i

PRAISE FOR THE FIRST EDITION

‘Critical Thinking:The Basics covers the basics engagingly and without

unnecessary jargon, provides lots of well-chosen examples and prac-
tice exercises, and, unlike most critical thinking textbooks, explains
the theoretical background to the recent rise of critical thinking in
education.’
Justine Kingsbury, University of Waikato,
New Zealand

‘This is an excellent contribution to the growing literature on criti-

cal thinking.The discussion on causal reasoning, heuristics and biases,
framing, social power, metacognition, and self-deception are timely
and valuable. Critical Thinking: The Basics helps to bring together the
literature on the psychology and sociology of human judgement
with the philosophical art of argument analysis.’
Ted Poston, University of South Alabama,
USA

‘This text is loaded with references, explanations, history and exam-

ples from the development and enhancement of the understanding
of critical thinking. It offers a coverage of the basics of critical think-
ing and also provides a sense of how the field and individual expecta-
tions of critical thinking are changing.’
Charles Blatz, Professor Emeritus at
University of Toledo, USA.
ii
iii

CRITICAL THINKING

THE BASICS

Critical Thinking: The Basics is an accessible and engaging i­ntroduction to

the field of critical thinking, drawing on philosophy, c­ ommunication, and
psychology. Emphasising its relevance both to academic literacy and to deci-
sion-making in a range of contexts, this book introduces and explains the
knowledge, methods, and skills needed to identify and avoid poor reasoning,
reconstruct and evaluate arguments, and engage constructively in dialogues.
Topics covered include:
• The relationship between critical thinking, emotions, and the psychology
of persuasion
• The role of character dispositions such as open-mindedness, courage, and
self-knowledge
• Argument identification and reconstruction
• Fallacies and argument evaluation.
This second edition has been revised and updated throughout, and includes
an additional chapter on the relationship between critical thinking and
emotions. There are also new sections on concepts such as the ‘experiential
mind’ and ‘need for cognitive closure’, and contemporary examples drawn
from issues including conspiracy theories, the pandemic, and misuses of
social media.
With updated and expanded discussion questions/exercises and sugges-
tions for further reading at the end of each main chapter, this book is an
essential read for students approaching the field of critical thinking for the
first time, and for the general reader wanting to improve their thinking skills
and decision-making abilities.

Stuart Hanscomb is Senior Lecturer in Philosophy and Communication

at the University of Glasgow, UK.With Benjamin Franks and Sean Johnston
he is author of Environmental Ethics and Behavioural Change, also available
from Routledge.
iv

THE BASICS
The Basics is a highly successful series of accessible guidebooks
which provide an overview of the fundamental principles of a sub-
ject area in a jargon-free and undaunting format.
Intended for students approaching a subject for the first time,
the books both introduce the essentials of a subject and provide
an ideal springboard for further study. With over 50 titles span-
ning subjects from Artificial Intelligence to Women’s Studies, The
Basics are an ideal starting point for students seeking to under-
stand a subject area.
Each text comes with recommendations for further study and
gradually introduces the complexities and nuances within a subject.
AMERICAN PHILOSOPHY BIOETHICS (SECOND EDITION)
NANCY STANLICK ALASTAIR V. CAMPBELL

ANIMAL ETHICS EASTERN PHILOSOPHY (SECOND EDITION)

TONY MILLIGAN VICTORIA HARRISON

ARTIFICIAL INTELLIGENCE PHENOMENOLOGY

KEVIN WARWICK DAN ZAHAVI

EVOLUTION ATHEISM
SHERRIE LYONS GRAHAM OPPY

FOOD ETHICS EMOTION

RONALD SANDLER MICHAEL BRADY

PHILOSOPHY (FIFTH EDITION) PHILOSOPHY OF MIND

NIGEL WARBURTON AMY KIND

GLOBAL JUSTICE METAPHYSICS (SECOND EDITION)

CARL DEATH AND HUW L. WILLIAMS MICHAEL REA

HUMAN GENETICS (SECOND EDITION) FREE WILL (SECOND EDITION)

RICKI LEWIS MEGHAN GRIFFITH

LOGIC (SECOND EDITION) CRITICAL THINKING (SECOND EDITION)

J.C. BEALL STUART HANSCOMB

Other titles in the series can be found at: https://www.routledge.

com/The-Basics/book-series/B
v

CRITICAL THINKING

THE BASICS

Second Edition

Stuart Hanscomb
vi

Cover image: © Tim Ewing, DeviantArt.com/timemit

Second edition published 2023
by Routledge
4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
605 Third Avenue, New York, NY 10158
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2023 Stuart Hanscomb
The right of Stuart Hanscomb to be identified as authors of this work has been
asserted by them in accordance with sections 77 and 78 of the Copyright, Designs
and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised in
any form or by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without
intent to infringe.
First edition published by Routledge, 2017
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Hanscomb, Stuart, author.
Title: Critical thinking : the basics / Stuart Hanscomb.
Description: Second edition. | New York, NY : Routledge, 2023. |
Series: The basics | Includes bibliographical references and index.
Identifiers: LCCN 2022060892 (print) | LCCN 2022060893 (ebook) |
ISBN 9781032163017 (hardback) | ISBN 9781032162997 (paperback) |
ISBN 9781003247944 (ebook)
Subjects: LCSH: Critical thinking.
Classification: LCC B105.T54 H363 2023 (print) | LCC B105.T54 (ebook) |
DDC 160—dc23/eng/20230320
LC record available at https://lccn.loc.gov/2022060892
LC ebook record available at https://lccn.loc.gov/2022060893
ISBN: 978-1-032-16301-7 (hbk)
ISBN: 978-1-032-16299-7 (pbk)
ISBN: 978-1-003-24794-4 (ebk)
DOI: 10.4324/9781003247944
Typeset in Bembo
by codeMantra
vii

CONTENTS

List of Boxes xi
Preface and Acknowledgements xiii

Introduction 1
0.1 What is critical thinking? 3
0.2 Critical thinking and the spirit of philosophy 6
0.3 The historical and academic context of critical
thinking as a discipline 12
0.4 Alternative approaches 16
0.5 Critical thinking and ethics 17
0.6 This book’s approach (1): Overview of contents 19
0.7 This book’s approach (2): The use of examples 20
Further reading 23

1 Rationality and cognitive biases 25

1.1 Rationality 25
1.2 Heuristics and biases 30
1.3 Framing 46
1.4 Conclusion: Argument and rhetoric 48
Exercises 49
Further reading 49
viii

viii CONTENTS

2 Critical thinking and emotions 51

2.1 The nature of emotions 52
2.2 Emotion and rationality 54
2.3 Emotion and reflection 62
2.4 Arguments appealing to emotion 64
2.5 Emotion and dialogue 68
2.6 Conclusion 69
Exercises 70
Further reading 71

3 Critical thinking and dispositions 72

3.1 Some core critical thinking dispositions 74
3.2 Guidelines for constructive dialogues 90
3.3 Conclusion 92
Exercises 92
Further reading 93

4 Arguments and argument reconstruction 95

4.1 The structure of arguments 96
4.2 Propositions (statements) 97
4.3 Argument reconstruction 98
4.4 Summary 112
Discussion question 113
Exercises 113
Further reading 114

5 Argument forms and fallacies 116

5.1 Deductive, inductive, and plausible arguments 116
5.2 Argument forms and fallacies 121
5.3 Critical questions 123
5.4 Summary, and the wider analysis of argument forms 131
Exercises 131
Further reading 132

6 Arguments, social power, and message source (Part 1) 133

6.1 Arguments from expert authority 137
6.2 Arguments from position to know (information power) 145
6.3 Legitimate, reward, and coercive power 153
6.4 Summary 163
Exercises 163
Further reading 163
ix

CONTENTS ix

7 Arguments, social power, and message source (Part 2) 165

7.1 Arguments appealing to character 165
7.2 Appeals to referent power and ad populum arguments 175
7.3 Summary 181
Exercises 181
Further reading 182

8 Causal arguments, generalisations, arguments from

consequences, and slippery slope arguments 184
8.1 Causal arguments and causal fallacies 185
8.2 Generalisations 201
8.3 Arguments from consequences and slippery slope
­arguments 210
8.4 Summary 218
Exercises 218
Further reading 219

9 Arguments from analogy 221

Exercises 232
Further reading 232

10 Further fallacies 234

10.1 Affirming the consequent (and denying the antecedent) 234
10.2 Circular arguments 237
10.3 False dilemma 239
10.4 Perfectionist fallacy 241
10.5 Red herrings and equivocations 243
10.6 Summary 244
Exercises 244
Further reading 245

Conclusion 246

Glossary 251
References 261
Index 267
ii
xi

BOXES

0.1 What is an argument 4

0.2 Definition of critical thinking 23
5.1 Basic critical questions 125
5.2 Necessary and sufficient conditions 126
5.3 Basic critical questions and some
associated sub-questions 126
6.1 Forms of social power 135
ii
xiii

PREFACE AND
ACKNOWLEDGEMENTS

The second edition of Critical Thinking: The Basics contains a new

chapter (on emotions and rationality), several smaller additions (such
as a developed discussion of dual process theory and the inclusion of
need for cognitive closure), numerous new examples and exercises,
and updated references and recommended readings.
For comments and suggestions, motivation, and general support,
contributing to one or both editions of this book, my thanks go
to Benjamin Franks, Martin Connor, Julia Weckend, Paul Hartog;
several anonymous reviewers; University of Glasgow’s T&C, A-R-T,
and CTC students; Kristen Layne, Kate Shenton-Ross, Hui Wang;
Tony Bruce and Adam Johnson at Routledge; Carolyn Dodds; and
students and staff at NHL Stenden in Leeuwarden and the Professors’
Association for Students’ Services (PASS) in Pakistan, for discussions
about drafts of the emotions chapter.
And to Tim Ewing (at DeviantArt.com/timemit) for the cover
image.
ii
1

INTRODUCTION

My husband says I’m argumentative. He’s wrong though, and here are
three reasons why … (Sacha T. Burnstorm, pers. comm.)

Eating breakfast in the kitchen, I glance up from some sobering read-

ing and my eye is caught by a sticky note. Scribbled on it is a scene
from the living room: ‘Please stop chewing the remote control’ says
Mum. ‘It’s not my fault, it’s a habit!’ shouts back our youngest. Not
bad for an 8-year-old, and hopefully a line he’s picked up from Bart
Simpson or Sponge Bob rather than me.
There are worse habits. The article I’m reading is about risks posed
by misinformation, bullying, addiction, and the incessant exposure of
harmful content to young people via social media. The world is still
trying to make sense of this phenomenon, and probably will be for
some while. It is hard to find comparisons for the multilayered moral
and legal issues it raises, but with varying degrees of success many are
trying. There are analogies with how risk of addiction is increased by
slot machines intricately designed to keep people playing; the argu-
ment that the liability of social media companies should be in line
with newspapers and manufacturers; evidence of cover-ups reminis-
cent of cigarette makers in the 1960s, and the prediction that it will

DOI: 10.4324/9781003247944-1
2

2 INTRODUCTION

be parents that force further legislation on the issue as Mothers against

Drunk Driving did in the 1980s. There is a very long way to go with
this debate, and arguments will continue to be presented to us, or gen-
erated by us, that are more or less strong, more or less well motivated.
I scribble a lot of things on sticky notes, scraps of paper, the kitchen
wall – exchanges like the one above, information from podcasts, titles
of songs from the radio. Sooner or later though I need to transfer
them to somewhere they won’t be lost. If a few have accumulated
this is a chore, and in a small but significant way I’m conflicted. The
right thing to do is to push through and make the effort, but an alter-
native is to convince myself, in that moment, that these fragments are
not worth saving. Laziness, armed with self-deception, takes on truth.
It’s an example of what’s known as cognitive dissonance (something
I’ll say more about in the next chapter). I’m fairly wise to it now, but
it has led to some poor decisions, and will again if I’m not vigilant.
Poor reasoning and the absence of reasoning on occasions where
it would serve us well is commonplace. Life is typically fast-paced
and mistakes will happen, but even when things are slowed down
(for example, carefully drafting an email rather than a spontaneous
dialogue) we think erroneously in predictable ways.
Adding to our vulnerability is that bad arguments are often
­persuasive – entertaining even – impeding our ability and motiva-
tion to put them in their place. Professional persuaders (working in,
say, politics or marketing) know two important things:

1. That our critical thinking capabilities are not what they

might be.
2. That certain forms of persuasive communication make us less
likely to pay attention to, or even look for, poor reasoning, and
are therefore more likely to win us round to their point of
view.

In many contexts, influencing one another through arguments based

on dubious, partial, or irrelevant claims is usually not deliberate, but
whether intended or not, poor arguments can be remarkably effec-
tive if they encourage our cognitive and emotional biases.
3

INTRODUCTION 3

0.1 WHAT IS CRITICAL THINKING?

This is a book about how to avoid reaching the wrong conclusion
in everyday, professional, and academic contexts. The fundamental
subject matter of critical thinking is the reasoning we apply in a wide
variety of circumstances, and its aims are twofold:

1. To improve our ability to reason and generate strong arguments.

2. To improve our ability to assess the strength of the arguments
used by others.

Since we should assess our own arguments by the same standards we

use to assess the arguments of others, then these aims are very closely
aligned. Also, a substantial part of our overall argument on an issue is
an assessment of the arguments of others. As the nineteenth-century
philosopher John Stuart Mill put it:
When we turn to … morals, religion, politics, social relations, and the
business of life, three-fourths of the arguments for every disputed opin-
ion consist in dispelling the appearances which favour some opinion dif-
ferent from it. (1962, p.163)

ARGUMENTS

Although later chapters will provide more technical and detailed

information about the structure and components of arguments, the
basic concept of an argument is quite straightforward. There is little
disagreement among academics as to what they are, and below are
some examples of definitions:
An argument is an attempt to prove or establish a conclusion. It has two
major parts: a conclusion and the reason or reasons offered in support of
the conclusion. (Ennis, 1996a, p.2)
An argument consists of one or more statements that are used to pro-
vide support for a conclusion. (Halpern, 2003, p.183)
By ‘argument’ we mean a claim, together with one or more sets of
reasons offered by someone to support that claim. (Johnson and Blair,
2006, p.10)
4

4 INTRODUCTION

BOX 0.1 WHAT IS AN ARGUMENT?

These definitions state that an argument is comprised of:
1. A claim being asserted that we want other people to believe is
true.
2. Reasons offered in support of this claim, through which we try to
rationally convince other people that this claim is true.

Here is your initial piece of critical thinking terminology. While the

claim being asserted is simply referred to in critical thinking as a
conclusion, a reason offered in support of a conclusion is known
as a premise. Someone might assert that we should not eat meat.
We ask for their reasons and they say that it is better for our health,
for the environment, and is less cruel to animals. Whether or not we
think this is convincing, they have unquestionably presented us with
an argument. That we should not eat meat is the conclusion, and the
health, environmental, and animal welfare benefits are the premises.
This meaning of argument, then, does not refer to bad-tempered
disagreements or rows. If we are having a quarrel with someone, then
arguments (in the sense of conclusions with supporting reasons) will
be put forward or implied, but the two uses of the word are clearly
very different.
The context in which arguments are put forward is, however, one
of disagreement. The reason for offering an argument is usually to
provide the other person with reasons for believing something which
you want them to believe, but which they do not currently believe.
Not all conversations involve arguments, but many do. We talk to
each other for lots of reasons – to entertain and be entertained, to
inform and find out information, to offer and seek explanations – but
an important purpose of conversation is to express our view on an
issue. Often when we do this the other person will ask why we hold
this view, and our response will usually take the form of an argument.

DEFINING CRITICAL THINKING

Central to critical thinking is the process of identifying what the

argument is that is being put forward, and determining whether
5

INTRODUCTION 5

the premises justify accepting the conclusion. The broader aim is to

improve our deliberations and, ultimately, our decision-making. In
this vein Robert Ennis, author of one of the most influential text-
books on the subject, defines critical thinking as:
Reasonable reflective thinking focused on deciding what to believe or do.
(1996b, p.166)

Ennis’s reference to ‘doing’ as well as ‘believing’ underlines how it is

very much an applied area of study; it intends to teach knowledge
and skills that have a clear practical application.
American philosopher John Dewey is often credited as the origi-
nator of critical thinking as a field of study. In his book, How We
Think, Dewey defines what he calls ‘reflective thinking’ as:
Active, persistent, and careful consideration of a belief … in the light of
the grounds which support it... (1910, p.6)

‘In some cases,’ he says,

a belief is accepted with slight or almost no attempt to state the grounds
that support it. In other cases, the ground or basis for a belief is delib-
erately sought and its adequacy to support the belief is examined. This
process is called reflective thought; it alone is truly educative in value.
(ibid, p.2)

Dewey’s choice of words – ‘active’, ‘persistent’, ‘careful’ – indicate the

emphasis critical thinking places on a responsible and effortful mar-
shalling of our thoughts. Instead of passively accepting what appears
right on the surface, or what we have always believed, we actively
consider whether something should be believed or not. The ‘stream
or flow’ of ideas that pass through our mind becomes a ‘chain, or
thread’ (ibid, p.3) – we impose on them an order that allows us to
assess whether they should be believed or acted upon.
How, then, do we bring order to our unruly thoughts and become
better critical thinkers? There are three aspects to this:

1. Theoretical knowledge about arguments, rationality, and all of

the other elements covered in books like this one.
2. Practise – a lot of practise – in applying this learning to examples
of arguments and exchanges between people that involve argu-
ments and counterarguments.
6

6 INTRODUCTION

3. Self-reflection on how we form our beliefs and how we inter-

act with others.

Advice with respect to how this book can facilitate the second of these
can be found at the end of this chapter (and I would urge you to read
this section carefully, since to get the most from this subject it is vital to
combine what textbooks offer with external experiences and materials).
With respect to the third point, although critical thinking is about
avoiding errors in our reasoning and assessing the quality of argu-
ments, in a very important sense it is also about us. It is about us as
human beings, vulnerable to biases in our thinking, to moods and
emotions that can cloud our judgement, and to character dispositions
that entrench these tendencies. And it is about us as individuals with
a desire to know more about our strengths and weaknesses and how
to eliminate, mitigate, and improve upon them.
This book, then, is about how we can improve our thinking so that
we become better disposed to making good judgements, and it is also
therefore about self-knowledge: greater awareness of the aspects of
psychology and character that are relevant to constructive delibera-
tion and improved decision-making. Put another way, critical thinking
can contribute to self-improvement, and knowing about the self can
help improve critical thinking. This book considers the importance
of critical thinking as part of the wider question of how to live well.

0.2 CRITICAL THINKING AND THE SPIRIT

OF PHILOSOPHY
Critical thinking’s historical origins can be identified in two founda-
tional features of Western philosophy: (1) commitment to truth (even
in the face of social and political pressures to remain ignorant); and
(2) the individual’s development of virtues associated with wisdom
and sound judgement (with self-knowledge among the most promi-
nent). In what is often taken to be the seminal work of modern phi-
losophy, René Descartes opens his Meditations (originally published
in 1641) with the following resolution:
It is some time ago now since I perceived that, from my earliest years,
I had accepted many false opinions as being true, and that what I had
7

INTRODUCTION 7

since based on such insecure principles could only be most doubtful and
uncertain; so that I had undertaken seriously once in my life to rid myself
of all the opinions I had adopted up to then, and to begin afresh from the
foundations. (1968, p.95)

Descartes is expressing not just a desire for truth, but recognition that
the search for it must involve self-examination. In the eighteenth
century, Enlightenment thinking championed reason over supersti-
tion and tradition in politics and ethics as well as science. Reason was
regarded not just as the route to the truth, but as a virtue that benefits
the individual and society. Through reason comes progress, and to be
willing to apply rational thinking to all issues is to take responsibil-
ity for one’s destiny; to grow up. In his 1784 essay An Answer to the
Question:What is Enlightenment? Immanuel Kant emphasised freedom
from ‘self-incurrent tutelage’ – the ‘resolution and courage’ to think
for ourselves ([1784] 1963, p.3).
These are examples of great philosophers and scientists challeng-
ing established knowledge, such as religious dogma and ancient sci-
entific assumptions. Most of us are not going to be the authors of
world-changing theories, but the lesson from philosophy for most
individuals concerns the attitude we take to our own lives and the
issues that define them. Basic to this is the importance of thinking
through our beliefs rigorously, testing them out where possible, and
taking responsibility for them. In this vein, the nineteenth-century phi-
losopher Søren Kierkegaard observed in The Present Age:
There are handbooks on everything, and generally speaking education
will soon consist of knowing letter-perfect a larger or smaller compen-
dium of observations from such handbooks, and one will excel in propor-
tion to his skill in pulling out the particular one, just as the typesetter
picks out letters. ([1846] 1979, p.104)

His complaint concerned how, in an age proud to define itself as

scientifically and socially progressive, individuals come to see them-
selves as embodying this progress without themselves needing to put
the work in; without, as Cardinal Newman put it a few years later in
The Idea of a University ([1852] 1982, p.101) ‘making the objects of
our knowledge subjectively our own’. This is a product of concen-
trated thought about aspects of our world, whether novel or taken
for granted; and it is to understand the value of questions like: ‘What
8

8 INTRODUCTION

exactly does this claim mean?’, What can we infer from it?’, ‘How do
I know it is the truth?’, ‘Why is it important to know about?’, and
‘What difference might knowing about it make to my life?’ While
critical thinking is not proposing the impossible task of understand-
ing everything, it is proposing that there are dispositions or attitudes
we can develop that make us less susceptible to error; that enable us
to ask the right questions and, perhaps most importantly, to have a
reflective awareness of what we, as individuals, do and do not know.
Features of what is now called critical thinking are foundational to
the mood and practice of modern philosophy, but the prototypical
critical thinker is the ancient Athenian, Socrates (469–399 BCE). He
did not write anything, but his ideas, personality, and philosophi-
cal approach, as presented in Plato’s dialogues, are hugely influential.
Socrates (like his pupil Plato, and Plato’s pupil Aristotle) was con-
cerned about the level of ignorance he encountered in those around
him, including teachers and statesmen. More importantly, he was
exercised by people’s ignorance of their own ignorance. For example,
in the Apology, after engaging a politician with a reputation for wis-
dom in dialogue he concludes:
It is only too likely that neither of us has any knowledge to boast of; but
he thinks he knows something that he does not know, whereas I am quite
conscious of my ignorance. … I am wiser than he is to this small extent,
that I do not think that I know what I do not know. (Plato, 1993, p.42)

He then reports on speaking with a poet and a craftsman, and in

both cases finds that because they know about their particular art
or skill, they then assume that they ‘have a perfect understanding of
every other subject’ as well (ibid, p.43). This is an unwarranted gen-
eralisation which implies a lack of appropriate humility, or at least
self-knowledge.
For Socrates, this attitude was created and indulged by improper
education; notably an overemphasis on the practice of rhetoric.
Through being effective in various forums (such as politics and law),
rhetoric had tended to relegate and obscure both the nature and
value of truth, and the frames of mind needed to attain it. Rhetoric
(or oratory) in ancient Greece was the art of persuasive public speak-
ing, and can be more broadly defined now as the art of persuasive
communication. Socrates viewed it as the ‘knack’ of winning over
9

INTRODUCTION 9

uninformed audiences by ‘pandering’ to them; presenting versions of

reality which are superficially pleasing at the cost of true understand-
ing. In the Gorgias he compared rhetoric to cookery, which,
puts on the mask of medicine and pretends to know what foods are best
for the body, and, if a doctor or a cook had to compete before an audience
of children … with the job of deciding which of them is the better judge of
wholesome and unwholesome foodstuffs, the doctor would unquestion-
ably die of hunger. (Plato, 2004, p.32)

In this analogy the people of Athens are the children (a prelude to

Kant’s ‘tutelage’ metaphor), the orator is the confectioner, and the
philosopher is the doctor. The weakness in the comparison is that,
unlike children, the people of Athens should know better; they are, to
an extent, allowing themselves to be taken in by the more palatable
illusions of the speechmakers. Prominent among these illusions is the
belief that their opinions are the right ones, especially if supported by
those who deliver a version of them with confidence and eloquence.
Attaining true wisdom (rather than the superficial appearance of
it) requires the disquieting knowledge of our ignorance and a com-
mitment to ‘examining life’ that is hard, if ultimately rewarding. For
Socrates, being able to think independently and competently is a
fundamental component of living well and being happy, and he saw
himself as a living reminder of this fact. Famously he gave the state
little option but to kill him for his troubles, but in defence of himself
he argued in the Apology ‘if you put me to death, you will not easily
find anyone to take my place’. He depicted his city as
a large thoroughbred horse which because of its great size is inclined to
be lazy and needs the stimulation of some stinging fly; and all day long
I never cease to settle here, there, and everywhere, rousing, persuading,
reproving every one of you. (1993, p.54)

We do not need to be agitators in this way to be critical thinkers, but

the horse and fly metaphor can also represent our attitude to ourselves –
we need to be our own horse fly. Critical thinking is a form of ‘rousing’,
helping to improve our error-prone habits of thought and action.
The Socratic method is dialogue: the exchange of views – of expla-
nations and arguments – between two (or more) people. A lot will
be said about the nature and importance of dialogues in Chapter 3,
10

10 INTRODUCTION

but aspects of its significance for Socrates will be briefly discussed

here. A dialogue can involve people (often just two) with opposing
views trying to convince each other of the rightness of their particu-
lar view; or it can be an attempt to persuade an uncommitted person.
In Socrates’s dialogues we usually find him questioning someone who
believes they know something with the aim of testing the firmness of
the grounds of that belief.Whilst Descartes in his Meditations engaged
in a kind of internal dialogue to establish what he should and should
not accept as true, the Socratic method is interpersonal.
Socrates also saw himself as a ‘midwife’, assisting the person,
through philosophical discussion, to realise the status of their beliefs.
The vital point of the midwifery metaphor is that the individual is
not handed the truth (for example, as something to be learned by
rote), but is encouraged to reason their way to it. Most of us will
know from experience what this kind of clarity feels like: we remind
ourselves of why we know what we know, or alternatively we dis-
cover that we do not know what we thought we knew, and through
reasoning come to appreciate our ignorance. It must be said that read-
ing many of Plato’s dialogues can leave us with the impression that
Socrates did not always help people achieve this, but the principle
stands and remains basic to many educational practices. Rather than
passive absorption, dialogue is active, and rather than dependence on
the wisdom of another, it emphasises independence of thought. As we
have seen, these are basic to Dewey’s definition of critical thinking.
There is an asymmetry in many of Socrates’ dialogues – he the
sceptical enquirer, the other the person claiming to know something.
Teachers and peers can play this role, but very often dialogues are sym-
metrical, with both parties having diverging convictions about the mat-
ter at hand. Under these conditions the risk is that competition and
emotion interfere with truth-seeking.That aside, holding a constructive
dialogue is not simply a matter of common sense. For these reasons, as
we will see in the next section, strains of critical thinking have placed
great emphasis on understanding and promoting the art of dialogue.
The spirit of philosophy, as exemplified by Socrates, can be sum-
marised in terms of four archetypes:
• The critical thinker: questioning assumptions about the way the
world is.
11

INTRODUCTION 11

• The seeker: of ultimate truths (even if this is the realisation that

there can be no such truths), and of happiness or an otherwise
fulfilling life.
• The reflector: self-understanding with respect to what one wants,
what one wants to want, and the tendencies we have that impede
our chances of achieving these insights.
• The rebel/agitator: challenging the assumptions and practices of
others with the aim of improving society.
Socrates was all of these things, and other philosophers will embody
all or some of them. The perspective of this book is that critical
thinking as a discipline should not be isolated from these other char-
acteristics.Whilst it is not my view that that being a critical thinker is
the sole component of a fulfilling life, it is nevertheless vitally impor-
tant to that end. As already highlighted, self-understanding will be
improved by the approach I am taking, and other virtues seen as basic
to human fulfilment are often associated with effective critical think-
ing. The ethical and democratic value of being a critical thinker will
be explained soon and also addressed throughout the book.
Before moving on to outline modern approaches to critical think-
ing, I will say something about the status of rhetoric in the later
dialogues of Plato, and in the writings of Plato’s pupil, Aristotle.They
came to recognise that rhetoric is something to be studied – an art
or skill (technê) – and not merely natural eloquence enhanced by rote
learning and imitation. Part of learning rhetoric is to understand the
ways in which people are susceptible to persuasion – what we now
refer to as the psychology of persuasion or the psychology of
influence. Many of Aristotle’s insights in The Art of Rhetoric are sup-
ported by modern research, including how emotions function to bias
judgements, the persuasive power of examples and metaphors, and
the importance of the speaker’s authority and likeability (or at least
the appearance of these things). From the critical thinker’s perspec-
tive, an appreciation of the ways in which we can be unconsciously
influenced matters (1) because it allows us to be on our guard against
these processes, and (2) because of how these influences are able to
explain some of the errors we make in our judgements.
As we shall learn about in Chapter 2, it is now well known in psy-
chology and communication that arguments alone are rarely enough
12

12 INTRODUCTION

to make us change our behaviours. Plato and Aristotle knew this

too. Aristotle emphasised the value of rhetoric as compensating for
the limited intellect or attention span of audiences (1991, pp.75–6),
and in the Phaedrus Plato recognises that knowledge of psychology
(‘the nature of the soul’) has a part to play in changing minds. Both
maintained, however, that the most persuasive speakers are those who
also have some depth of knowledge of the topic they are speaking
about. Moreover, if the philosopher is going to offer any degree of
approval of the practice of rhetoric, consideration must be given to
the ends to which it is put: the improvement of society or merely the
self-advancement of the speaker. In short, rhetoric is a topic worthy
of the attention of philosophers.

0.3 THE HISTORICAL AND ACADEMIC CONTEXT

OF CRITICAL THINKING AS A DISCIPLINE
As we have seen, the spirit of philosophy and the spirit of critical
thinking are closely related, but the analysis of arguments and reason-
ing is specific to the branch of philosophy known as logic. Logic is
primarily interested in what is called ‘validity’ (see Chapter 5 for a
discussion of this concept), or the rules that govern our assessment
of inferences. An inference is the step we make in reaching a conclu-
sion based on certain premises. A simple example would be inferring
that someone is married because they wear a wedding ring or, with
greater certainty, that an object is considered (by some) to be a work
of art because it is on display in an art gallery.
If logic in general is the study of arguments and reasoning, infor-
mal logic is the study of arguments and reasoning as employed in
real-life contexts, rather than in abstraction. It is interested in the
kinds of arguments people use in their professional and daily lives,
how effective these are, the mistakes that are made, and how we can
avoid making these mistakes. To an extent it is possible to repre-
sent these arguments in schematic (abstract) forms, but a more com-
plete understanding of them requires contextual knowledge about
the topic under discussion, the audience the argument is directed
at, and certain characteristics of the person presenting the argument.
It is these considerations that informal logic seeks to make explicit,
and in so doing it offers a systematic approach to understanding
13

INTRODUCTION 13

and evaluating everyday arguments. This involves skills concerning

comprehension and interpretation – working out what the person’s
overall point is and the reasons they are providing (or implying) in
support of it – and of assessing the strength of the inference. Both
comprehension and assessment involve an understanding of typi-
cal forms of reasoning and typical mistakes in reasoning (fallacies);
an understanding that is enriched by a range of empirical consid-
erations, most notably the psychology of cognitive biases and their
motivational underpinnings.
It was traditional logic’s disinterest in this wider context that led to
informal logic asserting itself as a highly distinctive sub-discipline in
the mid-to-late twentieth century. Early examples of this paradigm
are Stephen Toulmin’s The Uses of Argument (1958), C.L. Hamblin’s
Fallacies (1970), and Michael Scriven’s Reasoning (1976). Of greatest
influence, however, have been Ralph Johnson and Anthony Blair,
whose textbook Logical Self Defense was first published in 1977, and
who launched the first journal devoted to the discipline – Informal
Logic Newsletter – in 1978.1
Also in the latter part of the twentieth century the more explicitly
interdisciplinary field of argumentation developed. Of particular
importance was the attention it paid to dialogues: their categori-
sation and an analysis of the norms and rules that should govern
them if they are to be constructive. Here the work of philosopher
Douglas Walton and the theory of pragma-dialectics (developed
at the University of Amsterdam by Frans van Eemeren and Rob
Grootendoorst in the 1980s) have been influential in their analyses
of the components and rules of what they term ‘critical discussions’ –
dialogues in which participants with differing views on an issue pre-
sent arguments to one another in the hope of reaching a resolution.
These join a number of theories which emphasise the interpersonal
and wider social aspects of argumentation: ‘rhetorical argumentation’
(Perelman and Olbrechts-Tyteca, 1969), ‘feminist argumentation’
(Nye, 1990), ‘coalescent argumentation’ (Gilbert, 1997), and ‘coop-
erative argumentation’ (Makan and Marty, 2001).These approaches
to argument and reasoning have closely aligned informal logic with
rhetoric and applied communication, and it is noteworthy that many
have sought to create alternatives to competitive (‘adversarial’) mod-
els of dialogue.
14

14 INTRODUCTION

More recently there has been growing interest in the links between
critical thinking and psychology, and in particular the now extensive
and quite well-understood range of biases that can affect judgement
and decision-making. If there are identifiable kinds of widespread
mistakes in reasoning that are consistently revealed in informal
­argumentation – for instance seeing causal relationships where they
do not exist, making hasty generalisations, or determining the quality
of an argument by reference to irrelevant attributes of the arguer –
then it is reasonable to expect that there are underlying cognitive
biases that predispose us to make these errors. Researchers in the
fields of philosophy and psychology have tended to follow parallel
paths in their investigations of reasoning, but convergence is now
happening that is substantially enriching the interdisciplinary field of
critical thinking (e.g., Halpern, 2014).
The motivation of critical thinking scholars is to make logic use-
ful and have direct, practical application to daily life so that someone
who took a course in it would not only be better at logic, but a better
deliberator on worldly affairs, and show improvement in general aca-
demic abilities. Similar desires inspired the education-oriented ‘criti-
cal thinking movement’ in the US in the 1980s. Prime movers such
as Richard Paul and Robert Ennis (who were also instrumental in
developing informal logic) were critical of the standards and standing
of critical thinking in American school curricula and sought, with
some success, to raise its profile and generate materials and tech-
niques for its teaching. ‘Our overemphasis on rote memorization and
recall of facts’ says Richard Paul:
does not serve us well. We must exchange our traditional picture of knowl-
edge and learning for one that generates and rewards active, independent,
self-directed learning so that students can gather and assess data rigor-
ously and critically. We need to abandon methods that make students pas-
sive recipients of information and adopt those that transform them into
active participants in their own intellectual growth. (1995, p.45)

Along with significant emphasis on its role in a healthy ­democracy,

Paul stresses the importance of critical thinking dispositions.
‘Weak’ approaches to critical thinking teach only the skills of argument
analysis, but a ‘strong’ approach seeks to foster habits of thinking and
dispositions (character traits) – such as reduced e­ gocentricity – that
have a deeper, more holistic effect on the individual’s development.
15

INTRODUCTION 15

To use higher education scholar Ronald Barnett’s expression (1997),

students become ‘critical beings’ rather than just people able to argue.
Paul (1984a) also sees dialogical context – arguments encountered
and analysed along with counterarguments in unfolding debates
and discussions – as crucial for nurturing this ‘strong’ conception of
critical thinking. Having one’s beliefs subject to open-minded and
respectful questioning, along with a critical stance towards the posi-
tions of others, encourages personal insight and responsibility. It is
part of what Dewey called a ‘community of inquiry’, and ‘schools
that model themselves on such a community,’ says educationist
Deanna Kuhn,
foster not just the acquisition of knowledge, but the acquisition of reason
and judgement – the … [essential condition] for participation in a democratic
society, as well as for realization of a fulfilled individual life. (1991, p.298)

In The Skills of Argument (1991) – an empirical investigation into

people’s explanations of various social phenomena (such as failure
at school) – Deanna Kuhn theorises increasingly sophisticated ways
of understanding the status of knowledge and how it is achieved.
‘Absolutist’ thinkers presume that there are clear answers agreed
upon by the experts, while ‘multiplist’ thinkers recognise divergent
views but assume there is no way of reconciling them. Faced with
this barrier, they are liable to fall back on whatever first-hand experi-
ence they have for explaining the phenomenon in question because,
according to this perspective, what you believe in the end comes
down to a matter of subjective opinion. ‘Evaluative’ thinking, on the
other hand, sees how progress can be made by assessing divergent
opinions towards a conclusion based on the relative strengths of
competing ideas. This perspective is correlated with critical think-
ing, and it is easy enough to see why. If you think that knowledge is
simply waiting to be discovered by those with the skills or experi-
ence, or that it is just a matter of opinion and that multiple opinions
have ‘equal legitimacy’ (ibid, p.184), then argumentation is not likely
to be seen as valuable. Conversely, a deepening appreciation of the
complexity and limitations of branches of knowledge, including the
uncertainty inherent in the social sciences and how they inform pol-
icy, helps establish the value of argumentation. And as Kuhn rightly
points out, ‘people must see the point of argument, if they are to
engage in it’ (ibid, p.201).
16

16 INTRODUCTION

0.4 ALTERNATIVE APPROACHES

This book largely works within the tradition set out in the previous
section. At its centre is the analysis of arguments and the processes of
argumentation, and both skills and dispositions are regarded as impor-
tant. Sharing the spotlight with cognitive biases and the motivations
that underpin some of them is still relatively unusual, but they are
linked in very important ways, and strongly aligned with the book’s
emphasis on the value of critical thinking for self-understanding.
Before moving on, and to help explain this approach, I will briefly
highlight some alternative emphases within the discipline.The previ-
ously mentioned basic definition of critical thinking – ‘reasonable
reflective thinking focused on deciding what to believe or do’ (Ennis,
1996b, p.166) – leaves a lot of scope for what precisely this should
include. I don’t think there could be an approach that neglects the
analysis of arguments in some form or other, but there is then a long
list of other subject matter that you may or may not find in various
textbooks. This includes argumentation and dialogue, dispositions
(or epistemic virtues), problem-solving, scientific methods, cognitive
biases, and very occasionally ‘critical theory’.
My exclusion of scientific methods is linked to the debate about
how transferrable critical thinking is, and its connection to specifi-
cally academic skills. My view is that while a general sharpening of
perceptual and analytic skills, and in particular a critical (enquiring,
open-minded) frame of mind will certainly result from a well-taught
critical thinking class, being appropriately critical within one’s sub-
ject is something that must largely emerge from the teaching of that
subject. Scientific methods are always taught along with specific nat-
ural and social science disciplines, and there is no reason why quite a
philosophical perspective cannot be taken within this context.
Learning about methods can thus segue into the philosophy of sci-
ence, and while there are clear connections between the philosophy
of science and critical thinking, these are not strong enough to make
it core to the discipline. Where the inclusion of scientific method is
more likely is in an approach to critical thinking that is primarily
geared to academic skills; a perspective that is nested in an assump-
tion about the academic generalisability of critical thinking skills.
In my experience, however, the main transferrable value of critical
thinking comes from its potential for broader personal transformation.
17

INTRODUCTION 17

There is an intimate connection between the knowledge and skills it

teaches and hones, the dispositions that motivate their use in a wide
variety of situations, and increased insight into the person we are and
want to be. In short, we are making decisions all the time, whether
in academic, professional, civic, or personal contexts, and this disci-
pline lets us see how poor some of these are and what we can do to
make changes. Understood in this way, the four areas I pay attention
to – arguments, cognitive biases, dispositions, and dialogues – are
deeply revealing of how judgements are reached and decisions are
made. They are also, as will be repeatedly demonstrated in this book,
strongly interconnected.
Critical thinking is, then, often framed in terms of improving
our decision-making. Problem-solving is, on the other hand, more
peripheral because it extends the discipline into creative thinking and
includes a set of distinctive techniques. Only occasionally it is explic-
itly included in critical thinking textbooks, but at the same time
there is no question that the more standard content will contribute
to our problem-solving abilities.
‘Critical theory’ – a post-Marxist paradigm which analyses ideas,
behaviour, and institutions in terms of power structures – is some-
times confused with critical thinking, but has very little in common
with it. Because critical theory seeks to improve democracy through
exposing the hidden or overlooked reaches of power within a culture,
there is an inherent suspicion of appearances and a practical applica-
tion that it shares with critical thinking. It rests, however, on a set of
sociopolitical insights, theories of knowledge, and specialised methods,
and in this respect is much more at home among other wide-ranging
philosophical ideas (like postmodernism). For most, then, a school of
thought like this is too different to warrant a close association with
critical thinking.That said, there is a fringe that stretches the discipline
in a way that encompasses critical theory (see Davies, 2015), and the
section below (on ethics) has a flavour of this way of thinking.

0.5 CRITICAL THINKING AND ETHICS

In this section I will address two issues associated with the ethi-
cal consequences of critical thinking. The first is linked with its
protective function and concerns critical thinking’s potential for
18

18 INTRODUCTION

empowering individuals faced with the agendas of others, and the

need to be appropriately self-determining.Value is placed on know-
ing ourselves and choosing the direction of our lives to the extent
that this is possible. Arguing for the virtues of critical thinking in this
respect can quickly elide into a generalised view of what constitutes
an ideal education in a modern liberal democracy.
Harvey Siegel offers a further reason for the ethical value of criti-
cal thinking in education – ‘democratic living’. Democracies, he says:
rely for their health and well-being on the intelligence of their citizens. …
such intelligence, if it is to truly be of benefit, must consist in part of the
skills, attitudes, abilities and traits of the critical thinker. It is not simply an
intelligent citizenry, but a critical one, which democracy wants. (1988, p.60)

The argument goes that democracy is a good thing, and to main-

tain this good thing, individuals need to be able to deliberate effec-
tively when engaging with the fast-moving, multifaceted, multivoiced,
bias-laden debates that are now basic to modern democratic living.
Although clearly informed by knowledge of a range of issues and dis-
ciplines, being a democratic citizen is not itself a specialist subject to
be learned as one would learn chemistry or history. The judgements
a democratic citizen needs to make, however, are significantly assisted
by the knowledge, self-knowledge, and know-how of critical thinking.
The second ethical consequence of critical thinking I would like
to discuss is less favourable and concerns its place in the increas-
ingly instrumental approach to education that is dominant in many
Western countries. When discussing its ethical qualities, Siegel says:
Critical thinking is no rubber-stamp friend of the status quo; indeed it is an
enemy of the unjustifiable status quo. (ibid, p.55)

There is a sense, however, in which it has been co-opted into an

‘unjustifiable status quo’ within education. If you read the rhetoric of
most universities in English-speaking countries there is surprisingly
little mention of the value of learning for its own sake – of finding
out about and being fascinated by the world – and a lot about how
students will graduate with transferable skills, be highly employable,
and be ‘equipped for the workplace’. If we are to understand the
modern aims of higher education through the lens of what are called
‘graduate attributes’ then the emphasis on learning for the sake of
19

INTRODUCTION 19

something else – being an effective team worker, problem solver,

communicator – is striking.
For our purposes the important point is that some variation of
critical thinking appears on virtually all these lists and is thus impli-
cated in an instrumentalist turn in higher education. If you are not
worried about this development then there is no ethical issue, but if
you are then critical thinking needs to be careful how it situates itself
in this setting.
My view is that this is a genuine concern. Transferable skills are
certainly important, and of course many students will want to know
where their degree might lead and how it can take them there, but it
is a matter of balance. We need to avoid the implication that univer-
sity is only for a certain type of person; not all students will have or
want a clear idea of future directions and are primarily at university to
learn things about subjects they care about. But perhaps more impor-
tantly, for most students, love of learning and deep immersion in the
subject matter of their degrees will be a possibility even if they enter
higher education with a broader instrumental agenda. The trouble
with this hefty emphasis on graduate attributes and measuring the
value of a degree in terms of employability or financial rewards is that
the ‘finding out about the world’ aspect is liable to be diminished or
even forgotten. Understanding learning for the sake of a career and
all that this implies is easy for most people to understand and accom-
modate, but the value of a deep engagement with the knowledge an
academic discipline brings is more abstract and not always so easy to
truly appreciate. So, in the context of global capitalism and the mar-
ketisation of higher education, it needs to be preserved.

0.6 THIS BOOK’S APPROACH (1): OVERVIEW

OF CONTENTS
Chapter 1 seeks to explain rationality with a strong emphasis on the
psychology of judgement and decision-making. Chapter 2 is about
the relationship between rationality and emotions, and the role of
emotion in argumentation. Chapter 3 concentrates on critical think-
ing dispositions and the nature of constructive dialogues. Chapter 4
is an in-depth look at the nature of arguments and explains the art
of argument reconstruction – setting out the essential structure and
20

20 INTRODUCTION

content of arguments expressed in natural language so that they can

be better understood and responded to. Chapter 5 explains deduc-
tive, inductive, and what are called ‘plausible’ arguments; defines and
explains fallacies, and introduces ‘critical questions’ as a method of
interrogating arguments. Chapters 6–9 are concerned with a range
of argument forms and their associated fallacies. These include argu-
ments from authority, ad hominem arguments (those attacking the
person rather than what the person has to say), causal reasoning, gen-
eralisations, and arguments from analogy. In each case my analysis will
include discussion of their persuasive (or rhetorical) potential and the
cognitive and motivational biases that predispose us to them; their
relevance to certain critical thinking dispositions (and the absence
of these dispositions); and their impact on attempts to generate con-
structive dialogues. Chapter 10 explains some further fallacies that
are important to know about, and the book’s final chapter includes
some reminders of how critical thinking is not just about assessing
the arguments of others, but how we construct and reflect upon our
own arguments as well.

0.7 THIS BOOK’S APPROACH (2): THE USE OF

EXAMPLES
The examples used in critical thinking books usually involve general
subject matter – statements and arguments that do not require special-
ist knowledge to understand. Up to a point the strength of an argu-
ment is determined by the form it takes – an argument containing a
contradiction is very likely to be a weak one, as is one that appears
to jump to a conclusion without providing sufficient premises – but
mostly it is the combination of the form of the argument employed
and the context and subject matter of that argument that decides its
quality. This means that in order to do critical thinking properly we
need to know something about the issue being discussed. Topics that
fall under ‘current affairs’ and ‘general knowledge’ thus serve two pur-
poses: first their familiarity allows for a deeper understanding of the
kind of work that critical thinking undertakes; and second they can
serve to illustrate how you should go about applying critical think-
ing to the subject matter and situations (academic, professional, civic,
personal) that you have significant knowledge of and responsibility for.
21

INTRODUCTION 21

With this in mind I will make a few points about the examples and
exercises found in this book:

1. For the various argument forms discussed I will provide multi-

ple examples so that, hopefully, those readers not familiar with
some will be more familiar with others, and where necessary rel-
evant background information will be provided.These examples
will come from a combination of academic and non-academic
sources, but in all cases I have tried to make them as interest-
ing or as topical as I can. (I should also point out the unusual
approach to referencing used in this book. As a general rule, texts
relevant to the ideas I am discussing will be shown as in-text
citations with the full information found in the reference list at
the end of the book. Texts that serve only as examples of these
ideas are fully referenced via the chapter endnotes.)
2. Since context, including dialogue, is so important for the
proper understanding and assessment of arguments, then stu-
dents and other readers will gain far more from applying their
learning to extended examples rather than brief and isolated
ones. An upshot of this is that critical thinking is best learned
through materials and experiences that no textbook by itself
can hope to provide.

However, remedies are not so hard to find:

1. If you are learning critical thinking in university or at school

then you and your teacher can select your own examples.
These could come from subjects being studied alongside criti-
cal thinking, or perhaps more profitably from the current affairs
of the place and time you find yourself in. Students of critical
thinking should get into the habit of seeing patterns of argu-
ment and reasoning errors in the world around them (and in
ourselves), and being tasked with finding examples is an ideal
way of helping to establish such a habit.
2. A similar point goes for those of you interested in critical
thinking but who are not currently in formal education. Apply
what you learn here to the information and conversations you
encounter at work and in other areas of your life.
22

22 INTRODUCTION

3. Extended arguments and argument-based narratives can be

found in various media, including documentaries, books, essays,
opinion pieces, speeches, and works of fiction. I make refer-
ence to many such examples in this book, but there are endless
possibilities, and I can very happily guarantee that the ideas
and skills you will learn from these pages will give you insight
into, and the tools to analyse, any text you choose that involves
argumentation and rhetoric. Each chapter (aside from this one
and the conclusion) lists some exercises at the end, but this
always assumes the additional recommendation of applying this
learning to examples selected with your context or interests in
mind.

Perhaps unexpectedly the extended text I have found to be the most

effective for teaching and learning critical thinking is the film Twelve
Angry Men (written by Reginald Rose in 1957, originally as a TV
drama, but later made into a film by Sidney Lumet). Since I will
make recurring reference to it in the book, I will briefly explain the
story. Set in New York, a jury of twelve men have to reach a unani-
mous verdict on a murder case in which a 16-year-old is accused
of killing his father. If he is found guilty, he will face capital pun-
ishment. On the surface the case seems open and shut – the boy is
clearly guilty – but one member of the jury (Juror 8, played by Henry
Fonda) is unwilling to return a guilty verdict before there has been
at least some discussion of the case. He persuades the other eleven
men to revisit the evidence, and as they do it starts to become appar-
ent that things are not as clear-cut as they had seemed. The ensuing
drama is, in effect, an extended piece of argumentation in which
Juror 8 battles with the prejudices, poor reasoning, and underdevel-
oped dialoguing skills of most of the rest of the jury. In the process,
countless arguments are presented by both sides, and many of these
are representative of the argument forms and fallacies that will be
discussed in this book.
Twelve Angry Men provides richness of context and in so doing
brings the subject matter of critical thinking to life. It allows for
a detailed assessment of the characters’ arguments, biases, disposi-
tions, and ability to engage with constructive dialogue. Reasoning,
reasonableness, and their opposites are put in sharp focus by highly
23

INTRODUCTION 23

recognisable behaviours, and all this in a situation where life and

death pivot on certain protagonists’ abilities to think critically. It’s old,
it’s black and white, it’s heavy on dialogue, light on ‘action’, but it is a
great film by any standards and one I urge you to watch as a teacher
or student of critical thinking.
I shall conclude this opening chapter with an extended definition
of critical thinking as explored in this book:

BOX 0.2 DEFINITION OF CRITICAL

THINKING
The aim of critical thinking is to make us better deliberators and
decision makers through knowledge, techniques and a frame of
mind that:
a) Help us identify the sorts of questions we should be asking
before making significant decisions.
b) Help us to construct, assess, and improve our own arguments.
c) Teach us about the pitfalls associated with reasoning terms of:
i. Weak arguments and their associated psychological
biases;
ii. Features of unconstructive dialogues; and
iii. Dispositions that make us prone to poor reasoning and
unconstructive dialogues.

FURTHER READING

• Good primary sources for understanding the Socratic spirit and

his critique of rhetoric are Plato’s Apology and Gorgias (multiple
editions are available). A good secondary source is David Melling’s
Understanding Plato (1987), Chapter 5.
• For an excellent article on the place of critical thinking in educa-
tion, including an important perspective on its relationship with
discipline-specific content, see R.T. Pithers and Rebecca Soden
(2000).
• A very good critical thinking textbook that also has a strong psy-
chological angle is Diane Halpern’s Thought & Knowledge (2014).
24

24 INTRODUCTION

• For a discussion of the wider contexts of argumentation, including

chapters on non-European and feminist perspectives, see Deborah
Berrill (1996), and Martin Davies and Ronald Barnett (2015).

NOTE
1 Known as just Informal Logic since the mid-1980s.
25

RATIONALITY AND
COGNITIVE BIASES

Reflective thinking is always more or less troublesome because it involves

overcoming the inertia that inclines one to accept suggestions at their
face value. (John Dewey, 1910, p.13)

Learning to think critically makes us more rational. Traditionally

the discipline of critical thinking does this through an analysis of
arguments, including forms of fallacious (or erroneous) arguments
that we typically encounter and generate in argumentation. It is also
interested in what causes us to make the kinds of mistakes we do, and
part of the answer comes from an understanding of the psychology
of judgement and decision-making. This chapter is an introduction
to aspects of this field of study, and its subject matter will remain
relevant throughout the book. Here it will be addressed under three
main headings: Rationality, Heuristics and biases, and Framing.

1.1 RATIONALITY
In an investigation of critical thinking, two forms of rationality are
important to distinguish: ‘cognitive’ and ‘practical’ (or ‘functional’).

DOI: 10.4324/9781003247944-2
26

26 RATIONALITY AND COGNITIVE BIASES

Cognitive rationality refers to the grounds on which we hold our

beliefs. Beliefs that are inconsistent with the weight of evidence
presented, or which contradict other beliefs held by the individual,
are prime examples of cognitively irrational beliefs. For instance,
my Uncle Vance’s refusal to accept that his wife Eliza was having
an affair despite her acting out of character, displaying signs of guilt,
‘staying late at the office’, coming home smelling of aftershave, and
the reports of friends who’d seen her out with another man, is a
likely case of cognitive irrationality. Someone’s proclamation that it
is always morally wrong to break the law and then insisting that ille-
gally assisting a suicide can be morally justified would also appear to
be cognitively irrational; so too my Aunt Eliza’s claim to hate all pets,
while evidently loving my brother’s dog.
Practical rationality refers to the alignment between our desires or
goals on the one hand, and our decisions and actions on the other.
A decision is rational in this sense if it is consistent with achieving
our goals. If Annie wants to pass her exam it is not rational to forsake
the necessary revision for a night on the Jägermeister with her pals;
but if Josie has finished her exams, wants to unwind with friends, and
likes Jägermeister then such a night could well be a rational decision.
So, Annie could have the cognitively rational belief that she needs to
revise in order to pass her exam, but because of some form of denial,
or perhaps weakness of the will, make the practically irrational deci-
sion to go out with friends.
Practical rationality works best where there is self-understanding –
knowing what our wants are, and how these wants are prioritised –
along with an accurate appreciation of how the world is in relevant
respects. In this sense it clearly benefits from cognitive rationality, but
there is no necessary connection between the two. The belief that
driving a manual car does not require any practice when you’ve only
ever driven automatics, justifies, in the practical sense, accepting a
manual rental car on arriving at the airport, but my friend Sacha now
appreciates that this conviction was cognitively irrational.

TWO SYSTEMS

Cognitive psychologists will agree that decisions are influenced

by different forms of information processing, and from our own
27

RATIONALITY AND COGNITIVE BIASES 27

experiences we know the difference between reasoned deliberation

and quicker decisions that involve little or no concentrated thought.
There is less agreement, however, on how we categorise these
modes of thought; for example, whether there are discrete systems
of fast and slow thinking, or a continuum running from automatic
­decision-making through to decisions based on the abstract reflec-
tions and controlled attention.
The more common view – found in what are called dual process
theories – is that humans have two ‘modes of thinking’, typified
by what Amos Tversky and Daniel Kahneman (1974) call System
1 and System 2 thinking. I will work with this terminology in
this book, but it should be acknowledged that a more nuanced cat-
egorisation (or a continuum) could turn out to be more accurate
(see Further reading at the end of this chapter). For our purposes,
the general distinction between automatic and reflective thinking
that the System 1/2 model points to is uncontroversial. And it is
this general distinction that matters for understanding how critical
thinking can provide insight into, and help us manage, the ways we
process information towards making decisions. They are described
in this way:
System 1 operates automatically and quickly, with little or no effort and no
sense of voluntary control.
System 2 allocates attention to the effortful mental activities that
demand it, including complex computations. The operations of System 2
are often associated with the subjective experience of agency, choice and
concentration. (Kahneman, 2012, pp. 20–1)

Most of our thinking is guided or influenced by System 1, which

is well adapted to quick decision-making amidst routine activities
in familiar environments. The more that System 1 can handle, the
more energy we have for mustering System 2 to attend to the unfa-
miliar and to tasks that require concentrated attention. Among the
examples of System 1 thinking offered by Kahneman are routine
driving on an empty road, understanding straightforward sentences,
and familiar arithmetic such as 2 + 2. In contrast System 2 is required
for more precise or unexpected driving manoeuvres, and for more
complex or unfamiliar sentences or problems to be solved. All critical
thinking goes on in System 2.
28

28 RATIONALITY AND COGNITIVE BIASES

These two ways of engaging with the world should be familiar

to us from everyday experiences. In System 1 we are in the flow of
things, letting our previous experiences guide us through the situa-
tions we encounter. In System 2 we step out of the flow and become
analytical and reflective.The importance of dialectics for Socrates and
Plato was that it maintains our reflective concentration. Rhetoric, on
the other hand, panders to the forms of quick thinking that are basic
to System 1.
The biases and heuristics that typify System 1 thinking are
explored below, but the work of Tversky and Kahneman, and even
more so that of Seymour Epstein (2014), also identifies its emo-
tional and motivational drivers. Epstein’s version of System 1 – the
‘­experiential system’ – is strongly linked to emotions and will be
further explored in the next chapter. For now, I will highlight a few
of its other features:
• It operates with concrete ideas, including recalled experiences,
and is highly responsive to images; for it, ‘seeing is believing’. The
‘rational system’ (the equivalent of System 2), by contrast, rec-
ognises the potential for finding truth in abstraction.
• The experiential system is ‘hedonistic’; strongly motivated by
pleasurable feelings.
• Among the things that feel good are being right and being
certain. This helps explain why the experiential system (or
System 1) tends towards crude categorisations (such as black
and white thinking) and, importantly, the conviction that the
beliefs it holds are ‘self-evidently valid’. It seeks to simplify the
world and filter information in a way that conforms to exist-
ing ideas, including ideas we have about our worth (usually
positively biased ones). The ‘rational system’ (the equivalent of
System 2), by contrast, if relatively unrestrained from the influ-
ence of the experiential system, is directly motivated by argu-
ments and evidence.
• For this reason, the rational system can change its beliefs rapidly;
the experiential system, on the other hand, changes slowly.
To illustrate a couple of these ideas, recall from the previous chap-
ter (Section 0.3) Deana Kuhn’s observations of how certain styles
of thinking are ‘inextricably tied to an individual’s own personal
29

RATIONALITY AND COGNITIVE BIASES 29

experiences’. Exploring people’s reasoning on the topic of why chil-

dren fail at school, one response to the question, ‘Could someone
prove that you were wrong?’, a revealing response is:
I suppose intellectually someone could give me statistics that would show
that the position I hold reflects only a minority of reasons why children
fail and therefore prove I was wrong. But I would probably not change my
opinion. It’s the result of lifelong personal experience and quite frankly, I
think it is right. I think it reflects the reality of the situation as I’ve experi-
enced it. (Kuhn, 1991, p.182)

Demonstrated here is not just the profound division between these

two systems’ grounds for holding beliefs, but also their relative power.
The experiential system is dominant, and it is stubborn – what it
believes is believed because it is regarded as self-evidently valid.
Strong objective evidence opposing a belief might be acknowl-
edged by the rational system, but this is often overridden by its
counterpart. To change a strongly held conviction a truth needs to
be experienced as well as (or sometimes instead of) reasoned out.
‘Experienced’ means consistently observed, lived-through, contex-
tualised encounters that, crucially, the person feels as well as cogni-
tively processes.
Suggested in this brief dip into Epstein’s theory is the way that the
experiential system is guided by basic human needs, and thus how
both it, and indeed the ideas of Tversky and Kahneman, provide a
window on to these motivations. Among those identified by Epstein
that have clear relevance to the biases and heuristics considered
below are the need for a stable and coherent set of beliefs about the
world, for self-esteem, trusting relationships with others, and pleasur-
able feelings (see Epstein, 2014, Ch.3).
The other aspect hinted at is the pervasive presence of System 1
or the experiential system (to keep things simpler, from here on I
will mostly use the language of System 1/2 rather than Epstein’s
terminology). Even concentrated thinking is prone to intrusions that
will influence our reasoning in ways that we are unaware of: infor-
mation more easily recalled because of its vivid nature rather than its
frequency, for example; or stereotypes substituting for more careful
categorical definitions; or the distorting effects that even mild emo-
tional responses can have on our judgements.
30

30 RATIONALITY AND COGNITIVE BIASES

1.2 HEURISTICS AND BIASES

System 2 is slow, and reasons in ways that critical thinking seeks to
understand and improve. System 1 is quick and employs what are
called heuristics.A heuristic is tool for quick decision-making; a rule
of thumb that is applied to certain types of situation that, although
lacking precision, makes better than chance judgements. Every day
we encounter a vast range of circumstances that require decisions,
and there is simply not the time to think all of them through in
detail. In Daniel Kahneman’s words, a heuristic ‘substitutes an easier
question for a difficult one’ (2012, p.105). Deliberation is often not
needed where things are routine and predictable, but some situa-
tions demand our effortful concentration. Ideally there would be the
right balance between these types of decision so that we have the
energy we need when we need it, but that is not the way it tends to
be. Instead decisions are prompted regarding what we attend to, and
this leaves a number of situations that would benefit from System 2
engagement in the hands of System 1. System 1 will then get them
right sometimes, and not others, but either way it seems that we need
to rely on it to take up the slack.
There are a couple of ways in which knowledge of heuristics and
biases is important for enhancing critical thinking:

1. Understanding the various heuristics we use can improve

our self-knowledge and motivate us to become less vulner-
able to their negative influence. It can, for example, contrib-
ute to a form of self-awareness – referred to in Chapter 3 as
metacognition – in which we are more alert to their likely
presence in certain situations and therefore able to employ
strategies for avoiding or mitigating their bias.
2. They can themselves be educated by the practice of critical
thinking. We have, for example, an authority heuristic that
inclines us towards those beliefs and decisions that are held and
advocated by apparent experts on the matter at hand. However,
our criteria for detecting expert authority in someone can be
more or less crude, and learning about both what constitutes
such authority, and the ways in which we tend to be tricked
into seeing authority where it does not exist, can help educate
31

RATIONALITY AND COGNITIVE BIASES 31

our quick as well as our slow responses. It can, in other words,

make our heuristics more intelligent.

