ME 2.3.7.2) is to

the Platonic notion that the judgement of the person with a capacity for higher
pleasures is to be accepted, on the ground that he or she has experienced both kinds
of pleasure, whereas the person with a capacity for lower pleasure alone has too narrow
a base for judgement: ‘who can tell that the philosopher’s constitution is not such as to
render the enjoyments of the senses, in his case, comparatively feeble?’
On the face of it, this seems a reasonable objection. But its force is somewhat
diminished by the fact that many intellectuals seem able to engage in paradigmatically
lower pleasures with at least as much gusto as sensualists.
Another objection is more serious, and had already been made by a number of
writers.30 Sidgwick claims that Mill faces a dilemma (GU I, 174–5; II, 177;
FPM 262–3; ME 2.1.1.4). Either he can remain a hedonist, in which case all differences
of quality must be resolved into those of quantity, since a higher pleasure must be more
valuable because, and only because, it is more pleasurable. Or he can say, as indeed he
does, that a higher pleasure is more valuable in so far as it is more noble. But this is to
abandon hedonism in favour of a view which allows for ideal, non-hedonistic goods—
in this case, nobility.
There is in fact a way out of this dilemma for Mill.31 He had inherited from the
empiricist tradition a conception of pleasure according to which its value at any time
depends only on its pleasantness, and its pleasantness only on the intensity of the
pleasure and its duration. Logically, Mill is permitted to claim that qualities such as
nobility also increase pleasantness, thus allowing for the possibility of higher pleasures
without giving up hedonism. It has to be admitted that this strategy does face some
serious objections. First, why expect that properties such as nobility will indeed
increase pleasantness—indeed that pleasantness will vary in proportion to the degree
of nobility? Many experiences which might be described as noble seem, if anything,
painful (giving birth, for example). Second, if nobility can affect value in this way
indirectly, why should it not do so directly, without mediation through pleasure?
Mill’s position so understood does seem to be something of an unstable—if never-
theless hedonistic—compromise between hedonism and the view that there are ideal
goods.
What response might Sidgwick make to the philosophy-of-swine objection? Mill
appears to accept that there are so-called ‘discontinuities’ of value, such that some finite
amount of a higher pleasure cannot be compensated for by any amount of lower,
thus putting the experience of Beethoven and that of the beer on entirely different

29
Mill, Utilitarianism, 2.5.
30
See e.g. T. H. Green, Prolegomena to Ethics (Oxford: Clarendon Press, 1883), 167–78.
31
See R. Crisp, Mill on Utilitarianism (London: Routledge, 1997), 31–5.
40 RO G E R C R I S P

footings.32 Sidgwick allows the possibility of such discontinuities, but claims not to
have detected any (FPM 262; ME 2.2.1.1), asserting also that ordinary prudential
reasoning rests on the assumption that there are none. As he recognizes (ME 2.2.1.1
n. 1), discontinuities of value introduce the mathematics of infinity into value theory,
and few would accept, for example, that to avoid the smallest risk of extreme agony we
should be prepared to accept a very great finite amount of moderate pain below the
alleged threshold of discontinuity. Nevertheless, because he does leave conceptual
room for discontinuity, this response to the objection is open to Sidgwick. What he
would probably do, however, is again appeal to his debunking account of common-
sense morality, arguing that the principles on which the objection rests (that art is
especially valuable or noble, for example, and valuable in itself independently of its
consequences) are themselves grounded on the hedonistic value which their adoption
promises.
One of the most influential objections to hedonism in modern times revolves
around the example of the experience machine. The most famous statement of the
objection is by Robert Nozick:
The Experience Machine. Suppose there were an experience machine that would give you any
experience you desired. Superduper neuropsychologists could stimulate your brain so that you
would think and feel you were writing a great novel, or making a friend, or reading an interesting
book. All the time you would be floating in a tank, with electrodes attached to your brain . . .
Would you plug in? What else can matter to us, other than how our lives feel from the inside? 33

This form of the objection applies to any theory of well-being which includes the
‘experience requirement’ that well-being consists only in the quality of our mental
states, understood independently from the external world. Since hedonism is such a
theory, it has to confront the objection.
The objection has most force in cases in which authenticity seems significant.
Imagine that I am drinking a cold beer. It is not implausible for a hedonist to claim
that the value of the enjoyment in this activity would be the same if I were in fact
plugged into an experience machine. But now imagine that I am writing a novel, and
enjoying it. According to the hedonist, the level of well-being of someone on an
experience machine which ‘copies’ my veridical experiences and replays them to that
person is, in this respect, equal to mine. This seems highly counter-intuitive. Don’t
creativity, genuineness, authenticity, knowledge, truth, accomplishment, and many
other goods matter independently of pleasure?
The experience-machine objection, then, essentially amounts to the traditional
claim that there are non-hedonistic or ideal goods. So, although Sidgwick never
confronts it explicitly, what he says about ideal goods provides us with the opportunity

32
Mill, Utilitarianism, 2.5. The term ‘discontinuity’ is Griffin’s; see Well-Being, 85–9.
33
R. Nozick, Anarchy, State, and Utopia (Oxford: Blackwell, 1974), 42–3. See also Nozick, The Examined
Life (New York: Simon & Schuster, 1989), 104–8.
 PLEASURE AND HEDONISM IN SIDGWICK 41

to construct a response on his behalf. As we have already seen, he draws attention to the
fact that ideal goods are productive of pleasure, and claims that common sense approves
such goods roughly in proportion to that productiveness. Because of his acceptance of
the consensus principle, however, Sidgwick appears committed in the case of the
experience machine to suspension of judgement, since many reflective thinkers accept
that the ideal goods illuminated by the objection are genuine and there is no plausible
account available of how such thinkers could be mistaken. Sidgwick may indeed claim
that they are mistaken, believing that items valuable because productive of pleasure are
valuable in themselves and misled by the fact that pursuing these items as if good in
themselves may—because of the paradox of hedonism—be the most effective way to
advance the balance of pleasure over pain. But this claim would be a mere hypothesis,
which though it may be open to empirical verification has not been so verified or even
properly tested.
The final objection to hedonism I shall discuss is not so much to the theory itself as
to its practicability. Sidgwick’s discussion of ‘empirical hedonism’ (esp. ME 2.3; see
also U 6) is a brilliant and detailed inquiry into whether one would be able to make
decisions on the basis of a hedonistic account of value. In recent years many of
Sidgwick’s doubts have been borne out by empirical research in psychology and related
disciplines, and it is almost certainly fair to say that some of his claims would be worth
testing further.34
Sidgwick says that the utilitarian idea of the ‘greatest happiness’ assumes compara-
bility across all pleasures and pains: ‘that every kind of feeling has a certain intensive
quantity, positive or negative (or perhaps zero), in respect of preferableness or desir-
ableness, and that this quantity can be known; so that each can be weighed in ideal
scales against every other’ (U 6; see ME 2.2.1.1).35 This claim is an exaggeration, as
Sidgwick himself would probably have accepted. It is not inconsistent to claim that
there is a possible state of the world in which happiness is at its greatest, but that we
have absolutely no idea how to bring about this state. What Sidgwick means is that the
notion of maximum happiness can be put fully to work only on the assumption of
commensurability.
Sidgwick does not set out to discuss whether this assumption is correct. But he does
claim that it is not verified by experience (that, of course, is not to say that it is falsified:
ME 2.3.6.1), and that various empirical objections can be made to it. First (U 6; ME
2.3.5.1–4; 2.3.6.2; 2.3.7.1–4), though we often do compare pleasures and pains, the
results of those comparisons depend on our mood, state of satiety, or other qualities and
capacities at the time, and we have no reason to think that there is a ‘neutral’ vantage

34
See e.g. D. Kahneman, E. Diener, and N. Schwartz (eds.), Well-Being: The Foundations of Hedonic
Psychology (New York: Russell Sage Foundation, 1999).
35
Hurka has pointed out to me that this passage provides further evidence of Sidgwick’s internalism about
pleasure. If the intensity of a feeling is determined by the subject’s judgement of that feeling’s desirability, the
intensity will be definite only if the judgement is definite. But few if any of us (including Sidgwick, of course)
make definite judgements about the degree of value of our feelings.
Exploring the Variety of Random
Documents with Different Content
 Uxellum, Dumfrys in Nithsdale.
Uxellum, rightly placed by Baxter, the r. Nyth, Nithisdale, or
Dumfries.
Thus I have recounted the names of places contained in this
excellent map, to the number of 250; whereof 100, marked in this
catalogue thus *, are wholly new, or ill-placed by former writers. The
reader versed in these kind of inquiries, will find no small number of
them; to his judgement I leave them: as to me, the finding fault
with others endeavours is very disagreeable. This I may say; it sets
us right in abundance, wherein before we had no guide but
conjecture, from similitude of names: as, for instance, Uxella, placed
in some great authors at Lestwthiel, Cornwall, is in Somersetshire,
viz. at Barton, where the Roman road called Foss crosses the river, a
little north of Ilchester. Many more might be specified, where only a
map can properly direct us.
I must take notice of another use in our map. In the province of
Brittania Prima are two Venta’s; but till now we could not ascertain
them both: the map shows us, one is Wimborn minster, the other
Winchester: the former is on the river Alauna, seen plainly in
Blandford, being the ford over the Alauna; Llaunford, in the Belgic
pronunciation: called now Allen river. Our author calls Canterbury,
Cantiopolis, though before we knew no other name it had than
Durovernum: but the modern name of Canterbury seems derived
from the former; and the termination favours our author’s
observation, in another part of his history, of remains of Greek
traders preserved in some places; of which several more instances
may be given.
I extend my inquiries here, on Richard of Cirencester’s map, no
further than our island of Britain; leaving that of Ireland to those
that have proper opportunities.
Nor shall I pretend to assign places in Scotland, any further than
the map directs me; but leave them too to those that have proper
opportunities of inquiry, in that kingdom.
 III.