Since they systematically privilege some potentially relevant informa-

tion over other potentially relevant information, heuristics are by their
nature biased ways of thinking. However, their bias can be justified to
the extent that it has the practical advantages mentioned. The mean-
ing of bias here is thus equivalent to ‘tendency’ or ‘leaning’. There are
cognitive biases, however, which do not have an obvious heuristic
value. These can take the form of predictable mistakes in fast or slow
reasoning that are no more than logical errors, or distortions of reality
that are the product of deeper motivations. The meaning of bias in
this case is more like a systematic error. Unfortunately, the language
used to identify them is not particularly consistent, and nor is it always
that clear when a particular bias has a heuristic function. For these
reasons I will treat them in a similar fashion, and the remainder of
this section is devoted to descriptions of some of the main heuristics
and biases that help explain our vulnerability to fallacious arguments.

THE CONFIRMATION BIAS

The confirmation bias (sometimes referred to as the ‘myside bias’)

is a strong cognitive default in which we seek out, attend to, and
remember evidence and arguments that confirm our current beliefs
at the expense of those that disconfirm them. Scientists know that
in order to keep a theory alive, finding evidence which is consistent
with it is only part of the story.We must also know what disconfirm-
ing evidence would look like, and seek this out as well. Everyday
thinking, however, tends to neglect this requirement. Philosopher
Bertrand Russell observed:
If a man is offered a fact which goes against his instincts, he will scruti-
nise it closely, and unless the evidence is overwhelming, he will refuse to
believe it. If, on the other hand, he is offered something which affords a
reason for acting in accordance to his instincts, he will accept it even on
the slightest evidence.1
In her empirical study of informal reasoning, Deanna Kuhn found
that participants’ previously expressed theories about why some
32

32 RATIONALITY AND COGNITIVE BIASES

individuals drop out of school significantly limited their interpreta-

tion of new information on the topic. Despite being aware of a wider
range of causes, new data was typically comprehended in a way that
confirmed people’s pre-existing beliefs (‘it says pretty much the same
thing that I think’). Kuhn says:
Although such subjects have reflected, probably accurately, on the rela-
tion between one theme contained in the evidence and their own beliefs,
they have imposed their own beliefs on the evidence to such an extent
that it has prevented them from accurately representing what the evi-
dence in fact consists of. (1991, p.231)

A lot has been written about echo chambers in recent years, and it
is fairly straightforward to see how being overly exposed, through
our news and social media preferences, to people who share our
views can only exaggerate the effect of the confirmation bias (see
also social proof, below).
Unfortunately, confirmation bias is not limited to informal
­reasoning – the academic reporting of research findings is prone to it
as well. In theory though, avoiding it in this context is relatively sim-
ple. One method, especially common in medicine and other areas of
applied science research, is known as a ‘systematic review’. This puts
in place rules for unbiased selection of previous findings on a subject,
along with questions to guide their assessment. It is designed to avoid
bias caused by non-rigorous searches or a lack of open-mindedness
on the part of the researcher. Crucially, although the original research
of the scientist is peer reviewed in academic journals, the review part
of it (i.e., a summary of the findings to date) is not always subject
to the same scrutiny. The systematic review approach insists on peer
review of this aspect and thus requires transparency with respect to
the scientist’s method of selection and evaluation.2
In order to protect ourselves from the confirmation bias in eve-
ryday thinking we need to develop the habit of looking for disconfirm-
ing evidence. This is made all the harder, however, if we consider that
our worldviews are themselves partly the product of the other biases
and heuristics that will be discussed in this section. In other words,
the confirmation bias will amplify beliefs that are already under the
influence of a range of other distorting effects.
33

RATIONALITY AND COGNITIVE BIASES 33

THE SELF-SERVING BIAS

The confirmation bias is very common and hard to eradicate for

several reasons. One is that at its heart is the tendency towards, and
deep-seated desire for, the preservation of a consistent worldview.
Another is that it is supported by many of the other heuristics and
biases discussed in this chapter. Their simplifying logic is based on
coherence – making sense of new experiences in terms of exist-
ing frames of reference – and they are often under the influence of
motivated reasoning. A predictable world is comforting, and as a
point of pride we prefer our existing beliefs to be true rather than
false or questionable. Motivated reasoning, then, points not just to
the existence of biases, but to the work they are doing in making us
feel better about ourselves and the way the world is.With this under-
standing in mind, Richard Paul sees critical thinking as a corrective
to egocentric and sociocentric biases because:
There are deep seated tendencies in the human mind to reason in order
to maximise getting what we often unconsciously want. This typically
involves using cognitive and affective processes to maintain self-serving
or pleasant illusions, to rule out or unfairly undermine opposing ideas …
and otherwise to distort or “misinterpret” our experience to serve our
own advantage. (1984b, p.5)

The ‘pleasant illusions’ Paul refers to have become known as the self-
serving bias (Taylor and Brown, 1988): an automatic and systematic
tendency to overestimate those features of ourselves and the world
that are core to our sense of self-esteem and our motivation to suc-
ceed. These include our positive attributes (traits and abilities), the
amount of control we have over events that affect us, and how bright
our future will be in comparison to the futures of our peers. A gen-
eralised bias that serves these illusions sees us taking too much credit
for successes and not enough responsibility for our failures. And in
case you’re wondering, we also tend to self-servingly see ourselves as
less susceptible to cognitive biases than other people – what has been
termed the ‘bias blind spot’ (Pronin, Lin, and Ross, 2002).
Not everyone is subject to these self-serving illusions, but it seems
that the majority of us are. The implication of this research is that
we are not particularly good at dealing with reality square on, and
34

34 RATIONALITY AND COGNITIVE BIASES

in many respects it is an example of the capacity for self-deception

that has historically been seen as basic to the human condition.
One of the most comprehensive contemporary approaches to self-­
deception has been the psychotherapeutic analysis and categorisation
of ­ego-defences.These are ways in which we distort reality in order
to keep painful truths away from full consciousness, and one of the
most common is rationalisation. To rationalise is to provide a rea-
son for why something happened that (1) is not the real reason; (2)
we unconsciously know is not the real reason; but (3) does the job
of protecting us, at least temporarily, from the real reason. We protect
ourselves because the truth makes us uncomfortable. Often others
will know that we are rationalising because the explanation given is
not that plausible, or they have insight into our motivations. In the
moment, however, and perhaps for longer, the person rationalising
cannot afford to see the truth, and so they allow themselves to be
taken in by their contrivance.
Novels, dramas, and films are full of examples of rationalisations,
and because they are so common in everyday life, they are usually
quite recognisable. The plots of the UK comedy series Peep Show
are almost entirely based on absurd situations caused by rationalisa-
tions; and in stark contrast to his seemingly ultra-realist partner, the
character Martin Hart (played by Woody Harrelson) in the first series
of True Detective likes to take comfort from them. At one of his low-
est points, he justifies an extended affair with an attractive younger
woman by claiming that it is a release from the pressures of work and
therefore done for the benefit of his family. An almost saintly act of
self-sacrifice I’m sure we can all agree.

COGNITIVE DISSONANCE, COMMITMENT, AND CONSISTENCY

In Leon Festinger and James Carlsmith’s (1959) classic study of cog-

nitive dissonance, participants in a fake laboratory experiment con-
ducted a series of highly tedious tasks for a full hour. Once they
had completed this they were asked, after a (fake) debrief, if they
wouldn’t mind stepping in for an absent worker. That worker’s job is
to encourage the next participant (a fellow student) to stick around
because, despite what they had heard, the experiment is enjoyable.
Some were paid $20 to tell the lie, some $1, and others nothing.
35

RATIONALITY AND COGNITIVE BIASES 35

Most complied, and once the deed was done these participants were
then asked to rate the actual interestingness of the experiment, and
one group rated it much more highly than the others. You might
think it was the $20 group – perhaps because their elevated mood
caused by the not insubstantial payment had transferred to the expe-
rience as a whole – but it was the $1 group.The (now well-accepted)
reasoning goes that telling a lie to a peer for the sake of $1 is typi-
cally incompatible with a person’s self-image (as decent, honest, etc.);
$20, on the other hand, appears to be an acceptable trade-off. Self-
deception is thus enlisted by the $1 participants to preserve their
integrity, but as it is a psychological stretch to deny they had accepted
the money or told the lie, they instead subconsciously altered their
opinion on how engaging the task was.
This is an example of cognitive dissonance; a term coined by
psychologist Leon Festinger in the 1950s referring to the discom-
fort we feel when we become aware of an inconsistency among our
beliefs, attitudes, and actions. Faced with this tension, ego-defensive
strategies can be enlisted to help maintain a belief or attitude that is
important to us (often one that is associated with our self-image).
Rather than reappraise what’s important, we instead reinterpret other
features of the event so as to suppress the contradiction.
Cognitive dissonance is an aspect of a broad set of theories, now
well established, that point to a profound need for self-consistency.
Once we commit to something (a cause, a belief, a course of action)
‘we will encounter personal and interpersonal pressures to behave
consistently with that commitment’ (Cialdini, 2007, p.57). ‘Personal
pressure’ because most of us desire the character integration and
rationality that consistency is so central to; ‘interpersonal pressure’
because our need to rely on people being true to their word makes
us highly condemning of promise-breakers. For this reason, public
declarations or pledges are extremely powerful techniques for ensur-
ing that intentions are carried out.
The strength of this need for consistency is demonstrated by a
range of persuasion techniques which induce compliance in people
with respect to a trivial commitment, and then use this to leverage
more significant, but apparently consistent, behaviours. Known as the
‘foot-in-the-door’ technique, if I answer ‘yes’ to the telesales person’s
trivial question ‘Would you like a better deal on your phone?’, I’m
36

36 RATIONALITY AND COGNITIVE BIASES

now implicitly committed to the conversation (by being willing to

give an answer), and explicitly committed to being open to hearing
about the details of this ‘better deal’.
A more serious example of where small commitments can land
us is found in the Milgram experiments, discussed in Chapter 6. To
avoid being trapped by the consistency heuristic we need to be care-
ful what we commit to, both promises made and the views we hold.
In Chapter 8 (Section 8.2) we will consider how absolutist state-
ments and arguments can, through the need for consistency, force us
to defend unacceptable or implausible positions.

RECIPROCATION

Rules of reciprocation – a fair balance between what we receive

and what we give – are so important and pervasive among cultures
that they are often regarded by anthropologists as a universal norm.
Even if it is in the form of a ‘thank you’, a favour done merits some-
thing in return. Anyone who has not received the required murmur
of gratitude for holding a door open for someone else, or the raised
hand of acknowledgement from the driver of the car you have made
way for in traffic, will know how sensitive we are to this norm.
In Caveman Logic (2009, p.49), Hank Davis tells the story of a non-
religious friend whose wife underwent tests on symptoms that could
have meant a terminal condition. When he found out the condi-
tion was treatable, he immediately headed for the hospital chapel and
wrote ‘thank you’ in the message book. ‘He did not write “Thank
you God”’ Davis recounts. ‘He did not pray. He simply, as he put it
to me, felt an irresistible urge to say “Thank you” … A gift had been
received and some circuitry had been triggered in him.’
Persuasion specialists know that an unsolicited gift in a charity
appeal envelope, or as you step into a shop, will increase the likeli-
hood of you giving or buying. A variation of this tactic is known as
the ‘door-in-the-face’ technique. Unlike ‘foot-in-the-door’, ‘door-
in-the-face’ begins with a large request rather than a trivial one. A
refusal is expected though, and once this happens, the guilt we irra-
tionally but irresistibly feel at having to say no to someone makes us
much more likely to accept their next (more reasonable) offer. As
a naïve youth I was stopped in the street by someone who turned
37

RATIONALITY AND COGNITIVE BIASES 37

out to be from a local cult. We chatted a while and then he asked if

I wanted to buy one of the cult’s publications, for £10. That wasn’t
going to happen, but he swiftly produced a pamphlet for 50p, and
that was me, sold, and on my way.
Common to negotiations is what is known as a ‘reciprocal conces-
sion’. Early on in Twelve Angry Men Juror 8’s ‘not guilty’ vote means
that a stalemate has been reached: none of the other jurors seem will-
ing to talk about the case, but a unanimous verdict is needed. Juror
8 makes an offer: the jury will have another secret ballot, and if the
outcome is still 11-1 he will change his vote to ‘guilty’, but if there’s
one more vote for ‘not guilty’ then they must agree to discuss the
case further. From Juror 8’s perspective the riskiness of this offer is
plain to see, so the relatively small concession of agreeing to a ballot
is hard for the rest of the jury to say no to.3

THE REPRESENTATIVENESS HEURISTIC

The representativeness heuristic attempts to provide quick

answers to questions such as ‘What’s the probability of individual X
belonging to group Y?’, or ‘What’s the probability of event P being
the cause of event Q?’ It does this by accessing what a typical member
of group Y, or a typical cause of Q is like. If one of the answers avail-
able is representative of the type in question, this is what is picked.
So, if I enter a university laboratory to take part in an experiment
and I am expecting to be met by two people – the experimenter and
another volunteer – my representativeness heuristic will tell me that
the person in the white lab coat is the one running the experiment.
If asked whether someone whose hobby is fox hunting is more likely
to be politically left- or right-wing this heuristic will guide me to
answer ‘right’. If rain is forecast in the UK it will lead me to conclude
that it is more likely to be caused by a band of cloud moving in from
the west rather than the east.
However, as helpful as this aid to quick thinking can be, it can
also lead us to make some predictable mistakes. One concerns
­stereotypes – inaccurate generalisations about core features of a
particular category. Representativeness only has a chance of pro-
viding the right answer if our idea of the typical group member
or causal factor is accurate in the first place. The chimpanzee’s fear
38

38 RATIONALITY AND COGNITIVE BIASES

grimace, for example, looks a lot like a human smile, so short of the
necessary specialist knowledge this heuristic might well lead us to
conclude they are happy when quite the opposite is the case. (More
will be said about stereotyping when we discuss generalisations in
Chapter 8.)
The other kind of mistake caused by the representativeness heu-
ristic is the overlooking of other critical questions we should be
asking in certain situations. A famous example used by Tversky and
Kahneman (1983) is the ‘Linda Problem’, which goes like this:
Linda is 31 years old, single, outspoken, and very bright. She majored in
philosophy. As a student she was deeply concerned with issues of dis-
crimination and social justice, and also participated in antinuclear dem-
onstrations. Please [indicate] the most likely alternative:
a) Linda is a bank teller
b) Linda is a bank teller and is active in the feminist movement

The correct answer must be a) because b) contains the same infor-

mation along with another claim about Linda, making is necessarily
less likely. However, most of us immediately give answer b) because it
typifies the kind of person Linda seems to be in a way that just being
a bank teller very much does not. In this instance the representative-
ness heuristic causes us to overlook the conjunction rule (that the
likelihood of X + Y is smaller than X by itself), and thus to commit
what is known as the conjunction fallacy. Representativeness is also
behind the common fallacy of base rate neglect.

BASE RATE NEGLECT

Base rate neglect is a type of reasoning error related to the repre-

sentativeness heuristic that is distinct and widespread enough to war-
rant a separate entry. It refers to situations in which, when making a
judgement about a particular person or event, we overlook relevant
background information (it is sometimes known as ‘background rate
neglect’). Another famous example from Tversky and Kahneman
involves ‘Steve’ who (with one or two adjustments to the original)
we are told is:
quite shy, not especially sociable, but with a few close friends. He has a
need for order and structure and a passion for detail. (1974, p.1124)
39

RATIONALITY AND COGNITIVE BIASES 39

Asked if, for example, Steve is more likely to be a librarian or a sales-

person most of us will assume the former. However, this overlooks a)
the fact that there are many more salespeople than librarians, and b)
that there is nothing in this description of Steve that precludes him
from being a salesperson. Our representativeness heuristic causes us
to jump to the conclusion that he is a librarian because his personal-
ity is what we see as typical (or stereotypical) of this profession. In
doing this, though, we neglect other considerations that are relevant
to answering the question.
Christmas Eve in 2021, listening to the morning news I heard
the argument that there are more people in hospital with Covid
who have had the vaccination than who haven’t, therefore the vac-
cine does not work. It was correctly called out as base rate neglect
(though not in those words); in simple terms, no vaccine will fully
protect, so if a large enough proportion of the population have been
vaccinated then this is what we would expect. For example, if 80%
of people have been vaccinated and it provides 70% protection from
serious symptoms, and the other 20% have an 80% chance of serious
symptoms, then out of 100 people in hospital with Covid there will
be 24 who have been vaccinated and 16 who have not.
Base rate neglect describes the error of taking a statistic or other
piece of information out of its comparative context in such a way
that it makes it look more significant than it is. We neglect to ask the
critical question ‘What is the norm in this situation?’, and instead
make a judgement based on what is generally representative of high,
low, good, bad, and so on. A 100% price rise appears dramatic, and as
with the devastating increase in wholesale gas prices in 2022, nor-
mally is. However, a 100% rise in the price of lollipops from 10p to
20p (assuming your entire diet isn’t based on them) is probably not a
cause for concern because the base rate is so low.
Base rate neglect is quite a common error, and to an extent we are
used to people being challenged on it. In the middle of the 2017–18
Premier League season Mark Hughes was sacked by struggling Stoke
City. Hughes – a proven manager at this level who had led the club
well for years – wasn’t happy and made the argument (in the form
of a rhetorical question) ‘Who else will they get?’ He meant that the
meagre availability of qualified options had not be taken into con-
sideration by the board. Good managers are hard to find and Stoke
40

40 RATIONALITY AND COGNITIVE BIASES

were lucky to have him, and it turned out he was probably right.
Stoke were even worse under Hughes’s replacement and ended up
relegated.
Another twist on base rates is the fallacy of suggesting false back-
ground rates in order to minimise or normalise a situation that is
in fact not trivial or normal at all. In a speech after the Charleston
church shooting in June 2015, President Obama addressed just this
fallacy:
Let’s be clear: at some point, we as a country will have to reckon with the
fact that this type of mass violence does not happen in other advanced
countries … with this kind of frequency.4

THE AVAILABILITY HEURISTIC

The availability heuristic (also named and studied by Tversky and

Kahneman) judges the likelihood of an occurrence on how readily it
comes to mind. Availability is a fairly reliable measure of likelihood
because common occurrences are often those that present themselves
to us with greater ease. Communicating the reality, seriousness, and
urgency of climate change is challenging for several reasons, includ-
ing its relative invisibility among our everyday affairs. However, local
weather patterns, floods, and wildfires increase its availability because
they are not just visible, but vivid, dramatic, and happening now, and
thus provoke a more engaged response. Communicators in this field
are well aware of the importance of using these events to increase
and maintain the salience of climate change in government policy
and public discourse.5
However, the availability heuristic can also predict certain kinds
of mistakes. Early research demonstrated the biasing effect of news-
worthiness, vividness, and emotional impact on the perceived likeli-
hood of lethal events such as accidents or tornadoes (Slovic, 2000),
and contemporary literature connects it with exaggerated concerns
about, among other things, vaccines and terrorist attacks.6
A piece of advice given to young medics making diagnoses is ‘If
you hear hoof beats, think horses not zebras.’ More vivid and dramatic
causes will more easily come to mind, but these will not be repre-
sentative of their relative likelihood in comparison to more mundane
41

RATIONALITY AND COGNITIVE BIASES 41

explanations. In another example, an Ipsos survey in 2017 asked the

UK public to estimate the proportion of the UK that is ‘densely
built up’. The mean response was 47%, but the real figure is 0.1% – a
startling difference. One of Ipsos’s senior researchers suggested that
because the majority of the population live or work in towns and
cities this causes a skewed ‘mental image of the country’, 7 and other
commentators pointed to media representations and a pervasive nar-
rative of Britain as overcrowded or despoiled by industrialisation.8
Direct and indirect experience is thus typically one of urban life, and
this perception has a grossly exaggerated influence when people are
asked what the UK is like as a whole. It turns out that not only is this
a clear example of the availability heuristic’s potential to mislead, it
also has consequences. As commentators acknowledged at the time,
it will affect public attitudes on issues such as housing policy and
agricultural practices.

ANCHORING AND THE CONTRAST EFFECT

Anchoring and the contrast effect are examples of how the con-
textualizing of information can influence our perceptions and inter-
pretations. Anchoring happens when our judgement of a quantity
(or other value) is biased by an initial piece of information. Asked,
for example, to estimate the exact length of the Mississippi River
(which is around 2,300 miles), experimental participants who had
previously been asked if it was ‘more or less than 200 miles long’
guessed on average that it was around 1000 miles. By comparison, the
average estimate of participants who had previously been asked if it
was ‘more or less than 20,000 miles long’ was much larger – around
8,000 miles (McElroy and Dowd, 2007).
Anchoring is known to be a pervasive and powerful phenomenon
in a range of real-life as well as laboratory settings; so much so that
one of the few books written by the acclaimed psychology of per-
suasion academic and practitioner Robert Cialdini, is almost entirely
devoted to it (see Cialdini, 2017). It is most effective in situations
where attention is lower and people have limited knowledge of the
subject, but it can influence experts’ judgements as well. In nego-
tiations, for example, there can be an advantage in making the first
move because even a demand expected to be excessive will have a
42

42 RATIONALITY AND COGNITIVE BIASES

disproportionate influence over the range of prices (or whatever is at

stake) considered to be acceptable. In a different kind of case, when
two academics grade a student’s assignment a more objective assess-
ment is made if the second marker is ignorant of the first marker’s
grade.
The anchoring effect works with both System 1 and System 2
decision-making. In the former case it is automatically interpreted
as a hint towards what the right answer might be; and with the latter
its presence invites the confirmation bias. Subsequent deliberations
are skewed towards this anchor, prioritising reasons that confirm its
broad accuracy. It is also a good example of how the two systems can
interact. As concentrated thinking goes on in the foreground, behind
the scenes System 1 is searching for answers that are compatible with
the anchor, and these answers are more easily retrieved.
As with the example of academic second marking, in some situa-
tions we can choose to remain ignorant of possible anchors to avoid
their influence. Often, however, this is not possible. The exagger-
ated influence of first impressions in social encounters – discussed
in Chapters 6 and 7 under the halo effect – is also an example of
anchoring and is impossible to avoid. To reduce its effects, general
advice such as remembering its potency, being vigilant of System 1
intrusions into our deliberations, seeking the views of trusted oth-
ers, avoiding time constraints in decision-making, or undertaking
prior research where this is appropriate (as in formal negotiations)
might help. Overall though, there is a worrying lack of optimism
from researchers in this regard – it is an immensely pervasive and
powerful bias.
The contrast effect refers to the influence of nearby comparators
with contrasting qualities. Any parking ticket I have received in the
UK first declares that you must pay a substantial fine (say, £70), but
then goes on to state that if you pay within two weeks this will be
reduced to half that amount.Thirty-five pounds is still a lot of money
to pay for overrunning a meter by ten minutes, but it looks much
more reasonable when you have just been threatened with double
that amount. Another example comes from the late rock singer Mark
Lanegan’s autobiography Sing Backwards and Weep. He talks about how
much he disliked his name when he was young, but was able to come
to terms with it when he found out that he was almost called Lance.
43

RATIONALITY AND COGNITIVE BIASES 43

The contrast effect is commonly used as a sales and fundraising

tactic (‘If you cannot afford £100, could you then donate £10?’),
but it has a psychologically protective function as well. In a study
of the coping mechanisms of women diagnosed with breast cancer,
a downward comparison was typically found to be employed. For
example, those requiring a lumpectomy would be thankful they were
not like those facing a mastectomy; and those needing a mastec-
tomy in later life could at least be grateful they were not one of the
younger women who required this treatment.9 Finding a perspective
from which we can regard our situation as better than it might be is
a means by which the contrast effect can help us come to terms with
difficult or tragic circumstances. In his address to an interfaith vigil in
Newtown after the Sandy Hook shooting in 2012, President Obama
began by quoting from scripture which emphasises the temporary
nature of our Earthly lives in comparison to God’s eternal perspec-
tive. Under the circumstances it is hard to see what else he could have
offered by way of consolation.

AUTHORITY, LIKEABILITY, AND IDENTITY

When making quick judgements (in comparison to slow ones) we

are significantly more influenced by characteristics of the people
presenting arguments. In The Art of Rhetoric, Aristotle suggests three
means of persuasion: (1) pathos, referring to the emotions aroused in
the audience; (2) logos, meaning the speaker’s arguments; and (3) ethos,
the good character (or the apparent good character) of the speaker. In
recent research three features of the speaker’s (or writer’s) character
have been shown to be especially persuasive: the extent to which we
perceive them as an authority, how likeable we find them, and the
extent to which we identify with them.
Forms of authority will be considered at length in Chapters 6 and
7, but to illustrate the heuristic power of perceived expert authority,
consider an experiment investigating the effects of mood on how
we engage with arguments. It is known that being in a happy mood
will incline us to System 1 thinking more than a neutral mood will.
In this experiment a happy mood was induced in half of the partici-
pants (by rewarding them with money for a prior task), and the other
half remained in a neutral state. Everyone was then asked to listen
44

44 RATIONALITY AND COGNITIVE BIASES

to a speech about acid rain and to evaluate it in terms of how it had

changed their attitude towards the topic. Half the participants listen
to a speech containing strong arguments, half to one containing weak
arguments. Also, half were told that the speech had been written by
an expert, the rest were told it had been written by a non-expert.The
results showed that not only was there far less discrimination between
strong and weak arguments by the happy group, but that their beliefs
about expertise had a far greater influence of their assessment of
the speeches. For the neutral group, expert source had no signifi-
cant impact on how arguments were rated, whereas for the happy
group it led to a much higher reported attitude change (Worth and
Mackie, 1987).
The social psychology literature on likeability indicates a number
of factors that will increase our chances of liking someone, including
how physically attractive they are, their ability to make us laugh, how
much they like us, and perceived similarities. The latter blurs with
a third category of heuristic concerning message source – identity.
Short of time and information about a product or course of action,
if it is being promoted or modelled by someone who we identify
with – share a sense of place, common beliefs, and so on – this can
serve as a shortcut to deciding in its favour. If it works for them,
it might work for me too. Identity is basic to the idea of referent
power and plays a significant role in establishing trust. More will be
said about trust, referent power, and the role of authority, likeability,
and identity in argumentation and persuasion in Chapters 6 and 7.

SOCIAL PROOF

I used to hitchhike a lot in my twenties. Often I would be the only

person standing at a service station or a roundabout thumbing a lift,
but when there were other hitchers waiting as well I noticed how,
when a driver stopped to pick up one of us, other drivers would very
quickly stop as well.We could all be waiting for an hour with no one
taking the bait, then two or three cars would stop in rapid succession.
My suspicion was that social proof was the reason. Under condi-
tions of uncertainty, such as what to order from an unfamiliar curry
house, what an acceptable volume of speech is in a particular pub-
lic library, or indeed the appropriateness and safeness of picking up
45

RATIONALITY AND COGNITIVE BIASES 45

hitchers in modern Britain, the behaviour of other people becomes

an important (if you’ll excuse the pun) rule of thumb. Many of us
will be aware of the influence of customer reviews – and sellers’ prac-
tices of manipulating these – when buying things online. Similarly,
Spotify can’t directly tell you what’s worth listening to by a particular
artist, but it always knows what’s popular.
As some of these examples demonstrate, basing judgements on
social proof can be significantly unreliable. In the phenomenon
known as ‘groupthink’ (Janis, 1982), social proof contributes to a
range of problems affecting collective decision-making. Groupthink
describes a set of processes of social influence that cause a group of
otherwise rational and intelligent people to make very poor decisions.
Irving Janis’s theory was based upon mid-twentieth-century US for-
eign policy disasters such as the Bay of Pigs fiasco in Cuba, and the
escalation of the Vietnam War, but aspects of his detailed analyses can
be readily applied to group or committee decision-making in a wide
array of contexts. It is easy to see, for example, how the confirmation
bias can be exaggerated in group deliberations. If certain information
and points of view are shared by the majority of members, these will
tend to gain more traction in discussion at the expense of views held
only by lone individuals. It could be the case that, put together, these
individually held views amount to a strong counterargument, but a
collective confirmation bias obscures this potential.
Of particular relevance to social proof is the phenomenon of plu-
ralistic ignorance. Imagine you are sitting on a high-powered com-
mittee that needs to decide on an important and urgent matter, and
that you are feeling unsure about a piece of information, an argu-
ment, or a decision about to be made. If you are reluctant to admit
your ignorance, or to give the appearance of lacking conviction, then
you will quite likely (whilst maintaining a calm demeanour) look to
similar minded colleagues and try to ascertain their position on the
matter. You scan the body language of several, and since they show
no obvious signs of dissent, you go with what seems to be the major-
ity view. What, though, if they are feeling the same way and looking
to you and the others for the same signs? Under these conditions,
ignorance is mutually interpreted as assent to whatever direction the
deliberations are heading in. The result of these and other predic-
tors of groupthink (such as the presence of a dominant group leader
46

46 RATIONALITY AND COGNITIVE BIASES

intolerant of dissent, a strong sense of identity or similarity among

group members, or a crisis situation in which emotions are running
high) can be an inadequate evaluation of alternative courses of action
and more than likely a poor decision.

1.3 FRAMING
A generalised idea that can be applied to most examples of persuasive
communication is framing. The basic meaning of framing is cap-
tured by expressions like ‘frame of reference’, or ‘frame of mind’. In
communication it refers to a way of looking at things, of putting an
‘angle’ or ‘spin’ on a particular issue. In essence it involves emphasis-
ing some aspects of the issue at the expense of others.
Whereas some might say that my Uncle Vance is an alcoholic, oth-
ers would say he’s a ‘party animal’. Views on the ‘partygate’ scandal
that contributed to Boris Johnson’s declining credibility as UK prime
minister rested in part on whether boozy gatherings in Downing
Street during Covid lockdowns could be convincingly framed as
‘work meetings’. When Vladimir Putin characterised Russia’s actions
toward Ukraine in February 2022 as a ‘special military operation’
(rather than an invasion or a war) it was part of a frame that presented
the conflict as justified self-defence in the face of NATO expansion-
ism (rather than as unprovoked aggression).
Frames are selected by communicators to enhance the effective-
ness of messages, but the filters we automatically apply to informa-
tion also function as frames. All heuristics and biases frame situations
because they are systematically selective in what they lead us to take
from them. As we have seen, there are many ways communicators can
compose messages to take advantage of this.
There are two purposes, broadly speaking, to which framing is put:

1. It can serve the audience’s needs by helping to make sense

of something new by comparing it with something the audi-
ence is currently familiar with; a basic educational process. In
Naomi Klein’s 2014 book, This Changes Everything, 10 analo-
gies with the anti-slavery, women’s, and civil rights movements
form a central argument for why rapid and radical change to
the economic system that sustains climate change is in fact
47

RATIONALITY AND COGNITIVE BIASES 47

achievable. Alternatively, framing can provide a way of coping

with a situation. For example, I’ve always felt that it is healthier
for students to see exams as a ‘challenge’ rather than something
to fear or just to ‘get out of the way’. (For similar examples, see
the contrast effect, above.)
2. Or framing serves the communicator’s needs by deliberately
obscuring aspects of reality in order to gain an audience’s
approval. This is how propaganda operates: ‘successful ideolo-
gies,’ says philosopher Mary Midgeley, ‘commonly make their
impact by hammering at a single image, or small group of
images, which expresses one side of the truth so vividly that
they fill the reader’s imagination, making it hard to remember
that there is any other’.11

Things become more interesting when both sets of needs are rel-
evant. In his Letter from Birmingham Jail, Martin Luther King, Jr.
is attempting to convince an audience, otherwise sympathetic to
his cause, of the value of non-violent civil disobedience. He has
been accused of being an ‘extremist’, a label he at first rejects, but
then accepts with a caveat. He lists various radicals and lawbreak-
ers from history and the Bible – including Jesus, Paul, Lincoln,
and Jefferson – who are foundational to American values. He then
reframes the issue:
The question is not whether we will be extremist, but what kind of extrem-
ists we will be.12

He has selected his argument and the examples he uses with great
care; these are extremists for ‘love’ rather than extremists for ‘hate’. It
is by no means a balanced picture, but at the same time it will pro-
vide a potentially helpful angle for an educated audience to engage
with his argument. Unlike propaganda, it is part of a respectful
dialogue.
As this example also illustrates, an important method of framing
is the use of figurative language. Metaphors, analogies, and allegories
try to explain or illuminate some aspect of the world by compar-
ing it with something recognisable to an audience. Arguments from
analogy work in a similar way, and this will be the subject matter of
Chapter 9.
48

48 RATIONALITY AND COGNITIVE BIASES

1.4 CONCLUSION: ARGUMENT AND RHETORIC

Rhetoric, as explained in the opening chapter, is the art of persua-
sive communication. Arguments also aim to persuade, but a good
way to distinguish between the two is that whereas arguments
appeal to System 2, rhetoric appeals to System 1. As we have seen
though, arguments were one of Aristotle’s three routes to persua-
sion, and most rhetorical communication (such as rousing political
speeches and sales pitches) does involve arguments. However, the
quality of the rhetoric is assessed on how persuasive it is, rather
than how strong the arguments are, and weak arguments are not
discouraged so long as the intended audience is convinced by them.
It is for this reason that it is important to learn about fallacies (weak
arguments that have the appearance of being strong), and to under-
stand our cognitive biases and shortcuts as we have been doing
in this chapter. These two sets of information will help protect us
from forms of persuasive communication that keep System 2 asleep
in the back.
It should also be recognised that rarely (if ever) can argumenta-
tion be entirely free from rhetoric; from what I will call ‘System 1
candy’. Features of its content or source will be persuasive in ways
that do not directly relate to the essence of the point being made, and
so on all occasions we need to maintain vigilance. The next chap-
ter considers the unavoidable presence of feelings (emotion, mood,
affect) in argumentation, and the ways in which these bias our judge-
ments. In the previous section, on framing, we reviewed the impor-
tance of Martin Luther King, Jr.’s choice of analogies. The part on
authority, likeability, and identity is a reminder that all arguments
have a source and a medium, and these will rarely be persuasively
neutral (see, as well, the ‘halo’ and ‘horn’ effects, to be discussed in
Chapter 7).
When we are practicing argumentation rather than rhetoric it
would be wrong to use these features of arguments to gain an advan-
tage. An awareness of them, however, can be used more legitimately if
it means the difference between someone listening to your argument
and someone switching off, or being open to it rather than remain-
ing antithetical.
49

RATIONALITY AND COGNITIVE BIASES 49

EXERCISES
1. From what you have read in this chapter, draw up a list of things
we could do to protect ourselves from the negative effects of
heuristics and biases. This will include ways to prevent poor
decision-making in groups; protection from those who will
deliberately deploy persuasion techniques to influence us; and
of course protection from our own motivated reasoning.
Some recommendations have been explicitly included in the
chapter and some are implied, but there will be others that
you will be able to think of yourselves based on the chapter’s
content.
2. A good way to gain a better understanding of these biases and
how they operate is to explore their use in contextualized
examples of persuasive communication. Several are mentioned
in this chapter, including speeches by Martin Luther King, Jr.
and Barak Obama. These fall under the neutral definition of
rhetoric, but Twelve Angry Men (see this book’s Introduction)
is full of instances of how dangerous or unconstructive these
biases can be. Other examples can be found across subsequent
chapters, but it is also straightforward enough – and advanta-
geous since you can choose ones that are especially relevant to
your situation – to find your own.

FURTHER READING

• Further investigation of biases and heuristics can be guided by the

in-text references in this chapter; most notably Kahneman (2012),
whose focus is on judgement and decision-making; and Cialdini
(2007, 2017) who explores how heuristics are exploited by profes-
sional persuaders. For methods of debiasing see Larrick (2004),
and for an evaluation of dual-process theories of reasoning see
Osman (2004, 2013), and Evans and Stanovich (2013).
• For a review of the confirmation bias see Nickerson (1998). A lot
of articles have now been written that connect confirmation bias
and online echo chambers, including Modgil et al. (2021).
50

50 RATIONALITY AND COGNITIVE BIASES

• For a review of the anchoring effect see Furnham and Boo (2011);
and there’s an excellent discussion of it in Kahneman (2012, Ch.11).
• For an informative analysis of the components of groupthink see
Baron (2005).
• Largely for reasons of space the ‘exponential growth bias’ is not
covered in this book (it also has fewer connections to other biases,
dispositions, and fallacies than the ones I have included). However,
and not least because it helps explain responses to the Covid
­pandemic, it is worth exploring. There are quite a few academic
articles linking the two topics, and a good introduction can be
found in Steven Pinker’s book Rationality (2021, pp.10–12).

NOTES
1 Roads to Freedom (Routledge, 1996), p.116.
2 There is a lot of helpful literature in this area, but one good guide to con-
ducting systematic reviews in healthcare see Khan, K.S., Kunz, R., Kleijnen,
J., and Antes, G. (2003) Five steps to conducting a systematic review. Journal
of the royal society of medicine, 96(3), 118–21.
3 For an excellent (and entertaining) example of negotiation skills, including
the art of concession-making, see Better Call Saul, Season 1, ep.2 (approx. 18
minutes in, but best to watch from the start to appreciate what’s going on).
4 For a transcript see www.theguardian.com/us-news/2015/jun/18/obama-
on-charleston-ive-had-to-make-statements-like-this-too-many-times. Joe
Biden made a similar point after the Robb Elementary shooting in Texas in
May 2022.
5 See, for example, https://climateoutreach.org/reports/ipcc-communications-
handbook/
6 I won’t cite specific sources, but there are many and they are easy to find in
an academic search engine.
7 www.ipsos.com/en-uk/public-hugely-overestimate-how-much-land-uk-
densely-built
8 See, for example, www.bbc.co.uk/news/uk-42554635
9 Discussed in Brown (1986), pp.165–7.
10 Klein, N. This Changes Everything (Penguin, 2014).
11 Wisdom, Information and Wonder (London: Routledge, 1991), p.48.
12 Available at: www.africa.upenn.edu/Articles_Gen/Letter_Birmingham.html
51

CRITICAL THINKING AND

EMOTIONS

I think I am smart unless I am really, really in love, and then I am ridicu-

lously stupid. (Taylor Swift, The Guardian, 30.5.15)

Emotions are typically seen as being irrational or non-rational

­features of our psychology; as impediments to accurate ­perceptions
and objective judgements. There is a significant amount of truth
in this, but perhaps not as much as we are inclined to think. In
­contemporary research emotions are often viewed as forms of heu-
ristics, assisting our more immediate judgements, and they can also
be viewed as ‘intelligent’ in a deeper sense as well.
This chapter explores the relationship between emotions and crit-
ical thinking. ‘Appeal to emotion’ as a category of argument, or as
a type of fallacy, is sometimes discussed in textbooks. I will work
towards a discussion of this idea, but prior to that take a wider look at
the relationship between emotion and knowledge. Specifically, I will
(1) offer a broad account of the nature of emotions; and (2) consider
the relationship between emotions and rationality. Then, after pro-
viding some tools for assessing arguments dealing with emotive sub-
ject matter, I will consider the influence of emotions on dialogues.

DOI: 10.4324/9781003247944-3
52

52 CRITICAL THINKING AND EMOTIONS

2.1 THE NATURE OF EMOTIONS

Emotions are central to understanding human motivation and
behaviour. They are basic to how we encounter ourselves and
the world and thus inevitably have an influence on thinking and
­decision-making. They are also complex and conceptually slippery,
distinguishable to an extent from desires and other types of feeling
such as moods and sensations, but also tied in interesting ways to
these and other features of our psychology. Although a book like this
one cannot possibly do justice to the many ways in which an inves-
tigation of emotions is important for how we understand ourselves
and others, I can at least endorse psychologist Richard Lazarus’s
view that:
From an emotional reaction we can learn much about what a person
has at stake in the encounter with the environment or in life in general,
how that person interprets self and world, and how harms, threats and
challenges are coped with. No other concept in psychology is as richly
revealing of the way an individual relates to life and to the specifics of the
physical and social environment. (1991, p.7)

Quite often they are unpleasant to experience, but they always sig-
nify drama – a need is or is not being met, and thus desire, value, and
purpose are always in tow and encoded in these complex feelings.
The account of emotions I provide in this section should largely
align with our everyday experience of them, and is supported by
extensive psychological and philosophical research in this area. The
components of an emotional response have been unpacked in terms
of cognitive, physiological, and expressive aspects, along with asso-
ciated action-tendencies, but for our purposes vital aspects can be
categorised under cognition and feeling.

COGNITION

Emotions are tied to beliefs about the world: to experience fear we

must interpret a situation as threatening; to feel disgust, what is before
us is repellent in some way; pride results from a personally relevant
achievement; and so on. Crucially they are beliefs that matter to us:
something has happened, and this is appraised to be positive or nega-
tive in some way. Anger, for instance, is generally speaking caused by:
53

CRITICAL THINKING AND EMOTIONS 53

(1) something frustrating us; and (2) this being perceived as unfair
or unjust. So, if someone at work gets an undeserved promotion, or
when we witness or read about a legal or political injustice, some
degree of anger is expected.There is an intimate connection between
emotions and desires (and thus values): if we did not care then certain
things happening or not would not bother us and there would be no
emotional response. Emotions, then, imply judgements, and as will
be discussed in upcoming sections, these can be of the rapid System
1 variety, or the result of slower processing.
Along these lines relatively simple and recognisable situations can
be formulated for all emotions. And since responses come in degrees
of intensity, this we take to be related to the extent to which the
occurrence affects the person (e.g., tripping on a paving stone nor-
mally causes less embarrassment than forgetting your uncle’s new
girlfriend’s name who you’ve met twice before). Lastly, notice that
the cognitive element of emotions says nothing about what is in
fact going on in the world, only what we believe to be going on. As
psychotherapists well know, to understand the logic of someone’s
emotional response we need to understand what they care about and
how they interpret a situation in light of these cares.

FEELING

Few would dispute that emotions are feeling states, but there is much
more debate about what kind of feeling this is. Many emotions are
accompanied by felt physiological changes (e.g., body temperature,
the coursing of adrenalin), but when we talk about feeling a cer-
tain way towards something, we are also referring to a quality of the
thoughts that we have towards the object of our emotion. If I am
experiencing guilt because I still haven’t returned my friend Sacha’s
call this is more than a simple judgement of how I have wronged
them, it is a distinctive feeling that is inextricably linked to such a
judgement.
Moods share qualities of the feeling component of an emotion,
but usually without the cognitive aspect – there is no specific thing
that is their focus (at least not consciously). Affect is somewhat ill-
defined, but also important.Very broadly it refers to a wide range of
feelings – including sensations, moods, desires, and emotions – that
54

54 CRITICAL THINKING AND EMOTIONS

are experienced either positively or negatively, and are thus motivat-

ing, and involve some degree of arousal (sensory alertness).
Seymour Epstein (2014, Ch.4) provides two further categories of
feeling: ‘unresolved feeling states’ and ‘vibes’. The former are ‘pre-
cursors to emotions’; states of arousal such as anxiety or excitement
that are not currently about anything but prepare us to react emo-
tionally should circumstances change.Vibes are more subtle and easy
to ignore, but tuning into them can provide important guidance; a
feeling – that a certain course of action is right or wrong, say – that
is the product of familiarity and expertise. In that moment it holds
information (‘intuitive knowledge’) that has not so far been con-
sciously articulated.Where an urgent response is called for this might
be acted on immediately, but in other circumstances it helps steer our
deliberations.

2.2 EMOTION AND RATIONALITY

One question that is posed by philosophers is ‘Are emotions rational?’
A common answer is that they are rational if they are appropriate
and proportional responses to what has happened (or is believed to
have happened). There are plenty of occasions when we regard an
emotional response as rationally justified, or where it would be irra-
tional for a person not to feel a certain way. About grief, philosopher
Martha Nussbaum says,
if a person believes that X is the most important person in her life and X
has just died, she will feel grief. If she does not, this is because in some
sense she doesn’t fully comprehend or has not taken in or is repressing
these facts. (1990, p. 41)

In such cases the emotional response expresses, or contains, the nature

and significance of the event – there has indeed been a loss and it is
a terrible one for the person in question. The grief that is felt is pre-
cisely this realisation, and in this sense this grief embodies the truth.
In a situation where what we thought to be the case turns out not
to be (e.g., the dog isn’t missing after all, but is found in a secluded
corner of the garden having a quiet moment with its bone), then to
the extent that emotions are rational, the emotion would go away,
perhaps replaced by a different one (such as relief). Emotions, then,
55

CRITICAL THINKING AND EMOTIONS 55

can be said to be rational when they correspond with our desires and
beliefs.
This line of thought can, however, lead to the claim that it is not
the emotion per se that is irrational, but the belief that causes it.
As philosopher Robert Solomon once put it: ‘emotions are not
­irrational; people are irrational’ (2003, p.235). Emotions only appear
irrational because we fail to understand what the person is think-
ing, and often what we think – as the previous chapter helps to
­demonstrate – is irrational. For Solomon, the emotions we feel when
we are being irrational simply correspond to these biases, or other
erroneous beliefs. So, when a small failure is pointed out to a narcissist
it is his false belief that he is generally superior to others that causes
his excessively strong reaction of anger. However, the claim here is
that the emotion – the strong reaction – is not the culprit, but rather
the distorted belief which underpins it. Put in this way, the emotion is
neither rational nor irrational, but better understood as non-rational.
This is of immense importance, but it is not the end of the story;
emotions have a reputation for being irrational because emotions and
beliefs do not always correspond. Consider the ‘red mist’ of uncon-
trolled rage, phobias, obsessive jealousy, or the dizzily disproportion-
ate rollercoaster of romantic love. Our emotions appear to make us
think and do bizarre things. Very strong emotions can lead to what
has been called ‘cognitive incapacitation’; a situation in which we are
unable to think clearly. Strong emotions in particular are associated
with:
• Exaggerated (disproportionate) thoughts and behaviour, such as
our perceptions of the other as ‘perfect’ or ‘the one’ when under
the spell of romantic love. In this chapter’s epigraph Taylor Swift
appears to concur. Attention tends to be narrowly focussed and
judgements tend to be black and white.
• Inappropriate or muddled thoughts and behaviour.
We find an example of this second type of distortion in a scene
from Rose Tremain’s novel Music and Silence. Set in 1629, landowner
George Middleton is throwing a New Year’s Eve party, and he’s a
very happy and relieved man. He is deeply in love with his fiancée
Charlotte, and has recently recovered, against the odds, from an oper-
ation to remove gall stones. In the kitchen the waiting coachmen,
56

56 CRITICAL THINKING AND EMOTIONS

now sloshed, have consumed a batch of the guests’ mince pies, and
George
knows nothing of all this, but he would nevertheless approve it, because
there is nothing, on this night, of which he is able to disapprove. Even
those neighbours of whom he is not particularly fond. When he looks at
them hopping in a jig or endeavouring to bow gracefully in a minuet his
heart forgives them their futile and irritating habits, their habitual dis-
putatiousness, their past attempts to marry him to their ugly daughters.
Indeed, he finds that he loves them. He even loves their daughters. He and
Charlotte pass from table to table and hands reach out to them, and they
seize these hands with an unconcealed show of affection. ‘[Charlotte]’
says George, ‘you have enabled me to adore the world!’ 1,2

Recall Solomon’s view that ‘emotions are not irrational; people are
irrational’. Should we say that George Middleton is irrational to want
to survive surgery and to marry Charlotte; or to feel relieved after
the operation? If not, then what’s irrational? Presumably it is the
effects of the emotion itself; an example of what is called ‘transferral
of affect’ where the feeling expands beyond its origins and attaches
itself to irrelevant and inappropriate objects and events.
Less well known but of great importance for critical thinking is
that mild emotions and other affective states have distorting effects
as well. Experiments have shown that a person’s emotional frame
of mind will affect their subsequent judgements. For example, if
individuals are induced into a positive or negative mood and then
asked to make judgements on crime or political figures, their mood
will condition the judgements they make.3 This effect is sometimes
referred to as emotional framing, and is a valuable persuasion tech-
nique (for more on framing in general, see Section 1.3). If a speech-
maker finds an angle on a topic that will predictably induce, say,
fear in their audience (e.g., a terrorist threat), they can then use the
resulting mood to elicit the response they want to other issues that
would be less persuasive were the audience not already in an anxious
frame of mind. For example, in many countries immigration is a con-
troversial topic that is associated with a range of emotions, includ-
ing anger, fear, pride, and compassion. An anti-immigration politician
will want to play on anger and fear, and will gain an advantage if
they can introduce the topic to an appropriately primed audience.To
put it another way, emotions and moods embody a confirmation
57

CRITICAL THINKING AND EMOTIONS 57

bias; when under their influence we are, often quite unconsciously,

inclined to seek out features of our surroundings that confirm our
feelings.

EMOTIONS AS HEURISTICS

If emotions cause irrationality in the cognitive sense explained above,

they can also assist more habitual forms of decision-making by serv-
ing as heuristics (referred to sometimes as the ‘affect heuristic’);
as speedy intermediaries between what goes on and what we need
and want. As will be discussed at greater length in the next sec-
tion, System 1 processing and emotional responses are closely linked.
Instinct and conditioning are triggered by certain happenings and
emotions embody this information.
In recent decades Antonio Damasio’s (2000) ‘somatic marker’ the-
ory has been highly influential.The idea is that our initial response to
many situations is an emotional one – broadly positive or negative –
and this ‘feeling of what happened’ then serves to guide our atten-
tion and the judgements we make. Whereas in the last section this
was viewed as problematically biased and something which leaves us
vulnerable to manipulation, Damasio’s position is that, because they
are hard to ignore, they help to capture and then focus our attention.
In this sense they become a vital component of rational action. I am
in the kitchen cooking when I hear my youngest son screaming in
the garden. My fear-based response is immediate; I drop the spoon
and I’m out the back door to see what’s going on. When I see that
he’s not hurt my antenna quickly switches and I see his older brother
chasing him with something gruesome he has found in the wood-
shed. Previous experience has mellowed any irritation towards my
son’s extreme reactions, and instead empathy is triggered and suitably
understanding words and actions follow.
Notice a couple of things about this, quite typical, situation. First,
I do not have a great deal of control over my thoughts and actions,
which is fine when the range of causes and effects encountered are
familiar, but not so helpful when a situation has quite a few novel
features. Second, and in accordance is a basic theme of this book, it
is important to appreciate the kinds of situations in which emotions
as heuristics are valuable guides to behaviour, and those in which
58

58 CRITICAL THINKING AND EMOTIONS

they are less so. Discussing the heuristic function of feelings in what
is generally an excellent textbook on emotions Oatley, Keltner, and
Jenkins use this example:
many of the judgements we make are often too complex to review all the
relevant evidence. For instance, a comprehensive answer to the question
of how satisfied you are with your political leader might lead you to think
about current environmental policy, the state of health care, unemploy-
ment and inflation rates, what is being done about global warming …
Given this complexity of so many important judgements, we often rely
on a simpler assessment based on our current feeling, asking ourselves
‘How do I currently feel about this person?’ (2006, p.265)

It is quite possible that what they have in mind here is someone

being asked this question in a street survey, in which case one’s feel-
ings used as a heuristic might be a reasonable basis for a response. In
other circumstances, however – such as a focus group or when decid-
ing who to vote for – feelings should be applied far more cautiously.
For instance, the general warmth I might feel towards the erstwhile
leader of a particular Scottish political party can function as a starting
point for investigating what is behind this, and the same goes for my
antipathy towards other leaders and parties. The heuristic becomes
a hypothesis that can lead me to both look closely at track records
and policy promises, and also at my own prejudices. Overall though,
if I am willing to fully investigate voting options then I need to be
dealing with the evidence on its own terms rather than relying on a
system designed to rapidly find coherence with existing beliefs and
attitudes.
The affect heuristic uses our feelings about a situation to guide
decision-making, and very much captures this function of emotion
and other forms of affect. In Paul Slovic et al.’s words it:
appears at once both wondrous and frightening: wondrous in its speed,
and subtlety, and sophistication, and its ability to ‘lubricate reason’;
frightening in its dependency upon context and experience, allowing us
to be led astray or manipulated – inadvertently or intentionally – silently
and invisibly. (2007, p.1349)

Switching to System 2 processing is clearly a capacity we have, but we

are also able to educate our heuristics so that quick responses to situa-
tions are better informed.This is usually not straightforward – coming
59

CRITICAL THINKING AND EMOTIONS 59

to appreciate something intellectually rarely directly translates into

a long-term change in our feelings about it (or other heuristic
responses) without some further work involved – but it is certainly
achievable.
To help explain how I will return to the ideas of Seymour Epstein,
a psychologist responsible for a more wide-ranging dual process
­theory than Tversky and Kahneman’s, encompassing emotional intel-
ligence and links to mental health and personality theory. Epstein, as
we saw in the previous chapter, refers to the experiential system
and the rational system. The broad idea is similar to the System 1/
System 2 distinction – the experiential mind is fast and automatic,
the rational mind slow and deliberate – but several more character-
istics of the experiential system are explored in Epstein’s model. A
core one is that it is ‘emotionally driven’ (2014, p.8). This means: (1)
that it seeks emotions and other types of affect that feel good; and (2)
that it relies on them as sources of knowledge to help navigate the
world. Emotion, and affect more generally, is the primary heuristic
of the experiential system, but there is an important added emphasis
from Epstein on how ignorant we can be of its source, presence, and
effects. For example, emotions that are unpleasant to experience or
that signal something that our ego-defences are protecting us from,
can be supressed or misinterpreted.The result is an intrusion into our
reflective thought that can be especially confusing and damaging to
our judgements (ibid, Ch.5).
A short case study can be taken from rapper and journalist Loki/
Darren McGarvey’s autobiographical social analysis Poverty Safari
(2017). The book takes its title from his response to a proposed
­government-funded research project in which the artist and activist
Ellie Harrison was to spend a year not travelling outside of Glasgow
to assess how a local focus, with its reduced carbon footprint, would
affect her well-being and her ability to do her job. The project was
named the ‘Glasgow Effect’ (after a study on Glasgow’s notoriously
low life expectancy) and represented visually by a bag of chips. Partly
because of poor communication, some Glaswegians took offence to
the project, and McGarvey sought to make sense of this upset. He
explained it in terms of how it represented class divisions and thus
exposed a deep seam of anger and frustration among the working-
class community, but in his commentary, by his admittance, he made
60

60 CRITICAL THINKING AND EMOTIONS

no effort to find out anything more about Harrison or the project

beyond what he had seen on social media. He made various assump-
tions on the basis of Harrison being middle class: as essentially a
spectator (on a ‘poverty safari’), she couldn’t possibly understand the
lives of working-class Glaswegians, and could not be truly committed
to the causes she claimed to support. In his words, he made her into
a caricature that was ‘easy to dismiss’ (ibid, p.208). (This is a common
feature of bad arguments known as a straw man argument – see
Chapter 4.) At this point McGarvey says this about his motivation:
Beneath all the rhetoric about class, and the insight about cultural ine-
quality and social mobility, ran a river of pure resentment which coursed
through me like a drug. This resentment … had clouded my mind pre-
cisely at the moment when I believed I was thinking most clearly. (2017,
p.207)

In their book on cognitive-behavioural coaching, Michael Neenan

and Windy Dryden (2020, p.9) use the analogy of a lock and key.
‘Core’ unconscious assumptions are primed to be released so long as
the right key is used, and various superficial contours of Harrison’s
identity and project were a good enough fit for McGarvey’s
judgement.
This highlights both the effect the experiential system can have on
the rational system, but also how the rational system can be blind to
this influence. This was not a spontaneous rant from McGarvey but a
thought-out argument containing some important points. In part it
was the result of the worthy aim of helping the middle classes make
sense of aspects of working-class experience, and therefore of why
the Glasgow Effect was received in the way it was. But he was seri-
ously off the mark in his attack on Harrison in a way that could easily
have been avoided by doing some basic research. It turned out that
Harrison is a ‘renowned social activist’ (McGarvey, 2017, p.209) who
supports similar causes and methods to McGarvey. Her principles are
deep and consistent and reflected in her lifestyle.
In other words, Ellie Harrison is exactly the kind of person
McGarvey should be admiring, and this is what brings him up short
and provokes some considerable soul-searching. He admits that, as
an artist himself, he was resentful of the attention Harrison’s project
was receiving, and as legitimate as the point about how working-class
61

CRITICAL THINKING AND EMOTIONS 61

marginalisation explains its reception might be, this is not what moti-
vated his criticism of the project and its creator. At the time he was
not consciously aware of his reasons, and as a result the efforts of his
rational system were corrupted:
I had used righteous anger as a smokescreen to conceal something more
self-serving. I had used the ‘working class’ as a Trojan horse to advance
my own personal agenda. And I did all of this while believing myself to be
well informed and deeply virtuous, unaware of how personal resentment
was subtly directing my thinking. (ibid, p.209)

In sum McGarvey’s recounting of this episode demonstrates the effects

emotions can have on the rational system. He comes to acknowledge
that some of the assumptions supporting his argument, and how he
chose to construct it, were influenced by his experiential system in a
way that reveals a dislocation between the way he wanted to see his
self, and the full truth. Anger is empowering and therefore a conveni-
ent mask for thoughts and feelings associated with vulnerability (such
as envy, or a recognition of our ignorance). His examination of his
angry response revealed not just this function, but a range of other
defences as well. Dogmatism and the confirmation bias are also pro-
tective, and for McGarvey honest self-reflection resulted in a hard-won
humility and greater self-understanding. In a Socratic fashion he starts
to question what else he is wrong about. He apologised to Harrison.
Importantly, and as this example demonstrates, Epstein sees the
experiential system as something that can be influenced by the
rational system, and partly for this reason is able to contribute ideas
to cognitive psychotherapy as well as experimental psychology.
Metacognition (see Chapter 3), self-awareness more generally, and
the use of judgement in deciding when to ‘obey the promptings’
(Epstein, 2014, p.87) of the experiential system is important for effec-
tive decision-making, and for mental well-being more generally. To
be able to do this we need to know about the existence of the expe-
riential system – unlike the rational system ‘most people are unaware
they even have an experiential system, let alone being aware of its
manner of operation’ (ibid, p.10) – familiarise ourselves with its char-
acteristics (not just in the abstract but through personal reflection),
and attain the habit of noticing its influence on our thinking. These
are all aspects of what is broadly referred to as emotional intelligence.
62

62 CRITICAL THINKING AND EMOTIONS

The more emotions are entangled with other elements of the

experiential system (such as deep-seated motivations, or beliefs
resulting from strongly reinforced experiential learning), the less trac-
table they will be. There are many instances where argumentation’s
engagement with the rational system is enough, but the application
of critical thinking cannot always change minds, even where argu-
ments are strong. There is an immense literature now – in fields like
behavioural economics – on what is called bounded rationality, or
motivated reasoning. While keeping sight of its immense impor-
tance, we must acknowledge the limitations of arguments for chang-
ing minds. Even where engagement with the experiential system
is essential for changing minds and behaviours, the rational system
invariably needs to be engaged as well.

2.3 EMOTION AND REFLECTION

Emotion and affect are intelligent in so far as they are generated by
what is currently important to us, and they can also serve as remind-
ers of what we might otherwise overlook, or not look at closely
enough. It is in this respect that they play a valuable role in critical
thinking.
As we have seen, being in a particular affective state will attune us
to features of a situation that are consistent with that feeling: anger
to injustice, sadness to loss, fear to threat, and so on. Handled rightly,
emotions can assist critical thinking by directing and holding our
attention so that we are better able to attend to relevant details of
the premises we are confronted with. In philosopher Michael Brady’s
words, they ‘motivate a reassessment or reappraisal’ of the events that
have given rise to them (2013, p.14). For example, feeling compas-
sion motivates empathy such that a moving image or vignette engen-
ders a deeper and more persistent empathetic engagement with the
full story behind the argument. As a consequence, we are better able
to know whether the compassion is justified.
However, even if the compassion is justified, true stories are often
complex, and the risk is that only one dimension is revealed by the
emotion-led argument, no matter what the accuracy of one’s new,
emotionally motivated understanding. In September 2015 images
of the drowned 3-year-old Syrian refugee Alyan Kurdi featured on
63

CRITICAL THINKING AND EMOTIONS 63

news media across the world, and had a galvanising effect on attitudes
towards the Syrian refugee crisis in some countries.
The image of the dead body – dressed like any other little boy –
picked up on a beach and in the arms of a police officer, was unbear-
able. There was some controversy about papers printing the image,
but the Independent in the UK justified it in this way:
The Independent has taken the decision to publish these images because,
among the often glib words about the ‘ongoing migrant crisis’, it is all too
easy to forget the reality of the desperate situation facing many refugees.4

Indeed, so the emotion prompts a more empathetic engagement

with the severity of a situation that drives people to take such risks.
And this degree of understanding is a vitally important aspect of
what we need to know in order to reach an informed decision about
how Europe (say) can best respond to the crisis, and therefore what
a given individual should be doing – who to donate money to, who
to lobby, and so on. However, it is only one aspect, and the risk of
emotion-led reflection is that the salience it affords distorts the argu-
ment. How, then, are we to counter this risk of imbalance?
In his discussion of what he calls appeals to pity (compassion is
probably the better word though) Douglas Walton (1989, pp.204–5)
makes the important point that although the emotion is a rational
response to the situation, we need to take care to not let this make
us follow blindly any recommendations for action – or we might
add, for wider interpretations – offered by the source of the image
or story. The insights the emotion facilitate need to be placed in a
broader perspective that is viewed more coolly in order to make
the right decision. He later discusses ‘critical doubt’ (Walton, 1992,
pp.268–9), the ability to detach ourselves from what is currently
compelling and assume a less biased perspective motivated by the
desire for truth and objectivity. Underpinning this shift is a deep
appreciation that in most circumstances we should avoid decision-
making whilst experiencing strong emotions.
Part of the answer is developing an intellectual appreciation of the
biases associated with emotions. This is the intention of this chapter,
but it is also clear that the influence of the experiential system is
such that knowing in this abstract sense is often not enough to influ-
ence attitudes and behaviours. In support of a more comprehensive
64

64 CRITICAL THINKING AND EMOTIONS

understanding of ourselves we need a range of the critical think-

ing dispositions, and this is the subject matter of Chapter 3. These
include the ability to be flexible in our thinking in a way that helps
facilitate open-mindedness and metacognition. Dispositions are
not simply knowledge, but a combination of traits and skills that
must be learned through practice as well as reflection. In this way
they become part of our experiential system and thus equip us with a
more comprehensive and reliable insight when facing ‘hot’ decisions.
Amidst a torrent of events in which the rational system is too slow
or too incapacitated to help, our implicit familiarity with the terrain
of emotion and cognition employs its own form of intelligence to
guide thinking and behaviour.

2.4 ARGUMENTS APPEALING TO EMOTION

Critical thinking and argumentation textbooks take a wide variety
of approaches to emotion. Some ignore the topic completely or dis-
miss emotions as irrelevant, and where they are included, their rel-
evance is interpreted in different ways. Zackary Seech (1993) and Jill
LeBlanc (1998), for example, have some interesting comments on
‘emotionally charged language’, and Bowell and Kemp (2015) con-
sider it under ‘rhetorical ploys’. Occasionally authors will include a
category of arguments known as appeals to emotion (e.g., Douglas
Walton 1989, 1992), and largely these are discussed in terms of indi-
vidual emotions. In theory this could include any emotion you care
to name, but analysis is weighted towards those that tend to be associ-
ated with commonly used rhetorical techniques: appeals to fear, pity
(compassion), guilt, pride, and anger (or indignation).
Unlike the argument forms you will learn about in Chapters 6–9,
appeals to emotion do not have a distinct structure. For example,
arguments from expert authority involve claims like: ‘X is an expert
in the field of Y’ and ‘X thinks Z about Y’. An argument that evokes
an emotional response, however, can take any form at all, and they
can also piggyback on other types of argument. Slippery slope argu-
ments (see Chapter 7), for instance, gain much of their persuasive
traction from inducing fear in audiences.
Most arguments will involve emotions to some extent. If we care
to generate an argument, the chances are it will be on an issue we
65

CRITICAL THINKING AND EMOTIONS 65

care about, and what we care about has inherent connections with
emotional responses (anger with injustice, compassion with suffer-
ing, fear with threat, guilt with responsibility, and so on). Also, in a
dialogue it is usually the case that we stand to lose or gain something
depending on how the discussion goes – we may perceive it as a
win or a loss, or a mutual exploration of an issue – and in each of
these variations our relationship with other discussants is potentially
affected.
The matter becomes more complex when we consider that emo-
tion is aroused by both the content of an argument and by the way
it is delivered. Language known to express emotion and arouse it in
audiences has been referred to as ‘emotionally charged’ (or sometimes
‘emotionally forceful’). Examples include extreme descriptors such
as ‘awful’, ‘dangerous’, ‘radical’, ‘wonderful’, ‘awesome’; and absolutist
terms such as ‘totally’, ‘completely’, ‘the greatest’, ‘the worst’, ‘always’,
‘never’, ‘eternal’ (forms of expression Donald Trump is fond of). As
we have seen, emotions tend to deal in crude categorisations, a sim-
plified and polarised world in which it is easier to make decisions.
Different audiences will be moved by different words and phrases
associated with linguistic and cultural norms – often connected with
moral values – and a persuasion specialist will know what these are.
And as we also found out in this chapter, ‘charged’ doesn’t necessarily
mean obvious.The unconscious priming of audiences can take many
forms, including the employment of words that will bias readers’ or
listeners’ subsequent judgements.
We should also remind ourselves that a range of non-verbal cues
are associated with emotions, including facial expression, posture,
gesture, and paralinguistic aspects such as volume, pitch, and pace of
delivery. We are typically highly sensitive to these cues when used by
others, and they are notoriously hard for us to control. Even when
trying hard to conceal our feelings, non-verbal signals will leak from
us, perhaps subtly, but still noticeably, for attentive others.
The preceding exploration of emotion and cognition is impor-
tant preparation for formulating, receiving, and assessing arguments.
We are constantly vulnerable to error if we lack self-awareness and
allow strong feelings to bias our judgements. And this is especially
important since some of those who try to persuade us will seek to
undermine our critical thinking in just this way. With this in mind,
66

66 CRITICAL THINKING AND EMOTIONS

a key distinction we need to make when assessing the emotional

effects of arguments is between those arguments that generate emo-
tion through form and content that is not relevant, or not directly
relevant, to the core message of the argument; and those that generate
emotions only as a natural consequence of the subject matter under
discussion.
In 2014 Tania Clarence was sentenced to an ‘indefinite hospital
order’ (confinement to a psychiatric hospital) rather than prison after
killing her three children, all of whom suffered from the severely
disabling condition of spinal muscular atrophy. Details of the case
show that Ms Clarence had been under extreme pressure since their
births and was suffering from depression. It was a tragic case, and
defence QC Jim Sturman argued that a hospital order would be the
‘just and compassionate’ sentence and, importantly for our purposes,
that ‘anybody who reads the evidence cannot fail to be moved’.5
What is implied here, however, is not that being moved by the events
will interfere with our ability to reason well, but that being moved
is an appropriate response to the circumstances. The compassion we
feel is an accurate measure of the tragic nature of the case, and the
tragic nature of the case has a profound bearing on how it is under-
stood legally and morally. Put another way, when the defence lawyer
explains the circumstances of the killings he is offering premises that
are relevant to the conclusion (that Tania Clarence is not responsible
for her actions). These premises arouse compassion in us, but the
lawyer’s aim is not to use this emotion to get away with a poor argu-
ment, it is a natural consequence of the subject matter under discussion. The
lawyer’s reference to being ‘moved’ becomes a shorthand for a just
understanding of the case.
In Twelve Angry Men Juror 8 also employs an appeal to compas-
sion, but in this instance it is deliberately misleading. The other jury
members are reluctant to even discuss the case, so Juror 8 resorts
to a heartfelt description of the 16-year-old defendant’s tragic life
circumstance – born in a slum, mother died when he was young,
father served time in prison, time spent in an orphanage – claiming
this is a reason to at least give the possibility of his innocence some
consideration. It is quickly, and rightly, pointed out that the boy’s
upbringing is irrelevant, what matters is he had a fair trial and that
67

CRITICAL THINKING AND EMOTIONS 67

their decision is based objectively on the evidence provided: Juror 8

is called out on his poor argument. (It is worth pointing out here that
in this situation the compassion-inducing story serves a wider truth-
seeking purpose.The tone the story brings to the room seems to help
the two factions work through an impasse, and a more constructive
route is subsequently followed.)
As another kind of example, consider the difference between the
student who calmly asks for an extension for their assignment dead-
line because of a family-related crisis, and the student who burst
into tears in my office while telling me that he can’t submit his essay
because his computer broke. Of course, this second example might
not be designed to be manipulative, but even if it isn’t, I run the
risk of being softened up by it. By ‘softened up’ I mean the compas-
sion generated by the presence of someone in tears makes me overly
receptive to the unfortunate facets of his situation, and therefore
more likely to be biased in his favour when hearing his story and
making a judgement on an extension.
In order to better protect ourselves from the problematic influence
of emotions in arguments, we need to ask these questions:

Q1: Is this an argument that has generated strong emotions in me

(whether through its content, delivery, or both)?
Q2: Do the emotions arise as a natural consequence of the subject
matter of the argument, or are they disproportionate or incon-
gruent in some way?
For example, have they been generated by features of the inter-
action that are irrelevant to the argument such as a person’s
appearance, or partially relevant such as an engrossing story
used as an example? Have immediate or instinctive emotional
responses been evoked (such as by the sight of someone in
tears)? Have I overreacted to features of an argument because
of my own sensitivities or unresolved issues (as with the Darren
McGarvey example, above)?
Q2a: If the emotions are disproportionate or incongruent, it can also
be constructive to ask: Is the arguer attempting to manipulate
me by invoking strong emotions instead of presenting strong
arguments?
68

68 CRITICAL THINKING AND EMOTIONS

Q3: If emotions arise as a natural consequence of the subject matter

of the argument, are they representative of a balanced view of
the issue?
Q4: If so, is it the strength of the argument (including insights
assisted by the emotions evoked), rather than the feelings them-
selves (i.e., the affect heuristic) that leads me to accept or reject
the conclusion?

Each of these questions can be applied to ourselves as well in order

to facilitate reflection on how the content and delivery of our own
arguments might affect their reception.
This is an example of a set of ‘critical questions’, and as we explore
various argument forms in this book (from Chapter 6 onwards) you
will become familiar with lists like this. Their purpose is to help us
assess the strength of arguments which take a particular form.

2.5 EMOTION AND DIALOGUE

The following chapter (on critical thinking dispositions) concludes
with a list of rules for constructive dialogues. They include the will-
ingness to listen carefully and not misrepresenting other people’s
positions, avoiding attacking the arguer rather than their arguments,
and remaining civil. With each of these the presence of strong emo-
tions makes it much harder to achieve. Milder emotions can be con-
structive, but the broad lesson is the need to be highly vigilant of
ourselves and others in this respect. Some basic wisdom on the issues
is summarised below (inspired, in part, by insights from Seech, 1993;
LeBlanc, 1998; Walton, 1989, 2006; and Hargie 2017):
• Under the influence of strong emotions we are less good at listen-
ing, and less inclined to listen. The experiential system is geared-
up for quick judgements rather than concentrated deliberation.
• Strong emotions can be contagious: emotion begets emotion,
the experiential system engages directly with other experien-
tial systems, and the rational system is disempowered. In some
cases, other discussants will see this coming and disengage from
the discussion before it becomes heated. Aside from unproductive
outcomes, negative emotions can be stressful to experience and
69

CRITICAL THINKING AND EMOTIONS 69

generally disruptive, so many of us are inclined to avoid them.

From the previous example, Darren McGarvey observed too that
his ‘less adversarial’ attitude (2017, p.210) led to those of a different
view becoming more conciliatory towards him.
• Emotionally charged language can make the arguer seem extreme,
dogmatic, and perhaps out of control. It implies an all or nothing
conviction and even if we do not really mean it, it can commit
us to an extreme or inflexible position (Seech, 1993, pp.16–18). In
other words, emotion implies conviction, so emotionally charged
language may inadvertently lead us to defend our position more
dogmatically than we otherwise would, just so that we remain
consistent with our form of expression.
• Emotions are a distraction. For example, they may encourage us
to listen in order to be entertained (‘appreciative listening’) rather
than to be critical. Or if someone is angry we might become
angry back, or fearful, or focus on how to manage their anger
rather than progress the discussion. (Something to look out for
in ourselves and others: if someone doesn’t want to continue a
discussion but doesn’t want to admit this they can deliberately, or
unconsciously, use emotions for just this purpose.)
For these reasons and others, emotions are a minefield. In a milder
(contained, controlled) form they can communicate conviction that
can motivate constructive discussion, but mild or strong we can easily
be manipulated by them, and not just manipulated by the emotions
expressed by other people, but by our own emotions as well.
To keep things on track be prepared to take time out to calm down.
If someone else is becoming emotional, telling them to calm down
is often counter-productive, so a better approach is to be empathetic
and allow them to ‘ventilate’ (Hargie, 2017, p. 197) before continuing.

2.6 CONCLUSION
It is important for an approach to critical thinking that emphasises
cognitive biases, character dispositions, and constructive dialogues to
pay some attention to emotions. In this chapter I have attempted to
formalise some of the, often familiar, ways they are linked to irrational
thinking and biased perceptions, but also to the broad functioning
70

70 CRITICAL THINKING AND EMOTIONS

of the experiential system and its distinctive forms of intelligence.

Emotions have a heuristic function that can direct us to features of
situations that might otherwise be overlooked, and they can act as
reminders of how our cares are prioritised. Emotion, and other types
of affect, are basic to our psychology: treated with due caution, their
potential for impeding objectivity can be reduced, and their potential
for assisting us towards it embraced.

EXERCISES
1. In the introduction to this book I named the film and play
Twelve Angry Men as the best text I know for applying the ideas
of critical thinking. It is essentially a ninety-minute dialogue in
which the twelve jury members argue about a murder case. As
the name suggests, everyone gets angry at some point – even
the impressively temperate Juror 8 – and among many learning
opportunities it affords, it demonstrates the effects of emotions
on dialogue.
Juror 8’s appeal to pity was discussed in this chapter, but
there are countless further examples of emotion playing either
a constructive or (most likely) destructive role in the delibera-
tion. Find examples of:
• Emotion playing a constructive role
• Displays of emotion causing emotional responses in oth-
ers (with positive or negative outcomes)
• Emotion causing others to disengage
• Emotional expression reflecting negatively on the arguer
• A person’s emotional delivery forcing them into a more
extreme position than they might have otherwise settled
for
• Emotion as a distraction, or as a means for avoiding truth
• Constructive or unconstructive handling of emotions in
others
2. In line with what has been discussed in this chapter, monitor
your emotional responses in various contexts, and especially
dialogues that involve the exchange of arguments.
71

CRITICAL THINKING AND EMOTIONS 71

FURTHER READING
Many important sources are found in this chapter, all of which are
in the reference list at the end of the book. I must though highlight
Epstein (2014). Also, for a very good philosophy book on emotion
and rationality see Brady (2013); and an equally good psychology
source is the chapter ‘Emotions and Cognition’ in Keltner, Oatley,
and Jenkins (2013).

NOTES
1 Tremain, R. (2000) Music and Silence.Vintage, p. 299. Reproduced by permis-
sion of The Random House Group Ltd.
2 There’s a nice moment in Jonathan Franzen’s recent novel Crossroads that
presents a similar, though less euphoric, transfer of affect (see pp.44–5).
3 For an experiment along these lines, see DeSteno et al. (2004).
4 Withnall, A. (2015) [Article]. Available at: www.independent.co.uk/news/
world/ europe/if-these-extraordinarily-powerful-images-of-a-dead-syrian-
child-washed-up-on-a-beach-dont-change-europes-attitude-to-refugees-
what-will-10482757.html.
5 www.irishexaminer.com/world/arid-30645119.html
72

CRITICAL THINKING AND

DISPOSITIONS

It is not enough to have a good mind, rather the main thing is to apply it
well. (Rene Descartes, [1641] 1968)
My husband says I’m overly sensitive to criticism. BUT WHAT DOES
HE KNOW? WHO IS HE TO CRITICISE ME? HE’S LUCKY I DON’T
DIVORCE HIM! (Sacha T. Burnstorm, pers. comm.)

Learning about heuristics and biases is not the same thing as learn-
ing to think in ways that avoid their negative effects. Critical think-
ing is something that we need to do, not just know about. However,
the doing is particularly challenging because of the automaticity of
System 1 thinking. We need to develop the habit of critical think-
ing in order to counteract the powerful tendency to think quickly
in situations where thinking slowly would be more beneficial. The
skills of critical thinking thus aim to instil more constructive habits
of thought.
A further level is added, though, in which these skills are moti-
vated by dispositions to think critically. To be a critical thinker in
this sense is having what Ennis calls an ‘inclination’ to think critically.
According to this approach we don’t just ‘do’ critical thinking, but
‘become’ a critical thinker. ‘Becoming’ a critical thinker, however,
does not mean some cultish, full-blooded transformation in your
DOI: 10.4324/9781003247944-4
73

CRITICAL THINKING AND DISPOSITIONS 73

personality, as seems to be suggested by some theorists. Harvey Siegel

(1988, p.41), for example, says that ‘when we take it upon ourselves to
educate students so as to foster critical thinking, we are committing
ourselves to nothing less than the development of a certain sort of
person’. Learning critical thinking can certainly change the way we
approach our beliefs about ourselves and the world, and the ways in
which we make decisions, but this is usually about shifts in emphasis
and the nurturing of existing dispositions rather than the emergence
of dominant traits or attitudes. For each individual these tendencies
will mix in with the rest of their personality so that no two critical
thinkers will be recognisably ‘alike’ in any generalised sense. Instead,
what we would expect are similarities in certain dispositions they
exhibit in deliberations and other situations in which arguments are
exchanged.
Two further points should be highlighted before proceeding. The
first is that we must be careful not to see these dispositions as simply
enabling critical thinking in a practical sense, but as motivating it
as well. Critical thinking is valued, and should we find we are not
thinking critically on an occasion in which we should, then we are
emotionally affected by this, disappointed in ourselves.
The second point concerns the distinction between encourag-
ing dispositions that will tend to make us better critical thinkers,
and seeing critical thinking as a discipline that will foster these dis-
positions. Writings in this area tend to be framed in terms of the
former, but the latter is implied as well (see, e.g., Aberdein, 2020).
Therefore, answers to Robert Ennis’ question, ‘What are the critical
thinking dispositions?’ (1996b, p. 168) tell us: (1) which characteris-
tics we need to develop in order to have a readiness for, or be pre-
disposed towards, thinking critically; and (2) which characteristics
can be acquired as a result of learning critical thinking knowledge
and skills, and from learning about the personal and social value
of these things. These will of course be mutually reinforcing, but
from an educational point of view the second should be the primary
aim. Education will develop dispositions in students whether this
is intended or not, but this largely occurs as part of the intellectual
and social practices they are being inducted into, rather than as a
separate aim. For this reason, teaching critical thinking dispositions
74

74 CRITICAL THINKING AND DISPOSITIONS

in order to be a better critical thinker seems to put the emphasis in

the wrong place. Instead, the knowledge, skills, and values of criti-
cal thinking are taught, and we then expect certain dispositions to
develop out of this environment.

3.1 SOME CORE CRITICAL THINKING

DISPOSITIONS
There have been quite a few attempts to formulate lists of critical
thinking dispositions, including Richard Paul (1995, Ch.13), Robert
Ennis (1996a, pp.368–9), and Peter Facione (2006).The discussion that
follows is informed by several of these, and by other scholars interested
in what are known as ‘epistemic virtues’. Dispositions (or virtues) tend
to be highly interdependent, so that possessing one requires possessing
many others as well. Any list of critical thinking dispositions has the
potential, therefore, to be very long indeed. To avoid this, I will focus
on the ones that have the most direct influence: commitment to truth,
open-mindedness, modesty, self-knowledge, metacognition, and what
I am calling ‘dialogical dispositions’.Where appropriate, however, I will
also indicate related or complementary dispositions.

COMMITMENT TO TRUTH

According to Ennis (1996a, p.9), ideal critical thinkers will:

care that their beliefs are true, and that their decisions are justified; that
is, care to ‘get it right’ to the extent possible, or at least care to do the
best they can.
The critical thinker has a commitment to the value of truth, and thus
to the appropriate processes for reaching the truth. A ‘justified’ belief
here refers to one that is established through rational enquiry – the
use of reason and evidence. It is important to recognise that we can
care about truth, but not a commitment to rational enquiry as a
means of attaining it. Instead, we could regard notions like faith, or
uninformed feelings or intuitions as roads to truth. In some domains
(e.g., religious and spiritual beliefs) this might be appropriate, but
even here the critical thinker would need to provide an argument for
why these domains require a different type of knowledge.
75

CRITICAL THINKING AND DISPOSITIONS 75

One thing we need to be aware of, in ourselves and others, is the

desire to be right posing as the desire for truth. Many of the disposi-
tions discussed below have a bearing on this distinction. The desire
to be right implies competitiveness rather than a commitment to
truth, and it will leave us especially vulnerable to the confirmation
bias and other forms of motivated reasoning. Juror 4 in the film
Twelve Angry Men is the reasoned voice of the guilty vote in that he
is intelligent, calm, and willing to look at the evidence on its own
merits. But unlike Juror 8 he seems to also have excessive pride in
being right, and this need appears to dampen his inquisitiveness and
makes him unwilling to go to the lengths of Juror 8 in scrutinising
the arguments put forward by the prosecution.

OPEN-MINDEDNESS

Since seeking truth requires us to listen to the views and reasoning

of others, and an appreciation of the fallibility of our own beliefs
and convictions, then open-mindedness must be a fundamental dis-
position of the critical thinker. Open-mindedness is a corrective
to the confirmation bias. It does not mean that we should have no
opinion on an issue in order to deal with it fairly, but it does mean
that we are able to bracket – put aside – this opinion in order to
more objectively assess its worth. Instead of looking for premises
that support a conclusion already reached, we should be looking
closely at the premises that present themselves and then at the con-
clusions these might lead us to. This is what John Dewey calls the
‘attitude of suspended conclusion’ (1910, p.13), or as Johnson and
Blair (2006, pp.50–1) put it:
To engage in [argumentation] … is to admit in principle the possibility
that your premises do not constitute good grounds for your conclusion
(even though at the moment you think they do).
A good critical thinker understands the limitations of their individual
perspectives and the value of other perspectives – and thus the value
of dialogue – as a way of opening their mind. Often other people
will have positions and arguments that we had not thought of before
listening to what they are saying. Open-mindedness entails a will-
ingness to change one’s mind, either in the direction of another’s
76

76 CRITICAL THINKING AND DISPOSITIONS

view, or towards a new conclusion not previously considered by the

discussants. But, says physicist David Bohm in his book On Dialogue,
such communication can lead to the creation of something new only if
people are freely able to listen to each other, without prejudice, and with-
out trying to influence each other. … If, however, two people merely want
to convey certain ideas or points of view to each other, as if these were
items of information, then they must inevitably fail to meet. For each will
hear the other through the screen of his own thoughts, which he tends to
maintain and defend, regardless of whether or not they are true or coher-
ent. (2004, p.3)
Open mindedness is hard to achieve because we must bracket, not
just our current belief in some abstract sense, but also the conviction
that will typically accompanying it. In order to truly listen, we need
to be calm. And as already indicated, carefully attending to what the
other is saying is a fundamental requirement for critical thinking.
In the passage from Bohm just quoted, he also states that each
participant in a dialogue
has to be interested primarily in truth and coherence, so that he is ready
to drop his old ideas and intentions, and be ready to go on to something
different, when this is called for. (2004, p.3)
The critical thinker recognises that beliefs are often provisional; open
to being disconfirmed by subsequent evidence and argument. If
open-mindedness is a willingness to change our mind, then flex-
ibility of thought is the ability to do so. Both are challenging. In
the case of flexibility, the firmness of commitment that is needed
to act is psychologically opposed to the ‘openness to being wrong’
that critical thinking demands. For this reason, many other philoso-
phers writing in this field – such as Valerie Tiberius (2008, Ch.3),
and Douglas Walton (1992, pp. 267–70) – recognise and support the
value of trying to be as flexible as possible in this way. The critical
thinker must try to be highly tolerant of – function well in the world
despite – uncertainty.
Endurance is also required in circumstances where open-minded
decisions need to be made. As we know, System 2 thinking is
energy-sapping, slows us down, and is characterised by John Dewey
as ‘mental unrest and disturbance’ (1910, p.13). Facing dilemmas, we
need to have the fortitude to maintain what psychotherapist Irvin
77

CRITICAL THINKING AND DISPOSITIONS 77

Yalom (1980, p.312) calls ‘simultaneous ambivalence’. This is the

result of remaining clearly focussed on the for and against of both
(or all) the options available, rather than letting one of them domi-
nate our attention and that way incline us to a less troublesome, but
biased, decision.
A disposition of significant interest to this discussion is ‘need
for cognitive closure’. It has been thoroughly studied by person-
ality and social psychologists in recent decades (e.g., see Webster
and Kruglanski, 1994, 1997) and refers to a desire for clear-cut and
quickly reached answers to questions and problems. A person with a
high need for cognitive closure has low tolerance for ambiguity and
uncertainty when making decisions. Low tolerance can be useful in
situations that demand quick decisions, but can be problematic in
ones where complexity demands and time constraints allow for more
careful deliberation. It can lead to inappropriate urgency – ‘seizing’
on available answers; and once seized, to rigidly maintaining a posi-
tion for the sake of the closure it is providing.
Situational variables – especially time constraints – play their part in
predicting our need for closure, but in a discussion of critical thinking
dispositions its status as a feature of our character is of most relevance.
In this respect it is correlated with closed-mindedness, inflexibility,
and overconfidence; increased susceptibility to a wide range of biases
(including anchoring effects and the stereotypic judgements associ-
ated with the representativeness heuristic); and to aspects of social
interactions conducive to suboptimal dialogical behaviours such as
decreased empathy and elements of groupthink. Understood from
a dispositional perspective, the aim is to strike a balance between a
need for cognitive closure that is so strong it is unable to adapt to the
requirements of a situation, and at the other extreme a reluctance to
ever settle on a view or course of action.
Flexibility operates at another level as well. A vital feature of a
fulfilled life is substantial time spent being ‘unreflectively absorbed
by what we value’ (Tiberius, 2008, p.67). This is a different psy-
chological mode to critical thinking, and is at odds with it – one
will tend to get in the way of the other. And since both this type
of absorption and reflective thinking are so important we can see
how being ‘ready’ (to use Bohm’s expression) to move between the
two is important.
78

78 CRITICAL THINKING AND DISPOSITIONS

MODESTY

Modesty (or humility) is primarily understood as possessing accurate

perceptions of ourselves and the status of our beliefs in relation to
others. The exaggerated pride in being right we identified in Twelve
Angry Men’s Juror 4 is the opposite of modesty. This is a person for
whom it is important to see himself as better than others, whereas in
the modest or humble person we find a recognition of and desire for
equality. The upshot of modesty is less egocentricity in discussions,
and thus a greater chance of open-mindedness.
There is a further reason why modesty is important. Critical think-
ing is empowering in terms of the insights it provides, and how it
improves our ability to successfully interrogate the arguments of oth-
ers. Since these are not insignificant abilities, they have the potential
to create a sense of superiority in the learner. Modesty is a correc-
tive to this. It will incline us to realise that critical thinking is just
one among many important practices, and that virtually anyone can
become a better critical thinker if they commit themselves to learn-
ing it. Perhaps most important of all is the recognition that if people
are not schooled in this way it does not follow that: (1) they are
unintelligent; (2) they do not have worthwhile beliefs; or (3) they are
not worthy or respect. (For more on this kind of issue, see ‘Don’t be
a smartarse’, below.)

SELF-KNOWLEDGE

Knowledge of one’s self in part comes from an appreciation of the

strengths and frailties humans share, including the intellectual frailties
that are of interest to critical thinking. It also comes from the spe-
cific content of what we believe and feel. This will include relatively
superficial and practical knowledge, but also deep personal attachments
to values and worldviews (such as religious, political, and ethical beliefs)
that can be an impediment to open-mindedness.
These attachments are part of what it is to be a person, and though at
least on occasion we should be prepared to question them, the time and
place of such evaluation requires much care on behalf of the believer
and those they engage with. Important for critical thinking, however,
is that we are aware of what these deep-seated commitments are.
79

CRITICAL THINKING AND DISPOSITIONS 79

One reason is that it allows us a clear choice about what we do

and do not want to apply scrutiny to, and to put out there for criti-
cal appraisal by others. Also, foundational beliefs will permeate many
of our other beliefs, so recognising what our commitments are will
enable clearer explanations of why we hold the positions that we do.
(For further discussion of these ideas see Walton,1992, p.255.) Two
brief examples are someone’s views on abortion being premised in
their Catholic faith, and support for Brexit being an expression of a
belief in Britain’s inherent distinctiveness from continental Europe
(rendering arguments about its economic cost irrelevant).

METACOGNITION

Closely related to self-knowledge is the disposition to be metacogni-

tive. The most specific meaning of this term is an awareness of our
thought processes, exemplified by ideas like ‘alertness to loss of con-
trol of one’s thinking’, and ‘the impulse to stand back and take stock’
(Perkins, Jay, and Tishman, 1993, p.8). It is not about formal knowl-
edge of our cognitive biases, but rather the disposition to monitor
and assess the quality or mode of our thinking in different situations.
An important aspect of this disposition concerns knowing when
to think critically and when not to. For example, Perkins, Jay, and
Tishman refer to the importance of the ‘detection of complex think-
ing situations’ (ibid), which will include knowing when to switch
from System 1 to System 2 thinking.
David Bohm discusses the ‘blocks’ we have that make us una-
ware of some of the contradictions in our beliefs, and therefore less
open-minded. To help understand these blocks, he recommends we
develop sensitivity to our emotional responses to certain topics of
discussion. ‘If one is alert and attentive,’ he says,
he can see for example that whenever certain questions arise, there are
fleeting sensations of fear, which push him away from the consideration
of these questions, and of pleasure, which attract his thoughts and cause
them to be occupied with other questions. So one is able to keep away
from whatever it is that he thinks may disturb him. And as a result, he
can be subtly defending his own ideas, when he supposes that he is really
listening to what other people have to say. (2004, p.5)
80

80 CRITICAL THINKING AND DISPOSITIONS

With reference to ideas discussed in the previous chapter, we might

call this the development of an intelligent affect heuristic.
We should not lose sight, however, of how these considerations
are part of a broad sweep of activities and practices making up a life,
including ones where metacognition is simply not welcome or nec-
essary: painting or playing sport for pleasure, unselfconscious dancing,
and an evening with a Stranger Things boxset come to mind. Many
such occasions can be sought out and (reflectively) worked into the
composition of one’s life. It might require critical thinking and its
associated dispositions to determine and facilitate these happenings,
but many of the resulting experiences are then justifiably pressing
metacognition’s snooze button.

DIALOGICAL DISPOSITIONS

Critical thinking does not necessarily occur in the context of a dia-

logue, but it usually does. Written arguments (in academic journals,
opinion pieces in news media, social media forums, and the like)
are invariably a response to other arguments, and are responded to
in turn. Arguments presented in spoken, and especially face-to-face,
dialogue can be some of the most persuasive, and are of course found
in multiple professional, personal, political, and legal settings.We offer
arguments to convince others of our position; arguments provoke
questions and counterarguments, and via this process the open-
minded, self-aware person should be able to edge closer to truth
about the issue under discussion. Good quality dialogues are thus
profoundly important, and so the ability to conduct them construc-
tively has equivalent importance. In part this is a matter of knowing
and applying certain rules (see below), but there are also disposi-
tions that facilitate this process and embody its value and significance.
These include a genuine desire to listen to others’ positions, and a
desire to present your own position as clearly as possible. And this
means providing not just your conclusion, but the reasons supporting
that conclusion as well.
I will discuss some important dialogical dispositions – courage,
staying focused, respect for others, ‘deliberative wit’, and not being
a ‘smartarse’ – before providing some basic rules for constructive
dialogues.
81

CRITICAL THINKING AND DISPOSITIONS 81

COURAGE

One reason we might be reluctant to be clear about our grounds

for holding the beliefs we do is fear of these beliefs being cast into
doubt. This is one reason why an important dialogical disposition is
courage. Another is that critical thinking is about independence of
thought and thus taking responsibility for one’s convictions, and this
can sometimes mean standing alone in the face of various forms of
authority (see Chapter 6) or significant social pressure to conform
(see Chapter 7). Social media is of course notorious for unfiltered
and unjustified abuse, and university professors don’t always rise
above this. One academic researcher writes about how she came into
the 2015 UK general election as a floating voter who raised ques-
tions about the policies on all sides, but who was confronted ‘time
and time again’ by,
Posts from my peers packed full of expletives implying that I was bigoted
for even doubting the Labour or the Green economic approach.1
There are two types of reprimand that the person challenging a
group norm can face: one relating to the content of their opinion,
the other to the process of speaking out itself. Speaking out can be
perceived as problematic for various reasons, including ‘rocking the
boat’ (destabilising an established, possibly hard won, equilibrium). In
certain circumstances or at certain times critical thinking, or at least
its expression, is not appropriate, and avoiding ‘boat rocking’ could be
one of them. However, as the previous discussion of the characteris-
tics of groupthink demonstrates (see Chapter 1), this is often not a
judgement call that is easy to make.
Another consideration that affects motivation for critical thinking
is that it exposes us to the dark underbelly of existence. By this I do
not mean human deviousness or folly, but two of the fundamental
existential concerns that continually and inevitably haunt us: that
we must take responsibility for our decisions and that there is no
preordained order to our lives – i.e., no final set of truths or essential
self to be discovered. At a cultural level Kant (1963) meant some-
thing like this when he described the Enlightenment as an emer-
gence from an ‘tutelage’ that is ‘self-imposed’, not by an inability to
reason, but by a lack of ‘resolution and courage’. Critical thinking
gains much of its significance from a profound freedom that comes
82

82 CRITICAL THINKING AND DISPOSITIONS

with the understanding that existence has no ultimate answer or pur-

pose. Words like ‘active’ and ‘judgement’ serve as reminders that in a
very important sense human existence is what we choose to make
of it. There is a resoluteness, perhaps even an excitement, associated
with this recognition, but also an anxiety, and it is this anxiety that
can make critical thinking off-putting. It is strangely disquieting.

STAYING FOCUSSED

In dialogues it is very easy to become sidetracked, so the critical

thinker always tries to stay focussed on the overall point of the dis-
cussion. Losing focus can happen by accident, but it can also be the
result of tactics employed by an arguer who feels they are losing,
or who wants to end the discussion prematurely. Fallacies associated
with losing focus include ad hominem arguments (see Chapters 6
and 7), red herrings (see Chapter 10) and straw man arguments
(see Chapter 4).
Part of the art of staying on track is asking the right questions at
the right times. Ennis (1996a, pp.373–5) identifies distinct types of
questions:
‘Clarification questions’ like:
• ‘Would you say a little more about that?’
• ‘What do you mean?’
‘Main point’ questions like:
• ‘Let me see if I have this right. Is this your main point …?’
• ‘I’m afraid I don’t quite see what you’re driving at. Could you say
a little more about it?’
‘Reason-seeking’ questions, or requests, like:
• ‘Perhaps you could elaborate on why you believe that.’
And ‘relevance’ questions like:
• ‘How does that support the conclusion?’
• ‘Are you assuming that …?’
Other terms for ‘staying focussed’ might be persistence or persever-
ance, and it is noteworthy that John Dewey includes the former in
83

CRITICAL THINKING AND DISPOSITIONS 83

his definition of critical thinking: ‘Active, persistent, and careful con-

sideration of a belief … in the light of the grounds which support it’
(1910, p.6). Persistence is part of the courage discussed above, but it
can also be valuable when faced with an absence of clarity or deliber-
ate attempts to divert the discussion.

DELIBERATIVE WIT

The notion of deliberative wit, as explained by Scott Aiken and

Caleb Clanton (2010), does not refer to humour, but more to
‘having your wits about you’. In a dialogue it concerns timely
contributions and avoiding being dull, for example through lec-
turing others. Instead, the points others make are constructively
developed, and because you have pre-empted challenges to your
own arguments, your responses are more likely to be articulated
without the sort of delay that could see the discussion move on
without you.
Since, however, this takes us into quite a sophisticated level of
argumentation, it runs the risk of creating an imbalance between
discussants based on their experience. For this reason it needs to
be mitigated by the presence of other virtues – notably respect and
humility – ­considered in this chapter and also stressed by Aiken and
Clanton.
An example of an insight that the person with deliberative wit
might develop – to do with where to place your conclusion in an
argument – is found in Roger Fisher and William Ury’s book on
‘principled negotiation’:
In talking to someone who represents a construction company, you
might say, ‘We believe you should build a fence around the project within
forty-eight hours and beginning immediately should restrict the speed of
your trucks on Oak Street to fifteen miles an hour. Now let me tell you
why....’ If you do, you can be quite certain that he will not be listening to
the reasons. He has heard your position and is no doubt busy preparing
arguments against it. He was probably disturbed by your tone or by the
suggestion itself. As a result, your justification will slip by him altogether.
If you want someone to listen and understand your reasoning, give your
interests and reasoning first and your conclusions or proposals later. Tell
the company first about the dangers they are creating for young children
84

84 CRITICAL THINKING AND DISPOSITIONS

and about your sleepless nights. Then they will be listening carefully, if
only to try to figure out where you will end up on this question. And when
you tell them, they will understand why. (1991, p.29)

RESPECT FOR OTHERS

Among critical thinking scholars there is some disagreement about

whether respect, or care, for others should be counted as a critical
thinking disposition. Peter Facione (voicing the view of the majority
of scholars at the time) says:
Good critical thinking has nothing to do with any given set of cultural
beliefs, religious tenets, ethical values, social mores, political orienta-
tions, or orthodoxies of any kind. Rather, the commitment one makes as
a good critical thinker is to always seek the truth with objectivity, integrity,
and fairmindedness. (2006, p.11)
‘Integrity and fairmindedness’ can of course be ethical dispositions,
but what Facione has in mind here is their role in truth-seeking – in
other word as epistemic dispositions. Respect and concern for o ­ thers –
qualities that seem to be present in Twelve Angry Men’s Juror 8, but not
in master problem-solver Walter White (from Breaking Bad) – are not,
according to this view, part of the profile of the ideal critical thinker.
It is interesting that, while recognising it is not a defining charac-
teristic, Ennis still sees the need to include ‘care about the dignity and
worth of every person’ among his list of critical thinking dispositions.
His reason is it would serve as a ‘corrective’ against critical thinking’s
misuse, implying that it is a powerful ability that has the potential to
hurt (humiliate, disempower, oppress) others. However, I will argue
that: (1) this is not enough of a reason to include it as a core dis-
position; but also (2) there is another reason why we should regard
respect and concern for others as having particular importance for
critical thinking.
Critical thinking dispositions can indeed be used for unethical
purposes, but this is true of most sets of practice-related dispositions,
such as those relevant to being a good sports person or a good busi-
nessperson. The good businessperson is not the same thing as a good
person more generally conceived, but if we want them not to use
their talents and dispositions for ill-intent, or in a way that neglects
negative externalities (side-effects of an organisation’s activities that it
is not legally responsible for), then they need to also have virtues such
85

CRITICAL THINKING AND DISPOSITIONS 85

as compassion and respect for others. So, in a sense, we would add

this basic ethical disposition to most or all other lists of dispositions.
A general respect for the welfare and dignity of others is a disposition
we would hope to promote and instil in our children, whether or not
we are promoting critical thinking, and most professions these days
have ethical codes of conduct.
So, in this sense the ethical dimension is relevant, but Ennis has
not provided a reason why it should have special relevance for critical
thinking, and to this extent Facione and others have a point. I believe,
though, that there are some more specific reasons why care for others
should be seen as, if not fundamental, then as having greater importance
to critical thinking than it has to other practices (such as sport). These
reasons concern the functioning of constructive dialogues, and the first
of these I will initially express in terms of premises and conclusion:

Premise 1: Constructive dialogue is crucial for critical thinking.

Premise 2: Constructive dialogue is less likely if we do not have
concern for the welfare of the people we are in a dia-
logue with.
Conclusion: Therefore concern for the welfare of others is a dispo-
sition of an ideal critical thinker.

The initial premise has already been explained, but the second
requires some elaboration. One reason constructive dialogue is less
likely if we are not respectful towards other participants is that it
could provide an incentive for them to disengage. A person detecting
signs of disrespect might leave the discussion entirely, or be reluctant
to contribute their full energy, or to be entirely open about their
position and grounds for holding it.
This point brings us to a second argument for why concern for
others is so important for critical thinking. The reasoning here is less
about the functioning of the dialogical process, and more about our
ability to understand the positions that others hold. It has already
been established that open-mindedness is fundamental to critical
thinking, but this is not just a matter of being able to detach our-
selves from our commitments in order to objectively assess alterna-
tives, it is also the ability to really listen to those alternatives with the
right degree of attentiveness. Often the bases of people’s beliefs are
subtle and highly contextual, and in order to truly understand them
86

86 CRITICAL THINKING AND DISPOSITIONS

we need to be willing to devote time and energy, and a kind of self-

lessness, to others’ belief systems. John Stuart Mill (1962, p.164) felt
strongly about this:
Ninety-nine in a hundred of what are called educated men are in this
condition; even of those who can argue fluently for their opinions. Their
conclusion may be true, but it might be false for anything they know:
they have never thrown themselves into the mental position of those who
think differently from them, and considered what such persons may have
to say; and consequently they do not, in any proper sense of the word,
know the doctrine which they themselves profess.

Being properly informed on an issue, Mill goes on to say, necessitates

that we ‘attended equally and impartially to both sides’ (ibid). To be
willing to do this, I would propose, we need to have a prior respect
for the other as the holder of these beliefs. It is this respect that helps
motivate careful and sustained listening. Also, to repeat the point made
above, the person who suspects that this respect is not present will be
reluctant to fully divulge their beliefs and the reasons supporting them.
And since the whole point of critical thinking is to scrutinise beliefs, it
is even more important that this is carried out against a background of
trust: trust that others are doing it for the right reasons, and that they
are aware that beliefs do not exist independently of believers.

BELIEFS AND BELIEVERS

‘Argument, on this model,’ says Michael Gilbert about his theory of

‘coalescent argumentation’, ‘is among persons, not between theories’
(1994, p.112). Because many of our beliefs – and certainly many of
those worth debating – are personal, then to enter into argumenta-
tion dialogues can be to run a significant risk. It is more than just the
possibility of finding out you are wrong in a way that is analogous to
getting an answer on a test wrong; it is becoming doubtful of a belief
that is central to your values and commitments. To take this on board
can require quite a far-reaching re-evaluation of aspects of one’s life.
This is a risk we will be more likely to take if we feel that our partners
in dialogue are appreciative of this fact, and correspondingly motivated
to listen to us in a way that is underpinned by basic respect.
87

CRITICAL THINKING AND DISPOSITIONS 87

Sadly, the way that critical thinking is taught (and how its aims and
methods are communicated) too often runs counter to this attitude.
In his article ‘Argument is War … and War is Hell’ (1995) Daniel H.
Cohen is critical of the adversarial, combative way in which argu-
mentation tends to be understood. This, he says, risks creating ‘not
just able arguers, but argumentative arguers: proficient, pedantic and
petty’ (ibid, 180–1). Taking pleasure in argument for argument’s sake,
or seeing the aim as winning the argument rather than establishing
truth, is all part of the dispositional profile of an adversarial approach.
It is something that is readily apparent in the practice of formal
debates, and in the way court cases in many countries are conducted.
As many social media exchanges demonstrate, it can seep more
widely into a culture as well. An upshot, Deborah Tannen (1998, p.9)
points out in her book The Argument Culture, is that ‘you’re usually
not trying to understand what the other person is saying, or what in
their experience leads them to say it. Instead, you’re readying your
response: listening for those weaknesses in logic to leap on.’
In place of the ‘war’ metaphor, Cohen suggests various alternatives,
including collaborative frames such as brainstorming. In place of lis-
tening in order to defeat there is listening motivated by inquisitiveness.
In place of me against you, there is me and you trying to sort out a
problem, the solution to which we might not have been able to reach
alone, and which could end up being a hybrid or synthesis of our
initial, individual positions.
A side-effect of this attitude to dialogues can be a kind of inti-
macy that is profoundly rewarding. The careful, respectful listening,
and thus opening up to the other’s world, is one reason for this.
Another is that the sharing of ideas in a dialogue towards new, mutu-
ally generated, important insights is an excellent basis for bonding.2
Philosopher Bertrand Russell describes his meeting with novelist
Joseph Conrad in a way that demonstrates this potential:
We talked with continually increasing intimacy. We seemed to sink
through layer after layer of what was superficial till gradually both reached
the central fire. It was an experience unlike any other that I have known.
We looked into each other’s eyes, half appalled and half intoxicated to
find ourselves together in such a region. The emotion was as intense
as passionate love, and at the same time all-embracing, I came away
88

88 CRITICAL THINKING AND DISPOSITIONS

bewildered, and hardly able to find my way among ordinary affairs. (Cited
in Yalom, 1980, p.396)
This is an example of what has come to be known as ‘flow’: the
experience of focussed engagement with an intrinsically rewarding
task that you are skilled at and in control of, but which is suitably
challenging and provides immediate feedback. It is associated with
an experience of timelessness (being lost in the moment), non-
self-consciousness (the boundaries of the ego are more supple than
usual), and calmness. According to the concept’s originator – Mihaly
Csikszentmihalyi (2002) – it can be elicited by a range of activities,
including sport, creative work, and one-to-one social encounters.
A non-dialogical variation of flow is what Dewey (1910) calls
‘wholeheartedness’. In contrast to the perseverance needed to main-
tain concentration in some deliberative situations, wholeheartedness
is an intellectual absorption in a subject where ‘the material holds
and buoys [the] mind up and gives an onward impetus to thinking’
(ibid). It is complex, reflective thinking with its own momentum, and
in which questions and ideas arise ‘spontaneously’.
To return to the central point, care for the well-being of others is
an important disposition for a critical thinker to possess, but it should
also be apparent that the practice of critical thinking, if encouraged
in the right ways, can itself serve as a gateway to understanding and
compassion. Critical thinking can thus be motivated not just by
truth-seeking, but by connection and intimacy.

DON’T BE A SMARTARSE

The broad applicability of critical thinking – to personal, profes-

sional, and civic life as well as academic development – means that
a level of care is needed in how we use it that is distinct from other
disciplines. For example, in exercising one’s ability one must be aware
that this can, as Ennis (1996b, p.171) puts it, ‘intimidate and confuse’
others who have not had this training, or who are not otherwise
inclined to think in this way. This does not mean that we should not
engage them, or even ‘push’ them to do so, but it does mean that we
should do this sensitively.We need to be careful with the language we
use, and we need to not believe that we are in some sense superior.
89

CRITICAL THINKING AND DISPOSITIONS 89

For Ennis, then, the ideal critical thinker will ‘take into account
others’ feelings and level of understanding, avoiding intimidating
or confusing others with their critical thinking prowess’ (ibid). In
other words, don’t be a smartarse. Since their primary intention is
to entertain rather than to seek the truth, caution is needed when
assessing arguments in comedy routines.3 I am, however, going to
consider a story from comedian Stewart Lee. It is an example that
I believe is justified because the comedian’s own published reflec-
tions seem to make it clear that this was a real event and that this
is how he felt about it. During one of his shows, Lee is making a
point about intolerance, and the context is a cab driver who says
to him, out of the blue, ‘all homosexuals should be killed’. Lee asks
him for his reasons.
And then there was a pause, because he’d never had to go to the next
level of the argument … after a moment he said ‘Well, because homo-
sexuality is immoral.’
Offering the example of the ancient Greeks, Lee then explains to
him that ‘morality is not a fixed thing’ and therefore not the best basis
from which to argue this point. The cabbie’s response is:
‘Well, you can prove anything with facts, can’t you?’
Lee continues:
For a minute I went, ‘Yeah.’ And then I thought, ‘Hang on! That’s the
most fantastic way of winning an argument I’ve ever heard! ‘… I’m not
interested in facts. I find they tend to cloud my judgement. I prefer to rely
on instinct and blind prejudice.’4
As abhorrent as the cabbie’s stated view is, Lee is being a smartarse
because the argument he presents is bound to cause confusion. It
deals in historical facts and abstract concepts like moral relativism
that the cabbie probably will not be familiar with. Lee is talking
over his head, and so understood more charitably the response ‘you
can prove anything with facts’ should not be taken literally. Instead,
it should be interpreted along the lines of ‘I don’t understand what
you’re talking about’, or ‘displaying familiarity with technical terms
and giving the appearance of clever arguments can fool some people,
but not me’, or perhaps simply ‘I know when I’m being patronised.’5
90

90 CRITICAL THINKING AND DISPOSITIONS

A virtue of ‘deliberative friendliness’ captures the type of construc-

tive approach that would mitigate smartarsery. It has been defined as:
the willingness to entertain discussion in a manner that does not unnec-
essarily offend or alienate interlocutors. (Aiken and Clanton, 2010, p.415)
It does not directly refer to the respect for others previously discussed,
but to the style of one’s engagement.This should be critical but encour-
aging, rather than ‘quibbling and quarrelsome’, and resolutely not one
that ‘takes any argumentative failure on the other’s side to be evidence
of cognitive asymmetry between the two’ (ibid, p.415).
In summary, some important lessons from this branch of the dis-
cussion of dispositions are these:
• Dialogues can be valuable for reaching mutually satisfying and
edifying answers, rather than a win-lose framework.
• Because the bases of some of the beliefs we hold are complex or
hard-to-get-at, dialogues involving these issues need to be con-
ducted with respect, sensitivity, and tolerance (including for appar-
ent dogmatism).
• Increased or deepened knowledge of others will often be a result,
and can be an additional aim, of argumentation.
• Increased or deepened self-knowledge will often be a result, and
can be an additional aim, of argumentation.
• Careful and sustained listening is privileged as a skill, and is associ-
ated with dispositions such as open-mindedness and humility.
• There is a requirement for sensitivity, not just to the complex-
ity, subtlety, and distinctiveness of the positions others hold, but
to their style of thinking, vocabulary, and conversational norms.
These are not necessarily those of someone educated in critical
thinking, but this is not the same as being unintelligent or unin-
formed. And even if someone is these things, it does not follow
that they cannot be engaged in some level of argumentation.

3.2 GUIDELINES FOR CONSTRUCTIVE

DIALOGUES
In addition to this analysis of dialogical dispositions it might be help-
ful to provide a summary of rules of conduct (influenced by Ennis,
91

CRITICAL THINKING AND DISPOSITIONS 91

1996b; Walton, 2006; and the work of pragma-dialecticians such as

Eemeren and Grootendorst, 2004) that we should try to follow for
dialogues to have the best chance of success. The person who has
developed critical thinking dispositions will be inclined to these ways
of behaving, in which case these guidelines can serve as a kind of
summary of how an ideal critical thinker comes across when engag-
ing in argumentation. It can also, however, function as a standalone
checklist that has value regardless of underlying dispositions to con-
duct oneself in these ways.

1. The discussants should be allowed to speak freely – both when

expressing their views, and when being critical of the views
of other discussants. Where appropriate, however, due notice
should be paid to the timing and manner of contributions (see
‘deliberative wit’, above).
2. If a discussant is asked to explain their viewpoint – for exam-
ple, provide more clarity, or provide reasons for why they hold
that view – they must be prepared to do so. This is especially
important since premises are often implicit. (See Ennis’s list of
‘clarification questions’, above.)
3. Discussants have a duty to listen carefully to, and avoid misrep-
resenting, each other’s views.
4. Unless the issue is about the person or persons involved, per-
sonal attacks (ad hominem arguments – see Chapters 6 and 7)
should be avoided where possible. Often these are fallacious,
but even where they are not their emotive nature can cause the
discussion to descend into a quarrel.
5. Discussants should address each other in a civil manner.
6. Discussants should follow basic rules of conversation such as
turn-taking.
7. The discussion should usually only end when all parties agree
that they have said all that they need to say (including asking
for clarifications and explanations from others), and that all the
issues have been given due consideration, even if some of these
are postponed to a later time. Also, note that dialogues should
only start under at an least implicit agreement by all parties
that they want to enter into a discussion (or must enter into a
discussion, as in the case of a jury).
92

92 CRITICAL THINKING AND DISPOSITIONS

3.3 CONCLUSION
Dispositions are fundamental to critical thinking because the skills
taught in the next two chapters should be rightly motivated, and
because their effective employment will invariably require wise
judgement. While a wide range of dispositions or virtues can be seen
as relevant to critical thinking, this chapter has identified and dis-
cussed ones that are closer to its heart, such as open-mindedness and
commitment to truth. I also take the position that, since construc-
tive dialogues are basic to the discipline, dispositions relating to care
for others, as well as the practical facilitation of these interactions,
are of great relevance. Finally, the importance of self-knowledge is
an important theme from the opening chapters of this book: the
cognitive and motivational biases that typify human thought, and
the ways emotions can interfere with or enable rational decision-
making. All critical thinking dispositions and skills help us counter-
act or enhance these features, but the self-­knowledge derived from
reflection on our personal preferences, priorities and prejudices is
vitally important too.

EXERCISES
1. A particularly valuable exercise I have used in class involves
watching (and/or reading) Twelve Angry Men and assessing the
critical thinking dispositions of some or all the characters in
the play/film. The quality of the content of their arguments,
and especially the way they interact with one another, will
serve as clues to the presence or otherwise of dispositions and
behaviours that have been the subject of this chapter.The same
approach can of course be applied to characters from other
stories or from real-life contexts (such as political debates) as
well. A good example is Cathy Newman’s infamous interview
with Jordan Peterson in 2018 (originally Channel 4, now easily
locatable online).
2. A good way to develop our open-mindedness is through
what’s known as ‘counter-attitudinal advocacy’. This means to
write or speak in favour of a position that you do not hold
93

CRITICAL THINKING AND DISPOSITIONS 93

and/or to argue against a position that you do. Its effectiveness

with respect to changing minds is well-established in the psy-
chology and communication research (see Petty and Cacioppo,
1996, Ch.8), but its weakness is that you can only get people
to act in this way under certain circumstances. A critical think-
ing class is just such a circumstance though. For example, class
members’ views on contentious contemporary issues can be
sought and a debate organised in which the participants argue
counter-attitudinally.

FURTHER READING
Aside from the in-text references to the ideas of people like Ennis
and Facione, the following sources are particularly relevant to some
of the ideas discussed in this chapter:
• For an interesting discussion of flexibility see Tiberius (2008, Ch.3).
• For a review of research on need for cognitive closure see Webster
and Kruglanski (1997).
• A very useful source of advice for constructive dialogues can again
be found in Robert Ennis (1996a, pp.371–5), and for or a more
in-depth look at the nature of dialogues (including ‘Rules for a
Critical Discussion’) see Walton (2006, Ch.5).
• An excellent source on constructive conversations, including
politeness, is Deborah Tannen (1992); and her later work (1998),
directly about argumentation, offers a wide range of insights into
problems resulting from an excessive emphasis on combative
approaches to disagreements.

NOTES
1 Diana Beech, Attitude is Everything, Times Higher Education, 21.5.15.
2 Cohen and Miller’s (2016) notion of ‘compathy’ in argumentation shares sev-
eral features with what is being discussed here.
3 Although there are of course genuine hybrids, such as Comedy Central’s The
Daily Show, and the performances, writings, and podcasts of people like John
Oliver, Rachel Parris, and Matt Forde.
94

94 CRITICAL THINKING AND DISPOSITIONS

4 How I Escaped My Certain Fate: The Life and Deaths of a Stand-up Comedian
(London: Faber & Faber, 2011) pp.80–2.
5 I should point out that Stewart Lee is not usually a comedian who mocks any-
one vulnerable, such as inarticulate people. Much of what he says and seems
to stand for is actively promoting fairness and non-prejudicial attitudes. (And
of course, what’s astonished/angered him here is the cab driver’s prejudicial
statement.)
95

ARGUMENTS AND ARGUMENT

RECONSTRUCTION

A physician cannot treat a disease … properly without diagnosing it cor-

rectly. An attorney cannot advise a client properly without knowing the
precise and full particulars of the client’s situation. Nor can a reasoner
evaluate an argument properly without a precise understanding of what
the argument is. (Johnson and Blair, 2006, p.11)

In this chapter we turn our attention away from the arguer’s cog-
nitive biases, emotions, and dispositions, and towards the nature
of arguments themselves. The majority of the chapter is devoted
to explaining argument reconstruction: the practice of extracting
the essential content and structure of someone’s argument from
the everyday language in which it is expressed or implied. Here
we will encounter concepts and techniques such as ambiguity and
vagueness, straw man arguments, implicit premises, and the prin-
ciple of charity. Prior to this, I will provide a reminder of what an
argument is, and explain how the sentences which comprise the
premises and conclusions must be what are called ‘propositions’ or
‘statements’.

DOI: 10.4324/9781003247944-5
96

96 ARGUMENTS AND ARGUMENT RECONSTRUCTION

4.1 THE STRUCTURE OF ARGUMENTS

In the book’s Introduction it was explained how the identification
and analysis of arguments is at the heart of critical thinking. An
argument, you will recall, is comprised of:

1. A claim being asserted that you want other people to believe

is true.
2. Reasons offered in support of this claim, through which we
try to rationally convince other people that this claim is true.

The point of offering an argument is to convince someone else of your

point of view.This point of view forms the conclusion of the argument,
and information (statements) offered in support of this are called prem-
ises. An argument can have any number of premises, and in the tradi-
tions of formal and informal logic they are usually set out in this form:

Premise 1:
Premise 2:
Conclusion:

For the sake of brevity, a premise is abbreviated to ‘P’ and a conclu-

sion to ‘C’, so it becomes:

P1:
P2:
C:

This would be the structure of a simple two premise argument, such as:
Anything that intensifies racial discrimination should be outlawed.
Capital punishment intensifies racial discrimination. Therefore capital
punishment should be outlawed.