L ET us now proceed to his Itinerary; a truly invaluable monument!

From these two we may hope to obtain a complete knowledge of
Roman Britain.

CAPUT VII.
Our author calls these, Iters of his Diaphragmata, from their
similitude to the animal midriff, passing through the body from side
to side.
Rhutupis colonia, Sandwich, Richborough and Stonar castle, Kent,
is the first city, says our author, in the island of Britain, towards Gaul;
situate among the Cantii, opposite to Gessoriagum, the port of
Bononia, Boloign. Hence is the most commodious passage of ccccl.
stadia, or, as others will have it, xlvi. miles.
From that city Rhutupium, says he, is drawn the Roman way called
Guithlin-street, quite to Segontium, Caernarvon, through the space
of cccxxiv. miles, or thereabouts. Thus,
To Cantiopolis, which is also called Durobernum, stipendiaria,
Canterbury, Kent, x. miles.
Durosevum xii. Sittingburn, Kent.
XXV.
Duroprovis, stipendiaria, Rochester, Kent.
Thence, at xxvii. miles, it passes the Thames, and enters the
province Flavia, and the city of Londinium Augusta, London. Thence
IX.
To Sulloniagis, Suellaniacis, Edgeware, Middlesex.
 XII.
Verolamium, municipium, Verlamcester, or St. Alban’s. Of this place
were Amphibalus and Albanus, martyrs.
XII.
Forum Dianæ, Market street, near Dunstable, Hertfordshire.
XII.
Magiovinium, Dunstable, Bedfordshire.
XII.
Lactorodum, Stoney Stratford, Bucks.
XII.
Isannavaria, Isantavaria, Towcester, Northamptonshire.
XII.
Tripontium, Dowbridge, Stanford, Northamptonshire.
IX.
Benonis, Highcross, Cleycester, between Warwickshire and
Leicestershire. Here the road is divided: the one branch, the Foss,
goes to Lincoln; the other to Viriconium, Wroxeter, from Tripontium.
XII.
To Manduessedum, Mancester, near Atherston, Warwickshire.
XIII.
Etocetum Wall, by Litchfield, Chesterfield wall, Staffordshire.
XII.
Pennocrucium, by Penkridge, Staffordshire.
XII.
Uxoconium, Okenyate, Shropshire.
XI.
Virioconium, Wroxcester, Salop.
XXVI.
 Banchorium, Bonium, Banchor, Flintshire.
X.
Deva colonia, leg. vices. victrix Cretica, Westchester; the border of
Flavia and Secunda provinces.
XXX.
Varis, Bodvary by Denbigh on r. Clwyd.
XX.
Conovium, Aberconway, Carnarvonshire.
XXIV.
Seguntium, stipendiaria, Caernarvon.
Were I to recite all I have written upon this work, by way of
comment, it would amount to a large volume; yet some few remarks
I must make.
What all others call Durolenum our author names Durosevum,
which I affix to Sittingburn, favouring this reading: the distance
conformable.
Sulloniacis, or rather Suellaniacis, has its name from Suellan, or
Cassibelin, who fought Cæsar. I place it at Edgware, which has its
name from the agger, or high raised Roman way, Watling-street.
Here was Cassibelin’s usual residence: his oppidum, or military town,
which Cæsar stormed, was at Watford.
Forum Dianæ, a new name, was crouded into the roll of the
original Itinerary, where the intermediate distance, xii. miles,
between St. Alban’s and Dunstable, remained unaltered: therefore
the transcriber repeated the same distance erroneously.
I doubt not, the place is what we now call Market-street, a little on
this side Dunstable, upon the great road Watling-street. Here was a
fane, and forum, or portico, sacred to Diana; where a panegyre, or
fair, as we call it, was annually celebrated, to the honour of the
goddess, by the lovers of hunting, on the great festival sacred to her,
when stags were sacrificed: this was upon August 13, the hunters’
day, in the Roman kalendar.
I have no need to be ashamed in acknowledging an error incurred
in my juvenile travels, when we knew nothing of this work of our
author’s; for now I apprehend Durocobrivis is another name of a
town near this place: the modern name of Redburn proves it, which
means the same as Durocobrivis, the passage over the Redwater
brook.
Rotten row, Rowend, Flamsted by Forum Dianæ, names importing
high antiquity: Rotten row, just by Bremenium, Ruchester; again at
Dorchester, Oxfordshire: they relate to panegyres, or fairs.
Manduessedum, Mancester, on each side the Watling-street, was
walled about.
The vestigia of Benonis are at Claybrook.
Thus we have the whole length of the Watling-street, from Dover
to Caernarvon.

ITER II.
A Segontio, Caernarvon, Virioconium, Wroxcester, usque lxxiii.
miles, thus.
Segontium, stipendiaria, Caernarvon, Carnarvonshire.
XXV.
Herirus mons, Raranvaur hill by Bala, Merionethshire, by
Pimblemere.
XXV.
Mediolanum, Myvod, on Merway r. Montgomeryshire.
XII.
Rutunium, Rowton castle; Stanford, Watlesborough, west of
Shrewsbury.
XI.
 Virioconium, Wroxcester on the Severn, below Shrewsbury, under
Wrekin hill.
Caernarvon stands on the river Seint, Seient, Segont, said to have
been built by Constantine the Great. Nennius gives it the name Kaer
Kustenidh, for that reason: he probably made the Via Heleniana, in
honour of his mother, called Sarn Helen.
Herirus mons has its name from the eagles inhabiting the place,
Celtic.

ITER III.
From Londinium, London, to Lindum colonia. Lincoln, thus,
Londinium Aug. London.
XII.
Durositum, Romford, Essex.
XVI.
Cæsaromagus, Chelmsford, Essex.
XV.
Canonium, Kelvedon, Essex.
IX.
Camulodunum colonia, leg. gem. Mart. Victrix, Colchester, Essex.
VI.
Ad Sturium amnem, ad Ansam, Stretford street, Suffolk.
XV.
Combretonium, Bretenham, Stow, Combe, Suffolk.
XXII.
Sitomagus, Thetford, Norfolk.
XXIII.
Venta Cenomanorum, stipendiaria, Caster by Norwich, Norfolk.
XXVII.
 Icianis, Ixworth, Suffolk.
XX.
Camboritum, colonia, Chesterford, Cambridgeshire.
XX.
Durosiponte, Godmanchester, Huntingdonshire.
XX.
Durnomagus, Latio jure donatus, Dormancester, Caster by
Peterborough, Northamptonshire.
XX.
Causennis, Corisennis, Stanfield by Bourn, Lincolnshire.
XX.
Lindum colonia, Lincoln.
Iter VI. of Antoninus, a Londinio Lindum, goes quite a different
way from this; the one to the right, the other to the left of the
straitest way, the Hermen-street. Instead of our Durnomagus on the
northern, he mentions Durobrivis, Chesterton, on the southern bank
of the river Nen, a walled city: a bridge over the river, built since the
time of our Itinerary. And also
From Camboritum to Durosiponte, in this Iter of ours, and Vth of
Antoninus, I collect, the Roman city of Cambridge, Granta, was not
then in being.
I suppose, it was founded by Carausius, when he carried the
Carsdike from Peterborough to Cambridge, and made the road over
Gogmagog hill from Durosiponte, Godmanchester, to Camulodunum
colonia, Colchester; for all these Itineraries were made before
Carausius’s time.

ITER IV.
From Lindum, Lincoln, to the Vallum, the Roman wall, thus.
Lindum colonia, Lincoln.
 XIV.
Argolicum, Littleborough on Trent, Nottinghamshire.
XX.
Danum, Doncaster, Yorkshire, you enter Maxima Cæsariensis.
XVI.
Legolium, Castreford, Yorkshire.
XXI.
Eboracum municipium, formerly colonia, leg. vi. victrix, York.
XVI.
Isurium, Aldborough by Boroughbridge, Yorkshire.
XXIV.
Cataractonium, Latio jure donat. Cateric, Yorkshire.
X.
Ad Tisam amnem, Piersbridge, Durham county.
XII.
Vinovium, Binchester, Durham county.
XIX.
Epiacum, Chester in the street, Durham county.
IX.
Ad Murum, Newcastle, Northumberland.
XXV.
Ad Alaunam, flu. Alnwick, Northumberland.
XXX.
Ad Tuedam, flu. Berwick, Scotland.
LXX.
 Ad Vallum, Falkirk, Scotland.

ITER V.
From the Vallum, Falkirk, to Prætuarium, Patrinton. Vallum,
Antonini, Falkirk, Scotland.
. . . .
Corium, on the Watling-street, Romanhow, Korstonlaw.
. . . .
Ad Tines, Rochester on the river Tyne in Redesdale.
. . . .
Bremenium, stipendiaria, Ruchester, upon Watling street.
XX.
Corstoplium, Corbridge, Northumberland.
IX.
Vindomora, Ebchester upon Dervent river, Durham county.
XIX.
Vinovium, Binchester, Durham county.
XXII.
Cataractonium, Latio jure donatum, Cateric, Yorkshire.
XL.
Eboracum, leg. vi. Victrix, York.
VII.
Derventio, Stanford bridge, Yorkshire.
XIII.
Delgovicia, Wighton, Yorkshire.
XXV.
 Prætuarium, Patrinton, Yorkshire.