Set out formally it would read,

P1: Anything that intensifies racial discrimination should be

outlawed.
P2: Capital punishment intensifies racial discrimination.
C: Therefore capital punishment should be outlawed.
97

ARGUMENTS AND ARGUMENT RECONSTRUCTION 97

Another example of a simple argument is:

Isaac Newton was not a true natural scientist. This is so because anyone
who believes in the principles of alchemy cannot be a true natural scien-
tist, and Newton believed in the principles of alchemy.

P1: Newton believed in the principles of alchemy.

P2: Anyone who believes in the principles of alchemy cannot be a
true natural scientist.
C: Therefore Newton was not a true natural scientist.

Often arguments are more complex that this, involving many more
premises (some of which might be ‘implicit’), more than one conclu-
sion, and what are known as ‘sub-conclusions’. These variations will
be explored shortly.

4.2 PROPOSITIONS (STATEMENTS)

Premises and conclusions must be what are called propositions (or
statements). A proposition is a sentence that makes a claim about
something that can (in theory) be adjudged to be true or false. The
majority of the things we say are propositions, from ‘The primary
cause of the American civil war was slavery’ and ‘A proven liar
is unlikely to last long as the UK Prime Minister’ to ‘I’m feeling
hungry’.
Propositions are contrasted with other kinds of sentence, most
commonly:

Questions (‘Where did you leave the children?’)

Directives (‘Please check to see if they are okay.’)

Directives are instructions or orders like ‘Pass the remote control’, or

‘Take one tablet twice a day.’ Both questions and directives, you will
notice, cannot be true or false; they are not attempts to say things
about the world. Because arguments are all about establishing truth
and falsity then their components must be sentences that can be
evaluated in this way; in other words, they must be propositions.
When reconstructing arguments, part of the challenge can be con-
verting rhetorical questions into statements. If my wife asks me if I
would like to watch an episode (or two) of Game of Thrones there is
98

98 ARGUMENTS AND ARGUMENT RECONSTRUCTION

no statement implied in this, she is simply asking me a question to

which I need to supply an answer. If, however, as another character
is graphically dispatched, she asks ‘Is that really necessary?’ she is not
expecting a reply, but implying a proposition. Her rhetorical question
can be translated into a statement such as ‘Game of Thrones would be
just as good without quite so much gore.’

4.3 ARGUMENT RECONSTRUCTION

The skill of argument reconstruction and evaluation is one way in
which some of the core knowledge and skills of critical thinking are
applied and practised. In this section you will be learning about the
reconstruction part – what argument (or arguments) is (or are) being
put forward – and the rest of the book will be relevant to the evalua-
tion part: assessing the quality of the argument or arguments.
Argument reconstruction is a skill more than simply a demonstra-
tion of what you know, and like many skills it involves judgement
because:

1. In each case you will be applying it to new arguments, and the

forms these take across different contexts vary considerably.
2. Quite often what people are expressing is open to more than
one plausible interpretation.

You will become better at making these judgements the more you
become familiar with the subject matter of critical thinking, but
more importantly it is a matter of practice. As indicated, argument
reconstruction is primarily a skill. There are a number of guiding
principles and recommendations (set out below), but mastering it
requires doing it, and doing it really quite a lot. This can be tedious
for sure, but with perseverance your improvement will be very appar-
ent, as will the way in which this ability transfers itself to improved
questioning and comprehension in any part of life where arguments
are found.
The basic aim of argument reconstruction is simple enough: distil
the essence of the argument that is being made. It can be understood as
a form of summarising. Critical thinking has been defined as mini-
mising errors in our reasoning by improving our ability to generate
99

ARGUMENTS AND ARGUMENT RECONSTRUCTION 99

strong arguments and evaluate the arguments of others. Since for-

mulating our own arguments usually involves responding to the
arguments of others, then the quality of these evaluations is of funda-
mental importance. But assessment can only be effective if the argu-
ment has been properly understood in the first place.
The truth of this is perhaps obvious, but worth highlighting because
its intellectual obviousness can mask to how difficult it is to achieve
in practice. Many of the reasons behind this difficulty were discussed
in Chapters 1 and 2, and the interpersonal, ethical, and epistemologi-
cal significances of careful reading and listening were addressed in
Chapter 3. In the language of argumentation, a misrepresentation of
someone’s position so that it is negatively distorted or caricatured
is known as a straw man argument. Straw man (or straw person)
arguments are so named because it is easier to push over, or bring
down, a straw effigy than a real person. The metaphor represents the
difference between X’s real position and a weaker position ascribed
to them by Y that is easier to argue against.
For example, journalist and broadcaster Andrew Marr, in discus-
sion with Rachel Holmes, author of a recent book on Emmeline
Pankhurst, puts it to her:
At one point in the book you describe [Pankhurst] as a ‘wild, unconven-
tional, obdurate, upstart, ultra-left, feminist nightmare’.

Holmes responds:
I say that’s one version of how some people saw her. … As George Bernard
Shaw … said, there were only two views about her, either that she was
miraculous or that she was unbearable. That’s the unbearable version …

Marr’s question implies that this is Holmes’s view, which it clearly

isn’t. His intention is to spice up the conversation, and it can perhaps
be forgiven as a journalistic device in what an essentially low-stakes
and non-confrontational context.1
Journalism – and especially the broadcast variety – of course has
at least one eye on entertainment, but academic debates are far from
immune from straw manning. In her book Testosterone Carole Hooven
discusses some of the ways biological and evolutionary approaches
to explaining human behaviour are caricatured by their opponents.
They are, for instance, portrayed as rejecting any place for cultural
100

100 ARGUMENTS AND ARGUMENT RECONSTRUCTION

influence or choices made by individuals; or as condoning immoral

behaviours that have an evolutionary basis. In Hooven’s words, ‘Who
on earth thinks that?’ (p. 250).2 No one sensible from this school of
thought does, and nothing in their theory requires such back and
white thinking.
Sometimes a distorted version of an opponent’s position is delib-
erately employed, but the straw man label also applies to instances
where it is a mistake. From a dialectical point of view, being misrep-
resented is extraordinarily frustrating, and if due time or space is not
allowed for us to make corrections to what others have said then the
dialogical process is liable to be derailed.
Controversial topics (such as abortion or immigration) and situ-
ations where lines are sharply divided (such as party politics) breed
straw man arguments that are fuelled by emotion and competitive
urges. They often happen under cooler conditions as well though,
sometimes because of inattention to detail. In all cases the confir-
mation bias is implicated (see Chapter 1), and the discussion in the
previous chapter about open-mindedness and paying careful atten-
tion to both the words and the circumstances of the other is impor-
tant for avoiding unintentional straw man arguments. Most especially,
we need to avoid, where possible, hearing arguments second-hand.
As John Stuart Mill (1962, p. 63) says, we should ‘hear them from
persons who actually believe them; who defend them in earnest, and
do their very utmost for them’.
Also, a common occurrence is the opposite of a straw man
­argument – known as an ‘iron man’ argument. As you would imagine,
these are cases where we misrepresent a position – whether inten-
tionally or in error – as being stronger than it is.The result, or at least
the aim, is that it is easier to defend: ‘When my client said the victim
“was disappeared” your Honour, what he meant to say was that the
victim just disappeared, without explanation.’ 3
Critical thinking scholars are fairly consistent in listing the tasks
that need to be performed towards reconstructing arguments.4 In this
tradition the rest of this chapter will help you develop and refine this
skill under the following headings: Identifying premises and conclu-
sions; Rewording; Implicit premises and conclusions; and Argument
structure: Sub-conclusions and the reordering of premises.
101

ARGUMENTS AND ARGUMENT RECONSTRUCTION 101

IDENTIFYING PREMISES AND CONCLUSIONS

There are two methods for identifying the conclusion of someone’s

argument.The first is to read (or listen) carefully and ask yourself ‘What
point are they making?’ In dialogue this might require asking others
to clarify what they mean, and with written arguments the responsi-
bility lies with us to read, and if necessary reread, what is presented in
order to establish the overall conclusion. Similarly with premises, once
you have established the overall point it is a matter of working out
what is being said in support of that point. Not everything in the text
in question will be relevant to the argument, and often it will not be
presented in the best order or as clearly as it might be. As you will see,
part of the argument reconstruction process is to tidy up these aspects
leaving only what is (or what seems to be) relevant to the argument.
The second method is to look for premise and conclusion indi-
cators.These are words that often function as precursors to premises
and conclusions; for example, conclusions can often be identified by
the presence of terms such as:
• ‘Therefore’, ‘thus’, ‘so’, ‘hence’, ‘accordingly’, ‘in which case’
And premises can often be indicated by phrases like:
• ‘My reason is …’,‘My evidence for this is …’,‘This is so because …’
The two methods can be used together, but for several reasons I
am inclined to emphasise the first one. The first is that reading and
rereading a passage helps us to understand it more deeply, and not
be tempted by the shortcuts that premise and conclusion indicators
can provide. The second is that premise and conclusion indicators
are not always present. For instance, if the conclusion is at the start
of a passage – like the paragraph you are reading right now – it will
not be prefixed by an indicator like ‘therefore’. A further reason is
that we find premise indicators in explanations as well as arguments.
The difference between an argument and an explanation is that with
arguments the conclusion has not been agreed upon, whereas with
explanations it is agreed that something is the case, and someone is
then enquiring about what has led to this. So, I might explain that
I was late for the meeting because of a puncture; or that Pluto is
102

102 ARGUMENTS AND ARGUMENT RECONSTRUCTION

sometimes closer to the sun than Neptune because it has a more

elliptical orbit. Notice that the word ‘because’ is used here as well, and
with both arguments and explanations we are providing ‘reasons’.
Notice also that it is not always that clear whether something is an
explanation or an argument if we are lacking the necessary context.
In the latter example it could be that someone needs to be convinced
of the fact that Pluto can be nearer the sun than Neptune.

REWORDING

Part of the skill of argument reconstruction is to make what is pre-

sented both argument-appropriate and as concise as possible with-
out losing its meaning. Great care must be taken with this because
inattentive rewording of premises and conclusions can lead to the
misrepresentations of authors’ intentions.Where possible, the rule for
avoiding this is: if in doubt, retain the original words.
The three main aims of rewording are:

1. Translating information in the form of non-propositional

­sentences (such as rhetorical questions) into propositions.
2. Condensing the argument where necessary through:
i. using more efficient expressions and
ii. leaving out material that does not contribute to the prem-
ises or conclusions.
3. Clarifying ambiguous and vague sentences.

The passage below is from Martin Luther King, Jr.’s Letter from
Birmingham Jail. As you will recall from Chapter 1, the overall aim
of the letter was to justify King’s method of non-violent (but some-
times illegal) direct action against forms of white oppression in the
southern states of America. The letter, though published in the press
nationwide, was ostensibly written to members of the white clergy
in Alabama who, though supportive of his cause, were critical of his
methods. One of many arguments he makes is this:
In your statement you asserted that our actions, even though peace-
ful, must be condemned because they precipitate violence. But can this
assertion be logically made? Isn’t this like condemning the robbed man
103

ARGUMENTS AND ARGUMENT RECONSTRUCTION 103

because his possession of money precipitated the evil act of robbery?

Isn’t this like condemning Socrates because his unswerving commitment
to truth and his philosophical delvings precipitated the misguided popu-
lar mind to make him drink the hemlock? Isn’t this like condemning Jesus
because His unique God-consciousness and never-ceasing devotion to
His will precipitated the evil act of crucifixion? We must come to see, as
federal courts have consistently affirmed, that it is immoral to urge an
individual to withdraw his efforts to gain his basic constitutional rights
because the quest precipitates violence. Society must protect the robbed
and punish the robber.5

It is immediately evident here that King, Jr.’s argument employs a

series of rhetorical questions. Each though contains a relevant point,
and so they need to be preserved in the reconstruction in the form
of propositions. Some of his sentences are quite wordy and can be
condensed; and the last sentence seems to be a rhetorical restatement
of the conclusion that has already been expressed in the previous sen-
tence (and in the second sentence of the passage).Taking these things
into consideration, a reasonable reconstruction might be as follows:

P1: It is claimed that our peaceful actions must be condemned

because they precipitate violence.
P2: To condemn our peaceful actions in this way is like condemn-
ing someone (e.g., the robbed man, Socrates, Jesus) for doing
what is right (or what they are entitled to do) just because it
provokes bad/misguided people to do wrong.
P3: It is wrong to condemn the robbed man/Socrates/Jesus.
P4: The federal courts have consistently affirmed that it is immoral
to urge an individual to withdraw his efforts to gain his basic
constitutional rights because this precipitates violence.
C: Therefore it is illogical and morally wrong to condemn non-
violent protest because it precipitates violence.

This presents the essential structure of the argument (which is an

argument from analogy (see Chapter 9)) and an appeal to various
authorities (see Chapters 6 and 7), but by condensing some of the
premises this version quite possibly loses the full implications of
the examples he uses. For audiences familiar with the relevance of
Socrates and Jesus to the point being made, this version might be
104

104 ARGUMENTS AND ARGUMENT RECONSTRUCTION

adequate; and so too in a situation where the reconstruction aims

only to reveal the argument’s essential structure on an occasion
when all parties are looking at the original passage at the same time.
Otherwise the detail King offers is important to include, resulting in
the following reconstruction:

P1: To condemn our actions because they precipitate violence is

like condemning the robbed man because his possession of
money precipitated the evil act of robbery.
P2: This is like condemning Socrates because his unswerving com-
mitment to truth and his philosophical delvings precipitated
the misguided popular mind to make him drink the hemlock.
P3: And this is like condemning Jesus because His unique God-
consciousness and never-ceasing devotion to His will precipi-
tated the evil act of crucifixion.
P4: We do not condemn the robbed man/Socrates/Jesus.
P5: The federal courts have consistently affirmed that it is immoral
to urge an individual to withdraw his efforts to gain his basic
constitutional rights because this precipitates violence.
C: Therefore it is illogical and wrong to condemn the peaceful
actions of the civil rights movement.

Premises 1–3 could be grouped into a single, very long, premise, but
separating them in this way makes the argument easier to analyse. For
example, we might want to contest the strength of King’s analogy
with the robbed man, but be more satisfied with the Socrates and
Jesus analogies.
In summary then: questions and other non-statements must be
reworded if they are to form part of the reconstruction; but if in any
doubt as to whether significant rewording or condensing of sentences
might lose important content, then leave them in their original form.
The third aim of rewording is to clarify, where relevant and pos-
sible, ambiguous and vague sentences. An ambiguous sentence is one
that has more than one possible meaning, for example:
After the bar room brawl the tables were turned.

As this suggests, ambiguity, as the basis of puns, is a fertile source

of humour, but it can be cause serious problems as well. A former
105

ARGUMENTS AND ARGUMENT RECONSTRUCTION 105

colleague of mine came across this sentence in a child development

textbook:
19 per cent of US children are poor, rates that climb to 30 per cent for
Hispanic children, 32 per cent for Native-American children, and 34 per
cent for African-American children.6

This can imply that these ethnic groups are not US children, whereas
what the author intends to say is that ‘19 per cent of US as a total
population are poor’. (This has since been corrected.)
In his Guardian newspaper column Giles Fraser wrote a piece
called ‘Assisted suicide is the equivalent of a zero-hours contract
with life’. He argued that a recent trend for suicide in the UK is
a symptom of excessive individualism. Everything has become an
individual choice (rather than a community responsibility), includ-
ing when we die.
Maud lives round the corner from me in south London. She remembers
a time when everyone knew everyone else, and when there was genu-
ine community solidarity. Nowadays people come and go, she says, and
young people can’t be bothered with the elderly. She is often lonely. ‘Even
the doctor came round to see me and asked me if I wanted to commit
suicide,’ she says.7

Some very black humour flirts with this example because I do not
believe for a second that the doctor was offering Maud suicide as an
option (as he might offer her forms of medication or counselling).
Considering her loneliness or depression, I imagine he was trying to
ascertain if she was a suicide risk. In this example it is hard to know
whether it is the doctor, Maud, or Giles Fraser that has caused the
mix-up. Let’s hope it is Fraser, but whichever is the case what we are
presented with is ambiguous.
When teasing out the premises and conclusion of a person’s argu-
ment, ambiguous sentences must be disambiguated. In a dialogue we
can ask the person for clarification, but if they are not available for
questioning we can either:
• Do our best to suppose their intended meaning, applying the
‘principle of charity’ (see below); or
• Reconstruct more than one version of the argument, applying
alternative meanings of ambiguous sentences in each case.
106

106 ARGUMENTS AND ARGUMENT RECONSTRUCTION

If an ambiguous sentence is one which can mean more than one thing,
a vague sentence lacks precision. In many contexts an absence of preci-
sion is entirely appropriate so long as the communicator and their audi-
ence share relevant cultural and linguistic norms:‘give me a moment’,‘I’ll
be there soon’,‘just a few potatoes’. (The process of learning these norms
was highlighted a few years ago when my then 4-year-old asked me, in
all innocence and with no agenda, how long a ‘moment’ is.)
It has been suggested that the Intergovernmental Panel on Climate
Change (IPCC) should change its style of reporting probabilities to
the media because of a mismatch between the vague terms it uses and
what these are taken to mean by the general public. For example, in
statements like ‘Ice-free Arctic Ocean summers are very likely at levels
of global warming higher than 2°C’8 they use ‘very likely’ to mean
>90% certainty, but it has been found that the public’s understanding
of ‘very likely’ is >70% certainty.9 This represents a significant dis-
crepancy that will not help the already tough task of communicating
urgency and motivating climate-related behaviour change.
Vague statements can signal the beginning of a dialogue in which
a relatively unformed belief about an issue achieves sharper focus:
‘You say that it’s “very likely” that … But what exactly do you mean
by this?’ But in non-dialogical communication vagueness can be
problematic where it causes (1) misunderstanding (as in the IPCC
example), and (2) imprecision that is inappropriate for the context in
question, leaving us unable to respond until we know exactly what is
meant by, for example, ‘the train is delayed’; ‘the airport is quite close
to the city’; ‘their actions were borderline illegal’; ‘Lou was talking
about you the other day’; or ‘it was a fairly innocent flirtation’.
Unlike ambiguous sentences, vague statements can form premises
and conclusions in arguments where (1) and (2) above do not rep-
resent a problem (‘I’ve only got a couple more things to do, there-
fore I will be there in a short while’). Where (1) or (2) is a problem,
the issue is similar to the one caused by ambiguity. Vague sentences
become, in effect, ones that cannot be determined to be true or false,
and therefore cannot function as propositions. When reconstructing
arguments we therefore need to make these meanings more pre-
cise, either through an educated guess or through further research. As
with the rewording of ambiguous sentences, the ‘principle of charity’
should also be applied (see below).
107

ARGUMENTS AND ARGUMENT RECONSTRUCTION 107

IMPLICIT PREMISES AND CONCLUSIONS

Much of the time we do not explicitly say everything that is neces-

sary for the complete formulation of the argument being presented.
The most implicit an argument can be is when it is presented visually,
such as advertisements with minimal written text, or gestures like
the two-handed diving motion used by footballers to indicate to the
referee that a player went over rather too easily.10 Partial arguments
are most common though; a simple example might be:
Only extraverts are energised by social situations, so Susan is an extravert.

The whole argument would run:

P1: Only extraverts are energised by social situations.

P2: Susan is energised by social situations.
C: Therefore Susan is an extravert.

P2 might typically be missed out in natural language because it is

obvious to those listening that this is the implication, or it could
already be common knowledge. The Greek word for an argument
with a missing premise (or premises) is an enthymeme, and Aristotle
(1991, p.195) recommends their use where being fully explicit can
be unnecessarily wordy and have the effect of being boring, patronis-
ing, and in some cases harder to follow (similar to what is referred to
these days as ‘mansplaining’).
Conclusions as well as premises can be unproblematically implicit.
The context of the following letter to a newspaper was the UK
Government’s response to IS incursions into Iraqi Kurdistan in the
summer of 2014. It contains implicit premises and a conclusion
implied by a sarcastic question:
Prime Ministers have regularly used war abroad to distract from consti-
tutional matters or problems at home, as history shows. But it couldn’t
happen today, could it?11

A reasonable reconstruction might be:

P1: History shows that prime ministers have regularly used war
abroad to distract from constitutional matters or problems at
home.
108

108 ARGUMENTS AND ARGUMENT RECONSTRUCTION

P2: The current government is unpopular.

P3: The current government has involved us in another war abroad.
C1: The current government is using war as a distraction.
C2: Using war as a distraction can still happen today.

Often the reason for making implicit premises explicit is to highlight

an assumption that needs to be questioned. An assumption, in the
context of an argument, is a belief that has relevance to the argument,
but which has not been defended. Assumptions can be implicit or
explicit, and in many cases they do not need to be defended if they
are common knowledge or trivially true. Dangerous ones, though,
are those that are both implicit and need defending. Making these
assumptions explicit is therefore one of the most important reasons for
reconstructing arguments. For example, should we want to respond
to the letter above, it is important to make P2 and C1 explicit.

ARGUMENT STRUCTURE: SUB-CONCLUSIONS AND

THE REORDERING OF PREMISES

More complex arguments can involve what are known as sub-

conclusions (sometimes referred to as ‘intermediate conclusions’).
Sub-conclusions indicate that there are one or more smaller argu-
ments that contribute to a larger argument. In our reconstructions
it is important to identify and separate these so as to aid clarification
and evaluation.
Arguments with sub-conclusions have a structure along these lines:

Premise (1)
Premise (2)
Conclusion (1)
Premise (3)
Conclusion (2)
Premise (4)
Etc.
Overall conclusion

Arguments containing sub-conclusions are usually chained argu-

ments. In a chained argument each sub-conclusion forms a premise
109

ARGUMENTS AND ARGUMENT RECONSTRUCTION 109

for a further argument, and in such a circumstance it is crucial that

we put the premises in the right order in our reconstruction. Here is
an example of a chained argument:
Healthcare is basic to human welfare. Resources basic to human welfare
should be free at the point of use, and therefore healthcare should be free
at the point of use. Healthcare funded by private insurance is not free at
the point of use, so healthcare should not be funded by private insurance.

The correct reconstruction would be:

P1: Resources basic to human welfare should be free at the point

of use.
P2: Healthcare is basic to human welfare.
C1: Therefore healthcare should be free at the point of use.
P3: Healthcare funded by private insurance is not free at the point
of use.
C2: Therefore healthcare should not be funded by private insurance.

In order to establish C2 (healthcare shouldn’t be funded by private

insurance), C1 must first be established (healthcare should be free
at the point of use). If the two arguments were put the other way
around the overall argument would not make sense:

P1: Healthcare funded by private insurance is not free at the point

of use.
C1: Therefore healthcare should not be funded by private insurance.
P2: Resources basic to human welfare should be free at the point
of use.
P3: Healthcare is basic to human welfare.
C2: Therefore healthcare should be free at the point of use.

The best approach to deciding whether a conclusion is the main one

or a sub-conclusion is to ask what overall point the arguer seems to
be making. Then, the coherence or otherwise of your reconstruction
will help confirm whether this is right or not.
The passage below is a complex argument found in investigative
journalist and author Naomi Klein’s article ‘Golf Oil Spill: A Hole
in the World’, about the Deepwater Horizon disaster in the Gulf of
110

110 ARGUMENTS AND ARGUMENT RECONSTRUCTION

Mexico in 2010. Quoting from BP’s risk assessment prior to the spill
she says,
Best of all, should a major spill occur, there is, apparently, ‘little risk of con-
tact or impact to the coastline’ because of the company’s projected speedy
response (!) and ‘due to the distance [of the rig] to shore’ – about 48 miles
(77km). This is the most astonishing claim of all. In a gulf that often sees
winds of more than 70km an hour, not to mention hurricanes, BP had so
little respect for the ocean’s capacity to ebb and flow, surge and heave, that
it did not think oil could make a paltry 77km trip. None of this sloppiness
would have been possible, however, had BP not been making its predic-
tions to a political class eager to believe that nature had indeed been mas-
tered. Some, like Republican Lisa Murkowski, were more eager than others.
The Alaskan senator was so awe-struck by the industry’s four-dimensional
seismic imaging that she proclaimed deep-sea drilling to have reached the
very height of controlled artificiality. ‘It’s better than Disneyland in terms of
how you can take technologies and go after a resource that is thousands of
years old and do so in an environmentally sound way,’ she told the Senate
energy committee just seven months ago.12

A plausible reconstruction requires a sub-conclusion (as well as an

implicit premise):

P1: When presented with the industry’s four-dimensional seismic

imaging, Republican Lisa Murkowski the Alaskan senator pro-
claimed deep-sea drilling to have reached the very height of
controlled artificiality. She told the Senate energy committee
just seven months ago, ‘It’s better than Disneyland in terms of
how you can take technologies and go after a resource that is
thousands of years old and do so in an environmentally sound
way.’
P2 (implicit): A statement like the one in P1 is typical of a political
class eager to believe that nature had been mastered.
C1: The current political class is eager to believe that nature had
been mastered.
P3: BP erroneously claimed that should a major spill occur, there is
‘little risk of contact or impact to the coastline’.
P4 (implicit): The sloppiness of BP’s ‘initial exploration plan’ can
only be explained by a political class eager to believe that
nature had indeed been mastered.
111

ARGUMENTS AND ARGUMENT RECONSTRUCTION 111

C2: BP were allowed to become sloppy in their planning because

it had been making its predictions to a political class eager to
believe that nature had indeed been mastered.

Klein is relying on her claim about the eagerness of the current

political class to believe that nature had been mastered to support
her final conclusion that this is the reason for BP’s sloppy risk assess-
ment. Since she is not claiming that BP’s sloppiness is the cause of
this political attitude, then the two conclusions here must fall in the
order presented above.
Arguments that are not chained in this way are known as conver-
gent arguments. With convergent arguments separate claims serve to
reinforce a single conclusion, so the basic structure is simply:

P1:
P2:
P3:
Etc.
C:

A basic example might be:

Because it’s environmentally friendly, cycling to work is a good thing to
do. It also saves money, is good for your health, and helps you feel alert
and energised at the start of the day.

Reconstructed:

P1: Cycling to work is environmentally friendly.

P2: Cycling to work saves money.
P3: Cycling to work is good for your health.
P4: Cycling to work helps you feel alert and energised at the start
of the day.
C: Therefore cycling to work is a good thing to do.

Convergent arguments can be identified by phrases such as ‘Another

reason is this …’; or ‘Also in support of this …’, but these will not
always be present. A better test is to see whether the argument retains
its coherence no matter what order the premises come in.
112

112 ARGUMENTS AND ARGUMENT RECONSTRUCTION

THE PRINCIPLE OF CHARITY

It should be clear from the preceding discussions and examples that

argument reconstruction cannot always be that accurate. Our efforts
will often be provisional; the equivalent of saying in a dialogue ‘So
is this what you mean …?’ But since the person is not present to
consult, then the recommended language is more like ‘If this is what
X means’ or ‘If I’ve understood X correctly … then this is what the
argument looks like’. It indicates that you cannot always be sure that
you have grasped what the other person wants to say, and that the
true meaning of the argument and its intention rests with them. This
connotes the respect and modesty that were discussed in Chapter 3.
In this vein it is also important to attempt to reconstruct the best
(plausible) version of what we are presented with. In the choices that
we make concerning (1) the translation of non-statements into state-
ments; (2) the inclusion of implicit premises; (3) disambiguation; and
(4) making precision out of vagueness, we should aim to put the argu-
ment in the best light we can. This form of giving the benefit of the
doubt has been called the principle of charity, and the main reason
for employing it is summed up by Johnson and Blair in this way:
The idea is that since (normally) an author will be trying to make logical
arguments, it follows that if, in interpreting a passage, we reconstruct the
most logical argument we can make it out to contain, then that probably
will be the argument the author intended to make. (2006, p.15)

Enshrined in this principle is the truth-seeking disposition rather

than an attitude to critical thinking that sees it as a means of winning
or gaining superiority over others. To return to where we started in
this section, being motivated by winning will make us vulnerable to
generating straw man (and iron man) arguments.

4.4 SUMMARY
The art of argument reconstruction involves several skills: identify-
ing premises and conclusions; rewording (e.g., of relevant information
not originally provided in the form of propositions, or of ambigu-
ous or vague claims); teasing out implicit premises and conclusions;
reordering of premises and identifying sub-conclusions. All of this
should be carried out in the spirit of charity because the objectives
are truth, honest agreement, and constructive dialogues, rather than
113

ARGUMENTS AND ARGUMENT RECONSTRUCTION 113

point-scoring. Avoiding misrepresenting someone’s position (straw

man arguments) is of primary importance, and in this respect argu-
ment reconstruction is associated with the too-often overlooked dis-
position to carefully and respectfully read, or listen to, what others have to say.
The effective evaluation of arguments – the central theme of the rest
of this book – is only possible if we have properly understood them.

DISCUSSION QUESTION
The case is made in this chapter for straw man arguments always
being fallacious, but is this necessarily the case? Are there circum-
stances in which it is acceptable to distort someone else’s position in
order to make it easier to argue against? Two short articles that make
a case for this view are by Aikin and Casey (2011, 2016).

EXERCISES
Making use of the techniques explained in this chapter, reconstruct
the following arguments:

1. The following letter is in response to the UK Government

blocking a proposed energy saving public information cam-
paign in the face of increasing bills and possible gas shortages
in the winter of 2022:
When we have a medical problem we seek advice from a doctor.
If there is a legal problem we consult a solicitor. When there is a
national crisis – the energy crisis – surely the government should
issue guidelines to mitigate the impact of this on the population.
(Letter to the i newspaper, 10.10.22)

2. The use of food banks in some parts of the UK has seen a

rapid increase since the Tory government came into power
and introduced benefit cuts. Some Tories say that food banks
are part of the welfare state, but food banks are charities and
separate from the welfare state. They are a sign that the wel-
fare state is failing. (Adapted from MP Mhairi Black’s maiden
speech in the House of Commons (UK), 14.7.15.)

3. Recently I was knocked off my bicycle by a van coming out

of a side road. I was concussed, despite wearing a cycle helmet,
114

114 ARGUMENTS AND ARGUMENT RECONSTRUCTION

which was damaged by the impact of hitting the road. At the

hospital, doctors suggested that, without the helmet, I could
have died, or at least been in intensive care. And yet there
seems to be considerable resistance in some cycling groups to
any law requiring the wearing of helmets. Ministers have said
it would be impossible to enforce such a law, but couldn’t the
same argument apply to seatbelts? It is not difficult to see who
is wearing a helmet. In Australia, it is illegal to ride a bike or
for a child to use a scooter without a helmet. Cyclists have
accepted this law. Are Australians more law-abiding than we
are? Head injuries cost the NHS a considerable amount of
money. I fail to see why helmets are required for motorbikes
and not on bicycles, as the head injuries can be much the
same. (Letter to the Daily Telegraph, 8.4.11)

FURTHER READING
• For exercises on visual arguments see Morrow and Weston (2011);
and Lunsford and Ruszkiewicz (2010).
• For an instructive article about straw man and iron man arguments
see Aikin and Casey (2016); and for some important insights into
the motivations behind straw man arguments see Tannen (1998),
especially Ch.1.

NOTES
1 The good-natured and informative Start the Week, BBC Radio 4, 14.9.20.
2 Hooven, C. Testosterone (Cassell, 2021).
3 Saul Goodman knows a thing or two about iron manning his clients. Be
aware that the term ‘steel manning’ has also been coined; not as an alterna-
tive to ‘iron man’ but with the same meaning as the principle of charity (see
above).
4 Some good examples are Johnson and Blair (2006); Scriven (1976); Bowell
and Kemp (2015); and Morrow and Weston (2011).
5 Available at: www.africa.upenn.edu/Articles_Gen/Letter_Birmingham.
html
6 Berk, L. Infants, Children, and Adolescents, 7th Ed. (Pearson International
Edition, 2010), p. 74.
7 The Guardian, 29.8.15
115

ARGUMENTS AND ARGUMENT RECONSTRUCTION 115

8 IPCC (2018) Global Warming of 1.5°C, p. 258. Available at: www.ipcc.ch/

site/assets/uploads/sites/2/2022/06/SR15_Full_Report_LR.pdf
9 Painter, J. Climate Change and the Media (I.B. Tauris, 2013).
10 I am wary of using visual argument for reconstruction and evaluation
because the skill required is more one of interpretation than reconstruction,
but at the end of the chapter you will find some suggestions for good critical
thinking books that do go down this path.
11 Letter to The Guardian, 27.9.14
12 Klein, N. (2010) Golf Oil Spill:A Hole in theWorld [Article].Available at: www.
theguardian.com/theguardian/2010/jun/19/naomi-klein-gulf-oil-spill
116

ARGUMENT FORMS AND

FALLACIES

[Fallacies] are like bad habits. They are hard to break. (John Woods,
2013, p.5)

In this chapter we will begin to explore the various types of argu-

ments that have been identified by informal logicians and critical
thinking scholars, and take an initial look at the types of fallacies
(poor arguments) that are associated with them. There are numerous
ways of classifying arguments, and here I will explain two that are
important to know about: deductive, inductive, and plausible argu-
ments; and argument forms (or ‘schemes’).

5.1 DEDUCTIVE, INDUCTIVE, AND PLAUSIBLE

ARGUMENTS
DEDUCTIVE AND INDUCTIVE ARGUMENTS

In books on logic, and some books on critical thinking, you will find
discussion of deductive and inductive arguments. My view (and also
the view of Scriven (1976) and Johnson and Blair (2006), among oth-
ers) is that learning about this distinction is of limited value outside
DOI: 10.4324/9781003247944-6
117

ARGUMENT FORMS AND FALLACIES 117

of formal logic. However, because I do believe this categorisation

has some value for the effective communication of arguments, and
because they are frequently mentioned and discussed elsewhere in
the field, I will devote some space to them.
A deductive argument is one which attempts to provide a line
of reasoning in which the conclusion is necessarily deduced from the
premises, for example:

P1: Either the Earth has remained in a static state, or it has changed
form due to tectonic shifts.
P2: The Earth has not remained in a static state.
C: Therefore the Earth has changed form due to tectonic shifts.

A successfully structured deductive argument is known as a valid

deductive argument, and valid deductive arguments are what Douglas
Walton calls ‘airtight’ or ‘truth preserving’ (1989, p.115). If the premises
are true, the conclusion is guaranteed to be true. If a person accepts
the premises of a valid deductive argument, then they must also accept
the conclusion because the truth of the conclusion can be logically
deduced from the truth of the premises. The argument above claims
that there are only two options that could be true. If this claim is true,
and one option is then rejected, the remaining option must be the case.
Notice though that the validity of a deductive argument is entirely
to do with this logical relationship between premises and conclusion,
and has nothing to say about the actual truth of the premises. The
following argument is deductively valid:

P1: Donald Trump says that voter fraud lost him the 2020 presi-
dential election.
P2: Everything Donald Trump says is true.
C: Therefore voter fraud lost Donald Trump the 2020 presidential
election.

But since the second premise is not true, the truth of the conclusion
cannot be guaranteed. If presented with this argument we wouldn’t
challenge its logic, but rather the truth of its premises.
A valid deductive argument with true premises (and therefore a
true conclusion) is known as a sound argument, and it is the strongest
118

118 ARGUMENT FORMS AND FALLACIES

type of argument there is. If you can convince someone of the truth
of your premises, and the argument has a valid deductive structure,
then they have no option but to accept your conclusion.
An inductive argument on the other hand lacks this airtight
quality. An inductive version of the above argument might be:

P1: Either the Earth has remained in a static state, or it has changed
form due to tectonic shifts.
P2: As far as the evidence goes, it seems the Earth has not remained
in a static state.
C: Therefore the Earth has changed form due to tectonic shifts.

In this form the conclusion does not necessarily follow from the
premises – unlike valid deductive arguments it would not be a con-
tradiction to reject the conclusion whilst accepting the premises.
The reason why I and other critical thinking scholars are not
inclined to take this distinction too seriously, however, concerns the
ease with which an inductive argument can be turned into a deduc-
tive argument, and vice versa. As Michael Scriven says,
A slight juggling of the premises (by adding some unstated ones) and the
conclusions can always convert an inductive argument into a deductive
one without any essential loss of the ‘point of the argument’. (1976, p.34)
Take this argument:

P1: If the well is poisoned, then Lassie is probably dead.

P2: The well is poisoned.
C: Therefore Lassie is dead.

In this form it is inductive, but if we include the word ‘probably’ in

the conclusion it becomes deductive:

P1: If the well is poisoned, then Lassie is probably dead.

P2: The well is poisoned.
C: Therefore Lassie is probably dead.

Rather than use the first version we are better off using the second,
but in both cases we know that the strength of the argument depends
on the truth or falsity of the premises.
119

ARGUMENT FORMS AND FALLACIES 119

If, on the other hand we are presented with a deductive version

along these lines we are in the same position:

P1: If the well is poisoned, then Lassie is dead.

P2: The well is poisoned.
C: Therefore Lassie is dead.

We can accept that if the premises are true then the conclusion must
also be true, but this is trivial. What matters, again, is the truth or
otherwise of the premises, and we will demand (or seek) evidence of
this before we accept that Lassie is dead.
A more pressing concern for critical thinking is appreciating the
context in which an argument is put forward. Scientific discovery
is usually seen as a process whereby careful observations are used to
predict outcomes in similar situations towards developing a theory to
explain the phenomenon in question. However, no matter what the
theory’s current success is in predicting results, scientists recognise
that it is possible for it to be proven wrong by subsequent find-
ings. In this sense it has been known as an inductive approach; the
kind of deductive proofs available to mathematicians and logicians
do not apply to empirical investigations. Assessment of the inductive
approach is often linked to an analysis of generalisations, and more
will be said on this in Chapter 8.1

PLAUSIBLE ARGUMENTS

Understood in this way the deductive/inductive distinction is impor-

tant, not for distinguishing essentially different forms of argument,
but different realms of knowledge and their associated standards of
correctness. If maths and science exemplify deductive and induc-
tive reasoning though, how are we to understanding the many other
contexts of professional, personal, and civic life in which reasoning
plays a part? Many factors become relevant here, such as how impor-
tant the decision is, how urgent, whether the situation is simple or
complex, familiar or unfamiliar, the limitations of the expertise of the
people involved, and so on. It is very much this territory that critical
thinking is interested in, and that the rest of this book is devoted to
exploring.
120

120 ARGUMENT FORMS AND FALLACIES

In recent decades Douglas Walton (1995, 2001, 2006) has pro-

vided a characterisation of everyday argumentation in terms of
what are called plausible arguments (also known as ­‘presumptive’,
­‘defeasible’, or ‘abductive’ arguments).These are sometimes explained
as a third alternative to deductive and inductive arguments, but it
is perhaps more helpful to understand them as representing a style
of reasoning that functions quite efficiently for certain types of
­decision-making. Plausible arguments are characterised as:

1. Making claims based on what is reasonably or normally

expected in familiar situations.
2. In so doing, shifting the burden of proof to any claim that
contradicts the one being made.
3. But always recognising that the conclusion drawn is provisional
in nature; that is, open to being proven wrong should the case
in question turn out to be other than what would normally
be expected.

Plausible arguments are highly pragmatic. They operate in situations

where decisions need to be made fairly quickly, often in the absence
of precise information, but in circumstances that are generally famil-
iar. As Walton (2001, p.156) puts it, they ‘move a dialogue forward’.
Unknowns and unique features are acknowledged but, vitally, a deci-
sion of some consequence must be made. Put another way, plau-
sible arguments operate in situations in which we commonly find
ourselves. They can take a deductive or inductive form, but always
implied is something along the lines of ‘according to available data’,
‘according to available expertise’, ‘the committee are inclined to
believe that’, or ‘our best guess is’.
To put a slightly more hopeful twist on the previous example,
faced with a dog that seems under the weather, and a rumour that
the well that filled its drinking bowl is poisoned, you make a quick
decision based on the information available to you and rush her to
the vet. One way to set out this argument is:

P1: The well might be poisoned.

P2: Lassie usually drinks well water.
C: We better take her to the vet.
121

ARGUMENT FORMS AND FALLACIES 121

This is an example of an enthymeme, and could be shorthand

for either a deductive or inductive argument. A deductive version
might be:

P1: If your dog looks unwell and there is at least some reason to
believe its water source is poisoned then the best course of
action is to rush it to the vet.
P2: Lassie looks unwell and there’s a rumour that toxins from frack-
ing fluid have been polluting the local groundwater.
C: Therefore the best course of action is to rush her to the vet.

There will likely be details of this situation relevant to the conclusion

you reach that can be worked into the argument, but whatever its
content the formulation can always be deductive. That the conclu-
sion must be accepted if the premises are accepted should be now
be clear enough to other discussants, in which case any disagreement
will focus on the plausibility of the premises.
With this in mind, the primary trajectory of critical thinking has
been towards more helpful ways to categorise and assess plausible
arguments. The remainder of this chapter is an overview of this
approach.

5.2 ARGUMENT FORMS AND FALLACIES

The value of understanding the nature and prevalence of plausible
arguments is that it can sensitise us to the presence of a range of
argument forms – and their associated fallacies – that are commonly
employed in everyday decision-making. It is to the classification
and analysis of such arguments that many books on critical thinking
devote themselves, and this will be the focus of much of the rest of
this book too.
Examples of argument forms (sometimes referred to as ‘argument
schemes’) include ‘arguments from authority’, ‘arguments from anal-
ogy’, ‘causal arguments’, ‘generalisations’, and ‘ad hominem argu-
ments’. Argument forms like these will often be recognisable to
you, and this is to be expected since critical thinking is about how
we think and argue in everyday situations. It provides the tools for
assessing different argument forms, and reveals, among other things,
122

122 ARGUMENT FORMS AND FALLACIES

how psychology can explain our vulnerability to fallacious versions

of them.
The study of argument forms emerged from the study of fallacies.
Traditionally in logic, and then in informal logic, space was devoted
to a selection of named fallacies, derived initially from Aristotle’s
‘Sophistical Refutations’ (a section of his work on logic, the Organon).
In terms of practical reasoning, however, it became apparent that it is
not always fallacious to argue in these ways; not always irrational or
foolish to reach decisions on the basis of arguments from authority,
analogy, the character of the arguer, and so on. Establishing crite-
ria for determining the strength of an argument conforming to a
particular type became the focus of many critical thinking scholars’
efforts, and these criteria are primarily expressed in terms of what
are called critical questions. I will say more about these shortly, but
will first consider the notion of fallacies in more detail.
A fallacy is defined by Trudy Govier (2018, p.273) as a ‘mistake in
reasoning... that occurs with some frequency in real arguments and
that is characteristically deceptive’. A fallacious argument is one that
is not only bad, but bad in a way that:

1. Conforms to a recognisable type of error that we are prone to

make when constructing arguments.
2. Tends to create the (superficial) impression of being a good
argument.

Fallacies are, in short, poor arguments of certain kinds that are used
frequently and are liable to be convincing to those not thinking criti-
cally. A bad argument that conforms to a certain pattern but that is
not particularly deceptive is still a fallacy, but the ‘liable to be con-
vincing’ clause remains an important component of the definition.
One reason is that we often argue fallaciously without realising we
are doing so, indicating that, in the moment at least, we think the
argument is strong. And if we are deceived, some of our audience
are likely to be as well. A second reason is that, since there are many
fallacies, any given class or textbook must make choices about what
to include and what to prioritise. A sensible criterion for inclusion
is a fallacy being ‘characteristically deceptive’ since these will be the
more disabling ones in processes of decision-making.
123

ARGUMENT FORMS AND FALLACIES 123

This book has defined critical thinking in terms of the attempt

to avoid reasoning errors, so identifying fallacious arguments is
clearly central to its purpose. As the previous discussion has indicated
though, fallacies need to be distinguished from argument forms, and
this is made harder because, for historical reasons, they tend to share
the same names. You will still find books and teachers that refer to
arguments from authority, arguments from popular opinion, slippery
slope arguments, appeals to emotion, ad hominem arguments, and
so on, simply as ‘fallacies’. In nearly all cases, however, this is not the
right way to understand them, and therefore not the right terminol-
ogy to apply to them.
The approach of this, and many other, critical thinking textbooks
is to analyse argument forms, and (among other things) to determine
critical questions that should be applied to them in order to establish
how good a particular instance of that argument is. In this context
the word ‘fallacy’ takes on two, more subtle, meanings:

1. A particularly poor instance of an argument form is often called

a ‘fallacious ad hominem argument’, ‘fallacious argument from
analogy’, and so on. Alternatively, they can be called ‘weak’ or
‘poor’; the terminology does not need to be that precise.
2. There are some argument forms that are always weak or just
plain wrong, in which case the term ‘fallacy’ can be applied
more freely. A straw man argument is often unproblem-
atically just referred to as a straw man fallacy; and a circular
argument denotes a distinctively fallacious form of reasoning
(see Chapter 10).There are also fallacies that are so common or
distinctive that they acquire their own name despite being an
erroneous version of a particular argument form. For example,
‘confusing correlation and cause’ (rather than ‘fallacious causal
argument’) names a collection of fallacies that can be commit-
ted when constructing a causal argument (see Chapter 8).

5.3 CRITICAL QUESTIONS

‘Critical questions’ are the questions that are important to ask in
order to establish the strength of an argument. There are two kinds:
(1) general ones that can be applied to most arguments; and (2)
124

124 ARGUMENT FORMS AND FALLACIES

specific ones that are relevant to particular argument forms. Some

characteristics of critical questions to be aware of are these:

1. There is no definitive list either of the general or the specific

questions, and you will find some variation across different
textbooks.
2. Where a list is prescribed, judgement is needed on behalf of
the critical thinker to determine which of the questions to
apply, and whether some should have priority over others. As
always with critical thinking, the context of the argument is a
vital guide to how we go about analysing it.
3. Some critical questions will seem to be common sense, others
less so. Sometimes, by paying attention to them, we might be
doing little more than reminding ourselves of how we would
intuitively interrogate an argument, but there will also be
occasions where we will learn new interrogative techniques
and gain new insights into the nature of the argument form
under scrutiny.

GENERAL QUESTIONS

Perhaps the best starting place for critical questioning is to ask

yourself:
• What kinds of questions would you ask about any or most claims
in deciding whether or not to believe them, or at least find them
plausible enough to act on?
Or
• What kinds of questions would you ask about this particular claim
in deciding whether or not to believe it, or at least find it plausible
enough to act on?
These initial or ‘meta-questions’ are beautiful in their simplicity, and
they are the hallmark of a critical thinker. They imply appropriate
perspective and a willingness to take responsibility for selecting the
tools of one’s evaluation. (Before reading on, it is a worthwhile task
to apply your mind to the first of these questions and to start generat-
ing a list of basic critical questions.)
125

ARGUMENT FORMS AND FALLACIES 125

Ralph H. Johnson and J. Anthony Blair (2006) identify three basic

critical questions, otherwise known as the ‘Acceptability, Relevance,
Sufficiency’ criteria (generating the slightly unfortunate acronym
ARS). ARS criteria are briefly explained in Box 5.1.

BOX 5.1 BASIC CRITICAL QUESTIONS

1. Are the premises acceptable, i.e., likely to be true, or likely to be

accepted as true or plausible by the people one is attempting
to persuade?
2. Are the premises relevant to the conclusion the arguer is
­seeking to establish?
3. Are the premises sufficient for establishing the conclusion?

These questions are theoretically independent of one another. A

premise can be acceptable but not relevant, or relevant (if true) but
not acceptable, and even if all premises are acceptable and relevant,
they might not be sufficient by themselves to persuade us of the
truth of the conclusion. However, in practice, if a premise is ruled out
because it is not acceptable then its relevance is of no consequence,
and it cannot be sufficient for establishing the conclusion. For this
reason, acceptability should be the first question that is posed.
So, for example, it might be claimed that voluntary euthanasia
should not be legalised because all killing is wrong. This premise,
if true, will be relevant to the conclusion, and because euthanasia is
undoubtedly a form of killing, sufficient for establishing the conclu-
sion as well. However, many will counter that not all killing is wrong,
especially not a mercy killing carried out on the basis of someone’s
valid consent. In this case they will find the argument unconvincing
because the premise is unacceptable. Someone else might argue that
euthanasia is wrong because it contravenes a religiously derived pro-
hibition on suicide. In response to this it could be argued that reli-
gious prohibitions are not relevant in a modern secular society. Or it
could be argued that even though mercy killing is permissible, this is
a necessary, but not sufficient reason (see Box 5.2) for legalising euthana-
sia. Possible abuses of the law, and old and terminally ill people feeling
126

126 ARGUMENT FORMS AND FALLACIES

that they should die so as not to be a burden on relatives, are potential

consequences of legalisation that could outweigh the moral merits of
allowing people to choose to end their suffering.

BOX 5.2 NECESSARY AND SUFFICIENT

CONDITIONS
In logic, the concepts necessary and sufficient conditions are
important. A necessary condition for something is one that is
essential to it – such as having a ball (or ball substitute) being
necessary for playing a game of football. Having sufficient condi-
tions means, simply, that everything needed for something to be
the case has been provided or established. Sometimes some of
these conditions are also necessary conditions, but not always.
Playing football on a Tuesday night is sufficient to keep me hap-
pily occupied for an hour, but many other things could fulfil this
role as well.

Box 5.3 shows how each of these fundamental questions can be

connected to a range of possible sub-questions (many of which we
will look at in detail in subsequent chapters).

BOX 5.3 BASIC CRITICAL QUESTIONS

AND SOME EXAMPLES OF
SUB-QUESTIONS
1. Are the premises acceptable, i.e., likely to be true, or likely to
be accepted as true or plausible by the people the arguer is
attempting to persuade? For example:
• Are the claims being made clear?
• Is appropriate evidence cited in support of the claims being
made in terms of quantity and quality. For example, where
experts are cited, are they the right experts? Do they repre-
sent a consensus in their field?
• Is the author biased in a way, or to an extent, that should
make us question the objectivity of the evidence presented?
127

ARGUMENT FORMS AND FALLACIES 127

• Are any of the claims inconsistent with one another?

• Have opposing positions been accurately represented?
2. Are the premises relevant to the conclusion the arguer is seek-
ing to establish? For example:
• Has the appropriate kind of evidence been presented? For
example, what the general public think about who should
govern them is relevant to who should govern them, but what
they think the consequences of a 4-degree global temperature
rise will be is not relevant to what these effects will actually be.
• Where analogies are employed, are they similar enough to
the issue being discussed to allow us to reach conclusions
based on those analogies?
3. Are the premises sufficient in order to establish the conclusion?
For example:
• If the conclusion is a generalisation, is this one that can be
supported by the evidence provided?
• Has relevant information, or have relevant perspectives,
been over-looked?
• What alternative conclusions can be drawn from the evidence?

Returning briefly to deductive arguments, notice that a sound

argument is the strongest kind because it fulfils all the ARS criteria.
A valid deductive argument that is not sound (where at least one of
the premises in untrue) meets the relevance and sufficiency criteria,
but not acceptability.

QUESTIONS SPECIFIC TO PARTICULAR ARGUMENT FORMS

On a British political analysis TV programme in 2015 the possible

repeal of the fox hunting ban in Britain was being discussed.2 Rock
guitarist and animal welfare activist Brian May wanted to retain the
ban, arguing that it is not a sport and that it is cruel. Former front-
bench politician Michael Portillo agreed that it is not a sport but was
in favour of repealing the ban because the state should respect certain
traditions, even if viewed by others as distasteful or cruel. He is no fan
128

128 ARGUMENT FORMS AND FALLACIES

of bullfighting, but said that the argument would extend to allowing

this tradition to be maintained as well. At this point Brian May and
the host – experienced journalist Andrew Neil – retorted that the
same argument could apply to witch hunts (May) and bear baiting
(Neil). Portillo said that it couldn’t, that these were different, and at
this point the discussion had to end because they had run out of time.
What we have here is an example of an argument from anal-
ogy. They are commonly used and often quite weak, but even when
they are weak they can be significantly persuasive. They are, then,
important to analyse and will be the subject of Chapter 9, but for the
moment I am going to preview this discussion and use them as an
example of specific critical questions.
The basic structure of an argument from analogy is this:

P1: X is similar to Y.
P2: Z is true of Y.
P3: Therefore Z is also true of X.

In the case of Andrew Neil’s argument, X, Y, and Z can be filled in

like this:

P1: Fox hunting as a tradition is similar to bear baiting as a tradition.

P2: Bear baiting was banned.
C1: Therefore fox hunting should also be banned.

Four of the critical questions that can be applied to arguments from

analogy are these:

Q1: Is what is said of Y actually true (or plausible)?

Q2: Are there relevant similarities between X and Y?
Q3: Are there dissimilarities between X and Y that undermine the
similarities?
Q4: Can convincing counteranalogies be found?

Applied to Neil’s argument the likely answers – i.e., the basis for
further research, or what we would put back to him if we were in
Portillo’s shoes – are these:

1. This is true; bear baiting was a tradition in England until the

nineteenth century.
129

ARGUMENT FORMS AND FALLACIES 129

2. There are some relevant similarities; most obviously that both

involve harming animals (including the dogs in bear baiting)
as a necessary part of a practice from which people derive
recreational pleasure.
3. It would appear that there could be sufficient dissimilarities
to undermine the similarities. For the sake of this example it
is not appropriate to go into much detail, but the degree of
cruelty, the degree of human skill and courage involved, and
the victim having a sporting chance, are among several that
differentiate the two practices.
4. Portillo mentioned the counteranalogy of bullfighting, but this
is not as strong as it might be since there is also a lot of support
for banning this. Alternative comparisons are other traditional
forms of hunting (deer, grouse, etc.) that are less controversial.

The first three critical questions tend to be where our basic rational-
ity will take us when encountering arguments from analogy, but I
think that the fourth one is slightly less intuitive, and it is also the
hardest one to answer since it involves a lot of imaginative work.
However, even if all four are relatively obvious, familiarity with a list
like this – ­questions that have been broadly agreed upon by a range
of ­scholars – gives us some confidence that these are the angles we
need to pursue in the discussion.
Clearly this is just the start of a long debate (which is itself just one
part of the discussion around fox hunting), but knowing the basic
questions to pose once an argument from analogy has been identified
is a good start, and one that has the potential to lead the exchange
along constructive lines.
Returning to the ARS criteria, notice that critical question Q1
for arguments from analogy is a form of the acceptability criterion;
Q2 concerns relevance; and Q3 and Q4 question the sufficiency of
the premises.

CRITICAL QUESTIONS AND IMPLICIT ASSUMPTIONS

In the previous chapter, on argument reconstruction, we discussed

the importance of making implicit premises explicit. This helps
ensure that we do not overlook assumptions the arguer is making
that need to be justified. Identifying argument forms in reconstruc-
tions is also about making implicit assumptions explicit.
130

130 ARGUMENT FORMS AND FALLACIES

An argument commonly voiced during the 2015 Eurozone crisis

(caused by Greece defaulting on a major debt repayment) was that
if Greece leaves the single currency then other countries will follow,
leading to the collapse of the Eurozone. This is a version of a slip-
pery slope argument, one in which it is claimed that a relatively
small and containable happening will inexorably lead to a series of
events that ends in disaster (for more on this argument form see
Chapter 8). In the process of identifying something as a slippery slope
argument three assumptions are being made: (1) a specified sequence
of cause and effect is likely to happen; (2) if it does it will be beyond
the control of relevant parties to put a stop to; and (3) the final con-
sequence will be significantly negative. These then generate criti-
cal questions that examine the likelihood of the sequence of events
occurring, whether it really will be unstoppable once it has started,
and whether the final consequence is in fact disastrous.
In a similar way, the formal structure of many argument forms con-
tains a premise which makes an often implicit assumption explicit.
An argument from authority, for example (see Chapter 6), such as
Richard Dawkins says that the world would be a better place without
organised religion, therefore it is likely that the world would be a better
place without organised religion.
Would be fully reconstructed as:

P1: Richard Dawkins says that the world would be a better place
without organised religion.
P2: Richard Dawkins has relevant expertise on the history, politics,
and ethics of organised religion.
C: Therefore it is likely that the world would be a better place
without organised religion.

P2 is the implicit premise that, when made explicit, clearly reveals the
argument as an argument from authority. With respect to the guid-
ance on reconstructing enthymemes offered in Chapter 4, it might
well be unnecessary or inadvisable to fully reconstruct the argument
in this way in many situations, but it can certainly be helpful to the
process of making clear one’s critique of a position that relies on
Dawkins’ authority on this matter.
131

ARGUMENT FORMS AND FALLACIES 131

5.4 SUMMARY, AND THE WIDER ANALYSIS OF

ARGUMENT FORMS
This chapter began by exploring the nature and significance of
deductive, inductive, and plausible arguments. Emphasis on the latter
explains the value of analysing the wide variety of commonly used
argument forms (or schemes) and the role of critical questions
as a tool for evaluating them.
The argument forms and associated fallacies that will be discussed
in subsequent chapters have been selected because they are particu-
larly prevalent across a wide variety of contexts. Most of them are
also what I would call ‘interesting’ in the sense that an analysis of
them can open up deep and wide considerations about the human
condition and about oneself.They can, you might say, serve as a portal
into the themes underpinning this book: rationality and persuasion,
dispositions, and constructive dialogues. From a study of the argu-
ment forms we commonly use and the ways in which they can go
wrong we can learn a lot about, not just argument reconstruction and
evaluation, but our character and the way we interact with others. For
example, fallacies are, as John Woods puts it, ‘like bad habits … hard to
break’ (2013, p.5), and it is their link to biases and heuristics that helps
explain, not just why this is the case, but why certain argument forms
and fallacies are as frequently employed and as seductive as they are.
It is with this in mind that the argument forms and fallacies featur-
ing in Chapters 6–9 will be investigated in relation to the following
themes and headings:

• Description and basic structure.

• Critical questions used to guide our evaluation of them.
• Their relationship to heuristics and the psychology of persuasion,
or what might be called their ‘rhetorical power’.
• Dispositions they are liable to foster, or to be a sign of, and their
significance in terms of our pursuit of constructive dialogues.

EXERCISES
My view is that there is limited value in applying the ARS questions
to relatively unfamiliar and decontextualised arguments, so instead of
132

132 ARGUMENT FORMS AND FALLACIES

providing some of these (as I did at the end of the last chapter), my
recommendation is to select passages that involve arguments from
topics and texts you are familiar with and apply the ARS critical
questions to them. This could include:
• Other disciplines you are studying;
• Topical events and debates;
• Extended texts such as those recommended in this book;
• Essays and other assignments that you have written.

FURTHER READING

• Perhaps the most comprehensive directory of argument forms is

Walton, Reed, and Macagno’s Argument Schemes (2008); and a less
comprehensive but more discursive book covering the details of
a range of important argument forms and fallacies is Christopher
Tindale’s Fallacies and Argument Appraisal (2007).
• Two sources in which the Acceptability, Relevance, Sufficiency
criteria are employed explicitly and with clarity are Johnson and
Blair’s Logical Self-Defense (2006); and Hughes, Lavery, and Doran’s
Critical Thinking (2010).
• If you would like to know more about deductive and inductive
arguments some good sources are Bowell and Kemp (2015) and
Ennis (1996a).

NOTES
1 Also, certain forms of deductive argument, and their associated fallacies, fea-
ture in Chapter 10.
2 This Week, 2.7.15
133

ARGUMENTS, SOCIAL POWER,

AND MESSAGE SOURCE (PART 1)

Facts come wrapped in authority. (Charles Willard, 1990, p.13)

Arguments from authority are lame (Lee, Lifeson, and Peart, 1974; Scott,
Thistlethwaite, and Wallinger, 1983; Amos, 1992; Hütz, Ryabtzev, and
Lemeshev, 1999; Baker Jr. et al, 2014). (Sacha T. Burnstorm, pers. comm.)

Questioning the influence that power and authority have on our

beliefs and actions is close to the heart of critical thinking. As we saw
in the opening chapter, a leading motivation of Western philosophy,
as epitomised by the Scientific Revolution and the Enlightenment, is
to free ourselves from an uncritical frame of mind. Whether encour-
aged by social institutions or by our own dispositions, reclining in the
comfort of conventional wisdom and its authority figures is typically
seen as detrimental, both to individual well-being and to the wider
culture.
In the context of argumentation, a common way of deflecting
requests to justify one’s beliefs is to appeal to an authority. By doing
so we are implying: ‘Argue with them, not with me’, or ‘Do you
think you know better than authority X?’ In the former case we are
failing to take ownership of our belief, perhaps seeing ourselves as

DOI: 10.4324/9781003247944-7
134

134 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

part of a larger system with which the responsibility lies. The impli-
cation of the second is that there are reasons for thinking authority
X is indeed the right authority to base our beliefs on, and it is this
reasoning that turns out to be the most important consideration for
critical thinking.
Since we cannot become experts in all the fields that are relevant
to the decisions we make in our personal and professional lives, reli-
ance on authority is unavoidable. There are also plenty of circum-
stances where an unquestioning approach to authority is usually the
best approach. The crucial thing is for us to know why this is the
case; why X is the right kind of authority to help inform our beliefs
on a particular topic, or why circumstance Y (a battlefield, medical
emergency, restaurant kitchen during peak demand, etc.) is the place
and time not to question orders. Modern analyses of arguments from
authority recognise the limitations of an individual’s knowledge and
powers, and the critical questions they generate embody this under-
standing. In Johnson and Blair’s words, ‘we are urged to be filters of
opinions rather than sponges who soak them up indiscriminately’
(2006, p.167).
The flipside of appealing to authority in support of our positions
is attempting to discredit those who oppose our beliefs. This can be
because of their lack of expertise, their biased view on the subject,
inconsistency between what they advocate and how they behave,
or simply because of their undesirable character (or associations
with other undesirable characters). All these forms of argument are
known as ad hominem arguments, and are a main focus of the
next chapter.
This chapter is about arguments in which either characteristics of
the arguer, or people that the arguer refers to, constitute the main
premise. It will stretch its net wider than arguments that appeal to or
discredit expert authority, addressing several argument forms that rely
on what has been called ‘social power’. The concept of social power
is a broad one, but generally refers to the ability of one person (or a
collection of people) to influence another person (or collection of
people). By influence is meant a change in behaviour and/or beliefs
and attitudes.
In their widely employed analysis of the subject, social psycholo-
gists John French and Bertram Raven (1959) identified five types of
135

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 135

social power: legitimate power, expert power, reward power, coer-

cive power, and referent power. A sixth type – information power –
was later added by Raven (1965). Other theorists have suggested
further categories and sub-categories, including witness testimony
(which I am treating as a subtype of information power) and ethotic
power.1 These six original forms of social power, along with ethotic
power and witness testimony as a sub-category are explained in Box
6.1 below, and it is these categories that will structure this chapter.

BOX 6.1 FORMS OF SOCIAL POWER

French and Raven’s original categorisations

Expert power: The authority of someone who is an expert, and whose
expertise is relevant to the issue at hand. We would thus take seriously
the view of a well-respected scholar of twentieth-century European
social and economic history on the origins of the European Union,
but be more cautious about their pronouncements on ancient Egypt.
Information power: There are plenty of situations in which, though a
person is not an expert, they do nevertheless possess knowledge that
is pertinent and valuable to that situation. The blackmailer trades
on information power, as does my wife when she hides the channel
changer when the cricket is on TV. An important subdivision of infor-
mation power is witness testimony, referring to someone being pre-
sent at and observing a one-off event. In court cases the information
power of a witness is of course very important for establishing truth.
Legitimate power: The authority conferred upon someone by virtue
of the position that they hold (e.g., teacher, judge, parent, checkout
worker) or ‘some sort of code or standard accepted by the individual,
by virtue of which the external agent can assert his power’ (French
and Raven, 1959, p.265). (Note that this second aspect is important
but often overlooked – invoking human rights, whether or not you are
a lawyer, is also an example of legitimate power.)
Reward power: Rewards come in many forms, from pay (e.g., the
prospect of a bonus at work) to praise (e.g., compliments from a
friend on a new hairstyle). Anyone able to influence the behaviour
136

136 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

of others through rewards has reward power, which extends this cat-
egory beyond the positional and normative reach of legitimate power.
Coercive power: The power of someone who is able to punish someone
who does not comply with a request or command. It is the opposite of
reward power, but similar to it in so far as it can be held by someone
without legitimate power.
Referent power: In contrast to the relatively impersonal legitimate and
expert power, referent power is held by a group or person who you
personally connect to. They demonstrate values and other qualities
that are – or you would like to be – an important part of your identity.
Without the need for requests, orders, or threats, you thus have a
desire to conform to a group’s or a person’s norms, and in this way
those norms have power over you.

An additional form of social power

Ethotic power: The power of a person who is regarded, in a general
sense, as a ‘good’ person (or a virtuous or moral person). It is some-
one who, without reference to expertise, any position they occupy, or
specific codes of conduct they adhere to, is of good character and
serves as a role model.

‘Power’ is quite a general term that is flexible enough to be used in

the ways that French and Raven do. ‘Authority’ is a more specific
concept that tends to refer to an official or otherwise socially rec-
ognised sanctioning of someone’s power. The line ‘by the authority
invested in me …’ captures this objective quality, and it thus makes
sense to talk about ‘legitimate authority’, ‘expert authority’, and pos-
sibly ‘ethotic authority’, but less so to use it in relation to the other
kinds of power. Other critical thinking textbooks usually concentrate
on expert authority under the heading ‘arguments from authority’,
but because the perspective here is wider, both power and authority
become relevant (and sometimes interchangeable) terms.
Before moving on to discuss these in detail, I will add a quick
comment about the source or location of social power. Put briefly,
a person’s power does not simply reside in them independently, but
137

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 137

is actualised through being recognised or accepted by other people.

My mother might be an expert on parenting, but her advice won’t be
heeded by me unless I acknowledge this. This dependency explains
the importance of signalling power via, for example, uniforms and
titles.
To sum up this introduction, we will consider whether and when
it is justified to act, or to form beliefs, in accordance with sources of
power with reference to two broad types of argument:

1. Positive arguments that claim that because a particular person

believes or does X we should consider believing or doing
X. These include appeals to expert, legitimate, and eth-
otic authority; arguments based on information power
(including witness testimony) arguments based on referent
power (a species of which is known as the ad populum argu-
ment), and arguments that are based on coercion (known as
ad baculum arguments). Arguments from authority and power
can either be referring to the authority/power of the per-
son who is presenting the argument – the explicit or implicit
message that you should take them seriously because of what
they know, who they are, etc. – or they can be referring to a
third person.
2. Negative arguments – otherwise known as ad hominem
arguments – which claim the opposite: that because cer-
tain beliefs or actions are those of, or those associated with, a
particular person, this counts as a reason for not believing or
doing these things.These are sometimes called ‘poisoned well’
arguments, the metaphor referring to the source of the view
being compromised (by lack of expertise, biased views, poor
character, and so on) such that anything that springs from it
carries this contamination.

6.1 ARGUMENTS FROM EXPERT AUTHORITY

Since it is not possible to find out everything for ourselves directly
we have no choice but to include second-hand information in many
of our rational deliberations. The question of which information to
trust becomes the question of who to trust; the need for competence
138

138 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

in our assessment of expert advice. By ‘expert’ we typically mean

someone who has undergone substantial education or training. They
have considerable experience of working in their specialist field, and
the level of their knowledge is often verified by qualifications, influ-
ence, and peer approval.
The basic structure for any argument that offers proof or evidence
for a position by appealing to the expertise of the person or people
supporting that position is this:

P1: X is an expert in the domain of Y.

P2: X believes that Z in relation to Y is true.
C: Therefore Z is true (or Z is plausible).

Other argumentation and critical thinking textbooks vary in

the number of critical questions they list,2 but I will organise the
essential criteria for assessing this type of argument into four main
questions:

Q1: Is the authority cited an expert in their field?

Q2: If so, is this field relevant to the issue under discussion?
Q3: If so, is the authority cited to be trusted?
Q4: If so, is the strength of the conclusion reached appropriate?

Each of these, though, is associated with a number of impor-

tant sub-questions, and some of these will be discussed
below.

THE NATURE OF THE EXPERTISE

Is the authority cited an expert in their field, and is this field relevant
to the issue under discussion? If someone is a medical doctor and
they are giving their opinion on a common bodily ailment, then we
can say yes to Q1 and Q2. If they are making a claim about a psy-
chological condition, then we need to be more cautious. This much
is obvious, but if we are not concentrating we can easily be fooled.
Not every historian is an expert on Scottish history, and not every
Scottish historian is an expert on the Jacobite Risings, so if some-
one commenting on the Battle of Culloden is only introduced as a
139

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 139

‘historian’ or even ‘Scottish historian’, then we do not yet know if

we are in safe hands.
Expert knowledge tends to be highly specialised, and this is why
Q1 and Q2 are so important. However, we also need to recognise
that generalist and interdisciplinary writing and research has a couple
of vitally important functions:
• Most of the big issues affecting us (such as climate change, global
inequality, the coronavirus pandemic) require multiple discipli-
nary perspectives to understand and formulate responses to them.
• Generalist writers (such as journalists) are often better at commu-
nicating these issues to wider publics than are specialists.
There is, then, an unavoidable trade-off between specialist expertise
and a broader perspective, and we need to keep this in mind when
assessing the value of blogs, columns, books, and documentaries
which tackle big questions for non-specialist consumption (includ-
ing any influence they might have on policymakers). The books of
journalist and political activist Naomi Klein are good examples. Her
2014 work This Changes Everything combines climate science with
ethics, politics, and economics. Few people, including academics, are
expert in all of these areas, but it is entirely reasonable to argue that
someone needed to write this book, and that that someone needed
to be able to reach a large and broad audience. Klein’s lack of formal
expertise in climate science inevitably affects the book’s credibility,
but three factors help to limit its negative impact: (1) she is very
open about the efforts she has made in the years researching the
book to properly understand climate science and its implications;
(2) she makes continual reference to multiple sources to support
her arguments, and includes a large number of endnotes detailing
their origin; and (3) in her many years as a journalist and writer she
has gained a reputation for scholarly research and professional and
­personal integrity.
So, when asking questions about expertise we need to be ­sensitive
to the limitations of specialist knowledge.We also need people willing
to investigate more than one discipline and synthesise the findings of
a range of experts in pursuit of solutions to serious and highly com-
plex problems. Our evaluation of such efforts needs to be differently
oriented; recognising, for example, that despite some of their claims
140

140 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

being more provisional than in narrowly focussed research, this does

not necessarily detract from the larger picture they are presenting.

THE TRUSTWORTHINESS OF THE EXPERT

Indicators of an expert’s trustworthiness include the extent to which

their views on the issue are shared by their peers (consensus), the ways
in which they appear to be applying their knowledge to the problem
(judgement), and whether some form of bias might be compromis-
ing their objectivity. The question of bias will be considered under
circumstantial ad hominem arguments (below), and in this section I
will make some observations about consensus and judgement.

Consensus
In a radio programme3 interviewing the world famous entomologist
and founder of sociobiology, E.O.Wilson,Wilson defended his theory
of group selection over the selfish gene theory of cooperative behav-
iour.4 In this peak hour broadcast it was instructive to see how this
complex debate was defended by both Wilson and his rival Richard
Dawkins (who had also been interviewed on the subject). Most of
the listeners would lack the expertise to be convinced one way or
the other by the scientific arguments, so both relied significantly on
arguments from authority. Wilson made the point that mathemati-
cal modelling disproved the selfish gene theory. He had little option
but to spare us the maths, but did stress that his co-researchers were
two well-respected Harvard mathematicians: Martin Nowak and the
‘genius’ Corina Tarnita.
Beyond this, it should be added, Wilson went some distance to
explaining and defending his theory without reference to authority,
but Dawkins was quite explicit about needing to use an argument
from authority (something he said he generally dislikes doing) and
simply pointed out that a reply to Wilson’s paper elicited a damming
response in the form of a letter to the journal Nature (who published
Wilson’s article) from 140 eminent evolutionary biologists. His argu-
ment came down to numbers: yes, Wilson is a leading authority in
evolutionary biology, but on this issue he is vastly outnumbered by
other experts. Who are you going to trust? The first part of Wilson’s
141

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 141

response to this was to draw an analogy with a 1921 paper entitled

‘One-hundred physicists against Einstein’, which erroneously argued
against the general theory of relativity.
It is also noteworthy that the rationale for the letter sent to Nature
referred to above was to ‘keep non-specialists from wasting time’ on
the theory, and that wasted time was a genuine concern because of
‘Nowak’s fame and Nature’s prestige’.5 The non-specialists referred
to would presumably include academics from other disciplines (such
as psychology) and journalists, again underlining the risks associated
with generalist and interdisciplinary writing and research. In the
Klein case, discussed above, there appears to be enough of a consen-
sus among relevant scientists about enough of the points she makes
to maintain the thrust of her economic and political argument. But
in the case of Wilson’s theory, the non-expert would have to tread
very carefully indeed, and the letter was intended as a warning to
this effect; one that was perhaps seen as necessary because of various
heuristics triggered by the big names involved.

Expertise and judgement

A different aspect of the trustworthiness of experts concerns the
intellectual activity that contributes to their decision-making. For
example, I may have every reason to have confidence in the expertise
of my dentist, but if he recommends a certain (significant or drastic)
treatment on a day in which he appears to be particularly harried or
distracted it might well be sensible to not act on this until I see him
again, or until I seek a second opinion. This highlights how expertise
is not just about years of learning and practice, it is also about the
judgements made which apply this experience to the matter at hand.
Put another way, as a critical thinker our judgement on the expert
authority of someone includes our consideration of them as a criti-
cal thinker. What to look out for in this respect is broadly covered
by the critical thinking dispositions (see Chapter 3), and this can be
condensed into the following critical question:
Is the authority in question behaving in accordance with relevant critical
thinking dispositions? (For example, are they attending carefully to the
precise facts of the matter? Are they being open-minded rather than dog-
matic when considering options?)
142

142 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

In some situations an additional relevant question might be (Goodwin,

2011, pp.293–4):
What do they have to lose if they turn out to be wrong?

Of course, a learned practitioner might have developed some habitual

weaknesses affecting their judgement, in which case the direct (or abu-
sive) ad hominem argument can legitimately be employed against them.
This will be discussed further under ‘Arguments appealing to character’.6

THE STRENGTH OF THE CONCLUSION

The appropriate degree of confidence in an argument appealing to

expert authority hinges importantly on whether the conclusion con-
cerns a course of action, or whether it concerns what to believe. In
the latter case an element of caution is often appropriate, but in cases
where there is total consensus among experts the better approach is
perhaps one of acknowledging one’s distance: not ‘X is true’, but ‘I
have every reason to believe that X is true.’
We should now be applying this kind of language to our belief in
anthropogenic climate change, but even if the IPCC were less sure
than they are (for example, if it was 87% of scientists rather than 97%)
the argument for acting as if it were true has a firmer conclusion:

P1: A clear majority of scientists believe that human activity is a

significant cause of global warming, and that the consequences
of global warming will be extremely serious.
P2: A clear majority of scientists believe that immediate reductions
in carbon emissions will slow down global warming, leading to
less severe consequences.
C: Therefore we need to act now to reduce carbon emissions.

This example also highlights the value of an extra critical question

introduced by Christopher Tindale (2007, p.142):
What are the consequences of accepting what the authority says, or ignor-
ing what is said?

With climate change, if the experts are right then the consequences
will be catastrophic, so we ignore them at our peril. For most of us at
143

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 143

least, the same cannot be said for authoritative views on the surface
temperature of Venus, or the history of cricket.

EXPERT AUTHORITY: PSYCHOLOGY AND RHETORIC

It is unsurprising that psychologists have identified an ‘expertise

­heuristic’ which prompts us to jump to conclusions when expertise –
or the appearance of expertise – is associated with an idea or product
(Smith, Mackie, and Claypool, 2014, pp.248–9). In order to exploit
these heuristics, a professional persuader will need to know who the
trusted experts are for a particular audience, and they will need to
know what the potent signifiers of expertise (or competence) are.
Once experts are known, persuaders can play on our tendency
to erroneously generalise expertise (interrogated by Q1). For exam-
ple, Carol Vorderman, an engineering graduate and TV presenter,
originally made famous as the maths person on the UK game show
Countdown, has been employed to lend scientific credibility to fish
oil and other health products. Generalisation of this sort can be
partly explained by the halo effect, i.e., assuming, without evidence,
­further positive qualities in a person on the basis of a few known
positive qualities. (This discussed in more detail in Chapter 7, under
ethotic arguments.)
Rapid message delivery, steady eye contact, and even face shape
(Smith, Mackie, and Claypool, 2014) are among the many non-­
verbal and paralinguistic signs of expertise and trustworthiness. Owen
Hargie (2017, p.372) identifies the ‘Three Ts of expert power’: titles
(Professor, Doctor), threads (clothes, uniforms), and trappings (other
signs such as framed degree certificates, shelves of technical books
and journals, or specialist equipment). In a confessional moment in
the TV series Billions the high-powered attorney Chuck Rhoades
(played by Paul Giamatti) explains why he always wears a suit to
social functions: he’s short and relatively unattractive and doesn’t
want people asking, with respect to his glamourous wife, ‘What’s she
doing with him?’The suit confirms his status, including the extent of
his expertise and the legitimacy of his power.
A separate phenomenon regarding trust in experts concerns
­situations in which the scientific consensus is rejected by certain
people or sections of a population (climate change and vaccinations
144

144 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

are current examples). The puzzle is that the intuitive answer – that
such people are less scientifically literate or rational – is at best only a
small part of the explanation. Research is now filling in the gaps, and
some of this relates to a subsection of the cognitive biases and social
dynamics discussed in this book. To go deeper into this complex and
fascinating area there are references included in exercise 2 at the end
of this chapter.

EXPERT AUTHORITY, DISPOSITIONS, AND CONSTRUCTIVE DIALOGUES

Kant pointed to ‘laziness and cowardice’ as

The reasons why so great a proportion of men … gladly remain in lifelong
immaturity, and why it is so easy for others to establish themselves as
their guardians. … If I have a book to serve as my understanding, a pas-
tor to serve as my conscience … and so on, I need not exert myself at all.
([1784] 1963, p.35)
Why laziness can be a motivator to over-use expert appeals is
clear enough, and cowardice refers to their role in avoiding direct
accountability for our beliefs and decisions. If degrees of these
vices can make us too keen on authority-based arguments, then an
absence of modesty can have the reverse effect. The person lacking
respect for authority is typically as closed-minded at the person
with too much respect. In both cases, the fault lies with overreliance
on limited sources – on oneself in the case of the immodest person,
and on too few others in the case of the dogmatic traditionalist.
In the latter case, open-mindedness would usually reveal that there
are multiple authoritative points of view, and this knowledge alone
should mitigate any tendency to confuse authority with truth. A
balance is needed between excesses of independence and depend-
ency, mediated by dispositions such as courage, open-mindedness,
and modesty.
In terms of its effect on dialogues it has been claimed, reasonably,
that ‘to invoke authority is to abort debate’ (Willard, 1990, p.18).The
effect of an appeal to authority that is accepted by the other person
is that it shifts the burden of proof. A matter is disputed, neither
party is able to convince the other through other forms of argument,
but Discussant 1 says that he has it on good authority that what he
145

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 145

is saying is the case. If enough of the criteria for a strong appeal to

authority appear to be met, then the ball is now in Discussant 2’s
court. Discussant 1 is not only unconvinced by Discussant 2’s argu-
ment, he has an authority to support his view, and so at this stage
there is little more Discussant 2 can do but to research the views and
relevant attributes of the authority in detail and/or find their own
authorities in support of their view.
None of this is unconstructive so long as: (1) the argumentation
process really has gone as far as it can under the circumstances; and
(2) all participants understand the reasons why this is the case. In
this way, frustration is avoided and the door is open for the dialogue
to recommence. In an important sense, responsibility is shifted to
authorities, but these critical thinkers retain control over the situation
in a number of respects: they are responsible for finding the relevant
authorities; for understanding them as far as they can; and, impor-
tantly, for recognising that an appeal to authority is the right way to
proceed if any agreement is to be reached.
It should also be said that this information-seeking part of the
discussion process need not be combative in the way implied here.
There is no reason why all parties cannot jointly research the subject
in order to make progress. Indeed, a willingness to do this suggests
suitable open-mindedness and the collaborative process could itself
lead to a softening of biases.

6.2 ARGUMENTS FROM POSITION TO KNOW

(INFORMATION POWER)
The significance of information power is concisely expressed by Alan
Brinton in this way:
Sometimes another person has access to information or evidence
which we do not have, either because it is inaccessible to us or because
the demands of the situation make it impracticable for us to get it.
(1986, p.255)
A corollary of information power in argumentation is arguments
from ‘position to know’ (Walton, Reed, and Macagno, 2008). Being
in a position to know is to have access to facts, whereas expertise
refers to deep and broad knowledge of the field in which those facts
146

146 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

are located. Examples of position to know range from our local geo-
graphical and cultural knowledge (where the local library is, the best
place to find job ads, how much, if at all, should we tip bar or res-
taurant staff) and basic procedures and norms of the workplace (who
to send expenses forms to, how to address senior staff in an email),
through to the boss who withholds or distorts information gathered
at high-level committee meetings in order to maintain control over
subordinates.7 Altogether juicier examples include revelations about
love affairs, corruption, illegal parties, and other immoral or criminal
behaviours that are found in autobiographies, shared in confidence,
haplessly revealed in TV interviews, leaked by whistle-blowers, or
obtained through computer hacking.
Examples of position to know arguments are endless, and take
us directly to some deep issues in ethical and political philosophy.
Information power is employed by governments in undemocratic
acts of propaganda, by civil servants who keep government ministers
in the dark, and it exists in the unavoidable agenda-setting that goes
on in news media. Information power can also be used benignly,
and on occasion – like those who wish not to know the full details
and extent of their terminal medical conditions – we can justifiably
choose to remain ignorant.
An important consideration for the position to know argument is
the difference between expertise and experience. Contrast, for exam-
ple, knowing all about cancer from a scientific perspective and living
through cancer; or between knowing about heartbreak from a liter-
ary and psychological point of view and having your heart broken.
One way of appreciating the significance of this difference is in terms
of the knowledge that qualitative research often seeks out. In its quest
for expert knowledge it finds out about the first-hand accounts of
people who have had certain experiences. Alternatively think of how
news reporting typically combines the experiences of people at the
sharp end of, say, welfare cuts or health service reforms, with expert
economic or political analysis. It is important not to view accounts of
experiences as provisional or otherwise inferior to theoretical under-
standing. For a number of reasons – including contextual and ‘felt’
qualities – it offers a distinctive insight into an issue, rather than one
that is simply waiting to be replaced by a theoretical account.
147

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 147

The basic structure of position to know arguments is:

P1: X has access to knowledge about Y.

P2: X believes that Z in relation to Y is true.
C: Therefore Z is true (or Z is plausible).

The critical questions are similar to those of arguments from expert

authority:

Q1: Is appealing to position to know appropriate in these

circumstances?
Q2: If so, is X really in a position to know?
Q3: If so, is X trustworthy?
Q4: If so, is the strength of the conclusion reached appropriate?

Q1 is worthy of some elaboration, mostly in the form of sub-­

questions. First, we should ask whether this is information we are
able to find out for ourselves, and if so, question our motives for rely-
ing on others (which might be entirely reasonable). Second, though,
if someone or something is stopping us from finding out, does this
prohibition have legal, moral, or normative (such as position-based, as
in the case of a parent) legitimacy? If not, can we exercise legitimate
powers in order to obtain the information (such as the Freedom of
Information Act or a moral appeal)? But if so, is it worth the effort?
(Or is it, for example, better to remain ignorant, or pay someone to
inform us?)

Witness testimony
For certain kinds of news stories (including extreme weather events
and terrorist attacks) news media rely heavily on witness testimony.
Witness testimony differs from position to know in that it refers to
situations where people experience an unusual event first-hand, and
thus become valuable for establishing the truth about it. Position
to know as a corollary of information power implies a degree of
stability in the information’s availability so that it is available to any-
one who happens to be in a ‘position to know’. The eyewitness, in
148

148 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

contrast, happens upon something through luck (or ill luck) rather
than anything about them or their station in life. And that thing is in
important senses a one-off, rather than something we can (or would
want to) recreate.
As well as news reporting, eyewitness testimony can of course be
important in court cases (it plays a pivotal role in Twelve Angry Men),
the school playground (‘Who saw what happened to Rory’s lunch
box?’), and many other circumstances. It is, however, notoriously
unreliable in two interrelated respects:
• Mistaken beliefs: cognitive psychology has demonstrated that
there tends to be a substantial gap between what we confidently
believe we have seen and what actually happened.
• Lying: in circumstances were a person stands to gain from an event
to which they are the only witness, the temptation will be there
to distort or fabricate the truth. This can of course work at a sub-
conscious level as well (see rationalisation and cognitive dis-
sonance in Chapter 1), with the outcome of creating distinctly
convenient mistaken beliefs (‘The cat was already dead when I
ran over it’;The person I saw with that good looking stranger in a
dimly lit restaurant looked too young to be my wife’.)
In David Hume’s famous essay ‘On Miracles’8 motivations like this
are among the reasons he gives for why a sole witness to a supposed
miracle should never be believed. In short, it will always be more
likely that the person is mistaken or lying than that a law of nature
has been contravened.
Miracles aside, and despite these problems, witness testimony
remains important because it is sometimes all we have to go on. In
order for us to have greater confidence in witness accounts, the fol-
lowing critical questions can be posed (based on Walton, Reed and
Macagno, 2008, p.91):

Q1: Is what the witness says internally consistent?

Q2: Is what they say consistent with what is otherwise known
about the event, including what others have said?
Q3: Is the witness liable to be biased? (In Twelve Angry Men, for
example, it is suggested that one of the witnesses was moti-
vated by the chance to testify in court and feel important,
149

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 149

rather than his genuine knowledge of events surrounding the

murder.)
Q4: Is what they have said plausible? (In the case of miracles and
UFOs it is presumably not.) ]
Q5: Is the strength of the conclusion reached appropriate?

CIRCUMSTANTIAL AD HOMINEM ARGUMENTS

Arguments imputing bias in those who otherwise have expert or

information power are known as circumstantial ad hominem
arguments. Their basic structure is:

P1: X believes/advocates Y.
P2: X is biased in a way that could prejudice their beliefs/advocacy
of Y.
P3: We should disregard/be cautious about the views of someone
who could be biased in this way.
C: Therefore, we should disregard/be cautious about X’s views
on Y.

The UK’s now concluded (in 2022) Independent Inquiry into Child
Sex Abuse provides a noteworthy example. Provoked by the Jimmy
Savile scandal, it was a wide-ranging investigation that included gov-
ernmental level suspects stretching back at least to the 1970s. When
it was set up in 2014 it was proving so difficult to find someone not
linked to possible suspects for the role of chair that it was neces-
sary to bring in a judge from New Zealand (Lowell Goddard). In
this instance a series of strong circumstantial ad hominem arguments
righty delayed the process by over a year.9
Rather less convincing was the foundational accusation of the
documentary The Great Global Warming Swindle (Martin Durkin,
2007) that the scientists who have found evidence for anthropo-
genic global warming have, in fact, known for a long time that
their views have been disproved. They maintain their defence of
this view, though, in order to keep their lucrative positions on the
Intergovernmental Panel on Climate Change. (Their view has, of
course, not been disproved, and their position on the IPCC is vol-
untary and unpaid.)
150

150 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

The critical questions that can be applied to all types of ad homi-

nem argument are these:10

Q1. It what is claimed about the person’s ability, character, or cir-

cumstances true?
Q2. Is this attack relevant to the claim that the argument makes?
Q3. Where it is relevant, is the right kind of conclusion drawn from
it? (For example, quite often someone’s position is rejected
outright on the basis of an ad hominem attack whereas at most
it should be called into doubt.)

Towards the end of Martin Luther King, Jr.’s Letter from Birmingham
Jail11 is a circumstantial ad hominem attack that is harder to judge
than the previous two examples. Here he pricks the conscience of
America’s white moderates by offering a theory for their lack of sup-
port for his methods. After explaining his reasons for the necessity of
non-violent direct action he says:
I had hoped that the white moderate would see this. Maybe I was too
optimistic. Maybe I expected too much. I guess I should have real-
ized that few members of a race that has oppressed another race can
understand or appreciate the deep groans and passionate yearnings of
those that have been oppressed, and still fewer have the vision to see
that injustice must be rooted out by strong, persistent, and determined
action.
The ad hominem argument in this passage, reconstructed in a sum-
marised form, seems to be this:

P1: Most white moderates do not support our methods.

P2: Members of the oppressing race will typically not appreciate
the psychological condition of the oppressed.
P3: White moderates are members of an oppressing race.
C1: White moderates will typically not appreciate the psychologi-
cal condition of the oppressed.
C2: The white moderates are wrong not to support our methods.

Here the reasoning is that white moderates, unlike the oppressed

blacks, are not in a position to know.Their experience makes them
unable or unwilling to put themselves in the shoes of the people that
151

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 151

King, Jr. is leading, and as a result the right kind of belief or response
is not to be expected.
If we apply the critical questions, Q2 can be answered in the
affirmative since not being in a position to know (in the form of life
experience), if true, is relevant to King, Jr.’s conclusion. Whether it
is true though (Q1) is another matter; quite possibly it is, but this is
not something that can be answered easily from a layperson’s point
of view. Since much of the rest of the speech is dealing with the
white moderates’ arguments, then this ad hominem attack cannot be
rejected as simply crude or dismissive, and nor can the conclusion
he draws from it. If the entire message had been based around an ad
hominem argument of this type, however, then it would be fair to
regard his, quite firm, conclusion as inappropriate (Q3).

INFORMATION POWER: PSYCHOLOGY AND RHETORIC

We tend to be impressed by first-hand accounts and to overlook the

errors that actors and observers are prone to. Acting on them can be
the result of the availability heuristic (see Chapter 1), and one rea-
son for this – especially in the case of witness testimony – is that they
have a narrative quality to them that is absent from statistics and other
more abstract data. Reasons why stories can be so persuasive include
their being concrete, personal, and moving, and thus more memora-
ble.We find the details of particular human experiences engaging in a
way that abstract reasoning tends not to be, which is why documen-
taries and books with serious messages trying to reach wide audiences
are either told via stories, or mix personal experience with more
technical arguments. (The environmental films An Inconvenient Truth/
Sequel (2006/2017), The Age of Stupid (2009), and David Attenborough:
A Life on Our Planet (2020) are quite typical in this respect.)

INFORMATION POWER, DISPOSITIONS AND

CONSTRUCTIVE DIALOGUES

Critical thinking will teach us to be on our guard against the ten-

dency to rely on our own experiences and immediate responses to
situations. Not only are these prone to error, but they are subject
to the debilitating effects of overconfidence. However, like most
152

152 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

dispositions there is a balance to be sought, and due regard for power

and authority, in all the forms we have encountered them here, does
not rule out the possibility of rejecting them in favour of what we
‘know’ or even ‘feel’ to be right.
At the start of Twelve Angry Men Juror 8 struggles to articulate why
he feels that the guilty verdict is wrong, but he rightly allows this intu-
ition to motivate further thinking about the case, even in the face of
severe and (especially if we include the lawyers and expert witnesses)
authoritative opposition. In other situations, we might experience or
witness, or otherwise believe something that is so extraordinary that
we shouldn’t expect others who haven’t experienced the same thing
to believe us. The person thinking critically would know this limi-
tation, but not necessarily reject their belief. Miracles and religious
experiences serve as an example, and even Hume was not saying that
the person themselves should not believe what they have seen.
Another type of situation where personal conviction can be justi-
fied is one in which there is something inherently personal about
a particular belief. It could be an aesthetic conviction, a love for
someone, or a form of life that most others do not share or under-
stand, but which is not incompatible with critical thinking on the
grounds that the person has good reasons for why this is an area
which is not (or should not be) amenable to reasoned justification.
‘The universalizable does not’ says virtue ethicist Martha Nussbaum,
‘determine every dimension of choice;... there are silences of the
heart within which its demands cannot, and should not, be heard’
(1990, pp.39–40).
Beware though, because this is an area that has ‘HANDLE WITH
CARE’ stamped in large letters. The line between admirable integ-
rity and dogmatic refusal to subject one’s beliefs to scrutiny is thin,
and the latter too often hides behind the justification of the former.
As with other examples of this apparent rejection of critical ques-
tions that we will come across in this book, there are subtle forms
of justification which help establish the credibility of lonely deci-
sions like these. Ethotic authority is one, and another is simply the
person’s willingness to acknowledge how their situation can seem
peculiar to others; that by normal standards (standards that they oth-
erwise respect), what they are choosing would be regarded as wrong
or foolish.
153

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 153

6.3 LEGITIMATE, REWARD, AND COERCIVE

POWER
THE MILGRAM EXPERIMENTS

Stanley Milgram’s experiments in the 1960s measuring and analysing

obedience to authority remain influential and important. They are
quite well known these days, not only because the results are highly
surprising and sobering, but because the experiments themselves
are dramatic and involve electric shocks. Here I want to pay atten-
tion to the processes of power and authority – discussed at length in
Milgram’s book Obedience to Authority ([1974] 2005) – that explain
the results. They serve both as an excellent example of arguments
from various forms of authority and power, including legitimate,
reward, and coercive, and as an insight into the psychology of persua-
sion that is associated with appeals to authority.
Motivated by the high obedience levels among soldiers and Nazi
officials in the WWII death camps, and by the trial of Adolf Eichmann
in 1961,12 Milgram wanted to examine obedience behaviour in the
laboratory. Members of the public were asked to volunteer for a mem-
ory experiment at Yale University. If selected they would turn up at the
lab, the procedure would be explained to them, and they would meet
their partner, another (apparent) volunteer called Mr Wallace.
As one of these participants you would, seemingly randomly, be
assigned the role of ‘teacher’, with Mr Wallace as the ‘learner’. You
would read out a list of pairs of words to Mr Wallace, and then test
his learning by reading out the first word from each pair, followed
by four options from which he would try to choose the correct one.
The twist in this procedure is that every time Mr Wallace answers
wrongly, you are required to deliver an electric shock (the supposed
purpose of the experiment being to test the effects of pain in moti-
vating memory). A screen separates you and poor Mr Wallace, but
prior to the experiment starting you see him being hooked up to the
shock machine and strapped into a chair. In front of you is a row of
switches delivering shocks in 15V increments from 15V to 450V.You
are instructed to increase the level of shock every time the subject
gets a question wrong (15V, 30V, 45V, 60V, etc.).
Pretty quickly Mr Wallace – who is, by the way, a stooge and not
actually receiving shocks – starts to make mistakes and the shock level
154

154 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

duly increases. As the shocks become more severe, at various stages

he starts, for example, groaning (135V), demanding to be allowed to
leave (150–165V) (the cry of ‘Get me out of here!’ becomes continu-
ous amidst other responses), screaming in pain (270V), refusing to
answer questions (300V), violently screaming (315V), and then (if
you are still participating by this point) after around 330V Mr Wallace
falls silent.
As this develops, if you are a typical teacher you would plead with
the experimenter and question the wisdom of continuing. In response
to this predicted reluctance the experimenter has a series of carefully
scripted prods to encourage you to continue: ‘Please continue’; ‘The
experiment requires that you continue’; ‘It is absolutely essential that
you continue’, and finally ‘You have no other choice – you must go on.’
The question is: at what point do you refuse to continue adminis-
tering shocks? The surprising result of this basic experimental set up
was that 100% of subjects went as far as 300V, and 65% went to 450V.
Variations of the experiment carried out by Milgram and by other
psychologists in a wide variety of countries and cultures across sev-
eral decades show similar outcomes. The implication is that many of
us, under these conditions, would be unwilling or unable to disobey
commands to administer potentially lethal electric shocks to a stran-
ger under circumstance in which the punishment for disobedience
is no more than the experimenter’s disapproval. Milgram concluded:
if a system of death camps were set up in the United States of the sort we
had seen in Nazi Germany, one would be able to find sufficient personnel
for those camps in any medium-sized American town. (Cited in Blass,
2000, p.35)

Explaining Milgram’s findings

So, what’s behind this? Milgram offers a complex, multilayered
explanation that includes an evolved disposition towards obedi-
ence (because of the survival advantage of individuals who are able
to operate in hierarchies); social learning (hierarchical structures of
rewards and punishments in families, schools, and the workplace);
and a series of social norms that bind subjects to the experimental
situation once they’ve entered it.These ‘binding factors’ include their
voluntarily agreement to participate (commitment and consist-
ency); receiving payment for participating (reciprocation); and the
155

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 155

forms of expert and legitimate authority represented by science

and its methods:Yale University, the laboratory environment, and the
experimenter’s lab coat and calm, confident demeanour.
The power of these social norms and their antecedents may seem
trivial compared to the pain inflicted, but these experiments show
that they are in fact not trivial. They all contribute to a ‘barrier’ of
anxiety (Milgram, 2005, p.154) that the majority of subjects were
unable to surmount, leading Milgram to propose:
the conflict between conscience and authority is not a wholly philosophi-
cal or moral issue. Many of the subjects felt, at the philosophical level of
values, that they ought not to go on, but they were unable to translate this
conviction into action. (Milgram, cited in Blass, 2000, p. 35)

The view of Milgram and other researchers is that it is the combina-

tion of legitimate and expert power that makes disobedience so diffi-
cult. Legitimate power can be fairly easily understood and identified
in terms of positions held and principles stood for, but its reality is
quite fuzzy. It often encompasses information or expert power (the
person has their position because of what they know), and it invari-
ably comes with forms of reward and coercive power.
Numerous factors ensure the experimenter’s legitimacy: his man-
agement of the procedure; his control of the space (the lab); his rela-
tionship to the place (the university) in which the experiment takes
place; the epistemic and moral authority of science (which is signi-
fied by all the above); plus his demeanour and dress. The wording of
his ‘prods’ makes no direct reference to any punishments for not con-
tinuing, and in fact do not make a great deal of sense as arguments. If
you still have your wits about you that is.‘You have no other choice –
you must go on’ is easily countered, as demonstrated by one subject:
I do have a choice. [Incredulous and indignant:] Why don’t I have a
choice? I came here on my own free will. I thought I was on a research
project. But if I have to hurt somebody to do that, or if I was in his place,
too I wouldn’t stay there. I can’t continue. (Milgram, 2005, p.52)

The implication is that this is not what the subject signed up for;
that the experimenter no longer has legitimate power over him. If
the subject refuses to continue, all the experimenter can say is that
the experiment must come to an end, highlighting the limits of his
power. Nevertheless, prior to this the barrier of anxiety that so many
156

156 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

subjects were unable to overcome was the result of fear of the experi-
menter’s disapproval, which itself – Milgram plausibly speculates – is
in large part caused by a generalising of an internalised association,
beginning in childhood, between doing what’s right and the power
of legitimate authorities to reward and punish (ibid, p. 138).

LEGITIMATE AUTHORITY: STRUCTURE AND CRITICAL QUESTIONS

The general structure of an argument from legitimate authority

is this:

P1: X has legitimate authority in the domain of Y.

P2: X believes that action Z in relation to Y must be performed.
C: Therefore Z must be performed.

The main critical questions applying to arguments from legitimate

authority are:

Q1: Is the power in question legitimate?

Q2: Is this person currently in a position to exercise this power?
(This refers to circumstances in which someone tries to impose
their otherwise legitimate authority in the wrong domain, such
as Robert De Niro’s ex-CIA operant and possessive father’s
inappropriate spying on his daughter’s fiancé in the film Meet
the Parents.)
Q3: Are there other considerations that might override the per-
son’s legitimacy (for example the legitimacy of a competing
authority)?

The Milgram procedure is primarily vulnerable to Q3 – the experi-

menter’s otherwise legitimate and relevant authority is challenged by
the authority of established moral standards concerning compassion
and people’s rights.

COERCIVE POWER AND AD BACULUM ARGUMENTS

In argumentation coercive power is usually analysed in terms of what

are called ad baculum arguments. Ad baculum means in Latin ‘to the
157

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 157

stick’, and refers to arguments that threaten someone with harm of

some kind (inflicted by the arguer) if they do not do a certain thing.
They are sometimes referred to as ‘appeals to force’, and take this
basic form:

P1: If you do not bring about X then I will make consequence Y

will occur.
P2: Consequence Y is not in your interests.
C: You should bring about X.

An immediate distinction to make is that between threats and warn-

ings. In everyday linguistic usage these terms are fairly interchangeable,
but one important difference is relevant to our discussion. The man-
ner of our relationship to someone who potentially holds the power
of punishment over us is quite different if that power is legitimate.
A warning might more typically be seen as the expression of legiti-
mate power (coming from the law courts, one’s teacher, spouse, etc.),
whereas a threat implies someone taking the law into their own hands.
That said, criminal gangs like the Mafia will ‘warn’ people, but arguably
these organisations use this language precisely to create or sustain an
air of legitimacy. ‘Threat’ sounds thuggish, ‘warning’ sounds principled.
Legitimacy, then, determines one of the critical questions, and the
others address more practical considerations. The complete list is:

Q1: Is the person making the threat a legitimate authority acting

within the context of their power?
Q2: Where not legitimate, is the person making the threat able to
use the force they have threatened?
Q3: Is the person being threatened able to comply?
Q4: Is the person being threatened willing to comply (is it worth it
for them)?

Consider the difference between these ad baculum arguments:

ARGUMENT 1

P1: If you do not attempt to adhere to professionally accepted

standards, then we will make you redundant.
158

158 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

P2: You want to avoid being made redundant.

C: You should try harder to adhere to professionally accepted
standards.

ARGUMENT 2

P1: If you complain again about unpaid overtime, then I will see to
it that you are fired.
P2: You want to avoid being fired.
C: You should stop complaining about unpaid overtime.

It is easy to imagine a situation in which Argument 1 meets the crite-

ria implied by all four critical questions: the targets referred to have
been agreed upon and are generally recognised as achievable, the
person being threatened has a strong desire to keep their job, and
the person presenting the argument is a legitimate authority (rather
than, say, the person’s subordinate). It is also easy to imagine a situa-
tion in which any combination of them is not met, despite the broad
context being one of a target-led environment. For example: it has
not been agreed that performance in terms of targets is a grounds for
dismissal, it is only these grounds that could be the basis on dismissal
in this case, the targets are not reasonably achievable by anyone, and
the person has other jobs they can happily go to.
It is harder to imagine a circumstance in which Argument 2 satisfac-
torily answers Q1 (but it is certainly not impossible).This is a situation
in which the legitimacy of the threat is in question, and quite possibly it
is a case of bullying. Significantly though, we can still see how it could
be a strong argument even in the absence of legitimacy. If the arguer
is in fact able to have the person fired (perhaps through manufactur-
ing legitimate grounds), the person is able to stop complaining (which
is entirely possible), and they want the job badly enough (maybe it is
with a prestigious organisation), then this could be enough to accept
the conclusion. At this point though, this form of argument analysis
seems to let us down. There clearly is something wrong here, and just
what that is will be discussed shortly under ‘Constructive dialogues’.
Beforehand, I will briefly say something about what can be meant
by ‘harm’. Ad baculums primarily refer to physical threats, but they can
apply to psychological threats as well: ‘If you don’t accept my point of
view, I will disrespect or dislike you.’ Sometimes this implied threat is
159

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 159

unavoidable (in trying to convert a racist, for example), and is perhaps

implied in quite a few discussion topics. Open-mindedness shouldn’t
stretch to what is clearly false, implausible, or morally unacceptable, and
so argumentation has the power to begin or end relationships. Also,
however, a consequence of argumentation is that we find out who we
do and do not agree with, and sharing views, even where disagreements
remain, can be a determinant of liking and respect (see Chapter 3).

AD BACULUM ARGUMENTS: PSYCHOLOGY AND RHETORIC

Ad baculum arguments are a sub-category of what are known as ‘appeals

to fear’. An appeal to fear is any argument that uses the possibility of
a fearful consequence as a premise for accepting a certain conclusion.
As well as appeals to force these encompass events that are not subject
to official decree – disease, death, climate change – where it makes no
sense to question legitimacy; ‘If you continue to do this, this is liable to
happen’, rather than ‘If you continue to do this, I will make this happen.’
In both cases it is fear of a certain consequence that leads us to
accept the conclusion, and as we saw in Chapter 2, emotions tend
to distort our thinking. Once the fearful possibility is raised, critical
questions can be overlooked or our capacity to investigate them accu-
rately impaired. An extreme case is terrorism, the potency of which
relies on ‘an irrational tendency … to magnify unexpected and “mys-
terious” evils out of their true proportion’ (Narveson, 1993, p.154).
Governments and other political groups have used fear (of terrorists,
immigrants, etc.) and uncertainty to push people towards more con-
servative values or a greater acceptance of authoritarian leadership,
and a dramatic version of this exploitative strategy has been labelled
the ‘shock doctrine’.13 The critical questions to apply to the fear ele-
ment of ad baculum arguments – i.e., the general questions applicable
to all appeals to emotion – can be found in Chapter 2.

AD BACULUM ARGUMENTS, DISPOSITIONS,

AND CONSTRUCTIVE DIALOGUES

Ad baculum arguments can be strong arguments in so far as the prem-

ises can be true, and if we accept these premises then we must also
accept the conclusion. The main problem with them is the effect
they have on a dialogue, which is typically to end it.
160

160 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

In the case of threats or warnings issued legitimately, there can be

further deliberation about the nature of this legitimacy, but if it is
simply someone forcing you to do or believe something or else suf-
fer the consequences, then they are effectively ending the discussion.
This is what happens when peace talks collapse, and typically this is
what war is – an unwillingness or inability to live with a disagree-
ment that has proved unsolvable through dialogue.
Implicit ad baculums can occur via the disapproval we often signal
through non-verbal or paralinguistic cues such as facial expressions
and tone of voice. Even if what is being communicated verbally is
not aggressive, the asymmetric raised lip of contempt or inappropriate
sigh can let other discussants know that they risk losing your respect if
they do not agree with your position, and it runs the risk of an angry
retaliation. (This is very common in dialogues, and there are endless
examples of it in Twelve Angry Men.) This might be a calculated act of
intimidation, but it might also be unconscious and unintended; you
might feel this way, but do not want to licence this feeling, or commu-
nicate it, precisely because it will prematurely close down the dialogue.
In cases like this the dispositions of metacognition and self-control
(part of what Aikin and Clanton (2010) refer to as ‘deliberative tem-
perance’) become valuable for maintaining a constructive dialogue.
A final point to consider is that there is one type of dialogue in
which threats are accepted as legitimate – the negotiation.14 The
structure of a negotiation is for each side to make demands which, if
not met, will result in various sanctions (such as strike action, lay-offs,
no longer allowing your van to be used for transporting the band’s
gear). Opposing parties are effectively trading ad baculum arguments,
but in a way that is legitimised by a mutual, if often unstated, under-
standing of the rules of the game: a recognition that the purpose is
to reach a compromised agreement, and an expectation that reasons
will be given to explain each side’s demands.

MILGRAM AND DISPOSITIONS

In his discussion of individual responses to the experimental conditions

Milgram distinguishes between a ‘Professor of Old Testament’ ([1974]
2005, pp.49–50) who disobeyed orders on the basis of a different
authority, God; and a participant, referred to as Gretchen Brandt, who
161

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 161

he sees as having virtues which enable her refusal (ibid, pp.87–8).There

is an independence exhibited in Brandt’s behaviour that makes her a
good example of a critical thinker, and Milgram is clearly impressed by
what he sees as her ‘total control of her own action’ which ‘seems to
make disobedience a simple and rational deed’ (ibid, p.88).
Also, in most cases of disobedience a significant amount of cour-
age is required. Not only is it a matter of disobeying orders from an
authority whose position is still upheld by the situation, but of trans-
gressing broad social hierarchies. Among other things, to disobey is to
disrespect and to discredit, and it is therefore to embarrass the other
person (which will also embarrass us) (ibid, p.152). And not only do
we need courage to stand up to and delegitimise an apparent authority
but the anxiety that challenging and unfamiliar situations provoke can
severely disable our ability to think critically. We are more likely to let
ourselves be led by those – in this case the experimenter – who display
confidence; and we are more likely to spontaneously generate ration-
alisations for our continued obedience: e.g., seeing the experimenter
as fully responsible for the harm, or viewing the learner as deserving
of their discomfort because they are stupid or stubborn (ibid, pp.47–8).

MILGRAM, LEGITIMACY, AND FRAMING

Framing was discussed in some detail in Chapter 1. Milgram uses it

to explain how someone perceived as having legitimate power is in a
position to define the meaning of a situation, and in that way attain
greater control over people’s subsequent beliefs and behaviour (ibid,
pp.146–7). Drawing a comparison with propaganda, the trick with
obedience is to put people in a position in which they willingly go
along with commands because these commands are consistent with
an already accepted definition of the meaning or purpose of a situ-
ation. Subsequent instruction is primed to be interpreted through
the lens of scientific legitimacy and expertise. For the subject, what’s
going on could be seen as wrong, but the behaviour of the experi-
menter, who continues to represent this frame of reference, suggests
otherwise. In most instances this contributes to obedience, and only
in a minority of cases is the competing frame of independent ethi-
cal thinking able to purchase enough of a foothold to help leverage
disobedience.
162

162 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

As we have noted on a few occasions, of enormous importance

for effective critical thinking is the willingness and ability to think
beyond a situation or problem as it is immediately understood, or as
it is presented to us by others. As critical thinkers we must be pre-
pared to ask ourselves: ‘Is there another way of looking at this?’ And
usually there is.

REWARD POWER

Like punishments, rewards are a fundamental aspect of life, and

come in many forms (from pay to praise). Anyone able to influence
the behaviour of others through rewards has ‘reward power’. Often
rewards are traded as we negotiate with others (longer holidays for
a pay cut, school children working hard for praise, and so on). And
we can control our own behaviour through self-rewards such as giv-
ing ourselves ‘treats’ – a session on Minecraft, an episode (or two) of
Ozark, a biscuit even – for accomplishing unpleasant or difficult tasks.
The structure of this kind of argument and the critical questions are
similar to those of ad baculum arguments:

P1: If you bring about X, then you will receive Y.

P2: Y is something that you want.
C: You should bring about X.

Critical questions:

Q1: Can the person genuinely provide the reward in question, and
are they likely to?
Q2: Are you able to do what is required to receive the reward
offered?
Q3: Are you (or should you be) willing to pay the necessary price?

The relevance of these questions for Milgram’s subjects is clear

enough. If Milgram is right to suggest that the approval we seek
from the experimenter is conditioned by a lifetime of associating
approval with obedience, then Q1 is perhaps the most crucial. System
1 will happily, but erroneously, generalise early family and school
experiences to our relationship with the source of authority in this
situation.
163

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1) 163

6.4 SUMMARY
This chapter has examined the strength of arguments based on expert,
information, legitimate, reward, and coercive power. In each case they
can form the basis of strong arguments, but, as the Milgram experiments
demonstrate, the deep and complex motivations behind obedience and
compliance is such that an enormous amount of caution – assisted by
self-awareness, courage, and a willingness to take responsibility for our
beliefs and actions – is needed when confronted by authority-based
appeals. The next chapter continues the themes of social power and
message source, exploring ethotic authority, further types of ad homi-
nem argument, and arguments appealing to referent power.

EXERCISES
1. Since appeals to the forms of power and authority discussed in
this chapter are so common, a worthwhile exercise is to find sev-
eral examples of such appeals from news stories, opinion pieces,
political commentaries, exchanges on social media, etc. Apply
the appropriate critical questions to them, assess their poten-
tial persuasiveness, their impact on the dialogue, and what they
might reveal about the dispositions of the people who use them.
2. The focus of this chapter has been the application of criteria
for avoiding and guarding against weak appeals to authority.
A separate phenomenon is the chronic distrust of experts in
areas where there is scientific consensus, with climate science
and vaccinations as outstanding recent examples.There is now
a large body of literature that seeks to explain the psychologi-
cal and political causes of this: Levy, 2019, and Baghramian
and Croce, 2021 are excellent starting points, as is the web-
site of PERITIA, the EU project on public trust in exper-
tise: https://peritia-trust.eu. To initiate your own thinking on
the issue a helpful exercise is to hypothesise reasons based on
some of the ideas discussed in this chapter and elsewhere in
this book (especially Chapters 1–3, 7, and 8.)

FURTHER READING
As with other chapters, use the in-text references as a guide to fur-
ther reading on specific aspects of what has been discussed here. Also,
164

164 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 1)

for a concise and illuminating discussion of ad baculum arguments see

Woods (1995) ‘Appeal to force’.
In the second exercise above, you will find recommendations for
further reading on the question of why certain people or groups
reject scientific consensus.

NOTES
1 A modern advocate of this, drawing on Aristotle’s and Seneca’s ideas on vir-
tues and rhetoric, is Alan Brinton. (See, for example, Brinton, 1986)
2 For example, Johnson and Blair (2006, pp.168–72) have four, Douglas Walton
(2006, p.88) has six, and Christopher Tindale (2007, pp. 134–43) seven.
3 The Life Scientific, BBC Radio 4, 28.7.15.
4 Group selection is the idea that evolution works, at least in part, at the level
of the adaptive fitness of groups rather than genes. Thus, a group in which
individuals are more cooperative and willing to self-sacrifice for the sake of
the collective will fare better than one in which individuals are more selfish.
The selfish gene theory, on the other hand, states that cooperative tendencies
(reciprocal and kin altruism) are only selected to the extent that they benefit
the individual and those who share their genes.
5 Pennisi, E. (2011) Researchers Challenge E.O. Wilson over Evolutionary
Theory [Article]. Available at: www.science.org/content/article/researchers-
challenge-e-o-wilson-over-evolutionary-theory
6 My analysis of these issues has been significantly influenced by Heather
Battaly’s article ‘Attacking Character’ (2010).
7 You know who you are!
8 This essay can be found in numerous places, including Stump, E. and Murray,
M. (eds) Philosophy of Religion:The Big Questions (Blackwell, 1999).
9 See Mason, R. (2014) Fiona Woolf resigns as chair of government’s child
abuse inquiry [Online article]. Available at: www.theguardian.com/­
politics/2014/oct/31/fiona-woolf-resigns-chairman-child-abuse-inquiry.
(For other r­ easons Lowell resigned (or was sacked) in 2016, and was replaced
by Prof Alexis Jay.)
10 Adapted from Douglas Walton (2006, p.123) and Christopher Tindale
(2007, p.89).
11 Available at: www.africa.upenn.edu/Articles_Gen/Letter_Birmingham.html
12 Described and analysed in Hannah Arendt’s book Eichmann in Jerusalem:
A Report into the Banality of Evil (Penguin, 1965).
13 Klein, N. The Shock Doctrine (Penguin, 2008).
14 For an insightful discussion of this, and ad baculum arguments in general, see
Woods (1995).
165

ARGUMENTS, SOCIAL POWER,

AND MESSAGE SOURCE (PART 2)

God told me to invade Iraq. (President G.W. Bush, as reported in The

Independent, 7 October 2005)
What’s wrong with ad hominem arguments? Only a fool would ask a
question like that. (Sacha T. Burnstorm, pers. comm.)

The theme of this chapter and the previous one is characteristics of

the arguer, or those they refer to in support of their arguments.These
have been analysed via categories of social power, and the remaining
types – arguments making a broad appeal to character (ethotic argu-
ments) and referent power (the last from French and Raven’s original
list) – are our current focus. Complimenting this is a further explora-
tion of ad hominem arguments, and the introduction of appeals
to common belief or ad populum arguments.

7.1 ARGUMENTS APPEALING TO CHARACTER

The structure of ethotic arguments is set out by Walton, Reed, and
Macagno (2008, p.336) in this way:

P1: If X is a person of good (bad) moral character, then what X says

should be accepted as more plausible (rejected as less plausible).
DOI: 10.4324/9781003247944-8
166

166 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

P2: X is a person of good (bad) moral character.

C: Therefore what X says should be accepted as more plausible
(rejected as less plausible).

In the broadest sense then, ethotic arguments employ the moral char-
acter of the person upholding a position to help establish the plausi-
bility of that position. As you can see from the scheme above, this can
refer to good or bad aspects of that character, depending on whether
the arguer’s aim is to support or reject a particular conclusion. In this
section I am going to use ‘ethotic authority’ to refer to arguments
appealing in a positive way to moral character, and ‘abusive ad homi-
nem’ for arguments that appeal to it in a negative way.

ETHOTIC AUTHORITY

Direct (or abusive) ad hominem arguments are discussed in most

textbooks that deal with argument forms and fallacies, but ethotic
authority is often overlooked. On the face of it this is surprising, but
on closer inspection the main problem with offering an argument
from moral character is that it will often collapse into one of the
other categories of authority – typically expert or legitimate.
For example, if we are swayed by the empathetic nature of a friend
who is campaigning for an increase in the percentage of GDP spent
on overseas aid, any argument to this effect would likely identify a
form of moral expertise as the basis of believing its plausibility. Or if
we are impressed by the views of a modest colleague on the con-
tribution of a co-worker to a project this would be down to the
modest person’s accurate (non-self-serving) insight into collaborative
situations. Certain positions, such as judges, parliamentary speakers,
or being a parent (of more than one child), demand impartiality, so
the person who acts in accordance with this norm is assumed to be
instantiating a moral legitimacy inherent to the role, rather than it
necessarily being a reflection of their good character.
We are, however, sometimes influenced by a more holistic impres-
sion of someone we regard as a ‘good person’. Juror 8 in Twelve Angry
Men, Ellen Ripley in the Alien films, and Mikael Blomkvist in Stieg
Larsson’s novel The Girl with the Dragon Tattoo might be examples, as are
many other fictional heroes and heroines. In real life Oprah Winfrey,
167

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 167

Sir David Attenborough, Nelson Mandela, and Taylor Swift serve to

illustrate the point, and most of us will have examples from our own
circles. In cases where these people are cited in support of a claim, or as
making that claim themselves, their character increases the chances of
us agreeing with them. But the circumstances in which this can form
the basis of a strong argument and where the person’s being good is
not reducible to a form of moral expertise as previously discussed, will
be quite select. We can conclude though that positive ethotic argu-
ments are most clearly defined by situations in which, with no further
expertise implied, an example of a good person – one that is likely to
be agreed upon in this context by the people the argument is aimed
at – is used as a heuristic for the right thing to do or believe.
Most typically it will be cases in which we have very little to go on,
and where our acceptance of plausibility will be highly provisional.
A good example is near the start of Twelve Angry Men when Juror
8 is attempting to persuade the others that they should spend time
discussing the case even though they are all convinced of the defend-
ant’s guilt. Juror 9 is quite explicit when he argues that he’s willing to
accept the request because he is impressed by certain characteristics
of Juror 8 – notably his willingness to stand up to the majority –
which are not relevant to the case itself.
In his unusually detailed and engaging social psychology textbook
(and one of the few academic sources to discuss Twelve Angry Men)
Roger Brown (1986, p.234) endorses the practical necessity of this
type of reasoning under circumstances of unavoidable ignorance. ‘In
a good expressive discussion’, he says, ‘… information is transmitted
that is not part of any relevant argument. Each participant in some
degree expresses his … assertiveness, intelligence, education, fair-
mindedness, compassion, prejudice, cowardice, and so on.’
The critical questions relating to ethotic arguments are these:

Q1. Is the person someone who can reasonably be regarded as hav-

ing good character?
Q2. Is the good character of the person of the right kind to be
relevant to the situation in question (counteracting the halo
effect (see below))?
Q3. Is the weight placed on their character proportionate for the
situation in question?
168

168 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

Q1 is dependent on the prior question of what constitutes good

moral character. Invariably it will involve a collection of virtues that
are, to a degree, culturally specific, but the preceding discussion indi-
cates that in many cases it will encompass someone who exhibits
some of the core critical thinking dispositions.
Q2 prompts us to reflect on possible situational limitations on
good character; Juror 8 is the right beacon in a jury room, but not
necessarily on a sports field.
Q3 serves as a reminder that placing significant emphasis on eth-
otic arguments is often most appropriate in situations of uncer-
tainty. As is explained in the discussion of direct ad hominem
arguments (below), appealing to this kind of authority (in the
form of role models or charismatic leaders) can be motivated by a
desire to not take responsibility for our actions in situations where
we should be.