ITER VI.
From Eboracum, York, to Deva, Chester.
Eboracum, municipium, formerly a colony of legio vi. victrix, York.
IX.
Calcaria, Tadcaster, Yorkshire.
XXII.
Cambodunum, Latio jure dotatum, Alkmanbury, Yorkshire.
XVIII.
Maucunium, Mancastle by Manchester, Lancashire.
XVIII.
Ad Fines, between Maxima and Flavia, Stretford on Mersey,
Cheshire.
XVIII.
Condate, Northwich, Cheshire.
XVIII.
Deva, colonia, legio Cretica, vicesima, Valeria, victrix, West
Chester.

ITER VII.
From the port of the Sistuntii, Lune river mouth, to Eboracum,
York.
Portus Sistuntiorum, Lune river mouth, by Lancaster.
XXIII.
Rerigonium, Ribcester on the Rible, Lancashire.
VIII.
Alpes Pennini, Pendleton by Pendlehill, Lancashire.
 X.
Alicana, Shipton in Craven, Yorkshire.
XIX.
Isurium Brigantum, Brigantium, Aldborough by Burrough bridge.
XVI.
Eboracum, municipium, formerly colonia leg. vi. victrix.
This is the first Iter of Antoninus, which is deficient in our three
first stations; which are those between the two Prætentura’s,
therefore at that time out of the possession of the Romans.
We learn hence, York was a colony city of the vith legion, built by
them in the time of Hadrian, who probably then made, or finished,
the artificial canal called Carsdike, when he made the vallum.

ITER VIII.
From Eboracum, York, to Lugubalia, Carlisle.
Eboracum, formerly colonia, legio vi. municipium, York.
XL.
Cataractum, Cateric, Thornburgh, Latio jure donata.
XVIII.
Lataris, Lavatris, Bowes, Yorkshire.
XIII.
Vataris, Verteris, Brough on Stanmore, Westmorland.
XX.
Brocovonacis, Brocavum, Brovonacis, Whitley castle, Browham,
Westmorland.
XIII.
Voreda, Castle Voran on the Wall, Cumberland.
XIII.
 Luguvalia, Carlisle, Latio jure donata.

ITER IX.
From Lugubalia, Carlisle, to Pterotone, Inverness. Luguvalia,
Carlisle, Latio jure donata.
. . . .
Trimantium, Cannaby, by Longtown, Netherby, Langhoom castle.
. . . .
Gadanica, Colanica, Colecester.
. . . .
Corium, Corsford by Lanerk.
. . . .
Ad Vallum, Falkirk.
XII.
Alauna, Sterling, on Alon river.
IX.
Lindum, Cromlin castle.
IX.
Victoria, Kinkel upon Erne r. Latio jure donata.
IX.
Hierna, Perth, on Terne river.
XIV.
Orrea, Dunkeld.
XIX.
Ad Tavum, Brumchester, on Tay frith.
XXIII.
Ad Æsicam, Brechin, on S. Esk river.
 VIII.
Ad Tinam, Eshlie, on N. Esk.
XXIII.
Devana, Aberdeen.
XXIV.
Ad Itunam, Fyvie.
. . . .
Ad montem Grampium.
. . . .
Ad Selinam, Celnius fl. on Devern river.
XIX.
Tuæssis, Rothes, on the Spay.
XXVII.
Pterotone, Alata castra, Inverness, Latio jure donata.

ITER X.
From the boundary Pteroton, Inverness, through the length of the
island, to Isca Dumnoniorum, Exeter.
Pteroton, Alata castra, Latio jure donata, Inverness.
IX.
Varis, in Badenec on Findern river.
XVIII.
Tuæssis, Ruthvan on Spay.
XXIX.
Tamea, Castleton on Calder, in Aberdeenshire.
XX.
- - - Spittle, in Glenshire.
 IX.
In medio, Strumnic on Eric river.
IX.
Orrea, Dunkeld.
XVIII.
Victoria, Latio jure donata, Kinkel.
XXXII.
Ad Vallum Antonini, Falkirk.
LXXX.
Lugubalia, Latio jure donata, Carlisle.
XXII.
Brocavonacis, Penrith, Browham.
. . . .
Ad Alaunam, Lancaster.
LXVI.
Coccium, Latio jure donata, Bury and Cockley chapel, Lancashire.
XVIII.
Mancunium, Mancastle by Manchester.
XXIII.
Condate, Northwich, Cheshire.
XVIII.
Mediolanum, Chesterton by Newcastle, Staffordshire.
. . . .
Etocetum, Wall by Litchfield.
. . . .
Bremenium, Birmingham, Warwickshire.
. . . .
 Salinis, Droitwich, Worcestershire.
. . . .
Branogenium, Worcester.
. . . .
Glebum colonia, legio vii. Aug. Claudia, Gloucester.
XIV.
Corinium Dobunorum, Latio jure donata, Cirencester.
. . . .
Aqua Solis, colonia, Thermæ, Bath.
XVIII.
Ad Aquas, Wells, Somersetshire.
. . . .
Ad Uxellam amnem, Balsborough, Lydford, Barton on the Foss,
Somersetshire.
. . . .
. . . .
Isca Dumnoniorum, stipendiaria, Exeter.
This Xth Iter is the only remaining monument of the Roman power
in Scotland. I shall no further attempt an assignment of the present
names, than I am led to them by our map; but leave them to be
determined more precisely, by those who have an opportunity of
inquiring on the spot.