ETHOTIC AUTHORITY: PSYCHOLOGY AND RHETORIC

Someone possessing ethotic power will tend to function as a role

model. A role model is someone we seek to imitate, and this notion
is helpful for understanding a difference between adherence to rules
(a type of legitimate authority) and ethotic power. The focus of the
latter is the individual rather than the abstract rule; a good example
being the (now faded) fashion among young Christians for wrist-
bands with the slogan ‘What would Jesus do?’.
Ethotic authority has persuasive potency because role models pro-
vide a vivid and memorable means for beginning our deliberations
on how to respond to a range of situations. Referring to the Roman
philosopher Seneca, Alan Brinton says,
it is easier to recognise a virtuous person when we see one than it is to
give or understand and evaluate (or be moved by) abstract accounts of
the nature of virtue. (1986, p.253)
Like Jesus for Christians, the virtues these people embody are widely
relevant and we are capable of imaginatively applying them by putting
ourselves in the shoes of our ethotic heroine or hero: ‘What would
Juror 8 do?’; ‘What would Oprah do?’; ‘What would Blomkvist do?’;
‘What would Taylor do?’; ‘What would Grandma do?’
169

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 169

A common approach in marketing is to employ the ethotic quali-

ties of celebrities to enhance the appeal of products. The idea is that
the intelligence, sincerity, or good taste of people like Serena Williams
or George Clooney rub off on the product. The consumer makes an
implicit association between the celebrity’s assumed qualities and the
brand, and thus to the extent the celebrity acts as a role model for
the consumer, the brand becomes more appealing. (This process of
persuasion is the opposite of certain forms of guilt by association,
discussed below.)
A risk with ethotic authority, underlining the importance of the
second critical question, is what is known to social psychologists as the
halo effect.The name – credited to early twentieth-century psychol-
ogist Edward Thorndike – refers to our tendency to generalise from
a person’s known positive qualities to other positive qualities. Without
any supporting evidence, an intelligent person is more likely to be
judged a morally good person than a less intelligent person, someone
attractive is assumed to be above average in intelligence, and so on.
This seems to explain odd (and often disastrous) decisions to give
outstanding footballers and/or pundits (like Alan Shearer and Gary
Neville) managerial posts in top-ranking football teams with no prior
experience of expertise in this area. It is also related to the accurate
cliché that first impressions matter. System 1’s appetite for coher-
ence is such that the receipt of any information automatically frames
subsequent information. Which of these two people do you prefer?

Annie is intelligent, industrious, impulsive, critical, stubborn, and

envious
Shirley is envious, stubborn, critical, impulsive, industrious, and
intelligent

Chances are it is Annie, even though the same traits are listed for
both, and you are not told that they are listed in order of prominence.
This is taken from a classic set of experiments by Solomon Asch in
the 1940s;1 he proposed that the halo effect causes the first entry to
influence the interpretation of what follows so that ambiguous quali-
ties such as being critical or impulsive attain a positive or negative
spin depending on what comes before them. The net evaluation is
thus reliably skewed by the initial piece of information we receive.
170

170 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

It is easy to see how the halo effect makes us vulnerable to mak-

ing and accepting weak ethotic arguments. Its reverse – sometimes
known as the horn effect – is also recognised, and is relevant to the
psychology behind the persuasiveness of ad hominem arguments.

DIRECT AD HOMINEM ARGUMENTS

In the direct ad hominem argument it is aspects of the general

character of the person that are attacked in order to cast doubt on
their view. It is a ‘direct’ attack because, unlike the other types of
ad hominem argument, it is aimed squarely at personal character-
istics rather than what the person has done or said. However, it is
also, confusingly, indirect because the causal relationship between the
characteristic and what the person believes or states is more oblique.
The target of direct ad hominems can be failings in terms of criti-
cal thinking dispositions. In certain circumstances a person’s views
can rightfully be taken less seriously if they are dishonest, dogmatic,
arrogant, inflexible, and so on. The implication is that possession of
such flaws will compromise their knowledge or judgement, or their
willingness or ability to communicate the truth.
In other situations a wider range of dispositions can be relevant
to our decisions about what to do. For instance, the capacity for
compassion is unlikely to figure in how we assess someone’s views
on astronomy, but it probably will figure in our decision on who to
vote for as the leader of our country. Quite often though, direct ad
hominem attacks are abusive and have a whiff of desperation about
them. They are knee-jerk moves we resort to when under pressure
and when emotions are strong. (The critical questions applying to
the main forms of ad hominem argument, including direct ones, can
be found in Section 6.2 in the previous chapter.)

ETHOTIC ARGUMENTS, DISPOSITIONS, AND

CONSTRUCTIVE DIALOGUES

One form of direct ad hominem attack concerns the questioning

of someone’s motivations for appealing to authority in their argu-
ments. This is worth paying some attention to because it focusses on
171

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 171

the important critical thinking dispositions of a willingness to think

independently, and to take responsibility for one’s commitments. As
mentioned, appeals to authority can indicate due modesty and rec-
ognition of the quite enormous limitations on what we can know
first-hand, but they can also indicate a lack of persistence in securing
first-hand understanding, or they can be used to shield us from the
consequences of our actions. In dialogical terms, the individual opts
out and directs the questioner elsewhere.
In this chapter’s epigram we find the claim that George W. Bush
placed responsibility for the invasion of Iraq in God’s hands; the
Yorkshire Ripper did the same with respect to his murders; more
recently boxer Tyson Fury referred us to the Bible when asked to
defend his offensive remarks on homosexuality; and of course the
‘following orders’ defence was used by Eichmann and other war
criminals. In his experiments on obedience to authority Milgram
found that there as a ‘reduction in strain’ in subjects when it was
acknowledged that the experimenter would take responsibility for
the consequences of the procedure (2005, pp.161–2).
A more comprehensive and quite vivid example of abrogation of
responsibility is that of the cult member and other situations in which
a particular expert or leader has, for some, an aura of untouchability.
This is studied under ‘charismatic leadership’. Charisma is a form of
ethotic power that is notoriously hard to define but is associated with
‘special gifts’ and ‘hypnotic’ or ‘magical’ qualities to someone’s per-
sonality that inspire love and make people want to follow them (see
Weber, 1978, pp.242–3; Burns, 1978, Ch.9). In a documentary about
Margaret Thatcher one of her Private Secretaries (Caroline Slocock)
talks about how inside No.10 ‘felt more real than the outside world’,
‘everything was in colour, while everything outside was in black and
white’.2
In all these cases there is a comforting sense of security and cer-
tainty that comes with handing over responsibility to someone else.
In some situations (such as small child to parent) there is nothing
wrong with this, but in many others it can be a dangerous illusion.
With people in positions of power considered as charismatic they can
be all too willing to accept this responsibility, and history attests that
the two motivations combined can have tragic consequences.
172

172 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

Determining when an appeal to authority is wrongly motivated is

significantly context dependent, but the questions we need to ask are
clear enough. Firstly:
Is this a situation in which the person should take direct responsibility for
their belief (or action), or is it one in which it is reasonable for them to
refer us to a particular authority?
If the answer is that responsibility should be taken, then the space is
created for a further question along the lines of:
What does the person stand to gain from shifting responsibility for a deci-
sion to an authority?
If there is a plausible case for a questionable motive for the appeal
to authority, then we have the basis for a motivation-based direct ad
hominem argument.

Direct ad hominem arguments and prejudice

However, direct ad hominem arguments also come with an increas-
ingly familiar warning. Because they are aimed at the person rather
than what they are claiming they can carry all manner of prejudiced
assumptions concerning people’s credibility.3 Racial and other stere-
otypes are fallacious direct ad hominem arguments. When hearing or
making direct ad hominem arguments we should pause to consider
the critical question:
What is motivating this attack?
And this could of course be the start of a further direct ad hominem
argument, generated by us (and potentially aimed at ourselves). For
mitigating prejudice, dispositions associated with listening and respect
are fundamental. Also important is a certain form of metacognition
that is attuned to forms of bias that research in social cognition has
revealed to be both far more prevalent than we would imagine, and
often under our conscious radar (see Fiske (2005) for a review).

TU QUOQUE ARGUMENTS

Spouse: I really wish you’d be friendlier towards my family.

Other spouse: Oh, and I suppose you’re a model of tolerance towards
mine are you?
173

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 173

Sometimes referred to as ‘whataboutery’, this kind of interchange is

remarkably common. Instead of responding directly to a criticism,
we hit back with a version of the same criticism (tu quoque trans-
lates from Latin as ‘you too’). It is a peculiar piece of argumentation
because, on the one hand, it is typically highly reactive and defensive
(and thus damaging to the progress of the dialogue), but on the other
there is often something in it. That something is along the lines of
‘Don’t judge me by standards you don’t yourself uphold’, or ‘You’re
being a hypocrite.’You are saying that the person is not in a position
to judge, and therefore providing a reason for why you should not
have to respond to their accusation.
The strength of this type of argument is heavily dependent on
a range of contextual factors concerning the relationship between
the people involved, the type of behaviour under discussion, and the
motivation of the person who uses it. For example, the parent who
smokes twenty-a-day and who warns their teenage child against
smoking arguably is in a position to judge. They wish they had not
started as a teenager and now they can’t stop, and that is one of the
reasons they are passing on the benefits of their experience to their
child. If the teenager’s response is along the lines of ‘Who are you to
talk?’ the parent can punctuate a highly convincing answer with a
hacking cough.
More considered tu quoque attacks are personal in quite a seri-
ous way because inconsistency, or hypocrisy, is seen as a significant
character flaw. A strong tu quoque argument will not only damage the
other arguer’s position, it can have a broader impact on subsequent
arguments they present by making them seem less trustworthy.
Evaluations of tu quoque arguments will place a lot of emphasis on
the second critical question for ad hominen arguments:

Q2. Is this attack relevant to the claim that the argument makes?

Al Gore received criticism for flying the world to promote his eco
film An Inconvenient Truth (2006), but this is not relevant to the con-
clusion that he is hypocritical or otherwise lacks credibility. Al Gore’s
choosing not to fly might have sent some kind of message (as Greta
Thunberg’s carbon neutral sailing across the Atlantic to attend cli-
mate events in 2019 did), but not if his film is not making an impact
on publics and policymakers in the first place. Very likely it was his
174

174 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

presence in these places, plus the commitment to the cause that his
touring demonstrated, that made people pay attention.4
On the other hand, the Church of England criticising payday loan
companies, whilst at the same time investing in one of them (Wonga)
looks like a stronger tu quoque argument.5 The same goes for the list
of politicians and advisers who broke Covid lockdown restrictions in
the UK in 2020 and 2021 and were consequently (though notori-
ously not in all cases) forced to resign.
In the spousal dialogue at the top of this section things are more
complicated. It is possible that the husband’s response is entirely
defensive and that his counteraccusation is simply a means of deflect-
ing the discussion away from the fact that he does need to be friend-
lier towards his in-laws. It is also possible that, even if true of the
husband, the wife’s accusation is indeed mirrored by her own behav-
iour. Under these circumstances cool heads are needed (the very cool
heads that ad hominem arguments tend not to encourage). Other
things being equal, both parties should, ideally, reflect on their own
behaviour and reach a joint conclusion that (1) both need to improve
in this way and, importantly, (2) that one cannot reasonably expect
the other to do this unless they do it themselves.
Underpinning this is the need for reciprocation in so many social
encounters (see Chapter 1). Since we cannot expect perfection in
our moral and social behaviour, our standards tend to be relative to
the norms that surround us. In other words, if our peers demand
things of us, we expect them to demand the same of themselves.
It is, however, this same rule of reciprocation that can cause a group
(or couple-based) conspiracy of silence. In order not to provoke the
rows and fallings-out that criticisms tend to generate, and in order
to avoid the critical spotlight being turned in our direction, we can
find ourselves not criticising others when we should be. In terms of
personal development and group deliberations, tu quoque responses
could have a detrimental long-term impact because people become
unwilling to offer constructive feedback that would otherwise serve
us, or the group, well. It is easy to imagine the role that this kind of
response could play in groupthink; if a critical suggestion is met
with the messenger being shot, fewer messengers will be turning up,
a greater number of poor arguments will go unchallenged, and the
‘illusion of unanimity’ will be strengthened.
175

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 175

GUILT BY ASSOCIATION

Before moving on I will briefly mention a further variation of ad

hominem arguments known as guilt by association. In these cases
disreputable features of a person or organisation an arguer has some
connection with is transferred onto them. These arguments can be
strong if the association in question has an actual and relevant bear-
ing on the credibility of a person’s position, such as a male politician’s
membership of a club known to be misogynist. They can unfortu-
nately also be persuasive where there is no relevance (or even truth),
and are for this reason commonly deployed and dangerous.
I once saw a placard in a demonstration against ‘Obamacare’ which
read,
FIDEL CASTRO SAYS HE LIKES OBAMACARE. THE RIGHT DIRECTION
FOR AMERICA?
The argument is flimsy and vague, but the negative association would
have been powerful for many Americans.

7.2 APPEALS TO REFERENT POWER AND

AD POPULUM ARGUMENTS
The influence on our beliefs or actions originating from the commit-
ments and behaviour of people who we see as being ‘like us’ is called
referent power. Peer pressure is an example, but so is any attempt to
persuade us that involves reference to a group of people we identify
with. It relies on what is termed our social identity, the aspects of who
we take ourselves to be (including what gives us esteem) that derive
from the groups that we belong to (Tajfel and Turner, 1986).
In argumentation, referent power falls under the broader head-
ing of ad populum arguments, which are defined as arguments that
appeal to the fact that a position is generally held to be true as a reason
for accepting that it is true. In its simplest form it has this structure:

P1: Everyone (or a large majority of people) believes X.

P2: Whenever something is generally accepted as true there’s a
strong likelihood that it is true.
C: Therefore X is true (or X is plausible)
176

176 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

A variation that appeals specifically to referent power is:

P1: Everyone (or a large majority of people) belonging to my ref-

erent group Y believes X about subject Z.
P2: Whenever something is generally accepted by my referent
group Y about subject Z as true, there is a strong likelihood
that it is true.
C: Therefore X is true (or X is plausible).

On the face of it, ad populum arguments are the antithesis of critical

thinking; basing one’s beliefs on second-hand evidence from a group
of non-experts. However, they are not always poor arguments, as
an analysis of the relevant critical questions will demonstrate. These
critical questions are:

Q1: Is X, in fact, generally believed to be the case?

Q2: If so, is X a domain of knowledge where popular belief is rel-
evant to its truth?
Q3: If so, is the population whose beliefs are referred to the one
with the appropriate knowledge of X?
Q4: If so, is this group of people reliable? (For example, are there
signs that the views of the group in question have been cor-
rupted by processes like groupthink, or that they have reasons
not to be truthful?)
Q5: If so, is the conclusion reached on the basis of general belief
appropriate (e.g., suitably cautious)?

Q1 guards against a mistaken belief about what is common knowl-

edge. It is the kind of error that can result from someone’s confident
(and potentially intimidating) claim that ‘everybody knows that!’
being accepted too readily. Asking a selection of one’s friends and
peers, checking with opinion polls, or simply pausing for reflection
are ways to attain answers to this question.
Sometimes Q2 can generate a clear ‘yes’; typically concerning
aspects of life that most people who have been around for a while are
in a position to know. One example is local knowledge: that Old
Gregor at number 34 is a bit weird but essentially harmless; that the
dog at the building supplies outlet’s bark is worse than its bite; that
177

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 177

Morton’s fish and chips are the best in town. Other examples serve as
an extension of this idea; the practiced, and hopefully reflected-upon,
experience that justifies jury-based decision-making, consulting with
employees when deciding organisational codes of conduct, and the
democratic election of governments.
More often a carefully worded ad populum will give rise to a con-
clusion that is plausible, or at least indicates that a position is worthy
of a closer look. Some will be variations of position to know argu-
ments where we are told that something is true because everyone
who has had certain experiences (relating to their job or other life
experience) gets it. This can be the basis of an important argument
when those who are not in that position tend not to agree.
Or we might just answer ‘no’ to Q2, and religious belief is an
outstanding example. It could be true, but millions believing that it
is should not count as a reason for a neutral person to believe that it
is. Also, in science there are numerous cases where what the public
believes and what experts believe are clearly at odds. Climate change
denial among the American Christian right is a specific and serious
example, and the exposing of commonly held beliefs as myths is
common in popular science forums.6
Overall, we need to be aware that potentially momentous changes
can occur in scientific or moral knowledge which contradict but fail
to dislodge traditional beliefs.This inertia could be a simple matter of
ignorance and the time it takes for knowledge to seep into a culture,
but some beliefs remain common because they are comforting or
otherwise useful.
Q3 is covered by previous analyses of position to know argu-
ments. Q4 is directed at dogmatic, lazy, fearful, antagonistic, or com-
petitive group mindsets that create the dangerous biases and unreliable
communication that is so depressingly common on social media.The
self-fulfilling deliberations that impede critical thinking in Twelve
Angry Men serve as an example. We also need to be careful not to
conflate witness testimony type cases with straightforward examples
of ad populum position to know arguments, since immersion in situ-
ations can come with its own biases. We can have some sympathy for
‘You weren’t there, man!’ or ‘You can’t handle the truth!’ defences of
dubious military practices, but they are not going to be the last word
on the matter.
178

178 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

Twelve Angry Men is also helpful for illustrating the important dif-
ference between ad populum arguments that refer to actions and those
that refer to beliefs. As discussed in Chapter 1, as a bargaining tool
to persuade the other jury members to talk about the case, Juror 8
proposes that if he cannot change their opinion after an hour of dis-
cussion then he would change his vote to guilty. He is not suggesting
that he will have changed his mind about the case, but is offering,
rather, a practical solution. This is of course quite common; it is basic
to democracies and, in general, to situations in which going along
with the majority – despite not agreeing with them – can be the
best or right thing to do.These kinds of deliberation come with their
own critical questions (e.g., even in a democracy a case can be made
for it not always being right to go along with majority decisions), but
the main point here is that basing one’s action on the popular view is
very different from basing one’s belief on it.
‘Social dilemmas’ are situations in which a certain rational perspec-
tive can dictate that what we believe to be the right thing to do is
not what we should do, because we have reasons to believe that not
enough others will also be doing it. Either we do not want to be the
‘sucker’ who makes the sacrifice when others aren’t, or, as with reduc-
ing carbon emissions, if not enough people/countries commit then
the efforts of a few will not make any difference.There is not the space
to pursue this topic any further here, but there is a vast literature to seek
out if you are interested (see Further reading at the end of the chapter).
Some recent cases of social media based ‘mob pressure’ or ‘witch
hunts’ raise some intriguing questions concerning the power of ad
populums. The forced resignation of Nobel laureate Sir Tim Hunt
from his academic post at UCL in 2015 after making sexist com-
ments (in the form of a joke) at a conference is one example. UCL’s
decision was regarded as controversial, and some, including high-
profile astrophysicist and TV presenter Brian Cox blame a ‘trial by
social media’ for forcing their hand. Commenting on the story, an
editorial in The Guardian concludes that:
Twitter is loud, shouty and mainly male. It is rubbish at nuance, detail
or ambivalence but it is perfect for rushing noisily to judgment, some-
times … in a downright threatening way. The experience of UCL is another
warning that a Twitterstorm is a digital riot, and that is how it should be
treated.7
179

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 179

Widely held ethical views are rightly taken more seriously than sci-
entific ones because they are value judgements that emerge from a
culture; experts are describing our sensibilities as much as they are
shaping them. However, if we assume that social media is able to
amplify what appear to be popular moral views in a way that can have
this kind of impact, then the helpfulness of the critical questions for
guiding our analysis is plain. The answer to each of them is not obvi-
ous and requires not only an investigation of the details of the case,
but philosophical questioning about the appropriate role of popular
opinion in the making of moral and political judgements.

REFERENT POWER: PSYCHOLOGY AND RHETORIC

As the discussion of social proof in Chapter 1 shows, appeals to

popular beliefs and behaviours are known to be extremely persuasive;
its effects are immediate and powerful across a wide range of situa-
tions. Schultz, Noland, and Caildini (2007) measured the effects of
social proof – and in particular referent power – on energy consump-
tion by providing a sample of residents in San Marcos, California
with information comparing their household consumption with
the average consumption in their neighbourhood. This had an effect
on electricity usage referred to as a ‘rush to the middle’: those with
higher than average bills consumed less, and those with lower con-
sumed more.
This is further evidence for the power of social proof, but the
aim of the study was to test a way to persuade people to use less
energy, not bring everyone closer to the average. The method was
to add to this purely descriptive norm an ‘injunctive’ norm (indicat-
ing approval or disapproval), originally in the form of a smiley face
for lower than average consumption, and a sad face for above aver-
age. Combining descriptive and injunctive norms in this way led
to a significant decrease in average consumption. The effectiveness
of the injunctive norm can seemingly be attributed to the broad
legitimacy of environmental values (or at least of not being waste-
ful), but the adoption of this practice by a progressive company in
California called Opower serves as an example of referent power.
The injunctive norm method works, but not always for customers
with a non-progressive political leaning (Republican voters), some
180

180 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

of whom increased their consumption. Unlike liberal customers, they

will not identify with the overtly progressive approach of Opower or
the kind of people who might aim to reduce consumption, and per-
ceived social pressure towards doing so can create acts of defiance.8
The general understanding among social psychologists is that the
tendency to rely on referent power increases when we find ourselves
unsure how to behave (see Chapter 1). A variation of the Milgram
experiments involved three teachers (two stooges plus the subject). At
different stages the stooges rebel, resulting in the obedience rates for the
remaining (real) subject dropping dramatically (Milgram, [1974] 2005,
pp.117–22). In the San Marcos consumption study, since there is ambi-
guity around what constitutes an acceptable level of energy to consume,
it makes sense that we turn to those who are ‘like us’ for a benchmark.

REFERENT POWER, DISPOSITIONS, AND CONSTRUCTIVE DIALOGUES

The way we handle arguments from popular opinion or that carry

referent power is, in part, a function of our courage and of our willing-
ness to take responsibility for our commitments. Zachary Seech rightly
emphasises the pressure that can come from ad populum arguments, not
so much towards changing our minds, but silencing us through fear:
When someone says, ‘Everyone knows that!’ an implication seems to be
that any person who doesn’t possess this common information must be
especially dense or poorly informed. In the company of peers or … supe-
riors, many people will relinquish [i.e., not defend rather than necessarily
stop believing] even the most secure positions … Other familiar phrases
are … ‘No one seriously doubts that’ or ‘No one in his right mind could
doubt that’ or ‘No educated person would doubt that’. The latter two are
especially intimidating. (1993, p.134)
Referent power implies the threat of some level of rejection by the
group, and in this respect is also a species of ad baculum argument.
One of Christopher Tindale’s ad populum critical questions (that
seems to combine Q1 and Q2 above) is: ‘Is the … belief or practice
so widely known to be correct that the burden of proof would lie
with anyone who questioned it?’ (2007, p.107). This underlines the
dialogical implications of going against the grain. Like Juror 8, you
are faced with a group of people (or a representative of that group)
with nothing to prove, and who are more than likely annoyed that
181

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 181

you do have something to prove. Like Socrates, you are regarded as

an irritant. Returning briefly to social dilemmas, part of the rational-
ity for not acting on your beliefs, or possibly finding subconscious
methods for changing them (see cognitive dissonance), can be to
remain onside with your peer group or wider cultural location.9
On the other side of this, however, our critical thinking disposi-
tions are also important for handling ad populum arguments that are
personal in nature.There are ways in which other people can know us
better than we know ourselves, and if we are to gain insight then it is
important to be receptive to the, often painful, feedback they can pro-
vide. While it can sometimes be right to be sceptical of isolated com-
ments about our character or behaviour, to remain dismissive of the
possibility that we are indeed arrogant, self-centred, fickle, smelly (or
whatever the criticism might be) in the face of unanimous agreement
among disparate friends, relatives, and colleagues, tends to indicate a
failing of dispositions such as courage, humility, or open-mindedness.

7.3 SUMMARY
We saw in the previous chapter that appeals to authority are unavoid-
able, and handled with care they can be strong arguments. Even more
care is needed with ad hominem arguments – they can be lazy decision-
making shortcuts, and seductive for this and other reasons. However, a
person’s circumstances, behaviours, or general character can sometimes
be relevant to how we go about assessing their arguments. So long
as the conclusions we draw are proportionate – rarely is a personal
attack enough by itself for an outright rejection – then they are by no
means always fallacious. The same, from a positive perspective, applies
to arguments appealing to ethotic authority.The social proof and refer-
ent power that ad populum arguments court helps explain their appeal,
but they too are not always weak, especially in circumstances where we
are deciding what to do rather than what to believe.

EXERCISES
1. Assess the following tu quoque argument (from Roger Scruton,
On Hunting (London: Yellow Jersey, 1999), pp.139–40) by
applying the ad hominem critical questions to it (see Ch.6,
section 6.2).
182

182 ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2)

It puzzles me that [those who oppose fox hunting] should have

singled out an activity in which animals and humans, working in
happy companionship, are fully and magnificently alive, and in
which no suffering occurs that is not part of nature’s due. Do
the protestors trouble themselves, I wonder, over the factory
farms, where pigs and chickens are grown like vegetables for
the sake of their meat? One glance into these fermenting seas
of misery would cure people of the illusion that they live on mor-
ally respectable terms with the rest of nature. … Many who shout
and scream at the hunt happily eat the tortured limbs of battery
chickens. … [Factory farmed pigs] are served in the restaurant of
the House of Commons. And not one of those members who
parade their tender conscience over fox-­hunting has protested
over the crime.
2. Since ad hominem arguments of all varieties are so com-
mon, an enlightening exercise is to find several examples of
your own from political debates, opinion pieces, interviews
(from print or digital news, broadcasts, podcasts, social media,
etc.). Apply the critical questions to them, assess their poten-
tial persuasiveness, their impact on the dialogue, and what
they might reveal about the dispositions of the people who
use them.
3. Consider further circumstances in which common beliefs are
an appropriate criterion for establishing one’s own beliefs, or
as grounds for a certain course of action; and examples where
they are used in this way but should not be.

FURTHER READING
Make use of the in-text references as a guide to further reading on
specific aspects of this chapter, and a further recommendation is
Douglas Walton’s book Ad Hominem Arguments (2009).
To find out more about social dilemmas a classic early text is
Robert Alexrod’s The Evolution of Cooperation (Penguin, 1984); there
is an enlightening and accessible chapter on it in Peter Singer’s How
Are We to Live? (Oxford University Press, 1993); and for a good book
applying this dilemma to climate change see Stephen Gardiner’s The
Perfect Moral Storm (Oxford University Press, 2011).
183

ARGUMENTS, SOCIAL POWER, AND MESSAGE SOURCE (PART 2) 183

NOTES
1 Asch, S. (1946) Forming impressions of personality. The Journal of Abnormal
and Social Psychology, 41(3), 258–90.
2 For two excellent examples of charismatic leadership (as well as a range of
rhetorical techniques) see the character Andreas Wolf in Jonathan Franzen’s
novel Purity (Fourth Estate, 2015), p.262, and Forest Whitaker’s portrayal of Idi
Amin in the film The Last King of Scotland (Kevin MacDonald, 2007) – most
notably the section where Amin is persuading the young Scottish doctor to be
his personal physician (between approx. 24 and 36 minutes).
3 A very good book devoted to this subject is Miranda Fricker’s Epistemic
Injustice (Oxford University Press, 2007).
4 See Hoffman, A. (2011) Talking past each other? Cultural framing of skep-
tical and convinced logics in the climate change debate. Organization &
Environment, 24(1), 3–33.
5 www.reuters.com/article/wonga-church-stakesale-idUSL4N0PL5O420140710
6 For example, Discovery Channel’s Mythbusters, the Fortean Times column
‘Mythconceptions’.
7 The Guardian view on theTim Hunt affair: an explosive combination of science,
sexism and social media (2015) [Editorial].Available at: www.theguardian.com/
commentisfree/2015/jun/30/the-guardian-view-on-the-tim-hunt-affair-
an-explosive-combination-of-science-sexism-and-social-media
8 See Payne, O. Inspiring Sustainable Behaviour (Earthscan, 2012), pp.97–99.
9 Kahan, D. et al. (2011) The tragedy of the risk-perception commons: Culture
conflict, rationality conflict, and climate change. Temple University legal studies
research paper, 2011–26.
184

CAUSAL ARGUMENTS,
GENERALISATIONS, ARGUMENTS
FROM CONSEQUENCES, AND
SLIPPERY SLOPE ARGUMENTS

Since we are all such absolutists by instinct, what in our quality as students
of philosophy ought we to do about this fact? Shall we espouse and endorse
it? Or shall we treat it as a weakness of our nature from which we must free
ourselves if we can? I sincerely believe that the latter course is the only one
we can follow as reflective [people]. (William James, ‘The Will to Believe’)
Generalisations … never right, always fun. (Henry Rollins, Talk is Cheap
Vol.3)
If people start using slippery slope arguments, where will it end?
(Sacha T. Burnstorm, pers. comm.)

When decisions are made they are often justified in terms of the con-
sequences they are intended to bring about: making these changes to
the criminal justice system will decrease reoffending; making these
cuts to renewable energy subsidies will help reduce the budget deficit;
introducing video assistant referee technology (VAR) in football will
make key decisions more accurate and not overly impede the flow of

DOI: 10.4324/9781003247944-9
185

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 185

the game; eating less processed meat will make us healthier; and so on.
But of course we, and those we are trying to convince, need to have
reasons for believing these consequences will occur. Sometimes these
reasons take the form of arguing for the causal relationship between
various phenomena, and this involves two kinds of reasoning:

1. Establishing the likely causes of events that have happened in

the past.
2. Generalising from those events to future events, or to aspects
of the world that have not been directly observed or studied.

So, in deliberations about policies or everyday courses of action con-

sequences, causes, and generalisations are intimately linked. In
this chapter we will investigate each of these, beginning with causal
arguments and ending with a common, but sometimes problem-
atic, type of reasoning from consequences called the slippery slope
argument. Establishing cause and effect is basic to scientific and
historical research and thus knowledge of the methods these disci-
plines employ, and analyses of the strengths and weakness of these
methods, is fundamental to students’ and practitioners’ ability to
thinking critically in these areas. Critical thinking as a subject will
explore the basic forms of this kind of reasoning, and sometimes
discuss its association with the philosophy of science (see the end
of this chapter for further reading along these lines). A philosophi-
cal or critical thinking perspective seeks to do more than provide a
general introduction to scientific and academic methods though. By
teaching us (or reminding us) about some fundamental features and
assumptions in this area, it aims to refresh our thinking and make us
more open-minded in our approach to knowledge. In accordance
with the themes of this book there is particular emphasis on the sorts
of errors we tend make when reasoning about causes, generalisations,
and consequences, and their relationship to rhetoric, psychology, and
dialogues. From these insights we can hopefully gain a better reflec-
tive understanding of ourselves as deliberators and the dispositions
that can assist or impede these aspects of critical thinking.

8.1 CAUSAL ARGUMENTS AND CAUSAL FALLACIES

When I bought my first MP3 player and read the instructions for
turning it on, it said ‘hold down the button for 2–3 seconds and it
186

186 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

will switch on’. I did this, but nothing happened for about 6 or 7
seconds. Because I’ve not always had the best experiences with tech-
nology I assumed that (typically) my MP3 player was a bit slow. It
otherwise worked fine though, so this wasn’t much of a problem, and
I would learn to love it anyway. I just developed the habit of holding
the button down for this length of time. Several months later I was
distracted by something whilst going through this ritual and let go
after a couple of seconds.When I looked back at the screen a few sec-
onds later I was surprised to see that it was on. The penny dropped. I
had been assuming that turning it on meant that something needed
to appear on the screen, but clearly that is not the case; holding down
the button for 2–3 seconds is indeed enough to activate it, but noth-
ing actually appears on the screen until after 6 or 7 seconds.
Apart from a need to reassess my relationship with technology, what
this illustrates is how the quick assumptions we make about cause
and effect can be both erroneous and habit-forming. Establishing
the cause of something can be very difficult indeed, and of course
attempts to do this are basic to the theory of the natural and social
sciences and to the work of historians. Each discipline will have dis-
tinctive criteria for establishing the strength of causal claims, and they
will share some general principles as well.
In the natural sciences a causally closed system is assumed in
which all phenomena adhere to the physical principles governing
the behaviour of matter. In the social sciences, explanations range
from the physical to the cultural and are entwined with philosophi-
cal debates about the nature of the mind’s relationship with the brain
and the meaning and existence of free will. How, for example, can a
thought – conceived of as a non-physical thing – cause changes to
the physical world? If the mind is essentially a physical thing, and thus
part of a chain of cause and effect, is free will possible? How are rea-
sons for acting in certain ways different in kind from the mechanical
causes of natural phenomena?
These are metaphysical questions, but more closely associated with
critical thinking are epistemological puzzles. As David Hume ([1748]
2008) pointed out, we do not directly observe the cause of an event;
the process by which X brings about Y. We observe that my golf club
making contact with the ball is perfectly correlated with the ball
moving (if not always the distance and direction I want it to go), but
187

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 187

the physical laws that we assume determine this are inferred rather
than evident to the senses. Kant ([1781] 1929) argued that the very
possibility of having experiences requires certain necessary condi-
tions, including causality, but that the actual basis of our experience
of causation is itself unknowable.
It is this limitation on our knowledge, combined with both the
complexity of the world and what Kahneman (2012, pp.74–8) sug-
gests is an innate tendency to ‘see’ causes where they do not exist, that
makes causal reasoning particularly vulnerable to error. The causes of
phenomena like climate change, educational attainment, and crimi-
nal behaviour; or of events like the Reformation, the American Civil
War, and the fall of apartheid in South Africa are multiple, complex,
and hard to establish with certainty. As we have seen, we have a ten-
dency to jump to conclusions about straightforward cases of cause
and effect, and with complex phenomena an added problem is to
oversimplify our explanations.
The specialisation of academics can present us with a variation of
this problem, and increasingly collaboration between disciplines to
improve accuracy in establishing causes is encouraged. For example,
the current obesity epidemic in parts of the developed world falls
into the domains of (among others) health and nutrition, sociology,
politics, psychology, and economics. Similarly, practitioners from a
range of professions (social work, medicine, mental health, educa-
tion) will work together to implement solutions for child abuse pre-
vention, community and clinical care for people with mental health
problems, or managing end of life care.
Everyday reasoning presents similar challenges. What caused Aunt
Eliza to divorce Uncle Vance? Were they ‘simply’ growing apart? Was it
the influence of her sister-in-law? Was she never truly able to forgive
him for what happened in Düsseldorf? Was it all of these things com-
bined, or some of them, or something else entirely? This is clearly not a
matter for a research project or a team of practitioners, but concerned
friends and family will often be mulling over causal explanations and
presenting arguments to one another in an analogous fashion.
It is outside the scope of this book to delve further into the sub-
tleties of the causal explanations that different situations present or
that different disciplines trade in. It is though important to survey
some general principles of causal reasoning. The aim here is to seek
188

188 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

to clarify these through the description and analyses of causal reason-

ing errors. After this I will explore the dispositions and psychological
vulnerabilities that help explain what makes these errors so prevalent.
An initial distinction to make is between cases in which we have:

1. Multiple instances of one event being correlated with another –

such as tea being more flavoursome and it being made with
near-boiling water rather than slightly cooler water; or a
patient’s belief in the effectiveness of a drug and an increase
in the speed of their recovery. In such cases natural and social
scientific methods can usually be employed to observe or rec-
reate conditions in which the cause and effect underlying the
correlation can be adequately demonstrated.
2. Unusual or highly distinctive cases where causation needs
to be argued for more indirectly (such as world wars, or the
extinction of dinosaurs). In instances like these we can nor-
mally still find historical precedents or comparisons with sub-
sequent events – for example, the causes of the 2008 financial
crisis shared some features with the 1929 crash, and the causes
and consequences of the Salmon Rushdie attack in 2022 are
not entirely dissimilar to other assassination attempts – and
thus we find that inductive arguments from analogy (see
Chapter 9) are often employed to help establish the causes.

With this distinction in mind we can formulate the general structure

of causal arguments:

P1: (Either through direct observation or analogy we have confi-

dence that) X is correlated with Y.
P2: The ways in which a causal relationship can be confused with
a mere correlation have been acknowledged and discounted to
an extent that is reasonable through appropriate methods.
C: Therefore it is probable that X causes/caused Y.

CAUSAL FALLACIES

When attempting to establish the cause of a happening, we can be

presented with two broad kinds of relationship between it and its
189

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 189

possible causes. On the one hand it can be clearly preceded by an

event that could plausibly cause it to happen, and the more times we
observe this temporal sequence the greater the chance of the prior
event having something to do with its causation. On the other hand,
we may be looking at two or more happenings that occurred around
the same time, but with no clear temporal ordering.
In both cases causality will not be known without further inves-
tigation, but where there is a clear temporal relationship the specific
fallacy known as post hoc ergo propter hoc (or post hoc for short) is a risk.
This Latin term mean ‘after this, therefore that because of this’, and
captures the assumption that because one event has followed another,
then the second event has been caused by the first.
Without a clear sequence of events the direction of causality – if
indeed there is a causal connection at all – can also be questioned. I
find myself in a grumpy mood and blame the kids’ bad behaviour, but
can I be sure that it was their bad behaviour that caused my grumpy
mood rather than my grumpy mood that caused their bad behav-
iour? To further complicate things, once initiated the causal link can
work in both directions in a mutually reinforcing fashion. But what
if the two things are entirely unrelated and I am being too quick or
too lazy in my hypothesising? If I reflect, then perhaps it was Aunt
Eliza’s email about the divorce that put me in a bad mood, and come
to think of it, the kids were already behaving poorly when I picked
them up from school.
The main fallacies associated with causal reasoning fall under the
headings of overlooking coincidence, mistaking direction of cause,
overlooking a shared cause, multiple causes, the placebo effect, and
mistaking cause for correlation (the Sod’s law fallacy). These apply to
either one or both of the types of correlation outlined in the previ-
ous two paragraphs.

OVERLOOKING COINCIDENCE

If two events regularly occur together but are not causally related, the
chances are that they have a shared cause (see below). With singular
or unusual events there is a strong possibility of mistaking mere coin-
cidence for cause. My belief about my MP3 on button is an example
of this, as are many superstitions. Every so often in the news we hear
190

190 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

about ‘super-centenarians’ – people who have lived past 110: ‘What

is their secret?’ is often asked, but no one knows, and the replies are
clear cases of coincidences pretending to be causes. In a brief article
on the subject in the Fortean Times various super-centenarians told us
their secrets:1 Besse Cooper (116 years) said ‘I mind my own business
and I don’t eat junk food’; ex-postman Jiroemon Kimura (also 116)
put it down to ‘the sun’, but on other occasions said it was caused
by his diet (e.g., steamed fish), exercise, and early rising. Magomed
Labazanov (122) claimed it was ‘abstaining from alcohol, tobacco and
women’. Maybe some of these can contribute to a given individual
living past 110, but even if they do it would be prohibitively difficult
to test these causal claims. Or maybe, as David Attenborough once
said about living into his 90s, it’s ‘pure luck’.
At the other end of life, our youngest child was a week overdue
so we found ourselves being fed, and occasionally looking for, advice
on how to induce labour: raspberry leaf tea, bumpy rides, pineapple,
walks, sex, curry. Having no other pressing engagements my wife
(and I) spent an afternoon doing lots of these things, and the next
day she went into labour – AMAZING! Like the super-centenarians,
these recommendations are largely anecdotal; no one knows.
One other example is more serious because it identifies a type
of mistake made in formal scientific reasoning. The phenomenon
known as regression to the mean predicts that an extreme meas-
urement for any phenomenon will typically be followed by a less
extreme one. For example, when trying to understand pain and other
symptoms, these tend to occur in cycles – severe → less severe →
severe and so on – and there is a reasonable chance that people will
seek medical intervention when their symptoms are at, or close to,
their worst. If pain relief or other medication is then supplied this will
typically correlate with a reduction in symptoms, and consequently
in the belief that it was the medication that caused this reduction. For
this reason, clinical trials need to control for, not just the placebo
effect (which is considered below), but for regression to the mean
as well.
Less weighty examples are found in sport. As hard-to-please
cricket pundit Geoffrey Boycott was keen to point out, getting
too excited about the then young England batsman Ben Stokes’s
record-breaking 258 runs in the Second Test against South Africa
191

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 191

in January 2016 is premature; he has to consistently score high to

really prove his self. More recently, a year after Emma Raducanu’s
extraordinary victory at the US Open in 2021 caused great opti-
mism among British tennis fans, she was back to having to qualify
for major tournaments.

MISTAKING DIRECTION OF CAUSE

As indicated, correlated events observed on repeated occasions are

likely to be linked in some respect, and the mistake is to jump to
unwarranted conclusions about the nature of this link. However
plausible a causal connection seems, it is worth reversing it to see if
there’s another possibility. In fact, the more plausible the direction
of cause seems, the more revelatory a reversal can be. Group psy-
chotherapist Irvin Yalom (1980) describes the case of Eve, who was
reluctant to fully engage with the other members of the group, or to
discuss herself except in the vaguest of terms. Finally, she was con-
fronted with the negative effect this was having on the others, and
Yalom encouraged her to take more of a risk by making more hon-
est and direct comments. At this point she revealed that she was an
alcoholic, and said that the shame this led her to feel was the reason
for keeping herself hidden. Yalom, though, proceeded to reverse her
causal argument, suggesting,
she did not hide herself because she drank, she drank because she hid
herself. (ibid, p. 394)

Yalom’s theory about Eve was that her underlying loneliness, created
by an inability or unwillingness to connect with others, caused her
drinking. Primarily, the drink was a means of coping with her isola-
tion, rather than her isolation resulting from the shame associated
with her drinking.

OVERLOOKING A SHARED CAUSE

A third type of mistake in causal reasoning is to overlook a cause that

is shared by the correlated events. If religious people are happier on
average than non-religious people this could be because religion leads
to greater happiness, but it could also be that a third variable – for
192

192 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

example, religious families being more stable – results in both the

happiness and the religious commitment.
In a contrasting example, research in 2015 into the relationship
between teenage goths and mental illness2 was widely reported in
the news as
Goths three times more likely to suffer depression or self-harm

Although this headline does not necessarily state that being a goth
causes or contributes to depression, this is what is implied, presumably
because it makes for a better story. Read on and the issue of direc-
tion of cause is usually mentioned in press coverage, and the article
itself, though very cautious about causation, suggests that identifying
with the goth subculture does increase vulnerability to depression
and self-harm (rather than a reversal in which mental health issues
make it more likely that a person identifies as a goth). However, the
authors readily acknowledge that they are not in a position to rule
this out and, importantly, nor do they rule out shared causes such
as ‘stigma and social ostracism’ (a key aspect of the article that was
sometimes overlooked in the news). In other words, rather than ‘peer
contagion’ (hanging out with other goths) being the sole, or even
contributing, cause of these mental disorders, both the disorders and
the goth identification could have a shared cause in underlying con-
ditions such as social exclusion and bullying.
The visibility of the variables in question – in this case mental
disorders and being a goth – and the simplicity of certain causal
explanations (perhaps combined with certain stereotypes) is prime
System 1 candy. Shared causes are far less obvious and require more
imaginative hypothesising, and thus we are vulnerable to o ­ verlooking
them.
Another example is found in an assessment by John Oliver (in
HBO’s Last Week Tonight) of the conspiracy theory that 5G masts are
spreading Covid.3 The theory is based on a correlation between the
siting of these masts and high rates of coronavirus, but Oliver points
out that these transmitters tend to be sited in areas of high population
density, which is also where Covid outbreaks are most likely. High
population density is, then, the reason behind both the existence of
5G masts and increased rates of Covid, rather than 5G masts being
responsible for spreading the disease.
193

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 193

OVERLOOKING MULTIPLE CAUSES

System 1’s fondness for jumping to conclusions is abetted by a dis-

like of complexity. What’s been called the ‘narrative fallacy’ (see
below) is explained by a desire for coherence in our understanding
of the world, and this is achieved by ignoring gaps in our knowledge.
Unfortunately, few interesting things have simple causal stories, and
so critical thinking must alert us to the fallacy of oversimplifica-
tion of causes. For example, there is a widely held belief that it was
the fire of London in 1666 that ended the bubonic plague epidemic
in England that had started the previous year. Current thinking on
the issue, although still not settled, is that the full picture is more
complicated. Although the fire did kill many of the plague flea car-
rying black rats, the plague was on its way out anyway. Theories sug-
gest that this was partly because black rats were being displaced by
the relatively non-plague transmitting brown rat, and that the cold
weather that autumn (the fire happened in September) killed off the
remaining fleas.
Once again, the cognitive availability of the fire and the plague,
and the rough plausibility of one destroying the other, is likely to be
at least partly responsible for the myth’s perpetuation. But whether
this is the case or not, the attraction of simple causal stories is again
demonstrated. Further examples include the cause of the American
Civil War (which was not just about slavery), and the connection
to the popularity of conspiracy theories is explored below (see The
narrative fallacy).

THE PLACEBO EFFECT

The placebo effect can be defined as someone’s expectation that an

intervention (such as a drug) for a certain condition will work causes
the condition to improve, rather than the intervention itself being the
cause of the improvement.
Mistaking a placebo’s effect for the intervention’s effect forms a
separate category of causal reasoning error because it is not a coinci-
dence, nor a mistake about the direction of cause, nor an overlooking
of a shared cause, and nor does it have to be one among multiple
causes. Instead, the cause is very much connected with the mistaken
cause, but in a way that is peculiarly indirect. I say ‘peculiarly’ because
194

194 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

the actual cause (the patient’s expectation) is of a different order to

the supposed cause (a directly physiological one). With non-peculiar
cases of causation an indirect cause is typically part of a chain of
cause and effect, and the error would be to miss a link in the chain,
but overlooking a placebo effect is to overlook a different chain alto-
gether that has its own beginning in the patient’s psychology.
In medicine the placebo effect is a well-recognised (if not well-
understood) phenomenon that clinical trials need to be conscious of
in their design. For example, if half the patients in a trial are taking a
new drug and half the same drug as before, although all participants
will be told that they taking part in a trial, it is crucial not to tell them
which group they will be in.
It is also important that trials are ‘double-blind’, which means that
not only are the participants unaware of whether they are in the
intervention or placebo group, the experimenters are also unaware
of who’s who. The reasons for this are that:
• This knowledge can unconsciously bias the way experimenters
observe and interpret results.
• Experimenters’ expectations can be communicated unknowingly
to participants, which can then influence their experiences and
behaviour.
This latter phenomenon is known as a ‘self-fulfilling prophecy’,
which itself shares features with the placebo effect. Examples of this
are the ‘pygmalion effect’ and its opposite, the ‘golem effect’. In their
book Pygmalion in the Classroom ([1968] 2003) Robert Rosenthal
and Lenore Jacobson demonstrated how teachers’ expectations about
children’s abilities would unconsciously affect these children’s educa-
tional attainment. Take two groups of school pupils of equal ability,
tell one teacher that their group are predicted (by a fictitious test)
to make an intellectual ‘spurt’; tell the other nothing, and at the end
of a school year the performance of the first group will tend to be
significantly better than the control. One suggested reason for this
is that teachers with high expectations are more likely to praise the
effort children put into their work, reinforcing the belief that effort
pays off, and thus encouraging persistence. In contrast, teachers with
lower expectations focus more on correcting negative elements of
pupils’ work. The absence of feedback about effort leads to these
195

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 195

students attributing success and failure to ability rather than effort,

thus discouraging persistence (Cooper, 1979).
What overlooking the placebo effect and overlooking self-fulfilling
prophecies share is a failure to account for the psychological influence
of actors within certain contexts. Both practitioners and the pubic are
now reasonably familiar with these phenomena, but we can neverthe-
less be surprised by the potency of unconscious processes in affecting
physical and social change. For this reason, and perhaps because of their
relative invisibility, they are easy to overlook or dismiss.

MISTAKING CAUSE FOR CORRELATION (SOD’S LAW FALLACY)

We are often motivated to jump to conclusions and affirm causal

relationships between phenomena without sufficient reasoning or
evidence. However, we can also be motivated to do the reverse and
mistake cause for correlation. This could take the form of a
rationalisation in a situation where we want to avoid having to take
responsibility. For instance:
When I drove at 50 in a 30mph zone, that’s when I ran over a dog. Sod’s
law I suppose.

Sod’s law means that things tend to go wrong at the most inconven-
ient time, and is essentially about bad luck. It is Sod’s law that my
boiler breaks down the very month that car and house insurance are
due for renewal, and all those things in Alanis Morissette’s song Ironic
are examples of Sod’s law (rather than being ironic). In the driving
example it is not Sod’s law that you ran someone’s dog down, it was
because you were driving too fast and couldn’t stop in time.To make
yourself feel better you infer an unfortunate coincidence of events
rather than accepting a causal relationship between your speed and
the accident occurring. In Boris Johnson’s resignation speech in July
2022 he famously said ‘them’s the breaks’. He was implying it was
bad luck that ended his tenure as prime minister rather the persistent
erosion of backbench MP’s and eventually his cabinet’s trust caused
by, among other things, his dishonesty and sense of entitlement. If
nothing else this parting phrase was true to form.
A somewhat different set of motivations lie behind this next
example. In October 2015 the singer and activist Charlotte Church
196

196 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

appeared on the live UK political debate show Question Time, and

among other things suggested that climate change was partially
responsible for the Syrian crisis. A few days later The Sun (the UK’s
best-selling tabloid newspaper) found this hilarious, mocking the
statement under the sub-header: CHARLOTTE ON … SYRIA.
On the face of it this assertion does seem far-fetched, and coming
from someone perceived as a political non-expert it can easily be
portrayed as silly or naïve. However, she was referring to a serious
piece of research that indicated a plausible connection between these
things. It is known as the ‘food shock’ argument (extreme weather
such as droughts leading to globally significant food shortages which
have a disproportionate effect on poorer places, especially those reli-
ant on imports) and connects a food shock in 2011 to civil unrest in
the Middle East and the Arab Spring. This is then causally connected
with the rise of IS and the devastating effects of the Syrian civil war.
‘In a sense what we’re living with IS today came out of a spark that
came from food price rises,’ says Professor Tim Benton of the UK
Global Food Security Program and the University of Leeds.The link
to climate change is that it is making food shocks far more frequent
than they were (a one in a hundred years to a one in thirty years
phenomenon, and potentially a seven in ten years phenomenon in
2070), and perhaps more to the point in terms of Church’s message,
this kind of humanitarian disaster serves as a synecdoche (meaning
a small part of something representing the whole) for the kinds of
impacts climate change will have.4

CRITICAL QUESTIONS FOR CAUSAL ARGUMENTS

The critical questions that can be applied to causal arguments

­correspond to the headings above (overlooking coincidence, mistak-
ing direction of cause, overlooking a shared cause, multiple causes,
the placebo effect, and mistaking cause for correlation (the Sod’s law
fallacy):

Q1: Has a correlation been identified?

Q2: If so, how convincing is the causal reasoning that attempts to
explain the phenomenon in question?
197

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 197

This second critical question can then be answered in accordance

with the headings above:

Can the correlation be explained by coincidence?

Can the causal relationship be reversed?
Can the correlation be explained by a shared cause?
Have multiple causes been overlooked?
Can the correlation be explained by psychological phenomena
such as the placebo effect or a self-fulfilling prophecy?
Are there reasons for thinking that the possibility of a causal
relationship (rather than a mere correlation) has been dismissed
too quickly or for the wrong reasons?

CAUSAL ARGUMENTS: PSYCHOLOGY AND RHETORIC

Errors in causal reasoning are exacerbated by several features of our

psychology:
• We like to have explanations rather than remaining uncertain
about why things happen.
• The explanations we come up with, or more readily believe, will
typically be consistent with our present assumptions about how
the world is.
• As we have seen, seeking coherence is a fundamental motivator of
System 1 thinking, making it more likely that we will jump to
roughly coherent, but incorrect conclusions.
When our Halloween pumpkins begin to rot the first thing that
­happens is the jaw collapses, closing the mouth and giving it the
appearance of its pre-carved form. A few years back our then
4-year-old reasoned that it ‘wants to be a pumpkin again’ – i.e.,
rejecting the grotesque face I had carved. And like many children
of that age, he had surmised that the clouds rain in order to water
the plants. The need for causal explanations is evidenced by kids’
constant ‘why’ questions, but before abstract scientific reasoning
is a cognitive option for them, they attempt to make sense of the
world in accordance with what they do understand – conscious
intentions.
198

198 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

Two areas of research that make particularly interesting ­contributions

to our understanding of causal reasoning errors are the psychology of
superstition, and what’s been termed the narrative fallacy.

SUPERSTITION

In his investigations into trial-and-error learning (‘operant condi-

tioning’) in rats and pigeons, behavioural psychologist B.F. Skinner
conditioned superstitious behaviour. Most of Skinner’s experiments
investigated animal learning in situations in which there is a causal
relation between behaviour and a reward; a pigeon must learn to peck
a button when a light comes on in order to receive food, for exam-
ple. In his studies of superstition, however, this causal connection was
absent. Food was presented to the (hungry) pigeons at regular intervals,
and nothing the birds did could alter this. Despite this it was observed
that in most cases an action randomly performed just prior to the food
arriving would then be repeated. If the repetition coincided with the
next scheduled delivery of food, then that behaviour – turning around
in a certain direction, a ‘pendulum motion of the head and body’ for
instance – would be learned.5 Correlation is confused with cause, and
the pigeons’ behaviours have the appearance of a ritual.
While pigeons cannot fathom the underlying causal features of the
situation, humans as a species have, as we know, become very good
at this. Nevertheless, we are extremely superstitious, suggesting an
adaptive advantage to being sensitive to correlations, which in turn
causes a negative emotional response to the interruption of routines.
System 2 knows that the colour of your socks has no effect on how
well you will perform in the exam, but this won’t make System 1’s
anxiety at the prospect of wearing stripes rather than just black go
away without a struggle.
In his book Caveman Logic (2009) Hank Davis examines the
­widespread nature of superstitious beliefs and behaviours, and his
evolutionary explanation is that interpreting causality on the basis
of too little evidence had greater survival prospects than its reverse
(i.e., being overly conservative in such judgements).Why is this? One
answer is simply the value of heuristics, as discussed in Chapter 1:
inheriting and being disposed to learn a range of intellectual short-
cuts rather than a bias towards the slow analysis of causal relationships
seems to have been more adaptively advantageous for our ancestors.
199

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 199

Another answer – one that is directly pertinent to superstition – is

that feeling in control of situations that we are in fact not in control
of is beneficial to us. Several reasons have been suggested for why
this is, including the idea that the confidence this inspires, although
resulting in some mistakes, also has beneficial side-effects. One of
these is being motivated to put more effort into our endeavours, so
that when we can in fact influence outcomes, those outcomes are all
the more impressive.

THE NARRATIVE FALLACY

The coherency of a causal explanation – for example, that it is con-

sistent with known truths – is a necessary but far from sufficient con-
dition for the truth of that explanation. It is easy to concoct multiple
alternative explanations for a phenomenon, but the difficult part is
establishing which, if any, of these is true. Here again we encounter
our dislike of uncertainty, and our resulting tendency to jump to
unwarranted conclusions.
The narrative fallacy is a term coined by risk scholar Nassim
Taleb in his book The Black Swan (2007, Ch.6). It refers to our attrac-
tion to, and confidence in, coherent but simple causal explanations
at the expense of an appreciation of complex and unknown or ran-
dom variables. Also known as the ‘illusion of understanding’, it is a
prime case of System 1’s fondness for simplicity and coherence, but is
also motivated by our desire to make sense of the world rather than
acknowledge the limitations of our understanding (see Kahneman,
2012, Ch.19). This leads us to invent (or readily accept), and then act
upon, straightforward but inaccurate causal stories.
One account of the appeal of wildly implausible (or ‘unwarranted’)
conspiracy theories (e.g., that alien visitations are being covered up by
governments or the Covid epidemic was a hoax) is that they embody
a convenient or comforting idea that the world is ultimately open to
being explained.6 Instead of acknowledging the unpredictability (and
thus uncontrollability) that we should expect from an existence that
is vast, complex, and inherently devoid of overall purpose or plan,
some are drawn to theories of powerful, coordinated, covert intent.
The philosophically deep (and quite spooky) drama series The
Leftovers is an excellent case study of the circumstances, motivations,
and outcomes of the narrative fallacy. The situation could not be
200

200 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

more random – a huge event that is devastating for billions as 2% of

the world’s population (140m people) disappear without trace for (1)
no reason and (2) with no way to even begin to imagine what could
have caused it. The main upshot is a range of desperate theories,
including religious ones, that are an attempt to cope with the situa-
tion and its possible recurrence.
The ‘hindsight bias’ is a version of the narrative fallacy in which our
preference for simple causal explanations leads to the heartfelt excla-
mation ‘I knew it!’ when we knew nothing of the sort. In one exper-
imental design, participants assess the likelihood of various outcomes
before an event has happened, and then again after it has occurred.
The result is that people misremember their original guesses to bring
them more in line with the outcome; they believe they thought it
was more likely than they in fact did. In effect the outcome of an
event determines how we perceive its predictability, and Kahneman
(2012, p.204) sees the hindsight bias as especially damaging because
it leads to people being blamed for negative outcomes they could
not reasonably have been expected to predict (9/11 is one example
he uses). Worse still, this in turn leads to a tendency towards risk-
aversion in future decision-making.

CAUSAL ARGUMENTS, DISPOSITIONS, AND CONSTRUCTIVE

DIALOGUES

The narrative fallacy provides psychological insight into all of the

causal fallacies explored in this chapter, which share two essential fea-
tures: seeing causes where they do not exist, and overlooking com-
plexity. For the most part we like to be able to control or at least
predict our environments, but there are an awful lot of things that
affect us that are unknown. Combine this state of affairs with our
deep-seated tendency to see causal connections where they do not
exist, or to oversimplify, and you can see how vulnerable we are to
false claims about how buying this product, performing this behav-
iour, or voting for this party will make us happier.Towards protecting
ourselves we need to:
• Work towards greater clarity about what we can and cannot know,
control, or predict;
201

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 201

• Develop metacognitive awareness of the conflict between how

things are in terms of what we can control and predict, and how we
would like them to be (see previous references to the self-serving
bias and motivated reasoning);
• Develop metacognitive awareness of our proneness to superstition
and the narrative fallacy.
We need to learn to tolerate uncertainty and be more like Juror 8
in Twelve Angry Men. The illusion of control, like other self-serving
biases, is in opposition to modesty, understood as a realistic percep-
tion of our attributes and abilities. From a dialogical perspective,
encountering overconfidence in others can be demotivating and lead
to antagonism. Moreover, jumping to conclusions has an unfortunate
self-fulfilling tendency. The principle of commitment and con-
sistency (see Chapter 1) predicts that once we have taken a position
(such as ‘the defendant is guilty’), especially where this is made public,
it is hard to go back on. Being consistent is very important, imply-
ing as it does dispositions such as wisdom, trustworthiness, and not
being unduly influenced by peer pressure. If a conclusion is prema-
turely declared, therefore, revising it can come at a cost. If this cost
is (consciously or unconsciously) seen as too high, then (consciously
or unconsciously) we may choose to dig in rather than go where the
evidence should be taking us.
Dialogues will often be more productive if firm conclusions are
avoided too early on. It is far better to declare oneself as currently
agnostic until the issue has been further discussed, or to qualify one’s
position so that a change of mind does not lead to embarrassment
(or worse): ‘I’m inclined to this view on the matter, but this isn’t a
firm conviction and I’m open to being persuaded otherwise’; or ‘I
think this is probably true, but I’m not entirely sure and would like
to know more before reaching a conclusion.’

8.2 GENERALISATIONS
Causal arguments are often generalisations, but not all generalisations
are causal arguments. Another type involves arguing from qualities
that characterise a particular class of things (bicycles, flowers, pol-
iticians, scientific theories, vegetables, and so on) to the expected
202

202 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

qualities of individual instances of that class (the old but reliable

hybrid I ride to work on, geraniums, Jacinda Ardern, string theory,
runner beans). Perhaps the most recognisable version of a generalisa-
tion is the reverse of this; arguing from a single instance or a sample
of something to claims about those things as a whole: string theory
is tough to understand, therefore all scientific theories are tough to
understand.
In a similar fashion to causal arguments, generalisations are funda-
mental to scientific reasoning. A significant proportion of the meth-
odological knowledge and skills of scientists is about the legitimacy
of generalisations: what constitutes a large enough and representative
sample; understanding the degree of strength that a generalisation
holds; and the statistical methods appropriate for finding these things
out. At the end of this chapter you will find recommendations for
critical thinking textbooks which go into detail about generalisa-
tions, often with a philosophy of science perspective, but you should
also bear in mind that texts on natural and social scientific research
methods are also an important (and often overlooked) critical think-
ing resource.
Generalisations are most commonly classified as absolute (also
known as ‘universal’, ‘strict’, or ‘hard’), and non-absolute (also
known as ‘non-universal’, ‘inductive’, or ‘soft’). Absolute generalisa-
tions claim that all of a certain type of thing are a certain way (like
‘all countries have flags’), and non-absolute generalisations claim
that most or the majority of a type of thing are a certain way (like
‘most countries have land borders with other countries’). Non-
absolute generalisations can be subdivided into statistical and non-
statistical forms. The latter will use vague terms like ‘most’, whilst
the former quantify probabilities in terms of percentages and other
fractions.
The statistical/non-statistical division has relevance to a third type of
generalisation that aligns with plausible arguments (see Chapter 5).
As characterised by Douglas Walton (2006, pp.17–19), a defeasible
generalisation (full name: a ‘presumptive defeasible generalisation’)
is an attempt to apply a generally accepted belief to a particular
instance.
Unlike absolute and non-absolute generalisations they do not typ-
ically involve qualifiers like ‘all’ or ‘most’, and they certainly never
203

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 203

involve statistics. Their most important feature is their practical and

contextualised nature. They are saying that prior personal experi-
ence and common knowledge has established the right to presume
X about situation Y in circumstances where more rigorous investiga-
tion is not practical; ‘meat that is before its sell-by date and which has
been stored in the fridge is edible’, for example.
Defeasible generalisations are, then, inherently provisional; there
is a background assumption that exceptions to the rule are always
likely. Since, however, the burden of proof falls to those who want
to claim an exception in this instance, they risk perpetuating ste-
reotypes: seemingly plausible generalisations derived from common
knowledge that is inaccurate, or personal experience that is skewed
by cognitive biases. This will be addressed under ‘Sweeping generali-
sations’, below.
The general formal structures of arguments employing a generali-
sation are these:

FROM A SPECIFIC INSTANCE TO THE WHOLE:

P1: All (or most) observed instances of X have characteristic Y.

C: Therefore all (or most) instances of X will have characteristic Y

FROM THE WHOLE TO A SPECIFIC INSTANCE:

P1: The class of things X typically exhibit characteristic Y.

C: Therefore Z (which is a member of class X) will typically
exhibit characteristic Y.

The remainder of this section will explain and discuss the main fal-
lacies associated with generalisations: hasty generalisations and
sweeping generalisations.

HASTY GENERALISATIONS

In her research into everyday argumentation, Deana Kuhn studied

people’s reasoning on a couple of topics, including why people fail
at school. One participant illustrated a certain pattern of reasoning
when, after explaining failure in terms of laziness and peer pressure,
204

204 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

gives this response to the question: ‘How do you know this is the
cause?’
[Because] I see it around me, you know. I have friends who fail. They figure
it’s the right thing to do, and, you know, they just get lazy or want to hang
out with their friends. (Kuhn, 1991, p. 74)

This is a case of generalising from too few examples to a conclusion

about the whole and, as with many cases like this (we’re all guilty),
the too few examples come from our own personal experiences. It is
understandable why we do this; our experiences are vivid and concrete
and thus collude with the availability heuristic and have broad appeal
to the experiential system (see Chapter 2).They exaggerate the rele-
vance of the pattern in question, prompting us to jump to a conclusion.
Hasty generalisations are also known as the ‘fallacy of insufficient
statistics’ or as a ‘sampling error’. In social science generalising from a
sample to a whole population is basic to its methodology. Since testing
or observing the whole population is usually too time-consuming and
costly, sampling and statistical methods have been devised which are
quite reliable in their predictions. In the hands of those suitably trained,
generalisations in these academic and applied contexts – such as mar-
ket research and opinion polls – are not fallacious when appropri-
ately derived and when their limitations are known and accounted for.
Where this is not the case, hasty generalisations can occur. These take
two general forms: too small a sample, and an unrepresentative sample.
The previous example is problematic in both of these respects: the
people this person knows who have failed at school will not be a large
enough sample to base a prediction about the whole of America on
(although an appropriately sized sample is far smaller than we might
typically imagine – 1,000–2,000 people); and it is highly unlikely
that they would be representative in terms of the demographics they
represent (age, gender, race, region, social class, and so on).
A variation on unrepresentative samples is what is known as cherry
picking. This means that information or examples that conform to
your hypothesis or that present your position in a flattering light are
presented as evidence, while non-conforming or unflattering data
is ignored or supressed. We can cherry pick in a non-deliberate way
(unconsciously driven by the confirmation bias), but it can also be a
deliberate attempt to misrepresent the available data.The main differ-
ence between an unrepresentative sample and cherry picking is that
205

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 205

with cherry picking a more complete set of information is available

but not shared, whereas with an unrepresentative sample only a par-
tial data set exists (or in the failing at school type of example above,
is known to the person presenting the argument).
Superficial and one-sided communications such as advertisements,
company annual reports, CVs, or social media personal profiles will
cherry pick like there’s no tomorrow. This might be undesirable, but is
at least something that we expect to happen. In other contexts it is more
troubling. It has been a common tactic of organised corporate and polit-
ical denial of the health risks from smoking and global environmental
issues, and, as Ben Goldacre argues, is commonly used in research car-
ried out by advocates of alternative therapies and nutritionism.7

SWEEPING GENERALISATIONS

The other topic Deana Kuhn used to study people’s everyday rea-
soning skills was criminals reoffending. In one example a participant
accounts for reoffending in the following way:
Human beings are very much creatures of habit, and I don’t think that
there’s such a habit as committing a crime, but everything that leads up
to committing a crime is probably habit. (Kuhn, 1991, p.60)

Asked how they know the participant replies,

I’m not certain, but it just seems pretty obvious from all other spheres of
life, people are so set in their ways. (ibid)

In cases like this a broad assumption about a phenomenon is used

to predict the cause of a specific instance: human beings are ‘crea-
tures of habit’ therefore reoffending will be the result of habit. As the
beginning of a process of hypothesis forming there is nothing wrong
with this – habit is a major component of human behaviour and it is
reasonable to assume that it plays a part in reoffending – but if this is
the only cause suggested, then the arguer runs the risk of presenting
a sweeping generalisation.
An example of a sweeping generalisation in ethics might be:
Killing people is always wrong.

In the hands of a pacifist this is defensible, but more usually in a

conversation it is the work of System 1 and needs qualifying. There
206

206 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

are, for instance, many examples of killing that can be argued to be

morally acceptable, such as combatants in battle, euthanasia, capital
punishment, and self-defence.
We refer to a generalisation as ‘sweeping’ when the context requires
us to know about a particular instance rather than the class was a
whole. My knowing that Nissans are reliable cars has limited value
when deciding whether or not to buy this particular ten-years-old-
with-three-previous-owners Nissan. Knowing that philosophy grad-
uates tend to be better than average critical thinkers does not mean
an employer should assume this about the philosophy graduate they
are about to interview. The very notion of a stereotype encompasses
this kind of problem. Generalisations are important and useful in all
sorts of ways, but in the wrong context they become oversimplifica-
tions that, in the case of social categories such as race and gender, can
be highly offensive.
‘Stereotype’ also refers to generalisations that are inaccurate in the
first place. As we saw above, defeasible generalisations run the risk
of being simply off the mark. It is one thing to be open to exceptions
to the rule (which is crucial to the correct presentation and dia-
logical handling of these kinds of generalisations), but quite another
for the rule to be wrong in the first place. To say that people from
the Home Counties of England (the areas surrounding London) are
posh Tories might be intended as a defeasible generalisation, but is in
fact implausible because it is based on a gross oversimplification of
the people who live in this region. Stereotype therefore means two
things: reducing an individual to a generalised truth about a category
to which they belong; and oversimplifying and thus distorting the
nature of the category in the first place. The first of these is a form
of sweeping generalisation, and the second is a hasty generalisation.

CRITICAL QUESTIONS

In light of the above discussions, the main critical questions that can
be applied to arguments involving generalisations are these:

FOR ABSOLUTE AND NON-ABSOLUTE GENERALISATIONS:

Q1: Is the generalisation made based on a large enough sample?

207

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 207

Q2: If so, is the generalisation made based on a representative

sample?

FOR DEFEASIBLE GENERALISATIONS:

Q3: Is the generalisation plausible?

FOR ALL GENERALISATIONS:

Q4: Is this a context in which applying a generalised truth is

appropriate?

GENERALISATIONS: PSYCHOLOGY AND RHETORIC

We are fond of generalisations; they render a complex, probabilistic,

and hard to predict world seem a simpler place. For philosopher and
psychologist William James we are ‘absolutists by instinct’8, and in The
Art of Rhetoric Aristotle (1991) observed how generalisations appeal
to prejudicial tendencies in audiences. We like to make quick leaps
from particulars to the general, and in persuasive communication he
identifies the use of maxims as especially effective for providing gen-
eralised legitimacy to the audience’s limited experiences. Maxims are
broad truths communicated in a single sentence, and spoken by the
right person they carry a form of moral wisdom. Among Aristotle’s
examples are: ‘He is no lover that not always loves’, and ‘Do not, thou
mortal, harbour deathless anger’. Although leaving us pretty cold,
these would have been familiar to Athenian audiences, and familiar-
ity is crucial for their effectiveness. In modern Anglophone cultures
we recognise expressions like ‘Actions speak louder than words’ and
‘All things in moderation’ (or better still, Oscar Wilde’s ‘All things in
moderation including moderation’).
Generalisations are, then, strangely satisfying, and they can also be
quite entertaining. Spoken word artist and comedian Henry Rollins
observes how they are ‘never right, always fun’9; a particularly good
line since it is itself a generalisation that is not accurate (since some
generalisations can be right, and are, as we’ve just seen, not neces-
sarily fun). Humour is full of generalisations, and no doubt this is
in part because we enjoy the fantasy of a simplified world and the
momentary release it affords. But journalism and many other forms of
208

208 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

communication are also full of generalisations, and where the genre

in question is not just entertainment then our liking for how they
represent the world can be a serious impediment to critical thinking.
As cognitive-behavioural therapy (CBT) demonstrates, gener-
alisations do not necessarily lose their appeal even when they work
against the individual. Mental disorders such as depression and anxi-
ety are caused and maintained by people’s inaccurate (or ‘inefficient’)
beliefs. Originator of CBT Aaron Beck identified ‘basic errors’ of
thought in patients such as overgeneralisations and black and white
thinking, and among the ‘irrational beliefs’ listed by another prime
mover in this field, Albert Ellis, are absolutes and generalisations like:
One should be loved by everyone for everything one does.
One should be thoroughly competent, intelligent and achieving in all
respects.
Because something once affected one’s life, it will indefinitely affect it.
Because I had bad luck once, I will always have it.10

CBT seeks out unrealistic and inflexible patterns of thought that

have become automatic responses to situations and guides to behav-
iour. In a therapeutic context the aim is to help people replace these
with healthier, largely more accurate, ways of thinking, and the rela-
tive success of this form of intervention is evidence for the broad
benefits of critical thinking.
Of particular interest to CBT are causal generalisations. Harmful
assumptions, such as bad luck always being with us, involve errone-
ous beliefs about the nature of luck. People suffering depression are
more likely than others to wrongly attribute negative events to fea-
tures internal to them (such as personality traits), and positive events
to external causes. It is easy to see how biased generalisations of this
kind will take a cumulative toll on an individual’s self-esteem.

GENERALISATIONS, DISPOSITIONS, AND CONSTRUCTIVE DIALOGUES

Towards protecting ourselves against the tendency to be seduced

by generalisations, much of the discussion in the equivalent section
to this under ‘Causal arguments and causal fallacies’ is pertinent. In
addition, however, overgeneralisations are a form of excess that we
need to moderate and so temperance is also very important. Like
Oscar Wilde (and Uncle Vance) I would suggest that a life without
209

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 209

moments of excess is unfulfilling for many of us, but there is a time

and a place, and in most respects an excess of generalisations, tempt-
ing though they are, is not amenable to the pursuit of truth.
Comments made in the causal arguments section about declaring
your position too soon and the premature closing down of dialogues are
also relevant to generalisations. Making absolutist statements can force us
into having to defend them even when we come to realise their prob-
lems (a single counterexample is enough to defeat them). Probabilistic
statements, on the other hand – especially vague or ‘soft’ ones containing
words like ‘most’, ‘many’, ‘the majority’, or ‘usually’ –allow for flexibility
in the positions we arrive at without appearing inconsistent.
A form of defence we can be pushed into as a result of an ill-
thought-through generalisation is the ‘no true Scotsman’ fallacy. Also
referred to as an ‘ad hoc rescue’, it involves changing the definition
of the class of things in question so as to avoid an absolute generali-
sation being defeated by a counterexample. The name comes from
philosopher Anthony Flew’s original example of a Scotsman reading
about a brutal crime committed by an Englishman in the paper and
proclaiming that no Scotsman would do such a thing. The next day
he reads about a worse crime committed by a man from Aberdeen
(in Scotland), but instead of acknowledging the inaccuracy of his
generalisation, he insists that this person is ‘no true Scotsman’. In an
ad hoc fashion – meaning that it is created in order to deal with a
previously unconsidered contingency rather than being part of the
original planning – he has changed the definition of a Scotsman from
a (presumed) geographical one to one that specifies certain traits or
behaviours (Flew, 1985, p.49). The desperation of such a manoeuvre
is pretty clear, but in technical terms the arguer has re-established
the validity of the argument by arbitrarily changing the definition of
what he is arguing about.
In an article about conspiracy theories, Matthew Dentith makes
the point that sometimes they are true, but have come to be defined
as something that is false. He is arguing against this ‘no true Scotsman’
attitude, and is concerned that real conspiracies will be missed as a
result:
Conspiracy theories sometimes turn out to be warranted, although many
deny this by saying: ‘Ah, but then it’s not really a conspiracy theory, is it?’ …
We have been told that conspiracy theories are bunk, and so we treat
them as such. 11
210

210 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

8.3 ARGUMENTS FROM CONSEQUENCES AND

SLIPPERY SLOPE ARGUMENTS
Causal arguments and generalisations are basic to many deliberations
because they help us to determine the consequences of our decisions.
Decisions can be argued for because of the positive consequences
they will bring about, or they can be argued against because of fore-
seen negative consequences. In both instances, judgements are often
based on relative positive and negative consequences: we should do
X because it will bring greater benefits than Y; or we should not do
P because it will bring worse consequences than Q.
Thus, the basic structure of an argument from consequences is this:

POSITIVE FORM:

P1: Deciding X will bring about consequence Y.

P2: Consequence Y is better than the consequences that will arise
from alternatives to X.
C: Therefore we should decide X.

NEGATIVE FORM:

P1: Deciding X will bring about consequence Y.

P2: Consequence Y is worse than the consequences that will arise
from alternatives to X.
C: Therefore we should not decide X.

The critical questions applicable to arguments from consequences

are these:

Q1: Is this a situation in which the consequences of a decision are

the appropriate standard of evaluation?
Q2: How sure are we that deciding X will actually bring
about Y?
Q3: Is consequence Y clearly good/bad in the way that is being
claimed?
Q4: If so, is it better/worse than consequences arising from alterna-
tive decisions?
211

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 211

Q5: If so, is the right degree of certainty established in order for us

to make a decision to encourage/allow/prevent consequence Y
from occurring?

Regarding the use of drones in attacks on terrorist suspects, the fol-

lowing negative argument from consequences is made:
If targeted drone strikes become legitimized in this context, the need
to try other means first to quell the threat may be diminished. The risk
becomes that military leaders will bypass nonlethal alternatives, such as
apprehending alleged terrorists and continued surveillance, and move
straight to extrajudicial killing as the standard way of dealing with the
perceived threat of terrorism.12

The context of the discussion is just war theory, and one of the crite-
ria for deciding if waging a war is just is whether it is a last resort. If
drone strikes are considered ‘actions short of war’, then their becom-
ing acceptable could corrupt the ‘avoiding violence’ impetus behind
this criterion.The negative argument from consequences concerning
this use of drones can satisfy Q1 and Q3, and there is possibly a case
for an affirmative answer to Q2. However, because there are a num-
ber of other arguments for and against drone strikes and their rela-
tionship to just war theory (e.g., greater precision leading to fewer
civilian casualties, but also less likelihood of surrender since the other
side do not feel that they are fighting a real enemy) then Q4 is com-
plicated to assess. We should be aware though that if, on balance, Q4
does not weigh against drone strikes, then greater certainty is needed
in response to Q2.

SLIPPERY SLOPE ARGUMENTS

A type of argument from negative consequences that has received

a lot of attention from scholars of informal logic is the slippery
slope argument. A number of authors and textbooks will refer to
this as the slippery slope fallacy (e.g., Johnson and Blair, 2006; Bowel
and Kemp, 2009; Lunsford and Ruszkiewicz, 2010), but as Douglas
Walton (2015) and others have argued, slippery slope arguments are
not necessarily weak.
212

212 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

They share the general structure of arguments from negative con-

sequences, but with these features:
• They claim that by bringing about an initial, relatively inconse-
quential (and possibly benign) occurrence, such as a policy change,
a series of hard-to-stop consequences will follow.
• This series is hard to stop for two reasons: we have little or no
control over its unfolding once it begins; and (in some instances
because of the incremental nature of the change) there is a sig-
nificant degree of indeterminacy about when this loss of control
will actually occur.
• Once underway this sequence will then eventually lead to a final
serious, or even catastrophic, negative consequence.
The tendency to treat slippery slope arguments as fallacies is due
to the combination of speculative premises and a dramatic conclu-
sion.They are, however, quite common in deliberative reasoning, and
perhaps most conspicuously in debates about drug legislation and
euthanasia. In both cases it is claimed that small changes to legislation
that might, on the face of it, be acceptable (such as legalising cannabis
or voluntary euthanasia) should not be enacted because this will lead
to a cascade of uncontrollable consequences, the end point of which
is highly unacceptable. In the case of drug laws, it is claimed that the
links between cannabis and harder drugs (links that are not there
with alcohol) will result in a greater number of heroin addicts. In the
case of euthanasia, incremental changes in social attitudes towards
death and the rights of the dying – including greater pressure per-
ceived by the elderly to agree to terminate their lives even though
it is not what they really want – will result in cases of involuntary
euthanasia and in public acceptance of non-voluntary euthanasia.
Communicating the seriousness of climate change is challeng-
ing for several reasons, and one of them is that it has the form of a
slippery slope argument. The IPCC’s Sixth Assessment Report on
climate change impacts (published in 2022) continues to make the
case for urgent ‘concerted anticipatory global action on adaptation
and mitigation’ to avoid the serious and wide-ranging consequences
caused by changes (observed and predicted) to Earth systems. As a
result of global temperature rises, its multiple sections report and
predict ‘cascading’ and ‘compounding’ impacts, ‘unavoidable increases
213

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 213

in multiple climate hazards’, and, especially relevant for our purposes,

‘irreversible losses’ of ecosystems and the increased risk of reach-
ing ‘tipping points’.13 The long-term consequences of the current,
inadequate, level of response remains too easy to ignore for most,
but predictions like these indicate not just a series of highly negative
outcomes, but one’s that we can’t undo and that will themselves lead
to increasingly terrible consequences.
The critical questions associated with slippery slope arguments
share the ones for arguments from negative consequences, except
that Q2 will need to be applied to every step in the sequence of con-
sequences. Also, to evaluate the true slipperiness of the hypothesised
slope, a sub-question must be added:

Q2a: Are there good reasons for thinking that a loss of control over
consequences will occur during the sequence of cause and
effect?

This could mean loss of control over purely physical events as in

the climate change example, but also loss of control over subsequent
actions of policymakers, or changes in public attitudes.
Applying this to the climate change argument: although some
commentators highlight the benefits of climate change (such as
Arctic shipping lanes opening up), most of us will accept that the
predicted consequences of anything above 1.5°C rise will be very
serious indeed. The argument thus satisfies Q3, and it also satisfies
Q1 and presumably Q4. In response to Q2 the findings of the IPCC
show that it is highly likely that not cutting emissions will lead to
catastrophic outcomes, and it is of course greatly important that the
report references the details of the research which confirms these
predictions. With respect to Q2a, as we have seen, tipping points are
involved (the nature of which is also explained in the details of the
research), and these are, by definition, irreversible. When they will
occur is uncertain, such that continuing to emit greenhouse gasses to
the extent that we are is running a significant risk. Considering the
devastating consequences brought about by these tipping points, the
cautionary principle seems the only logical one to apply (Q5).
In contrast, arguments against the inevitable proliferation of onshore
wind farms that will result from allowing some to be developed in
214

214 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

certain locations less obviously satisfies Q2a. More convincing was

the case FIFA general secretary Jérôme Valcke’s made against intro-
ducing goal line technology in football in 2010.
If we start with goal line technology, then any part of the game and pitch will
be a potential space where you could put in place technology to see if the ball
was in or out, and then you end up with video replays. The door is closed.14

The door was later opened, and we have ended up with video replays.
It is (so far) not for throw-ins, and since many agree that on balance
the introduction of VAR is a good thing, it seems not to have landed
us with a catastrophic consequence (Q3).

RELATIVES OF THE SLIPPERY SLOPE ARGUMENT

Care should be taken when labelling something a ‘slippery slope

argument’, as other argument forms and fallacies bear a resemblance
to it.These are important (or important to know about) in their own
right, but differ from slippery slopes in important respects.

Complex causal sequences

Arguments from negative consequences that involve chains of cau-
sation, but no loss of control are not slippery slope arguments. The
consequences of climate change without the tipping points would
be just such an argument. And we should note that the difference is
not trivial since policymakers and other deliberators need to know
whether and at what points an unwanted series of events can be
halted. To label a situation a slippery slope is to sound an alarm bell,
and although in some cases this might be the aim, to misapply the
label can lead to misplaced fear or panic.
The excessive waiting times for ambulances in parts of Britain in 2022
(and possibly beyond) was the result of a complex sequence of causes.
There was no shortage of ambulances, but they were being held up
at hospitals because of overcrowded A&E departments. Overcrowded
A&E departments are caused by an absence of beds in other wards to
move patients to, and this was the result of there being nowhere for
patients to go because of a shortage of care workers in communities.
It has been further suggested that the lack of care workers is caused
215

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 215

by Brexit, but even if this is true the situation is not irreversible. For
example, if we need to rely on non-British people to be care workers
then there is a system for granting visas to fill these vacancies.

Argument from precedent

An argument from precedent is one in which it is said that we should
not do X, because if we did, we would have to allow Y and Z (etc.) to
happen as well. The overturning of Roe v Wade in the US Supreme
Court in June 2022 is a recent example. Among the arguments against
this ruling is the concern that it will open the legal door to the repeal
of laws regarding contraception and same sex marriage. As Kamala
Harris put it, since the court’s reasoning is that the right to abortion is
not ‘deeply rooted in the nation’s history’, then the same can be said
of other more recently adopted rights that ‘we thought were settled’.15
Worries about precedent-setting are more obviously forms of
arguments from analogy (see Chapter 9) than slippery slope argu-
ments. First, as with this example, detractors will often see some-
thing inherently troubling in the initial step, whereas with slippery
slope arguments the consequences of the initial step are perceived
as unproblematic or positive. Second, the plausibility of subsequent
developments depends on the strength of the analogy rather than
reasoning around cause and effect.

The continuum fallacy

The continuum fallacy is related to the sorites paradox. Sorites means
‘heap’ in Greek, and the paradox identifies the contrast between ideas
that we confidently use in natural language, but which on analysis
turn out to be unquantifiable. We can readily distinguish between a
heap of sand and a handful of grains, but the process of adding sin-
gle grains of sand to the handful and trying to establish the point at
which it becomes a heap appears to be impossible. One grain is not
going to be the difference, but this means that no matter how many
single grains we add (or take away if we are talking about moving
from a heap to a non-heap) the non-heap can never become a heap.
Nevertheless, we do still want to maintain that there is a meaning-
ful difference between a non-heap and a heap, just as (sadly) there is
216

216 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

a difference between bald and non-bald (adding or subtracting sin-

gle hairs creates the same kind of problem). We make this distinction
through (1) acknowledging the inherent vagueness of these concepts,
and (2) rejecting the idea that vagueness makes concepts meaningless or
impractical. So, the fallacy is to conclude that because this incremental
process cannot define where the line is drawn, there is in fact no line
at all; no real difference between the concepts in question. There are
countless examples of this kind of vagueness, including what counts as
reasonable flexibility around a designated bedtime (‘If I can have five
more minutes, why not six?’), saplings and trees, or, more seriously, the
point at which alcohol consumption impairs our ability to drive.
The relevance of the sorites paradox to slippery slope arguments
should be clear. If used to argue against the justifiability of rules
then it is usually a fallacy: allowing five minutes flexibility around
bedtime does not mean that six is acceptable. Despite the vagueness,
agreement can be reached both on roughly where the line should
be drawn and, very importantly, that a line must be drawn. The drink
driving limit in Scotland is 50 milligrams of alcohol in every 100
millilitres of blood. No doubt we are no less safe at 55 milligrams,
but quite understandably the law cannot see it this way, and someone
caught will be banned whether they have 55mg or 155mg.
Stronger sorites-type slippery slope arguments are found in situa-
tions where we are policing ourselves, and weakness of the will, per-
haps caused by addictive or obsessive tendencies, predicts a loss of
control. The alcoholic should simply not touch alcohol, the gambling
addict not set foot in a casino or access online betting sites. In his
autobiography, Ozzy Osborne acknowledges this side of his personal-
ity; in one example he recounts his teenage self ’s accelerated and very
dangerous abuse of methylene chloride when operating an industrial
degreasing machine. One sniff a day rapidly became ‘every … five
minutes … I couldn’t stop myself ’. One day he was found passed out,
an ambulance was called, and for his own good he was fired.16
In circumstances like these the reasoning ‘just one (or one more)
won’t hurt’, is potentially catastrophic. Structurally similar, but with
less severe consequences, there are occasions where I stop myself
watching the short (20-minute) comedy shows Community or It’s
Always Sunny in Philadelphia before bed because my love of them is
such that I know I’ll struggle to keep it to ‘just the one’.
217

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 217

SLIPPERY SLOPE ARGUMENTS: PSYCHOLOGY AND RHETORIC

In different respects, slippery slope arguments and their relatives

can be seen as being too persuasive or not persuasive enough. Their
power comes from the severity of the end point and the fear it pro-
vokes, and from the anxiety caused by the loss of control that results
in this end point. (For more on the effects of fear and other emo-
tions in argumentation see Chapter 2 and the section on ad baculum
arguments in Chapter 6.) We should note though that a slippery
slope argument in support of the status quo, such as not changing
the laws concerning drugs or euthanasia, is far more likely to be suc-
cessful than one requiring quite radical changes. Despite decades of
overwhelming evidence for anthropogenic climate change and its
negative consequences, the rate of technological and social change
still works at a ‘business as usual’ pace. Under these circumstances the
complexity of the slippery slope very likely assists our deep-seated
desire to ignore the problem, or rationalise it away.17
Applied to rhetoric then, there appears to be a simple message: if
campaigning to keep things as they are, the fear and anxiety provoked
by slippery slope arguments could well work in your favour. But
if campaigning for change, their complexity can provide an excuse
for many to disengage. Also, since it is a quite a well-used and well-
known argument form in public debates, the simple appearance of
a slippery slope argument (including its relatives) could lead to an
automatic dismissal of the issue or your stance on it.

SLIPPERY SLOPE ARGUMENTS, DISPOSITIONS,

AND CONSTRUCTIVE DIALOGUES

Because of their complexity, slippery slope arguments are often presented

in a shortened (or ‘compressed’) form.We might hear, for example:
In theory, freedom of speech is a good thing, but allowing any form of
extremist religious or political views – whether or not they advocate
­violence – will eventually lead to more of our teenagers and young adults
joining organisations like Oath Keepers or IS.

This is a controversial and emotive topic, and you can imagine this
argument against the preaching of extremist beliefs having some sup-
port in countries facing anti-government or terrorist threats. Before
218

218 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

assessing it though we would need to hear more from its source to

gain a clear picture of what they think would happen to these young
people if we continue to tolerate non-violent extremist views. Thus,
the skills of constructive dialogue and argument reconstruction are
required to fill in the various steps. Only once this effort is made can
the argument be properly evaluated.
Many slippery slope arguments do not stand up to scrutiny, but some
do, and because they deal in severe negative consequences then we
need to be careful about what we dismiss. Notwithstanding the pre-
ciousness of our time and our developing ability to read signs of unreli-
ability in sources, we need to be suitably open-minded, persistent in
our focus, and inquisitive in the presence of proposed slippery slopes.
Although we are unlikely to have direct access to the evidence for the
steps in question, we can at least adopt a provisional stance on their
plausibility which can then shape our subsequent attitude to the issue.

8.4 SUMMARY
This chapter has highlighted the links between consequences, causes,
and generalisations.We need to gauge and evaluate, as best we can, the
consequences of different decisions.To help us do this we must avoid
mistakes in (1) reasoning about causation, and (2) how we generalise
from what is currently known to future events and to what has not
been directly investigated. These errors are in part the consequence
of psychological frailties – such as a liking for absolutes, or proneness
to superstition and fear appeals – that are now quite well understood.
Improvements in our critical thinking can be developed, not just
from learning about relevant argument forms and fallacies, but from a
reflective appreciation of these frailties and by continuing to develop
appropriate dispositions and constructive dialogical practices.

EXERCISES
1. Conspiracy theories are an excellent subject matter to apply
critical thinking to. One exercise is to generate a set of critical
questions designed to assess their plausibility. In conjunction
with this it would be useful to (a) read some philosophical
analysis of conspiracy theories (such as Cassam, 2019), and (b)
219

CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES 219

think about them in terms of the cognitive biases and argu-

ment forms and fallacies discussed in this book – a remarkable
number of which turn out to be relevant.
2. The background to Martin Luther King, Jr.’s Letter from
Birmingham Jail has been discussed in Chapters 1 and 4, and
here I would like to direct you to a section of it that contains
many of the argument forms discussed in this chapter.

I won’t re-produce it here, but it is easily locatable online. The rel-

evant part starts in the 23rd paragraph (about two-thirds of the way
through) with the line ‘If this philosophy [of nonviolent direct action]
had not emerged …’ and continues into the next paragraph up to
‘channelled through the creative outlet of nonviolent direct action’.
Attempt to identify and reconstruct some of the argument you find
here and, using the appropriate critical questions, assess its strength as
best you can. Taking into consideration King, Jr.’s audiences (white
Southern clergy, politicians, and the American public in general), you
might also want to evaluate its rhetorical power.

FURTHER READING
CAUSAL ARGUMENTS

Deanna Kuhn’s book The Skills of Argument (1991) is a very interest-

ing investigation into how people formulate causal arguments and
associated generalisations. In it she theorises four stages of sophistica-
tion in causal reasoning.
Hank Davis’ Caveman Logic (2009) is primarily an argument for
the dangers of heuristic reasoning, but has a particular focus on causal
reasoning errors.

GENERALISATIONS

A very readable and insightful discussion of generalisations (includ-

ing prejudice and stereotyping) can be found in Michael Scriven’s
excellent book Reasoning (1976, pp.196–210).
Within the context of critical thinking books, one of the most
comprehensive treatments of generalisations and their relationship
to scientific method is Chapter10 of Robert Ennis’s book Critical
Thinking (1996a).
220

220 CAUSAL ARGUMENTS, GENERALISATIONS, ARGUMENTS FROM CONSEQUENCES

ARGUMENTS FROM CONSEQUENCES AND

SLIPPERY SLOPE ARGUMENTS

For a good recent overview of slippery slope arguments see Anneli

Jefferson (2014) ‘Slippery Slope Arguments’.And for a more advanced
history and analysis see Douglas Walton (2015) ‘The Basic Slippery
Slope Argument,.

NOTES
1 Fortean Times, Issue 305, Sept. 2013.
2 Bowes, L. et al. (2015) Risk of depression and self-harm in teenagers iden-
tifying with goth subculture: a longitudinal cohort study. Available at: www.
thelancet.com/psychiatry, Vol.2.
3 Season 7, Ep.18, July 2020.
4 McGrath, M. (2015) Global Warming Increases Food Shocks Threat [Article].
Available at: www.bbc.co.uk/news/science-environment-33910552, and Bailey,
R. et al. (2015) Extreme Weather and Resilience of the Global Food System
[Report]. Available at: www.cabdirect.org/globalhealth/abstract/20163135764
5 Skinner, B.F. (1948) ‘Superstition’ in the pigeon. Journal of Experimental
Psychology, 38, 168–72.
6 E.g., Brian, K. (1999) Of conspiracy theories. Journal of Philosophy, 96(1),
109–26.
7 See Goldacre, B. (2008) Bad Science. (London: Fourth Estate), pp. 97–9 and
165–70.
8 The Will to Believe in Selected Papers on Philosophy (New York: Dutton,1967),
p.110.
9 Talk is Cheap Volume 3 (Audio recording, 2004)
10 Mahoney, M. and Freeman, A. (eds), Cognition and Psychotherapy (New York:
Plenum, 1985).
11 Fortean Times, February 2015, p.39. To avoid this error Quassim Cassam
(2019) distinguishes between Conspiracy Theories (with caps) and con-
spiracy theories. The former are without any serious credibility and their
propagation linked to political agendas.
12 Brunstetter, D., & Braun, M. (2011). The implications of drones on the just
war tradition. Ethics & International Affairs, 25(3), 337–358.
13 IPCC (2022) www.ipcc.ch/report/ar6/wg2/
14 See, e.g., www.irishexaminer.com/sport/soccer/arid-30448950.html
15 www.theguardian.com/us-news/video/2022/jun/24/this-is-not-over-
kamala-harris-speaks-out-against-overturning-of-roe-v-wade-video
16 I am Ozzy (Sphere, 2009), pp.26–7.
17 See George Marshall (2014) Don’t Even Think About It. (Bloomsbury).
221

ARGUMENTS FROM ANALOGY

Argument by analogy is often the most powerful and compelling type of

argument we can use. (Michael Scriven, 1976, p.210)
Using an analogy to win an argument is like putting a stethoscope on
a cat and claiming that it’s a doctor. (Sacha T. Burnstorm, pers. comm.)

An analogy is an attempt to illuminate, explain, or make an argu-

ment about a certain thing (or situation, event, etc.) by comparing
it with something that shares relevant features with that thing (or
situation, event, etc.). Analogies are extremely common in commu-
nication, sometimes as forms of explanation, but also as forms of
argument.
At a critical point in the Covid pandemic in 2021 composer
Andrew Lloyd Webber compared refusing the Covid vaccine to
drink driving. In both cases he claimed that people know their deci-
sion is endangering the lives of others, even though it is not their
intention. He was making a moral argument rather than saying that
refusing the vaccination should be made illegal, but if the comparison
is convincing then the conclusion reached has serious implications:
if we agree that drink driving is wrong because it is selfish, then a
similar degree of disapproval could be directed at anti-vaxxers.1

DOI: 10.4324/9781003247944-10
222

222 ARGUMENTS FROM ANALOGY

A letter to the UK’s Sun newspaper (29.9.10) uses an analogy

to argue against slipping standards in the long-running soap opera
Coronation Street (or ‘Corrie’):
The unpleasant sight of Kevin and Molly naked in bed before the 9pm
watershed shows how Corrie’s standards sink lower all the time. Clark
Gable and Vivien Leigh sustained over three hours of passion in Gone with
the Wind without once getting their kit off. Corrie should learn from this.
The argument is that the programme should not need to use explicit
sex scenes to facilitate romantic plotlines, and the reason it gives is that
Gone with the Wind – a prime example of the genre – did not need to.
Metaphors are very similar to analogies, but tend to be briefer
comparisons which have a more literary purpose. They illuminate
some aspect of the world in a way that seizes the audience’s atten-
tion and stirs the imagination. As we will see, that analogies can also
have this effect is relevant to understanding their persuasive potential,
but primarily they should be understood as having an informational
rather than poetic function.When used as explanations or arguments
the relationship between the items being compared needs to stand
up to analytical scrutiny. If not then the explanation will mislead, or
the argument will be weak.

STRUCTURE AND CRITICAL QUESTIONS

An argument from analogy is one which claims that because some-

thing is true about Y, so it is likely to be true of X, because X, in
relevant respects, is comparable to Y.
The most basic way of representing the structure of an argument
from analogy is this:

P1: X is comparable to Y.
P2: Z is true of Y.
C: Therefore Z is also true of X.

A more detailed version is this:

P1: Y has features a, b, c, etc.

P2: X is comparable to Y with respect to features a, b, c, etc.
223

ARGUMENTS FROM ANALOGY 223

P3: Z is true of Y.
P4: Z is linked with features a, b, c, etc.
C: Therefore Z is also true of X.

Set in the 1970s, The Last King of Scotland (Keven MacDonald, 2007)
is a film about a young Scottish doctor (Nicholas) who, when volun-
teering in Uganda, ends up becoming Ide Amin’s personal physician.
Nicolas initially rejects Amin’s offer because he has come to Uganda
to help ordinary people, but Amin then argues that by helping the
president he will be helping the people by assisting the development
of a national health service. Nicholas again, politely, refuses. Amin’s
second argument is an analogy that can be reconstructed like this:

P1: I had no personal desire to become president, but the people

wanted it.
P2: You have no desire to become my physician, but the people
will benefit from it.
P3: Through a sense of duty I became president.
P4: Doing one’s duty is relevant to cases where taking on an
important position, whatever one’s personal feelings, is some-
thing that will benefit the people.
C: Duty dictates that you should serve as my personal physician.

The whole of this sequence (between approx. 24 and 36 minutes into

the film) is worth watching and analysing as an excellent example of
persuasive communication, and this particular dialogue serves as an
example of an argument from analogy.
Terminology tends to differ slightly between different books and
theorists, but for our purposes the issue or entity that the conclu-
sion refers to (that the argument is essentially about) is known as
the ‘primary case’ (in this instance whether Nicholas should become
Amin’s physician); what it is being compared with is the ‘analogue
case’ (Amin’s decision to become president), and the relevant feature
of the analogue case that the argument wants to claim is also true of
the primary case is the ‘target feature’ (the decision being based on
duty rather than personal preference).2
Just how arguments from analogy should be categorised has
been the subject of a lot of debate among philosophers and other
224

224 ARGUMENTS FROM ANALOGY

academics in recent decades. One important and generally accepted

distinction (first identified by Aristotle (1991, Ch.2.20)) is between
these two types:
• Real-life (or historical) analogies, where the comparison is with
historical events or other accepted truths that the audience
is familiar with. These are commonly referred to as inductive
analogies.
• Hypothetical analogies, where the comparison is with an
invented situation that draws out, in a clear (and often vivid) way,
the target feature(s) of the primary case.
Extended inductive analogies can sometimes form the entire
structure of lengthy books and documentaries. For example, the
Mark Achbar, Jennifer Abbott, and Joel Bakan documentary The
Corporation argues that the problem with large, publicly traded
companies is that by law they have no option but to behave like
psychopaths. The symptoms and consequences of psychopathy are
systematically analysed and applied to corporate behaviour, and the
conclusion is reached that just as psychopathic behaviour is socially
discouraged and legally neutered, so too should the modern form of
the corporation. As the title suggests, Andrew Jennings’ 2014 book
Omertà! Sepp Blatter’s FIFA Organised Crime Family makes the case
that the corruption in football’s governing body is best understood in
terms of the Mafia, and thus the appropriate alarm, moral disgust, and
other responses we have towards organised crime should rightfully
also be directed towards Blatter’s FIFA.
Allegories are fictional stories with clear symbolic meaning, some
drawing comparisons with specific real-world events or situations.
Often they are indirectly making an argument, or series of argu-
ments. A famous twentieth-century example is Arthur Miller’s play
The Crucible where the Salem witch trials represent McCarthyism.
And a good recent one is the 2021 film Don’t Look Up in which
motivated ignorance about the Earth’s imminent collision with a
comet stands in for political and media responses to the climate crisis.
The history of philosophy is full of analogies used as arguments
and explanations. In one of the most famous contemporary exam-
ples, Judith Jarvis Thomson argued that legally or morally denying
the right to an abortion in the case of rape is akin to a person being
225

ARGUMENTS FROM ANALOGY 225

surgically attached to a famous violinist against their will for nine

months because their blood and kidneys are the only thing that can
keep him alive. If we agree that this violates the person’s rights, so
the argument goes, we should also accept the right to an abortion.3
Another hypothetical analogy, well known in philosophy and the-
ology, is William Paley’s ‘watchmaker’ argument for intelligent design
(published in1800). In it he states that on encountering a watch for
the first time, its complexity and evident function (in comparison
to a stone) would give us every reason for believing that it has been
designed and constructed by an intelligent being for a particular pur-
pose. That being the case, he draws an analogy between what we
believe about the origins of the watch and what we should believe
are the origins of nature as a whole:
every indication of contrivance, every manifestation of design, which
existed in the watch, exists in the works of nature; with the difference, on
the side of nature, of being greater and more, and that in a degree which
exceeds all computation. (Paley, [1800] 2009, p.19)
Living things also function mechanically, and in a way that is ‘accom-
modated to their end’. In other words, body parts and so on are ide-
ally designed to fulfil their purpose, just as human-made inventions
are. The most reasonable belief, he concludes, is that nature is the
product of intelligent design – the work of God.
The critical questions we can apply to arguments from analogy
are these:

Q1: Is what is said of Y actually true (or plausible)?

Q2: Are there relevant similarities between X and Y?
Q3: Are there dissimilarities between X and Y that undermine the
similarities?
Q4: Can convincing counteranalogies be found (or in the case of
hypothetical analogies, be hypothesised?)
Q5: Is the conclusion drawn from the comparison of the appro-
priate strength?

In his Dialogues Concerning Natural Religion (published in 1779, several

decades before Paley’s book) David Hume employed these questions
to good effect against design arguments. We can grant that someone
226

226 ARGUMENTS FROM ANALOGY

coming across a watch for the first time (in contrast to a stone)
would rightly believe that it is not a random occurrence (Q1). (Note
that with hypothetical analogies we can challenge the belief that is
assumed in the analogue – so in Thomson’s analogy we might want
to claim that we do not have the right to disconnect our self from the
violinist as she thinks we would – but in real-life analogies the facts
of the analogue can be challenged as well as our supposed attitude
towards those facts.) Arguably there are relevant similarities between
the watch and the workings of nature (Q2), but Q3 begins to reveal
weaknesses in the argument. Despite their complexity and function-
ality, living organisms and their component parts are noticeably dif-
ferent from man-made artefacts, and in many respects (irregularities,
shapes, textures, and so on) share more with the contrasting stone
than with watches and other devices. Also, a simple dissimilarity is
that we can observe watches and other machines being designed and
built, but have no such empirical evidence for the origins of nature.
Q4 is problematic for design arguments since an alternative anal-
ogy for the creation of nature is the creation of living organisms at
a result of procreation. A designer is not an observable or necessary
feature of the continuation of animal and vegetable species, so why
should it be for the world as a whole? In Hume’s words,
The world plainly resembles more an animal or a vegetable, than it does
a watch or knitting-loom. Its cause, therefore, it is more probable, resem-
bles the cause of the former. The cause of the former is generation or veg-
etation. The cause, therefore, of the world, we may infer to be something
similar analogous to generation or vegetation. (Hume, [1779] 1990, p.87)

Of course, Hume’s argument would have been greatly strengthened

had he been writing after Darwin’s Origin of Species had been pub-
lished; in the Dialogues he can only assert that we have no conclusive
evidence either way. Without this evidence, to make a firm assertion
on the basis of either the machine or generation analogy is unwar-
ranted (an example of a circular argument, see Chapter 10). A fur-
ther argument of Hume’s against the design argument, therefore, is a
variation of Q5. Applying this to Paley we can say that he concludes
too much from his analogy.
A small piece of advice before ending this section: prior to apply-
ing these questions, it is wise to ask ourselves whether the analogy is
227

ARGUMENTS FROM ANALOGY 227

intended to serve as an argument, rather than to explain or illumi-

nate the primary case. This is not always easy to discern, and some
interpretive work can be required. In her recent book parodying a
manual for how corporations (in tobacco, pharmaceuticals, oil, etc.)
should go about denying scientific discoveries that threaten their
profits, Jennifer Jacquet comes up with a truly insightful analogy. In
the book’s appendix she includes a letter to her editor explaining
her findings, and in it she points out that viewing denial strategies –
challenging the problem, challenging the causation, challenging the
messenger, and challenging the policy – as a series of tricks does not
do justice to what they have been achieving. Instead, she likens the
approach to a casino, an environment that can make people feel at
ease and forget that they cannot possibly trust the people who are
running it to have their interests, or the truth, in mind. This serves
as an aid to explaining how corporate scientific denial operates, but
it might also be an argument for how to combat it. Just as there are
ways to avoid the illusory comfort casinos create, so there are simi-
lar ways to avoid the spells cast by corporate PR.4 If progress seems
likely to be gained from interpreting an analogy as an argument it is
worth pursuing. At the same time, however, we must be honest about
our uncertainties regarding the author’s intention, and of course
apply the principle of charity (see Chapter 4).

ARGUMENTS FROM ANALOGY: PSYCHOLOGY AND RHETORIC

The success of many analogies is dependent on the audience’s

familiarity with them; what people know, and perhaps have strong
feelings about. The familiar case guides the understanding –­
provides a way into the issue in question – and so presenting an
analogy can be seen as a form of framing (see Chapter 1). Analogies
thus have the potential to be highly persuasive. In a benign sense
they are persuasive precisely because they motivate us, through the
familiarity of the analogue or the inventiveness and narrative fea-
tures of the hypothetical scenario, to engage more fully with the
topic. In this respect analogies can be seen as having an educational
function, but less benignly they can also be distracting, play to our
prejudices, and ultimately lead us to accept conclusions that we
should not accept.
228

228 ARGUMENTS FROM ANALOGY

The clever persuader will employ analogies and other forms of

figurative language that have emotional resonance for the audience
in question. When making an argument from analogy our choice
of comparison can be partly determined by this consideration. For
example, in An Inconvenient Truth, Al Gore did not have to choose
Hurricane Katrina (and later in the film, 9/11) to make his point
about preparedness and responses to global warming, but he knew
that these appropriately memorable and moving events would help
prompt a US audience (in 2006 at least) to engage with his message.
When comedian and Catholic Frank Skinner responds to the chal-
lenge that religion is nonsense from his atheist friend (David Baddiel)
he chooses football – a game they both love – to make his point.
Football is central to many lives, yet seen from an outsider’s perspec-
tive this is not only unfathomable but also ‘promotes violence, idola-
try and tribalism’. Skinner’s point (not necessarily a very strong one)
is that coolly evaluating these commitments misses something vital
about how they can make people feel.5
Even where figurative language is not directly used as part of an
argument, its presence can still influence audiences. Aristotle (1991)
and modern research (see Sopory and Dillard, 2002, for a review)
provide us with the following criteria for evaluating the persuasive
power of analogies and metaphors:

1. They should be unfamiliar, and yet lead to immediate recognition.

Used in non-fiction prose, the appeal of novel metaphors
needs to be balanced with their clarity of meaning. They
should, in Aristotle’s words, ‘name things without name,
which on being spoken immediately reveal their affinity’
(1991, p.220). This is quite an art; effectively knowing what
is on the tip of your audience’s tongue, and then finding
the right words or image to capture it. When Marx said in
the Communist Manifesto, ‘You have nothing to lose but your
chains’ there was little doubt about what he meant, and the
word ‘chains’ serves to dramatically crystalise the ‘affinity’
between the conditions of the working classes and slavery.
2. They should be situated so as to maximise message comprehension
It is important to get the positioning as well as the content of
metaphors right. Using a single metaphor rather than multiple
ones and employing the metaphor early in the message can help
229

ARGUMENTS FROM ANALOGY 229

to provide a clear frame for engaging with the overall argu-

ment. This process is explained by the ‘superior organisation’
theory which sees metaphors as facilitating our understand-
ing and memorising of the arguments presented by providing
a ‘structure’ for linking them together that is more effective
than literal language. In other words, well-situated metaphors
allow us to make better overall sense of what we are hearing
or reading (Sopory and Dillard, 2002, p.387).

If this potential for enabling audiences to structure the arguments

they are hearing is to be actualised, then the metaphor must be care-
fully selected. As well as being aesthetically pleasing and novel (but
recognisable), it should be consistent with the message and rich
enough to illuminate as many of the arguments within it as possible.
A number of these criteria, including using a single metaphor,
are exemplified by two well-received political speeches from the
last decade, the topics of which remain highly relevant. In President
Obama’s moving and sympathetic address in Newtown, Connecticut,
shortly after the Sandy Hook school shooting in December 2012, the
only metaphor he used referred to the anguish all parents face when
confronted with the impossibility of protecting their children from
all eventualities:
someone once described the joy and anxiety of parenthood as the equiva-
lent of having your heart outside of your body all the time, walking around.

This is a vivid representation of vulnerability; something we care

about more than anything is continually exposed to circumstances
beyond our control. It would lead to immediate recognition by any
parent, and serves to structure the central argument of the speech –
how the nation must do more to provide the protection that parents
cannot (alluding to, among other things, gun laws).
Ex-UK Prime Minister Gordon Brown’s fiery pro-union speech
in Glasgow, just prior to the Scottish independence referendum in
September 2014, also employs a single metaphor. He referred to the
risks of an independent Scotland in terms of an ‘economic minefield’,
and an ‘economic trapdoor down which we go and from which we
might never escape’. These images chimed with the fear appeal that
was central to the ultimately successful ‘No’ campaign, and which
was a central theme of this address. 6
230

230 ARGUMENTS FROM ANALOGY

From these examples we can see that the lone, carefully selected
metaphor helps keep messages coherent and memorable. Further use
of figurative language would risk creating linguistic clutter that would
interfere with this aim. In both speeches, however, the metaphor
appears closer to the middle of the speech than the beginning, and in
both cases the reason is that unrelated content was (for good reasons)
positioned prior to this. In the Obama case the speech naturally begins
with condolence rather than arguments, and with Brown other argu-
ments (concerning, for example, national (British) pride) were more
appropriate as openers. If a metaphor functions as a joker, it makes
sense that Brown would choose to play his on the economic theme.

ARGUMENTS FROM ANALOGY: DISPOSITIONS

AND CONSTRUCTIVE DIALOGUES

If there is a disposition with particular significance for arguments

from analogy, then it is having a good imagination. Hughes, Lavery,
and Doran (2010, p. 213) describe them as ‘probably the most crea-
tive form of reasoning’, and Christopher Tindale asks us to ‘note how
much this argument scheme requires us both to delve into the con-
text and to use our imaginations’ (2007, p.198).
Imagination is important both for formulating appropriate analo-
gies for our own arguments and for formulating counteranalogies
when questioning other people’s arguments (see Q3). This imagina-
tive aspect makes arguments from analogy unusual. It contributes
to their persuasive potential, but it is also part of what makes them
difficult (and risky) to generate. In particular, it is hard to conjure
convincing analogies in the heat of the moment. Producing counter-
analogies is, however, less tricky since we can usually follow the lead
of the other arguer’s analogy. Someone attempting to justify the Iraq
war, for example, might draw an analogy with the successful (but also,
technically, illegal) humanitarian intervention in Kosovo in 1999. But
once faced with this form of argument it is relatively easy to find
counteranalogies in other unsuccessful and less well-motivated wars
such as Vietnam (wasteful, futile, a national embarrassment to the US)
and the Crusades (self-righteous, militarist and imperialist folly).
As the design argument example demonstrates, arguments from anal-
ogy tend not to be particularly strong. Where hypothetical analogies
231

ARGUMENTS FROM ANALOGY 231

have creative qualities, though, they are effective for generating dis-
cussion and catalysing stronger arguments and deeper understanding.
There is a cognitive reason for this, concerning stimulation of the
imagination, and a motivational reason as well. Several literary phi-
losophers, such as Søren Kierkegaard, Jean-Paul Sartre, Iris Murdoch,
and Martha Nussbaum, have written about the importance of stories
(including allegories) for depersonalising topics; their aesthetic distance
providing an engaging focus whilst helping to avoid some of the emo-
tional distortion that political and ethical debates can generate and that
frequently impede our real-life deliberations. Since the story is not ours,
we do not find ourselves caught up in the ‘vulgar heat’ of our personal
jealousies or angers or in the sometimes blinding violence of our loves.
(Nussbaum, 1990, p.48)

Hypothetical analogies can function in a similar way. Bruce Waller says

that Judith Jarvis Thomson’s ‘famous violinist’ analogy (see above), is
effective because it starts with a case sufficiently different from our everyday
experiences that our preconceptions do not distort our view; thus we can think
about what principles we hold without the heat and passion and entrenched
doctrines that swirl around the question of abortion. (2001, p. 215)

A lot of this section has emphasised the role of familiarity in effec-

tive analogies, but here Waller makes a good case for the important
function of unfamiliarity in many hypothetical analogies. Rather than
insight into what an audience knows and has strong feelings about,
the imagination that goes into the production of these analogies is an
example of philosophers (and other academics) at their most creative.

SUMMARY

Arguments from analogy are commonly employed and vary greatly

in their quality. At their worst (and most dangerous) their superficial
appeal misleads and contributes to an audience’s hasty acceptance of
an unwarranted conclusion, but at their best they can encourage us
to think hard about an issue and form the beginning of a productive
discussion. They are good for holding people’s attention, especially
when the analogue case is familiar and matters, or is at least inter-
esting. Inductive analogies are based on real events and hypotheti-
cal analogies on imagined ones. In both cases a well-made analogy
232

232 ARGUMENTS FROM ANALOGY

draws out relevant (‘target’) features of the primary case that support
the conclusion. A strong argument from analogy impresses on us the
plausibility that since something is true of the analogue case it will
also be true of the primary case. A weak analogy should leave us
questioning the relevance of the connection between the two, and
searching for counteranalogies that further expose this flaw.

EXERCISES
1. There are quite a few examples of brief and extended analogies
in this chapter (such as Covid vaccine refusal being equated
to drink driving; the scene from The Last King of Scotland; the
documentary The Corporation; Thomson’s ‘famous violinist’;
the film Don’t Look Up; and Frank Skinner’s religion-football
comparison) which have been explained or cited, but not
analysed. This you could do by applying the critical questions
and the criteria for the persuasiveness of figurative language in
the ‘Psychology and rhetoric’ section. It is worth looking out
for your own examples as well – arguments from analogy are
used very frequently indeed.
2. As mentioned towards the end of the chapter, generating and
countering arguments from analogy is one of the most imagi-
native aspects of critical thinking. There are various activities
you can do around this, and one example is for a contentious
topic that interests you (or one agreed on by the class or set by
the teacher), devise an argument from analogy in support of
your position. To add realism (and perhaps make the task a bit
easier), determine in advance the type of person or audience
you are trying to convince with your argument.

FURTHER READING
Three sources on arguments from analogy that will help develop
your understanding of the subject and of some of the disagreements
within it are these:
• Bruce Waller (2013) ‘Classifying and Analysing Analogies’.
• Christopher Tindale (2007), Chapter 10.
• Trudy Govier (2013), Chapter 11.
233

ARGUMENTS FROM ANALOGY 233

NOTES
1 www.bbc.co.uk/news/av/uk-57146619
2 The choice of terminology is closest to Trudy Govier’s (2013), but ‘target
feature’ is Hughes, Lavery, and Doran’s phrase (2010), p.215.
3 Thomson, J.J. (1971) A Defence of Abortion, Philosophy and Public Affairs, 1 (1),
47–66.
4 The Play Book (Allen Lane, 2022), p.172.
5 On the Road (Arrow Books, 2009), pp.42–3.
6 Both of these speeches are easily located for viewing online.
234

10

FURTHER FALLACIES

Building new carparks is something you’re either for or against. For me

it’s a grey area. (Sacha T. Burnstorm, pers. comm.)

The previous three chapters have been about argument forms, and
their associated fallacies, that warrant extended analysis and explora-
tion. The focus of this chapter is a selection of fallacies and rhetori-
cal techniques that could not be located among these analyses, but
which are also important to know about. These are affirming the
consequent (and denying the antecedent); circular arguments;
false dilemmas; the perfectionist fallacy; and red herrings and
equivocations.

10.1 AFFIRMING THE CONSEQUENT (AND

DENYING THE ANTECEDENT)
There is a type of argument – sometimes known by the Latin modus
ponens – that has this form:

P1: If X, then Y.
P2: X.
C: Therefore Y.
DOI: 10.4324/9781003247944-11
235

FURTHER FALLACIES 235

So, for example:

P1: If a celebrity is responsive to her fans and treats them with

respect, then she will remain popular.
P2: Taylor Swift is responsive to her fans and treats them with respect.
C: Therefore Taylor Swift will remain popular.

This is a form of valid deductive argument (see Chapter 5) that

includes what is known as a ‘conditional statement’ or ‘conditional
premise’ – if X is the case, then Y will also be the case.
The logic of this argument is easy to follow, but we are also quite
easily led to accept as valid a non-deductively valid version of it that
goes like this:

P1: If X, then Y.
P2: Y.
C: Therefore X.

So,

P1: If a celebrity is responsive to her fans and treats them with

respect, then she will remain popular.
P2: Taylor Swift will remain popular.
C: Therefore Taylor Swift is responsive to her fans and treats them
with respect.

In this version the information provided in P1 and P2 cannot ensure

the truth of the conclusion because there will be other reasons for
a celebrity remaining popular other than being responsive to their
fans and treating them with respect. The first clause in a conditional
statement is known as the ‘antecedent’, and the second clause is the
‘consequent’. In the modus ponens argument form the second premise
should affirm the antecedent, so the fallacious version is known as
affirming the consequent.
In a radio interview in 2015 Carlo Rovelli, discussing his book
about science theories, said (with some slight paraphrasing),
Science is beautiful … The theory of relativity is a work of art. This is
because, like Shakespeare or the Sistine Chapel, it makes us emotional.1
236

236 FURTHER FALLACIES

This appears to be an example of affirming the consequent:

P1: If something is a work of art, then it makes us emotional.

P2: The theory of relativity makes us emotional.
C: Therefore the theory of relativity is a work of art.

Many things can make us emotional in the way that Rovelli seems to
mean – such as walking in a mountainous landscape, or contemplat-
ing the night sky, or watching our team win an important match –
but these are not works of art.
A similar argument to modus ponens is modus tollens, which has this
structure:

P1: If X, then Y.
P2: Not Y.
C: Therefore not X.

If X is the case, then Y must also be the case; therefore if Y is not the
case then X cannot be the case:

P1: If you like all forms of music, then you must like jazz scat
singing.
P2: You don’t like jazz scat singing.
C: Therefore you don’t like all forms of music.

The fallacious version of modus tollens is known as denying the

antecedent:

P1: If X, then Y.
P2: Not X.
C: Therefore not Y.

Applying the same example:

P1: If you like all forms of music, then you must like jazz scat
singing.
P2: You don’t like all forms of music.
C: Therefore you don’t like jazz scat singing.
237

FURTHER FALLACIES 237

As unlikely as it seems, someone not claiming to like all forms

of music could still like someone singing improvised impressions of
jazz instruments (i.e., scat). A modus tollens argument should deny the
consequent rather than deny the antecedent.
This fallacy appears to be at work in the quite frequent assertion
that because famous environmentalists were invariably brought up
close to nature, then that is how we can assure everyone grows up to
be environmentally conscious:

P1: If you grew up close to nature, then you are environmentally

conscious.
P2: Person Z is not environmentally conscious.
C: Therefore person Z did not grow up close to nature.