ITER XI.
From Aquæ Solis, Bath, by the Julian street, to Menapia, St.
David’s.
Aquæ Solis, Thermæ, colonia, Bath.
VI.
Ad Alone, Olland near Kainsham, Gloucestershire.
 VI.
Ad Sabrinam, Aust upon Severn, Gloucestershire.
III.
Statio Trajectus, Tydenham or Chepstow, Gloucestershire.
IX.
Venta Silurum, stipendiaria, Caer Went, Monmouthshire.
IX.
Isca Silurum, colonia, leg. vi. Aug. Caerleon, Monmouthshire.
VII.
Tibia amnis, Caerdiff, Glamorganshire.
XX.
Bovium, Cowbridge, Glamorganshire.
XV.
Nidum, Neath, Glamorganshire.
XV.
Leucarium, Loghor, Glamorganshire.
XX.
Ad vigesimum lapidem, Narbath castle, on Clethy river,
Pembrokeshire.
XIX.
Menapia, St. David’s.

ITER XII.
From Aquæ Solis, Bath, to Londinium, London.
Aquæ Solis, colonia, Thermæ, Bath.
XV.
Verlucio, Lacock on the Avon, Wiltshire.
 XX.
Cunedio, Marlborough.
XV.
Spinis, Spene, Berkshire.
XV.
Calleba Atrebatum, Wallingford, Berkshire.
XX.
Bibracte, Madanhead, Bray, Braywick, Sutton Bray, Berkshire.
XX.
Londinium Aug. municipium, London.

ITER XIII.
From Isca Silurum, Caerleon, to Urioconium, Wroxeter.
Isca Silurum, legio ii. Aug. Caerleon, Monmouthshire.
IX.
Bultrum, Burrium, Bullium, Usk in Monmouthshire.
XII.
Gobannium, Abergavenny.
XXIII.
Magna, Old Radnor.
XXIII.
Branogenium, Worcester.
XXVIII.
Uriconium, Viroconium, Wroxeter near Wrekin, Shropshire.

ITER XIV.
From Isca, Caerleon, by Glevum, Gloucester, to Lindum, Lincoln.
Isca Silurum, leg. ii. Aug. Caerleon, Monmouthshire.
VIII.
Bullium, Burrium, Usk in Monmouthshire.
XII.
Blestium, the Old town, Herefordshire.
XI.
Ariconium, Kenchester, by Hereford.
XV.
Glevum, colonia, leg. vii. Aug. Claudia, Gloucester.
XV.
Ad Antonam, flu. Evesham, Worcestershire.
XV.
Alauna, Alcester, Worcestershire.
. . . .
Præsidium, Warwick.
. . . .
XII.
Vennonis, Cleycester, by Highcross, Leicestershire.
XII.
Ratæ Coritanorum, stipendiaria, Leicester.
XII.
Vennomentum, Ratcliff and Cosinton, on Soar river, Leicestershire.
XII.
Margidunum, Wilughby, Nottinghamshire.
XII.
Ad Pontem, Bridgford, Nottinghamshire.
VII.
 Crococolana, Colingham, Nottinghamshire.
XII.
Lindum, colonia, Lincoln.
Vernometum is sacra planities. A vast long tumulus here of an
Arch-druid. Coes is a priest; whence Cosington. Radcliff is the course
of the annual games, to his memory.

ITER XV.
From Londinium, London, by Clausentum, Southampton, to
Londinium again.
Londinium, London.
XLIV.
Calleba Atrebatum, Wallingford, Berks.
XV.
Vindonum, stipendiaria, Silchester, Hampshire.
XXI.
Venta Belgarum, stipendiaria, Winchester.
VI.
e
Ad lapidem, Mansbridge, Stoneham, Hants.
IV.
Clausentum, Southampton.
X.
Portus Magnus, Portchester.
X.
Regnum, Chichester.
X.
Ad decimum lapidem, Arundel, Sussex.
 X.
Anderida portus, Newhaven, Sussex.
XXV.
Ad Lemanum, fl. Old Romney, Kent.
X.
Lemanus portus, Lymne, Kent.
X.
Dubris, Dover.
X.
Rhutupium, colonia, Richborough, Sandwich.
X.
Regulbium, Reculver.
X.
Cantiopolis, stipendiaria, Canterbury.
Durolevum, Sittingburn, Kent.
XII.
Madum, Maidston.
XVIII.
Vagniaca, Sevenoak.
XVIII.
Noviomagus, Croydon.
XV.
Londinium Aug. London.
We here correct Antoninus in the distance between London and
Noviomagus xv. whereas in the other it is but x. Newington is a
remnant of Novantes on both sides the Thames: they first fixed at
London, called Trenovantum, being fortified by them.

ITER XVI.
From Londinium, London, to Cenia, Tregeny, Cornwall.
Londinium Aug. London.
XC.
Venta Belgarum, stipendiaria, Winchester.
XI.
Brige, Broughton, Hampshire.
VIII.
Sorbiodunum, Latio jure donata, Old Sarum.
XII.
Ventageladia, Vindocladia, Wimburn minster, Dorset.
IX.
Durnovaria, Dorchester, Dorsetshire.
XXXIII.
Muridunum, Moridunum, stipendiaria, Seaton, Devonshire.
XV.
Isca Dumnoniorum, stipendiaria, Exeter.
. . . .
. . . .
Ad Durium amnem, Ashburton, Devonshire.
. . . .
. . . .
Tamara, by Saltash, Devonshire.
. . . .
. . . .
Voluba, Fowey, Cornwall.
. . . .
. . . .
Cenia, Tregeny, Cornwall.

ITER XVII.
From Anderida, Newhaven, to Eboracum, York.
Anderida, Newhaven, Sussex.
. . . .
Noviomagus, Croydon.
XV.
Londinium Augusta, London.
XXX.
Ad Fines Trinobantes inter et Cenomanos, Roiston, Hertfordshire.
. . . .
Durolisponte, Duroliponte, Durosiponte, Godmanchester.
XXX.
Durnomagus, Latio jure donata, Caster by Peterborough.
XXX.
Corisennis, Stow green, Stanfield, Lincolnshire.
XXX.
Lindum, colonia, Lincoln.
XV.
In Medium, Kirkton in Lindsey, Lincolnshire.
XV.
Ad Abum, Wintringham, Lincolnshire.
VI.
 Pecuaria, Brough, Yorkshire.
XLVI.
Eboracum, York.

ITER XVIII.
From Eboracum, York, through the middle of the island, to
Clausentum, Southampton.
Eboracum, York.
XXI.
Legeolium, Legiolium, Castleford upon Calder, Yorkshire.
XVIII.
Ad fines, Brigantes inter et Coritanos, Gravesborough by
Rotherham, Yorkshire.
X.
. . . Chesterfield, Derbyshire.
X.
. . . Alfreton, Derbyshire.
XVI.
Derventione, Little Chester by Derby.
XII.
Ad Trivonam, Egginton upon Trent, Burton, Staffordshire.
XII.
Etocetum, Walls by Litchfield.
XVI.
Mansuedum, Manduessedum, Manceter, by Atherston,
Warwickshire.
XII.
Benonis, Cleycester by Highcross, Northamptonshire.
 XI.
Tripontium, Showel near Lutterworth, Leicestershire.
XII.
Isannaria, Towcester, Northamptonshire.
XII.
Brinavis, Banbury, Oxfordshire.
XVI.
Ælia castra, Aldcester by Biceter, Oxfordshire.
XV.
Durocina, Dorchester, Episcopi, Durinum, stipendiaria, Oxfordshire.
VI.
Tamese, Stretley on Thames, by Goreing, Berks.
XV.
Vindonum, stipendiaria, Silchester, Hants.
XLVI.
Clausentum, Southampton.
Thus we have finished this famous Itinerary, much more large than
that of Antoninus, contains many names of places not comprised
therein, and ascertains much more of the geography of Roman
Britain, of England, and Scotland: it is useful to recite an alphabetical
index of it, marking those places with an asterisc, not mentioned by
former writers, or not rightly assigned to the modern names and
places; and still leaving many to the diligence and acumen of future
writers.
* Ad Alaunam, flu. Alnwic.
* Ad Alaunam, Lancaster, Alone.
* Ad Aquas, Wells.
* Ad Alone, Abone, on Frome r.
* Ad Antonam, Evesham.
* Ad Abum, Wintringham.
* Ad Æsicam.
* Ad Decimum.
* Ad Durium amnem.
* Ad Fines, between Maxima and Flavia, Stretford on Mersey.
* Ad Fines Trinobantes inter et Cenomanos, Roiston.
* Ad Fines, Brigantes inter et Coritanos, Gravesborough by
Rotheram.
* Ad Itunam.
* Ad Lapidem, Stoneham.
* Ad Lemanum, flu. Old Romney.
* Ad Murum, Newcastle.
* Ad Montem Grampium.
Ad Pontem, Bridgford.
* Ad Sturium, Stretford street.
* Ad Selinam.
* Ad Sabrinam, Awst.
* Ad Tisam, Peirsebridge, Yorkshire.
* Ad Tuedam, flu. Berwick.
* Ad Trivonam, Burton on Trent.
* Ad Tines, Rochester on r. Tyne, Redesdale.
* Ad Tavum.
* Ad Tinam.
* Ad Uxellam amnem.
* Ad Vigesimum, Narbath C.
* Ad Vallum Antonini, Falkirk.
* Ælia Castra, Alcester by Biceter.
Agelocum, Littleburgh on Trent.
* Alauna, Alcester.
* Alata castra, Pteroton, Inverness.
* Alpes Pennini, Pendleton.
* Alicana, Shipton by Craven.
Alauna, Sterling.
* Anderida Portus, Newhaven.
Ariconium, Kenchester.
Aquæ Solis, Thermæ, colonia, Bath.

Banchorium, Bonium, Banchor.

Benonis, High cross, Clebroke.
* Bibracte, Madanhead and Bray.
Blestium, Blescium, Old castle on Escel r.
Bovium, Boverton.
Branogenium, Worcester.
* Bremenium, Bromicham, Mr. Baxter had knowledge of this town.
Bremenium, stipendiaria, Ruchester.
Brige, Braga, Broughton.
Brinavis, Branavis, Banbury.
Brocavonacis, Brovonacis, Browham.
Bullium, Burrium, Bultrum, Caerphylli C.

* Cæsaromagus, Chelmsford.
Calcaria, Tadcaster
Calleva Atrebatum, Wallingford.
Cambodunum, Latio jure donata, Alkmundbury.
Camboritum, colonia, Chesterford.
* Canonium, Kelvedon.
* Cantiopolis, Durobernum, stipendiaria, Canterbury.
Cataractonium, Cateric, Latio jure donata, Thornbury.
Cenia, Tregeny.
Clausentum, Southampton.
Conovium, Aberconwey.
Coccium, Latio jure donata, Burton by Lancaster.
* Combretonium, Bretenham.
* Corisennis, Causennis, Stow, Stanfield.
* Corium.
* Corstoplium, Corbridg.
* Condate, Northwich.
Corinium Dobunorum, Latio jure donata, Cirencester.
Crococolana, Colingham.
* Camulodunum, colonia, leg. gem. mart. xiv. Colchester.
Cunedio, Cunetio, Marlborough.

Danum, Doncaster.
Delgovitia, Wighton.
Derventio, Stanford bridge.
Derventio, Little Chester by Derby.
Deva, colonia, W. Chester, leg. xx. v. v. Cret.
Devana, Aberdeen.
Dubris, Dover.
* Durnomagus, Latio jure donata, Caster.
Durnovaria, Dorchester, Dorsetshire.
Durolevum, Durosevum, Sittinburn.
Duroprovis, stipendiaria, Rochester.
* Durolitum, Romford.
Durovernum, Cantiopolis, stipendiaria, Canterbury.
Durosiponte, Godmunchester.
Durocina, Durinum, stipendiaria, Dorchester, Episcopi,
Oxfordshire.

Eboracum, municipium, York, formerly col. leg. vi.

* Epiacum, Chester in the street.
Etocetum, Wall by Litchfield.
* Forum Dianæ, Market street.

* Gadanica, Colanica, Colecester.

Glebon, colonia, Gloucester, leg. vii. Claud.
Gobannium, Abergavenny.

* Herirus mons, by Bala.

* Hierna.

Icianis, Ixworth.
* In medio.
* In medium, Kirkton, Lindsey, Lincolnshire.
* Isannavaria, Towcester.
Isca Dumnoniorum, stipendiaria, Exeter.
Isca Silurum, colon. leg. ii. Aug. Caerleon.
Isurium, Aldborough.

Lactorodum, Stony Stratford.

Lataris, Bowes.
Legiolium, Casterford.
Lemanus Portus, Lymne.
Leucarium, Loghor.
Lindum, colonia, Lincoln.
Lindum in Scotland.
Londinium, colonia, Aug. London.
Luguvalia, Latio jure donata, Carlisle.

Madum, Madeston.
Magiovinium, Dunstable.
Magna, Old Radnor.
Manduessedum, Mancester.
 Mancunium, Mancastle.
* Margidunum, Wilughby.
Mediolanum, Myvod.
* Mediolanum, Chesterton by Newcastle.
* Menapia, St. David’s.
Muridunum, stipendiaria, Seaton.

Nidum, Neath.
* Noviomagus, Croydon.

* Orrea, Dunkeld.

Pecuaria, Brough.
Pennocrucium, Penkridge.
Portus Magnus, Portchester.
* Portus Sistuntiorum, Lune river mouth.
Præsidium, Warwick.
Prætuarium, Patrinton.
* Pteretone, Latio jure donata, Inverness.

Ratæ Coritanorum, stipendiaria, Leicester.

Regulbium, Reculver.
* Regnum, Chichester.
* Rerigonium, Ribchester.
Rhutupis, colonia, Sandwich.
Rutunium, Rowton.

* Salinis, Droitwich.
Segontium, stipendiaria, Caernarvon.
Sitomagus, Thetford.
Sorbiodunum, Latio jure donata, Old Sarum.