This allows room for the possibility of being environmentally con-

scious even though you experienced an entirely urban upbringing;
all it says is that being close to nature is a sufficient condition, not a
necessary one. The fallacious version mistakenly necessitates close-
ness to nature:

P1: If you grew up close to nature, then you are environmentally

conscious.
P2: Person Z did not grow up close to nature.
C: Therefore Person Z is not environmentally conscious.

10.2 CIRCULAR ARGUMENTS

A circular argument (otherwise known as ‘begging the ques-
tion’) is when prior acceptance of the conclusion of an argument is
required in order to establish one or more of the premises. An exam-
ple is expressed in this dialogue:

Jesse: I have psychic powers.

Simone: Really? How do you know?
Jesse: My psychic aunt told me, and only a psychic can tell if
you’re psychic.
Simone: How do you know your aunt’s really a psychic?
Jesse: Well, because I’m psychic of course.
238

238 FURTHER FALLACIES

Jesse is trying to argue that he is psychic on the basis of his psychic

aunt’s insight, but he can only know that his aunt is psychic on the
assumption of his own psychic powers. The trouble is his psychic
powers can only be established if the aunt is psychic, and so round it
goes. In a circular argument no grounds for accepting a premise are
offered beyond what is stated in the conclusion of the argument – a
conclusion that is, by its nature, reliant upon that very premise for
establishing its own truth.
We tend to recognise circular arguments when we encoun-
ter them, but articulating them can sometimes be difficult. Setting
them out in a formal structure is slightly tricky as well, but Bowell
and Kemp (2015) achieve this by joining two separate arguments
together like this:

P1: I have psychic powers.

P2: Only a psychic can tell if someone else is psychic, and I know
my aunt is psychic.
C1: My aunt is psychic.
P3: My aunt says I have psychic powers, and only a psychic can tell
if someone else is psychic.
C2: Therefore I have psychic powers.

Circular arguments are fairly common, and if you get the scent of
one it is worth pursuing it even if it takes some trial and error to
pin it down. If someone tries to prove God’s existence using the
Bible, for example, persistence can pay off because it is probably
circular:

Simone: How can we rely on the word of the Bible?

Jesse: Because the Bible is the word of God.
Simone: But how do we know God exists?
Jesse: Because it says so in the Bible.

Some independent grounds are needed to prove either the existence

of God or the veracity of the Bible, just as an independent justifica-
tion is needed for the psychic powers of either Jesse or his aunt.
A more subtle version of a circular argument – the sort of thing I
see in student essays occasionally – is seen in this example:
239

FURTHER FALLACIES 239

If all of the issues were to be correctly considered, it would be seen that

abortion is wrong.
The word ‘correctly’ signals the circularity. Imagine setting out reasons
why abortion is wrong, and then adding a premise to the effect that
anyone who understands these reasons correctly must conclude that
abortion is wrong. Someone who concludes otherwise cannot have rea-
soned correctly. The idea of ‘correct’ here is therefore defined in terms
of leading to the conclusion that abortion is wrong, and the conclusion
that abortion is wrong is only supported by this ‘correct’ reasoning.

10.3 FALSE DILEMMA

A false dilemma is an argument in which a range of options are
presented that either deliberately or accidentally under-represent
those that are actually available. It is sometimes known as a false
dichotomy, the difference being that ‘dichotomy’ means only two
choices are presented, whereas ‘dilemma’ can mean two or more. Also,
however, a dilemma indicates that the alternatives are unsatisfactory;
that we would rather not have to choose between them because
each has significant drawbacks. Better state healthcare and education
provision typically comes at the cost of higher taxes (and vice versa);
and as a pretty poor golfer, if I tee-off with a wood instead of an iron
I will be sacrificing accuracy for distance (and vice versa). If I am
offered tea or coffee, it is not usually a dilemma (aside from the fact
that having one means that I cannot, right now, have the other), but
we might call it a false choice if hot chocolate is also available but
not been presented as an option.
The basic argument has this structure:

P: Situation X allows for only these beliefs/courses of action.

C: Therefore your decision must be one of these beliefs/courses
of action.

But quite often this forms a sub-conclusion towards a further (pos-

sibly implicit) conclusion:

P2: Of the options available, these beliefs/courses of action should

be rejected.
240

240 FURTHER FALLACIES

C2: Therefore you should choose the remaining belief/course of

action.

We need to ask ourselves two critical questions:

Q1: Are the beliefs/options offered genuine alternatives? (For

example, are some not real or feasible, or if they are, are they
really incompatible with one another?)
Q2. Are the outcomes/options listed really the only ones
available?

As Taeda Tomić (2013) points out, taking a critical stance towards

arguments that present dilemmas not only allows us to consider fur-
ther alternatives, but to think creatively in terms of how apparently
incompatible options might be synthesised into a previously uncon-
sidered theory or solution.
Notice that setting up a false dilemma is a form of framing, of
presenting the world in a certain way, sometimes driven by a deliber-
ate agenda. It can be very powerful too; dichotomies play squarely
into our tendency to see situations in black and white, especially if
strong emotions are involved (see Chapter 2). We hear quite often
in the TV talent shows like The X Factor how it’s ‘all or nothing’ or
‘now or never’ for those auditioning, whereas in many cases it doesn’t
actually come down to a choice between superstardom and working
at McDonald’s.
In a decidedly less trivial example, an episode (called ‘Are you with
us?’) of the BBC radio series Fatwa about the reception of Salman
Rushdie’s novel The Satanic Verses, sees two interviewees from Muslim
backgrounds talk about the dichotomous view of their white lib-
eral friends at the time. The interviewees’ perspective was that nei-
ther the book nor the fatwa was acceptable, but the expectation was
that if you thought Rushdie was wrong to write the book, then
you also agreed with the fatwa (i.e., that he should die). This simple
and unsympathetic ‘for or against’ – a position which of course has
broader historical significances – was a profoundly alienating experi-
ence (‘shocking’, as one of them puts it).2
The reverse of the false dilemma is used in argumentation as well;
denying that an issue is black and white when in fact it is. In This
241

FURTHER FALLACIES 241

Changes Everything Naomi Klein argues that we really do have a clear

choice between the continuation of unregulated capitalism and cata-
strophic climate change (a ‘battle between capitalism and the planet’).
She is critical of the ‘fetish of centrism – of reasonableness, serious-
ness, splitting the difference’ 3 in other words, believing that we can
have both.
The implication of this, and similar arguments, is that it can be
convenient to think that two or more things that we want are not
incompatible with one another. Moreover, to complicate the choice
by arguing for a false dilemma could delay actions towards radical
change. Klein might not be right, but the point stands that many situ-
ations do indeed present us with a limited number of incompatible
choices. The important thing is to determine precisely what these
choices are, but wanting to arrive at this insight too soon makes us
susceptible to false dilemmas.

10.4 THE PERFECTIONIST FALLACY

The perfectionist fallacy has similarities with the false dilemma. In
both cases, lazy, black and white thinking plays a role in making us
vulnerable to some quite extraordinarily weak arguments. The per-
fectionist fallacy, in particular, is one that System 2 will recognise as
absurdly poor, but that seems to be such sweet temptation for System
1’s ‘instinct to absolutise’. The basic structure is:

P: Response X to problem Y will not provide a perfect solution.

C: Therefore response X should be rejected.

If this seems pretty dumb, then that’s because, in the vast majority of
cases, it is. For example, after mass shootings in the US we usually
hear a variation on this argument:

P1: If the US changes its gun control laws then people will still be
able get hold of guns.
P2: If people can get hold of guns, then shootings and massacres
will be a possibility.
C: Therefore there’s no point in the US changing its gun control
laws.
242

242 FURTHER FALLACIES

Implicit here – the assumption being made – is that:

P3: There’s no point in making changes unless those changes will

solve the problem completely.

What a strange thing to think, and yet it is remarkable how readily

we do think this kind of thing across a range of situations. On the
gun law issue, words of wisdom come from rock musician and sup-
porter of guns Johnny Van Zant (of Lynyrd Skynyrd):
I would like to see more rules on owning guns. People could still get them
underground, man. But you gotta start somewhere.4
Yes, you do have to start somewhere, and even if some lives are
saved by a modest change in the law, surely this is worth it. In a
similar fashion I have heard it suggested that there is no point in
lowering the speed limit in the UK to 50mph because ‘people will
still die on the roads’. Yes, BUT NOT AS MANY. There might be
some good reasons for not lowering the speed limit, but this is not
one of them.
In the closing part of his COP26 speech in 2021, Barrack Obama
seemed to be countering perfectionist tendencies when he under-
lined how any progress made with climate change negotiations
will be ‘messy’, ‘every victory will be incomplete’, but how ‘imper-
fect compromises’ will at least ‘move the ball down the field’.5 In
a similar vein, Frans Timmermans (Vice President of the European
Commission) compared progress towards emissions agreements
to a ‘marathon’. In both bases there is not an unhopeful mes-
sage, but any hope requires us to recognise that a perfect or tidy
solution to a global problem like this should not be a realistic
expectation.
That life in general is such a thing that perfect solutions are
rarely feasible, is something that System 1 seems to struggle with. A
motivation-­based explanation for this tendency is suggested by moral
philosopher Jonathan Glover. In a discussion of how the individual
should respond to global problems when the difference their actions
will make is so minimal, he says,
In many of the cases where it is used, the argument from the insignifi-
cant difference can be dismissed at once. If I can rescue a single person
243

FURTHER FALLACIES 243

from death or misery, the fact that there are many others I cannot res-
cue is irrelevant to the moral worth of doing this. Huge problems some-
times produce an irrational paralysis of the imagination. It is so terrible
to think of the poverty and starvation that will still exist in the world
whatever I do, that it is tempting to despair and do nothing. (Glover,
1986, p.126)

10.5 RED HERRINGS AND EQUIVOCATIONS

If you hear the term red herring in relation to someone’s argu-
ment (or your own for that matter) it means: (1) it’s not relevant to
the main point being debated, but (2) it has the appearance of being
relevant such that it is liable to shift the focus of the debate. As with
straw man arguments, this can happen by accident – in the moment
the arguer feels that the point is relevant – or it can be a deliberate
ploy to divert or obscure proceedings.
In May 2022 the Taliban ordered female TV presenters in
Afghanistan to wear face coverings. In one response to presenters’
protests that Islam does not require them to dress this way, the Taliban
said presenters could wear medical masks instead, just as they had
done during Covid. The implication is that their disagreement was
based on discomfort and clarity of communication, but this is a red
herring, designed to distract from the real issue of the growing ero-
sion of women’s basic rights and freedoms.6
A common way in which red herrings succeed in diverting the
argument away from its proper focus is through equivocation. This
is a fallacy linked to ambiguity, in which an alternative meaning
of a word or phrase is deliberately or accidentally employed across
premises and conclusions in a way that leads the argument off track.
In all cases a very poor argument is the result, but it can be one that
stalls the argumentation process because equivocations can be tricky,
or at least time-consuming, to unpick.
The exasperating ‘all lives matter’ response to the Black Lives
Matter movement and slogan – especially in the aftermath of George
Floyd’s murder in 2020 – is an example. Few would disagree that all
lives matter, but this neglects the context of this statement, which
is quite clearly one of discrimination in which Black lives are not
treated as equal. To overlook this might just be a case of ignorance,
244

244 FURTHER FALLACIES

but where the equivocation is deliberate it might also be an example

of precisely why this organisation exists.
In a lighter example, after a bad run of games in 2013 the then
Chelsea manager Jose Mourinho responded to a question in a press
conference asking if there was a crisis at the club by saying,
Crisis at Chelsea? What crisis? Syria is a crisis, we’ve just suffered two
bad results.7
By employing the contrast effect and the impression of an unex-
pectedly worldly answer from a football coach, Mourinho, true to
form, managed to deflect the question, but it is a clear case of equivo-
cation. The type of crisis the questioner is referring to and the one
that Mourinho invokes are radically different. If we are to accept his
argument then ‘crisis’ is a word that could never legitimately be used
in the context of sport, but that’s not the case; it simply has a different
meaning to crises that are fatally serious, urgent, ongoing, large-scale
humanitarian situations.

10.6 SUMMARY
This chapter has explained some common fallacies not discussed
elsewhere in the book and provided some reasons for why they are
important to know about. Lists of fallacies can be very long, and
there will never be a definitive set, so a book like this is inevitably
selective. If you explore other critical thinking textbooks you could
well uncover others that you think have equal or greater relevance to
the worlds in which you are responsible for making decisions.

EXERCISES
This is a good place to remind you of the recommendation in the
book’s Introduction that the best way to practise identifying and eval-
uating argument forms and fallacies is by analysing detailed examples
in which the context is familiar to you.
Aside from this, an exercise I use in my critical thinking classes to
enhance students’ understanding of these ideas is to provide a list of
argument forms and fallacies, and assign each student two or three to
research. Dividing the list among the class has various practical and
245

FURTHER FALLACIES 245

pedagogical advantages, including the chance for them to individu-

ally explore their assigned categories in greater depth, and a more
interactive learning experience.
There will be many ways to organise this, but my list consists of
as many of the argument forms and fallacies discussed in this book
as is practical, along with instructions along the lines of: (1) Describe
them in detail, explaining what makes them fallacious or weak; (2)
Provide at least two examples, one of which must be of your own
invention (and ideally on a topic you care about). Guidance is pro-
vided on good sources to consult (mostly critical thinking textbooks
available through the library, whether print or electronic versions).

FURTHER READING
Most textbooks on informal logic and critical thinking will dis-
cuss the fallacies found in this chapter (as well as those featured in
other parts of the book). Some good examples are Bowell and Kemp
(2015), Chatfield (2018), Hughes, Lavery, and Doran (2010), Johnson
and Blair (2006), LeBlanc (1998), and Sinnott-Armstrong (2018).

NOTES
1 Today programme, BBC Radio 4, 28.9.15
2 Fatwa – 6. Are You With Us?, BBC Sounds. Available at: www.bbc.co.uk/
sounds/play/m0002g88
3 Klein, N. (2014) This Changes Everything. Penguin
4 Classic Rock, December 2015, p.22
5 Available at: www.youtube.com/watch?v=D2ScHyXs50Q
6 Available at: www.reuters.com/world/asia-pacific/taliban-say-female-afghan-
tv-presenters-must-cover-face-air-2022–05–19/
7 Jefferson, B. (2013) Jose Mourinho: Chelsea losing to Basel and Everton isn’t
a crisis, Syria is a crisis [Article]. Available at: www.express.co.uk/sport/foot-
ball/430957/ Jose-Mourinho-Chelsea-losing-to-Basel-and-Everton-isn-t-a-
crisis-Syria-is-a-crisis.
246

CONCLUSION

Interviewer: What makes you unhappy?

John Malkovich: That people are so certain about things that you really
can’t be certain about. (Guardian Magazine, 10.3.18)
What would Davis do? (Sacha T. Burnstorm, pers. comm.)

As set out in the introduction, and as hopefully this book has demon-
strated, the aim of critical thinking is to make us better deliberators
and decision makers through a frame of mind and a set of knowledge
and skills that:
1. Help us to identify, reconstruct and assess the arguments of
others.
2. Help us to construct, assess and improve our own arguments.
3. Educate us in the pitfalls associated with reasoning in terms of:
i. Fallacies and their associated psychological biases;
ii. Features of unconstructive dialogues; and
iii. Dispositions that make us prone to fallacious reasoning
and unconstructive dialogues.

The goals of assessing the arguments of others and improving our

own arguments are closely related. This is partly because a significant
DOI: 10.4324/9781003247944-12
247

CONCLUSION 247

aspect of putting forward a position is to demonstrate its strength in

comparison to rival positions, but it is also because we should assess
our own arguments by the same standards we assess the arguments
of others. There is an understandable tendency when learning criti-
cal thinking to concentrate on reconstructing and evaluating other
people’s arguments at the expense of our own, and for this reason it is
this reflexive application that I would like to emphasise in the book’s
concluding pages.
In the opening chapters the relevance of critical thinking to self-
knowledge is quite apparent, and in this respect a book like this
brings us close to the spirit of philosophy as represented by Socrates.
Fundamental to the search for truth is to understand the strengths
and weaknesses of the very thing that is doing the searching – the
human mind. Generalised discussions about cognitive biases and the
dispositions that can help us manage them takes us so far, but the
examples and chapter exercises provided are also designed to encour-
age self-reflection. On the one hand this is part of the process of
deepening our understanding of these ideas. Through appreciating
their relevance to our particular strengths and weaknesses as revealed
through our own distinct contexts we can understand all the better
their power and prevalence. On the other hand, this reflection is valu-
able for the individual in and of itself. Self-understanding and the role
it plays in improving our lives is also part of the spirit of philosophy.
The chapters on argument reconstruction and argument forms and
fallacies are less obviously ‘about you’ in this sense, but they never-
theless are personal because, as stated, these skills and this knowledge
apply equally to our own arguments as they do to the arguments
of others. The continued focus on psychology, rhetoric, dispositions,
and dialogues may help to remind us of this, as should those exercises
which are clearly reflexive in nature.
In this spirit I will make some final points about how the skills and
knowledge of argumentation can act as a catalyst for self-reflection.
Specifically, I will list some important ways in which you can apply
this learning to academic, professional, civic, and everyday delibera-
tion and decision-making.

1. When constructing arguments, the guidelines for argument

reconstruction and evaluation can be adapted so that we ask
248

248 CONCLUSION

ourselves questions like: What is my conclusion? What are my

reasons for believing it? Are these acceptable, relevant, and
sufficient?
2. Be careful to make your language as clear and precise as pos-
sible; avoid ambiguity, vagueness, and equivocation.
3. Be watchful of the tendency to allow absolutist premises and
conclusions to appear in your arguments where they are not
appropriate. Most of the time they will not be. In other words,
commit with care. Lessons can be learned in this respect from
the section on generalisations in Chapter 8, but also from the
discussion of false dilemmas and the perfectionist fallacy in
Chapter 10.
4. If you are going to use authorities to support your conclusions,
apply the critical questions and make sure you are using them
for the right reasons. In the end you must take responsibility
for the positions you argue for.
5. Before evaluating other people’s arguments, make sure you
understand them.The disposition of open-mindedness and the
technique of argument reconstruction (along with the princi-
ple of charity) are not peripheral but central to critical think-
ing. In the opening chapters an important distinction is made
between critical thinking as a more detached evaluation, and
critical thinking as the spontaneous asking of questions that is
part of being absorbed in the ideas that one is reading or hear-
ing (what Dewey calls ‘wholeheartedness’). Since the confir-
mation bias is such a deep-seated tendency, we need to make a
sustained effort towards being better listeners.
6. Ask questions that help to ensure open-mindedness, such as
‘Is there anything important that I’m missing here?’ and ‘Have
I understood with sufficient depth the point of view of the
people I am arguing against (or supporting)?’ Get to know
your biases so that you are better able to prevent them from
undermining your objectivity in argumentation. Be your own
devil’s advocate.
7. Be on guard against overconfidence in the strength of your
arguments. Because of our self-serving bias we are inclined to
think that we are less susceptible than other people to the vari-
ous biases discussed in Chapter 1 (and throughout the book).
249

CONCLUSION 249

But we’re not. Whatever our self-serving bias tells us, we too
are prone to the self-serving bias.
8. Become acquainted with the ways in which emotions can hin-
der (and help) argumentation. And notice the kinds of argu-
ments that are especially emotive, such as ad hominems, ad
baculums, and slippery slopes.
9. Consider the communication needs and limitations of your
audience. Make judicious use of enthymemes, take care with
your vocabulary, and avoid alienating others through smartars-
ery. Where permitted by the subject matter, choose authorities
and analogies that will be recognisable and appealing to your
audience.
10. If we listen carefully to the views of others, we might find that
we agree on more than we think we do; that disagreements are
caused by misunderstandings, or are superficial. To get there
we need to be willing to ask questions, and not with a view
to tripping people up. Thinking critically about beliefs and
courses of action means that our evaluations can be positive as
well as negative. To be oriented to the negative is a bias, and at
its worst it can make us ‘argumentative’. Don’t be argumenta-
tive, be a critical thinker.
ii
251

GLOSSARY

ad baculum argument (appeal to force) An argument that

involves a threat, such as: Accept X, or I will ensure that negative
consequence Y will occur.
ad hominem argument An argument that attacks the arguer
rather than their argument. A negative features of the arguer –
such as their character (‘direct’ or ‘abusive’ ad hominem),
behaviour that is inconsistent with what they are advocating (‘tu
quoque’), a biased perspective (‘circumstantial’ ad hominem),
or some association with people or institutions of questionable
character, etc. (‘guilt by association’) – is given as a reason for
us to reject their argument (no matter how strong that argument
otherwise is).
ad populum argument An argument that claims that, because the
majority of people hold a certain belief or behave in a certain
way, we should also hold that belief or behave in that way.
affect heuristic Basing a judgement or decision on how we feel
about a particular person, situation, etc., rather than on a more
reasoned assessment.
affirming the consequent A fallacious version of a valid deduc-
tive argument known as modus ponens.
252 GLOSSARY

ambiguity and vagueness An ambiguous sentence is one that has

two or more possible meanings. A vague sentence is one that is
imprecise.
analogy An attempt to illuminate, explain, or make an argument
about a certain thing (or situation, event) by comparing it with
something that shares relevant features with that thing (or situ-
ation, event).
anchoring This is the biasing effect of being exposed to prior rel-
evant information when making a judgement.
appeal to emotion A form of argument in which a particular
emotion (such as fear, compassion, or guilt) is instrumental in
establishing the conclusion.
argument A claim we make that we want others to accept
as true, combined with reasons supporting this claim. The
claim is known as the conclusion, and the reasons are called
premises.
argument forms Commonly used types of argument (such as
arguments from analogy, ad hominem arguments, causal
arguments, arguments from consequences, and so on)
that share a general structure and are often associated with cer-
tain types of fallacy. They are sometimes known as ‘argument
schemes’.
argument from analogy An argument in which a situation (X)
is compared with a supposedly similar situation (Y). In the case
of Y we know, or would believe, that Z is the case, and because
of Y’s similarity to X, it is concluded that Z is also true (or likely
to be true) of X.
argument from authority An argument in which we are
encouraged to accept the conclusion on the basis that a relevant
authority (such as an expert) endorses that position.
argument from consequences An argument of the form: If X
happens, then consequence Y will occur. (The consequence can
be positive or negative.)
argument reconstruction The practice of extracting the essence
of an argument from the everyday language in which it is
expressed or implied, and formulating it in terms of appropri-
ately ordered premises and conclusions.
argumentation The exchange of arguments in dialogues.
 GLOSSARY 253

ARS criteria The assessment of arguments in terms of the accept-

ability of their premises; the relevance of their premises to the con-
clusion; and whether the premises are sufficient for establishing
the conclusion. See also critical questions.
assumption In the context of an argument, an assumption is a
belief that has relevance to the argument, but which has not been
defended.
availability heuristic The quick rule by which we judge the
likelihood of an occurrence based on how readily it comes
to mind.
base rate neglect (or sometimes ‘background’ rate neglect)
A fallacy referring to situations in which, when making a judge-
ment about a particular person or event, we overlook relevant
background information.
burden of proof If a belief is taken to be generally established,
the norm among a group of people, or an agreed default posi-
tion, then this tends not to be a position that requires defend-
ing if challenged. Instead, the burden of proof falls to the
challenger.
causal arguments Arguments that claim that correlated events
(X & Y) have a particular causal relationship (such as X causes Y).
causal fallacies A range of arguments in which a mistaken claim
is made about a causal relationship; such as overlooking a shared
cause, misunderstanding the direction of cause, or overlooking
multiple causes.
cherry picking fallacy Information or examples that conform
to the arguer’s hypothesis or that present their position in a par-
ticularly flattering light are presented as evidence, whilst non-
conforming or unflattering data is ignored or suppressed.
circular argument (begging the question) An argument in
which we must accept the conclusion in order to accept one
or more of the premises. This is to argue ‘in a circle’ because, of
course, the conclusion can only be established once the truth of
the premise(s) is established.
circumstantial ad hominem argument See ad hominem
argument.
coercive power The social influence resulting from a person’s
ability to punish the actions of others.
254 GLOSSARY

cognitive dissonance The discomfort experienced as a result of

an inconsistency between one’s beliefs, or between one’s beliefs
and one’s behaviour. See also commitment and consistency.
commitment and consistency A heuristic based on the need for
beliefs and behaviours to be consistent with prior commitments.
conclusion See argument.
confirmation bias The largely automatic tendency to seek out,
attend to, or remember information that confirms existing beliefs
at the expense of information which disconfirms them.
constructive dialogue A dialogue is a discussion between two or
more people, and dialogues that aim to establish truth, solve a
problem, or resolve a dispute can be carried out in ways that are
constructive or unconstructive. A constructive dialogue is guided
by a set of (often implicitly held) rules that help ensure the dom-
inating presence of critical thinking.
contrast effect The effect that nearby comparators with con-
trasting qualities have on our perception and interpretation of
information. For example, if you are trying to impress someone
with your singer-songwriter abilities, don’t do it after they’ve just
been listening to Bob Dylan.
critical questions The questions that are important to ask in order
to establish the strength of an argument.
critical thinking disposition See disposition.
deductive argument An argument which attempts to provide a
line of reasoning in which the conclusion follows logically (or
necessarily) from the premises, meaning that if the premises are
true, the conclusion must also be true. A successful deductive
argument is known as a ‘valid deductive argument’.
defeasible generalisation A type of generalisation that attempts
to apply a generally accepted belief to a particular instance in the
context of a dialogue.They are regarded as provisional and form
premises in plausible arguments.
dialogue See constructive dialogue.
direct ad hominem argument See ad hominem argument.
direction of cause Where two events appear to occur simultane-
ously there is a danger of mistaking the direction of cause; in
other words, determining which event is the cause and which
event is the effect.
 GLOSSARY 255

disposition A tendency in a person that inclines them to think,

feel, and act in a certain way (similar to an attitude, character
trait, or virtue). A critical thinking disposition is one that
inclines a person to be a good critical thinker.
dual process theories Psychological theories that identify two
distinct modes of processing information, typically one that is
automatic and fast, and one that is deliberative and slow. See also
System 1/System 2 thinking and heuristic.
ego-defences Ways in which we distort reality in order to
keep painful truths away from full consciousness. See also
rationalisation.
enthymeme An argument with a missing premise or premises,
usually for the sake of brevity, with the arguer confident that the
audience is aware of what has been omitted.
equivocation This is a fallacy linked to ambiguity, where an alter-
native meaning of a word or phrase is deliberately or accidentally
employed across premises and conclusions in a way that leads the
argument off track. See also red herring.
ethotic power (ethotic authority) The social influence of some-
one who is regarded as a broadly ‘good’ or ‘virtuous’ person.
experiential system See System 1 and System 2 thinking.
expert power (expert authority) The social influence held by a
person by virtue of their expertise (or perceived expertise) in a
certain area.
fallacy Fallacies are weak arguments of certain types that are used
frequently and are liable to be convincing to those not thinking
critically.
false dilemma/false dichotomy/false choice A set of fallacies
in which the options available in a particular situation have been
misleadingly represented.
framing In communication, framing refers to a way of looking
at things; of putting an ‘angle’ or ‘spin’ on a particular issue. It
involves emphasising some aspects of a concept, policy proposal,
product, and so on, at the expense of others.
generalisation (absolute and non-absolute) Absolute (also
known as ‘universal’, ‘strict’, or ‘hard’) generalisations are
those that claim that all instances of a certain category of
thing have a certain quality, and non-absolute (also known as
256 GLOSSARY

‘non-universal’, ‘inductive’, or ‘soft’) generalisations are those

that claim that only ‘most’ or ‘the majority’ of instances have this
quality.
groupthink A set of processes of social influence (such as col-
lective overconfidence and pluralistic ignorance) that cause a
group of otherwise rational and intelligent people to make very
poor decisions.
guilt by association See ad hominem argument.
halo effect Our pronounced tendency to generalise known posi-
tive features of a person to other, unknown, aspects of them. The
opposite of this is the horn effect; illegitimately generalising
negative characteristics.
hasty generalisation Where a general truth is assumed on the
basis of insufficient evidence. A sweeping generalisation is one
in which a generalised claim is accepted as true, but where rel-
evant exceptions are overlooked.
heuristic: A mental tool for quick decision-making; a rule of
thumb that is applied to certain types of situation that, although
lacking precision, makes better than chance judgements. (See
also System 1 and System 2 thinking.)
horn effect See halo effect.
inductive argument In contrast to a deductive argument, the
conclusion of an inductive argument does not necessarily follow
from the premises, so it is possible for the premises to be true and
the conclusion false.
informal logic The study of arguments and reasoning as employed
in real-life contexts, and often viewed as synonymous with criti-
cal thinking.
information power A type of social influence based on the
information a person has access to (e.g., as a result of their
job, or people they know). A sub-category of this is witness
­testimony – the knowledge resulting from first-hand experi-
ence of an unusual event.
legitimate power Social influence based on the position someone
holds, or moral and legal principles their decisions and actions
uphold.
metacognition This can sometimes simply mean self-awareness,
but a more specific usage refers to our awareness of the mental
 GLOSSARY 257

processes (e.g., thoughts and emotions) we are currently experi-

encing with a view to regulating them.
mistaking cause for correlation Otherwise known as the ‘Sod’s
law fallacy’, in which a person overlooks a causal relationship
between two events and instead declares them to be a coinci-
dence. Where the effect is negative it might be interpreted as the
kind of bad luck that is typical of certain circumstances, hence
being ‘Sod’s law’.
motivated reasoning Biased reasoning that serves to create and
preserve beliefs conforming to how we want the world to be
rather than how it actually is.
narrative fallacy This can be understood as a psychological basis
of the oversimplification of causes, and refers to our attrac-
tion to, and confidence in, coherent but simple explanations at
the expense of an appreciation of complex, unknown, or random
factors.
necessary and sufficient conditions A necessary condition is
one that must be fulfilled for something to be the case (belong
to a particular category, cause something to happen, etc.); and a
sufficient condition (or set of conditions) is one that is enough
to make something the case. A sufficient condition can also be
necessary, but it doesn’t have to be.
need for cognitive closure A measure of the extent of a person’s
desire for clear-cut and quickly reached answers to questions and
problems.
overlooking shared cause The fallacy of looking for a causal
relationship between correlated events instead of considering the
possibility that the correlation is the product of a shared cause.
oversimplification of causes The fallacy of assuming a more
straightforward causal explanation for a phenomenon than is
actually warranted. In many cases multiple causal factors need to
be accounted for. See also narrative fallacy.
perfectionist fallacy The claim that unless a perfect solution to a
problem (such as gun crime or death on the roads) can be found
it is not worth intervening (through, for example, new policies
on gun control or reducing speed limits).
placebo effect Beneficial medical outcomes resulting from a
patient’s belief that the treatment they are receiving will improve
258 GLOSSARY

their health, rather than from any direct physical effects of the
treatment itself.
plausible argument A style of argument employed in delibera-
tions and other dialogues in which generally accepted assump-
tions act as the basis for reaching conclusions.They have practical
advantages, but are recognised as provisional in nature. See also
burden of proof, and defeasible generalisation.
pluralistic ignorance A socially based form of ignorance in which
indecision or inaction results from the mutual misinterpretation
of uncertainty as calmness or confidence.
position to know argument An argument based on informa-
tion power (but excluding witness testimony).
premise See argument.
premise and conclusion indicators When identifying and
reconstructing arguments, certain words and phrases can help to
indicate the presence of premises (e.g., ‘the reasons for this are’)
and conclusions (e.g., ‘therefore’).
principle of charity When reconstructing arguments, the princi-
ple of charity is the best reasonable interpretation of ambiguous,
vague, or otherwise unclear sentences or arguments.
proposition (statement) A sentence that (at least in theory) can
be adjudged to be true or false. These are contrasted with, for
example, sentences that are questions or directives (requests,
orders). All premises and conclusions in an argument must be
propositions.
psychology of persuasion The study of the psychological prin-
ciples underlying persuasive communication.
rational system See System 1/2 thinking.
rationalisation A type of ego-defence in which we uncon-
sciously provide a false reason for why something happened
because the real reason (which the rationalisation masks) causes
us psychological pain or discomfort
reciprocation A heuristic linked to the norm of equitable
exchange of favours, goods, and so on.
red herring An irrelevant argument with the potential to divert
other arguers away from the issue being discussed because
of its apparent relevance and often emotive content. See also
equivocation.
 GLOSSARY 259

referent power The form of social influence held by a person or a

group that we identify with.
regression to the mean A phenomenon whereby outstandingly
good or poor results or performances are typically followed by
more average ones. Fallacious thinking can result from this if we
base beliefs and decisions on a single extreme outcome.
representativeness heuristic A simple rule whereby quick
answers to questions such as ‘what’s the probability of individual
X belonging to group Y’ or ‘what’s the probability of event P
being the cause of event Q’ are provided by referring to what a
typical member of group Y, or a typical cause of Q is like.
reward power The social influence of someone who is able to
reward the actions of others (through gifts, treats, praise, bonuses,
etc.).
rhetoric The art of persuasive communication.
self-serving bias: An automatic and systematic tendency to dis-
tort features of ourselves, others, and the world in a way that bet-
ter satisfies our needs for self-esteem, a coherent and predictable
world, etc.
shared causes: See overlooking shared cause.
slippery slope arguments Arguments in which it is claimed that
an initial, relatively inconsequential or benign, consequence will
lead to a series of further consequences that (1) are largely out
of our control, and (2) will end in a disastrous final consequence.
social proof A heuristic in which the behaviour of other people
is used as a guide to what to believe or how to act.
sound argument A valid deductive argument that also has true
premises.
straw man argument An argument in which the position of an
opponent is misrepresented in such a way that it makes it easier
to argue against.
sub-conclusions Conclusions reached on the way to an overall
conclusion in a larger argument. The conclusions of compo-
nent arguments function as further premises in the larger
argument.
sweeping generalisation See hasty generalisation.
system 1 and System 2 thinking Tversky and Kahneman’s
System 1 is the ‘fast thinking’ that is reliant on heuristics. System
260 GLOSSARY

2 is the sort of ‘slow thinking’ that we associate with focussed

attention, problem-solving, and critical thinking. Epstein’s expe-
riential and rational systems are similar, but the experiential sys-
tem is regarded as feeling-led and is part of a more wide-ranging
theory of human motivation and personality.
System 1 candy A term I am using to characterise certain fea-
tures of arguments or other persuasive messages that encourage
System 1 (heuristics-based) decision-making in situations where
System 2 ought to be employed.
tu quoque See ad hominem argument.
vagueness See ambiguity and vagueness.
valid deductive argument See deductive argument.
witness testimony See information power.
261

REFERENCES

Aberdein, A. (2020) Eudaimonistic Argumentation. In From Argument Schemes to

Argumentative Relations in the Wild (pp.97–106). Springer, Cham.
Aikin, S. and Clanton, J. (2010) Developing Group-Deliberative Virtues. Journal of
Applied Philosophy, 27(4), 409–24.
Aikin, S. and Casey, J. (2011) Straw Men, Weak Men and Hollow Men.
Argumentation, 25, 87–105.
Aikin, S. and Casey, J. (2016). Straw Men, Iron Men, and Argumentative Virtue.
Topoi, 35, 431–40.
Aristotle (1991) The Art of Rhetoric. Penguin.
Baghramian, M. and Croce, M. (2021). Experts, public policy, and the ques-
tion of trust. In The Routledge Handbook of Political Epistemology (pp.446–57).
Routledge.
Battaly, H. (2010) Attacking Character: Ad Hominem Argument and Virtue
Epistemology. Informal Logic, 30(4), 361–90.
Barnett, R. (1997) Higher Education: A Critical Business. Open University Press.
Baron, R. (2005) So Right It’s Wrong: Groupthink and the Ubiquitous Nature of
Polarized Group Decision Making. Advances in Experimental Social Psychology,
37, 219–53.
Berrill, D., ed. (1996) Perspectives on Written Argument. Hampton Press.
Blass, T. (2000) Obedience to Authority: Current Perspectives on the Milgram Paradigm.
Erlbaum Assoc.
Bohm, D. (2004) On Dialogue. Routledge.
Bowell, T. and Kemp, G. (2015) Critical Thinking: A Concise Guide (4th Ed.)
Routledge.
262 REFERENCES

Brady, M. (2013) Emotional Insight. Oxford University Press.

Brinton, A. (1986) Ethotic Argument. History of Philosophy Quarterly, 3(3), 245–58.
Brown, R. (1986) Social Psychology. Free Press.
Burns, J.M. (1978) Leadership. Harper and Row.
Cassam, Q. (2019) Conspiracy Theories. Polity Press.
Chatfield, T. (2018) Critical Thinking. Sage.
Cialdini, R. (2007) Influence:The Psychology of Persuasion. Collins.
Cialdini, R. (2017) Pre-Suasion. Random House.
Cohen, D. (1995) Argument is War … and War is Hell: Philosophy, Education,
and Metaphors for Argumentation. Informal Logic, 17(2), 177–88.
Cohen, D. and Miller, G. (2016) What Virtue Argumentation Theory Misses: The
Case of Compathetic Argumentation. Topoi, 35(2), 451–60.
Cooper, H. (1979) Pygmalion Grows Up: A Model for Teacher Expectation,
Communication and Performance Influence. Review of Educational Research,
49(3), 389–410.
Csikszentmihalyi, M. (2002) Flow. Rider.
Damasio, A. (2000) The Feeling of What Happens: Body, Emotion and the Making of
Consciousness.Vintage.
Davies, M. (2015) A model of critical thinking in higher education. In Higher
Education: Handbook of Theory and Research (pp.41–92). Springer.
Davies, M. and Barnett, R., eds. (2015) The Palgrave Handbook of Critical Thinking
in Higher Education. Springer.
Davis, H. (2009) Caveman Logic. Prometheus Books.
Descartes, R. (1968) Discourse on Method and the Meditations. Penguin.
DeSteno, D., Petty, R.E., Rucker, D.D., Wegener, D.T., and Braverman, J.
(2004). Discrete Emotions and Persuasion: The Role of Emotion-Induced
Expectancies. Journal of personality and social psychology, 86(1), 43–56.
Dewey, J. (1910) How we think. D.C. Heath & Co.
Eemeren, F. van, and Grootendorst, R. (2004). A Systematic Theory of Argumentation:
The Pragma-Dialectical Approach. Cambridge University Press.
Ennis, R. (1996a) Critical Thinking. Prentice Hall.
Ennis, R. (1996b) Critical Thinking Dispositions: Their Nature and Assessibility.
Informal Logic, 18(2&3), 165–82.
Epstein, S. (2014) Cognitive-Experiential Theory. Oxford University Press.
Evans, J.S.B. and Stanovich, K.E. (2013). Dual-process theories of higher cogni-
tion: Advancing the debate. Perspectives on psychological science, 8(3), 223–41.
Facione, P. (2006) Critical thinking:What it is and Why it Counts. Measured Reasons
LLC.
Festinger, L. and Carlsmith, J. (1959) Cognitive Consequences of Forced
Compliance. Journal of Abnormal and Social Psychology, 58, 203–10.
Fisher, R. and Ury, W. (1991) Getting to Yes: Negotiating an Agreement Without
Giving In. Random House.
 REFERENCES 263

Fiske, S. (2005) What’s in a Category? In A. Miller (ed.) The Social Psychology of

Good and Evil. Guildford Press.
Flew, A. (1985) Thinking About Thinking. Flamingo.
French, J. and Raven, B. (1959) The Bases of Social Power. In D. Cartwright (ed.)
Studies in Social Power. Institute for Social Research.
Furnham, A. and Boo, H.C. (2011). A literature Review of the Anchoring Effect.
The Journal of Socio-Economics, 40(1), 35–42.
Gilbert, M. (1994) Feminism, Argumentation and Coalescence. Informal Logic,
16(2) 95–113.
Gilbert, M. (1997) Coalescent Argumentation. Erlbaum.
Glover, J. (1986) It makes No Difference Whether or Not I Do It. In P. Singer
(ed.) Applied Ethics. Oxford University Press.
Goodwin, J. (2011) Accounting for the Appeal to the Authority of Experts.
Argumentation, 25, 285–96.
Govier, T. (2018) Problems in Argument Analysis and Evaluation. Windsor Studies in
Argumentation.
Govier, T. (2013) A Practical Guide to Argument (7th Ed.). Wadsworth.
Halpern, D. (2003) Thought & Knowledge (4th Ed.). Lawrence Erlbaum.
Halpern, D. (2014) Thought & Knowledge (5th Ed.). Psychology Press.
Hamblin, C. (1970) Fallacies. Methuen.
Hargie, O. (2017) Skilled Interpersonal Communication (6th Ed.). Routledge.
Hughes, W., Lavery, J., and Doran, K. (2010) Critical Thinking (6th Ed.). Broadview
Press.
Hume, D. ([1748] 2008) An Enquiry Concerning Human Understanding. Oxford
University Press.
Hume, D. ([1779] 1990) Dialogues Concerning Natural Religion. Penguin.
Janis, I. (1982) Groupthink (2nd Ed.). Houghton Mifflin Co.
Jefferson, A. (2014) Slippery Slope Arguments. Philosophy Compass, 9/10, 67–80.
Johnson, R. and Blair, A. (2006) Logical Self Defense. IDEA.
Kahneman, D. (2012) Thinking Fast and Slow. Penguin.
Kant, I. ([1781] 1929) Critique of Pure Reason, trans. N. Kemp. Macmillan.
Kant, I. ([1784] 1963) What is Enlightenment? In L. White Beck (ed. and trans.)
On History. Macmillan.
Keltner, D., Oatley, K., and Jenkins, J. (2013) Understanding Emotions (3rd Ed.).
Wiley Global Education.
Kierkegaard, S. ([1846] 1979) Two Ages, ed. and trans. H.V. Hong and E.H. Hong.
Princeton University Press.
Kuhn, D. (1991) The Skills of Argument. Cambridge University Press.
Larrick, R. (2004) Debiasing. In D.J. Koehler and N. Harvey (eds.) Blackwell
Handbook of Judgment and Decision Making. Blackwell.
Lazarus, R. (1991) Emotion and Adaptation. Oxford University Press.
LeBlanc, J. (1998) Thinking Clearly. Norton.
264 REFERENCES

Levy, N. (2019) Due deference to denialism: Explaining ordinary people’s rejec-

tion of established scientific findings. Synthese, 196(1), pp.313–27.
Lunsford, A. and Ruszkiewicz, J. (2010) Everything’s an Argument (5th Ed.)
Bedford/St.Martins.
Makan, J.M. and Marty, D.L. (2001) Cooperative Argumentation: A Model for
Deliberative Community. Waveland Press.
McElroy, T. and Dowd, K. (2007). Susceptibility to Anchoring Effects: How
Openness-to-Experience Influences Responses to Anchoring Cues. Judgment
and Decision Making, 2(1), 48–53.
McGarvey, D. (2017) Poverty Safari: Understanding the Anger of Britain’s Underclass.
Luath.
Melling, D. (1987) Understanding Plato. Oxford University Press.
Milgram, S. ([1974] 2005) Obedience to Authority. Pinter and Martin.
Mill, J.S. (1962) Utilitarianism, On Liberty, Essay on Bentham. Fontana.
Modgil, S., Singh, R.K., Gupta, S., and Dennehy, D. (2021) A confirmation
bias view on social media induced polarisation during Covid-19. Information
Systems Frontiers, 1–25.
Morrow, D. and Weston, A. (2011) A Workbook for Arguments. Hackett.
Narveson, J. (1993) Morality and Violence. In T. Regan (ed.) Matters of life and
death (3rd Ed). McGraw-Hill.
Neenan, M. and Dryden, W. (2020) Cognitive Behavioural Coaching (3rd Ed.).
Routledge.
Newman, J.H. ([1852] 1982) The Idea of a University. University of Notre Dame
Press.
Nickerson, R.S. (1998) Confirmation bias: A ubiquitous phenomenon in many
guises. Review of general psychology, 2(2), 175–220.
Nussbaum, M. (1990) Love’s Knowledge. Cambridge University Press.
Nye, A. (1990) Words of Power: A Feminist Reading of the History of Logic. New
York: Routledge.
Oatley, K., Keltner, D and Jenkins, J. (2006) Understanding Emotions (2nd Ed.).
Blackwell.
Osman, M. (2004) An Evaluation of Dual-Process Theories of Reasoning.
Psychonomic Bulletin and Review, 11(6), 988–1010.
Osman, M. (2013). A case study: Dual-process theories of higher cognition—
Commentary on Evans & Stanovich (2013). Perspectives on Psychological Science,
8(3), 248–52.
Paley, W. ([1800] 2009) Natural Theology. Cambridge University Press
Paul, R. (1984a) Critical thinking: Fundamental to education for a free society,
Educational Leadership, Sept., 4–14.
Paul, R. (1984b) Teaching Critical Thinking in the Strong Sense: A Focus on
Self-Deception, World Views, and a dialectical Mode of Analysis. Informal
Logic, 4(2), 2–7.
 REFERENCES 265

Paul, R. (1995) Critical Thinking. Foundation for Critical Thinking

Perelman, C. and Olbrechts-Tyteca, L. (1969) The New Rhetoric: A Treatise on
Argumentation. University of Notre Dame Press.
Perkins, D., Jay, E., and Tishman, S. (1993) Beyond Abilities: A Dispositional
Theory of Thinking. Merrill-Palmer Quarterly, 39(1), 1–21
Petty, R. and Cacioppo, J. (1996) Attitudes and Persuasion: Classic and Contemporary
Approaches. Westview.
Pinker, S. (2021) Rationality. Allen Lane.
Pithers, R.T. and Soden, R. (2000) Critical Thinking in Education: A Review.
Educational Research, 42(3), 237–49.
Plato (1993) The Last Days of Socrates, trans. H. Tarrant and H. Tredennick.
Penguin.
Plato (2004) Gorgias, trans. C. Emlyn-Jones and W. Hamilton. Penguin.
Pronin, E., Lin, D., and Ross, L. (2002) The Bias Blind Spot: Perceptions of Bias
in Self Versus Others. Personality and Social Psychology Bulletin, 28(3), 369–81.
Raven, B. (1965) Social Influence and Power. In I.D. Steiner and M. Fishbein
(eds.) Current Studies in Social Psychology. Holt, Rinehart & Winston.
Rosenthal, R. and Jacobson, L. ([1968] 2003) Pygmalion in the Classroom.
Carmarthen, UK: Crown House Publishing.
Schultz, P., Nolan, J., Cialdini, R., Goldstein, N., and Griskevicius, V. (2007)
Constructive, Destructive, and Reconstructive Power of Social Norms.
Psychological Science, 18(5), 429–34.
Scriven, M. (1976) Reasoning. McGraw-Hill.
Seech, Z. (1993) Open Minds and Everyday Reasoning. Wadsworth.
Siegel, H. (1988) Educating Reason. Routledge.
Sinnott-Armstrong, W. (2018) Think Again: How to Reason and Argue. Pelican
Books.
Slovic, P. (2002) The Perception of Risk. Routledge (Earthscan).
Slovic, P., Finucane, M.L., Peters, E., and MacGregor, D.G. (2007) The affect
heuristic. European journal of operational research, 177(3), 1333–52.
Smith, E., Mackie, D. and Claypool, H. (2014) Social Psychology (4th Ed).
Psychology Press.
Solomon, R. (2003) Emotions and Choice. In R. Solomon (ed.) What is an
Emotion? (2nd Ed.). Oxford University Press.
Sopory, P. and Dillard, J.P. (2002) The Persuasive Effects of Metaphor: A Meta-
Analysis. Human Communication Research, 28(3), 382–419.
Tajfel, H. and Turner, J. (1986) The Social Identity Theory of Intergroup
Behaviour. In S. Worchel and W.G. Austin (eds.) Psychology of Intergroup
Relations (pp.7–24). Nelson-Hall.
Taleb, N. (2007) The Black Swan. Penguin.
Tannen, D. (1992) That’s Not What I Meant! Virago.
Tannen, D. (1998) The Argument Culture. Ballantine.
266 REFERENCES

Taylor, S.E. and Brown, J.D. (1988) Illusion and Well-Being:A Social Psychological
Perspective on Mental Health. Psychological bulletin, 103(2), 193–210.
Tiberius,V. (2008) The Reflective Life. Oxford University Press.
Tindale, C. (2007) Fallacies and Argument Appraisal. Cambridge University Press.
Tomić, T. (2013) False Dilemma: A Systematic Exposition. Argumentation, 27(4),
347–68.
Toulmin, S. (1958) The Uses of Argument. Cambridge University Press.
Tversky, A., and Kahneman, D. (1974) Judgment under Uncertainty: Heuristics
and Biases. Science, 185(4157), 1124–31.
Tversky, A. and Kahneman, D. (1983) Extensional Versus Intuitive Reasoning:
the Conjunction Fallacy in Probability Judgment. Psychological Review, 90(4),
293–315.
Waller, B.N. (2001) Classifying and Analysing Analogies. Informal Logic, 21(3),
199–218.
Walton, D. (1989) Informal Logic. Cambridge University Press.
Walton, D. (1992) The Place of Emotion in Argument. Penn State University Press.
Walton, D. (1995) A Pragmatic Theory of Fallacy. University of Alabama Press.
Walton, D. (2001) Abductive, Presumptive and Plausible Arguments, Informal
Logic, 21(2), 141–69.
Walton, D. (2006) Fundamentals of Critical Argumentation. Cambridge University
Press.
Walton, D. (2009) Ad Hominem Arguments. University of Alabama Press.
Walton, D. (2015) The Basic Slippery Slope Argument, Informal Logic, 35(3),
273–311.
Walton, D., Reed, C., and Macagno, F. (2008) Argumentation Schemes. Cambridge
University Press.
Weber, M. (1978) Economy and Society. University of California Press.
Webster, D. and Kruglanski, A. (1994) Individual differences in need for cognitive
closure. Journal of personality and social psychology, 67(6), 1049–62
Webster, D. and Kruglanski, A. (1997) Cognitive and Social Consequences of the
Need for Cognitive Closure, European Review of Social Psychology, 8(1), 133–73.
Willard, C.A. (1990) Authority. Informal Logic, 12(1), 12–22.
Woods, J. (1995) Appeal to Force. In H. Hansen and R. Pinto (eds.) Fallacies. Penn
State University Press.
Woods, J. (2013) Errors of Reasoning: Naturalizing the Logic of Inference. College
Publications.
Worth, L. and Mackie, D. (1987) Cognitive Mediation of Positive Affect in
Persuasion. Social Cognition, 5(1), 76–94.
Yalom, I. (1980) Existential Psychotherapy. Basic Books.
267

INDEX

ad baculum arguments 137, 156–60, argumentation 13–16, 22, 24–5, 44,

162, 164, 249 48, 83, 86–7, 90–1, 99, 133, 145,
ad hominem arguments 134, 137, 157, 159, 175, 217, 240, 243, 247–8
140, 165, 182, 249 Aristotle 8, 11–12, 43, 48, 107, 122,
ad hominem arguments, 207, 224, 228
circumstantial 149–51 ARS criteria see fundamental critical
ad hominem arguments, direct (or questions
abusive) 142, 166, 168, 170, 172 assumptions 7, 10–11, 60–1, 108,
ad populum arguments 137, 165, 129–30, 169, 186, 197, 203, 208,
175–81 225, 242
affect heuristic 57–8, 68, 80 authority 43, 133–4, 248
affirming the consequent 234–6 authority (as a heuristic) 43–4
ambiguity 104–5, 243, 248 authority, arguments from 130,
analogy, arguments from 103, 128–9, 134, 136
215, 221–31 availability heuristic 40–1, 151, 204
analogy, hypothetical 224–5
analogy, inductive 224 base rate neglect 38–40
anchoring 41–2, 50, 77 basic critical questions (ARS criteria)
argument forms 121–2, 131 125–7
argument reconstruction 98–113, 247 biases see heuristics
arguments, definition of 3–4, 96 Bohm, D. 76, 77, 79
arguments, structure of 96–7 burden of proof 120, 144, 180, 203
268 INDEX

causal arguments 185–8 206–7; for information power 147–

causal fallacies 188–96 8; for legitimate authority 156;
chained arguments 108–9 for reward power 162; for slippery
charisma 171 slope arguments 213–14
cherry picking 204–5 critical theory 17
Cialdini, R. 41, 49 critical thinking, definition 4–6,
circular argument 123, 226, 237–8 23, 246
coercive power 135–6, 137, 156 critical thinking dispositions see
cognitive-behavioural therapy 208 dispositions
cognitive dissonance 2, 34–6, ‘critical thinking movement’ 14
148, 181
cognitive rationality see rationality Damasio, A. 57
coincidence 189–91 deductive arguments 117, 127, 131, 235
commitment and consistency 34–6, defeasible generalisation 202, 206
154, 201 deliberative friendliness 90
conclusion 4, 75, 80, 96, 101 deliberative wit 83, 91
confirmation bias 31–2, 42, 45, 56, 61, denying the antecedent 236
75, 100, 204 Dewey, J. 5, 10, 25, 75, 76, 82, 88, 248
confusing correlation and cause 123 dialogical dispositions 80–90
contrast effect 42–3, 47, 244 dialogue 9–10, 13, 47, 65; emotion
consequences, arguments from and 68–9; ad baculum arguments
210–11 and 159–60; arguments from
conspiracy theories 199–200, 209 analogy and 230; causal arguments
constructive dialogue, guidelines for and 200–1; ethotic arguments
a 90–1 and 170–2; expert authority and
continuum fallacy 215–216 144; generalisations and 208;
convergent arguments 111 information power and 151–2;
courage 81, 161, 181 referent power and 180–1; slippery
counter-attitudinal advocacy 92 slope arguments and 217 direction
Covid 46, 174, 192, 199, 221, of cause 191
232, 243 dispositions 8, 14, 16–17, 64, 72–7, 80,
critical questions 20, 38, 68, 122–32; 84–5, 90–2, 131, 168; ad baculum
for ad baculum arguments 156–7; arguments and 159–60; arguments
for ad hominem arguments 150–2; from analogy and 230; causal
for ad populum arguments 176–7; arguments and 200–1; ethotic
for arguments from analogy 225–7; arguments and 170–2; expert
for arguments from consequences authority and 144; generalisations
210–11; for causal arguments and 208; information power
196–7; and emotion in arguments and 151–2; Milgram and 160–1;
67–8; for ethotic authority 167–8; referent power and 180–1; slippery
for expert authority 138; for false slope arguments and 217
dilemmas 240; for generalisations dual process theory 26–9, 49
 INDEX 269

echo chambers 32, 49 generalisations 185, 201–10, 248

ego defences 59; see also Gilbert, M. 13, 86
rationalisation ‘graduate attributes’ 18–19
emotion, appeals to see emotion, groupthink 45–6, 50, 77, 81, 174, 176
arguments from guilt by association 169, 175
emotion, arguments from
64–8 halo effect 42, 143, 167, 169–70
emotion and dialogue 68–9 hasty generalisation 203–5, 206
‘emotional framing’ 56 heuristics 25, 28–33, 36, 43–4, 46, 49,
emotional intelligence 61 51, 57–9, 68, 70, 72, 131, 141, 143,
‘emotionally charged language’ 167, 198, 219
65–6 higher education see ‘graduate
emotions, nature of 52–4 attributes’
emotions and rationality 54–7 hindsight bias 200
emotions as heuristics 51, 57–62 horn effect 170
Ennis, R. 5, 14, 72–4, 82, 84–5, 88–9, Hume, D. 148, 152, 186, 225–6
91, 93, 132, 219 humility see modesty
enthymeme 107, 121, 130, 249
Epstein, S. 28–9, 54, 59–61 identity 44
equivocation see red herring identifying premises and
ethics 17–19 conclusions 101
ethotic arguments 166 illusion of understanding 199
ethotic power (authority) 135–7, 152, imagination 230–2
165–70 implicit assumptions 129–30
experiential system 28–9, 59–64, 68, implicit premises and conclusions
70, 204; see also Epstein, S. 107–8
expert power 135, 137 inductive arguments 118, 131
expert authority 134–5, 137–45, 155 informal logic 12
explanations 101–2, 222, 227 information power 135, 137,
exponential growth bias 50 145–147, 151, 155
inquisitiveness 60, 71, 218
fallacies 13, 20, 22, 48, 82, 116, ‘iron man’ argument 100–12
122–3, 131, 211–12, 216, 234,
244–5 Johnson, R. and Blair, J.A. 3, 75, 95,
false choice see false dilemma 112, 116, 125, 134, 211
false dichotomy see false dilemma judgement 6, 14, 19, 49, 53, 61, 82,
false dilemma 239–41, 248 92, 98, 124, 141–2, 170
familiarity 227, 231
figurative language 47, 228, 230 Kahneman, D. 27, 49, 187, 199–200
flexibility 76 Kant, I. 7, 9, 81, 144, 187
‘flow’ 88 King, Jr. M. L. 47–8, 102–3, 150–1,
framing 46–7, 161, 227, 240 219
270 INDEX

Klein, N. 46, 109–11, 139, 141, 241 Paul, R. 14–15, 33, 74

Kuhn, D. 15, 28, 31–2, 203–5, 219 perfectionist fallacy 241–2, 248
placebo effect 190, 193–5
legitimate power 135, 155–7, 161 plausible arguments 119–20, 131
Letter from Birmingham Jail see Plato 8–13, 28
King, M.L. pluralistic ignorance 45
likeability 43–4 position to know, argument from see
listening 75–6, 85–7, 90, 248 information power
commitment to truth 74–5 post hoc ergo propter hoc 189
practical rationality see rationality
maxims 207 pragma-dialectics 13
McGarvey, D. 59–61, 69 premise 4, 96
metacognition 30, 61, 64, 79–80, 160, premise and conclusion indicators
172, 201 101–2
metaphors 47, 222, 228–9 precedent, argument from 215
Milgram experiments 36, 153–6, principle of charity 105–6, 112,
160–3, 171, 180 227, 248
Mill, J.S. 3, 86, 100 propositions (statements) 97
mistaking cause for correlation see
Sod’s law fallacy rational system see Epstein, S.
modesty 78, 112, 144, 171, 201 rationalisation 34, 148, 161, 195, 217
modus ponens see affirming the rationality 25–6
consequent reciprocation 36–7, 154, 174
modus tollens see denying the red herring 243–4
antecedent referent power 44, 136–7, 175, 179
motivated reasoning 33, 49, 62, regression to the mean 190
75, 201 representativeness heuristic
multiple causes 193 37–8, 77
respect for others 84
narrative fallacy 193, 198–200 reward power 136, 155, 162
necessary and sufficient conditions rewording 102
125–6, 199 rhetoric 8–9, 11–13, 28, 48; ad
need for cognitive closure 77, 93 baculum arguments and 159;
negotiation 37, 41–2, 83, arguments from analogy and 227;
160, 242 causal arguments and 197; ethotic
‘no true Scotsman’ fallacy 209 authority and 168; expert authority
Nussbaum, M. 54, 152, 231 and 143; generalisations and 207;
information power and 151;
open-mindedness 75–6, 78, 90, 92, referent power and 179; slippery
144–5, 159, 181, 218, 248 slope arguments and 217
overconfidence 151, 201, 248 Russell, B. 31, 87
 INDEX 271

self-fulfilling prophecy 194 System 1 thinking 53, 57, 162, 169,

self-knowledge 6–7, 9, 78–9, 193, 197–8, 205, 241–2
90, 247 ‘System 1 candy’ 48, 192
self-reflection 6, 247
self-serving bias 33–4, 201, 248–9 Tannen, D. 87, 93
shared causes 191–2 Tiberius,V. 76–7, 93
Siegel, H. 18, 73 Tindale, C. 142, 180, 230
slippery slope arguments 64, 130, 185, trust 86, 143
210–18, 249 trustworthiness 140–1, 143, 201
smartarse, don’t be a 88–9 truth, commitment to 74–5
social power 135–6 tu quoque arguments 172–4
social proof 44–6, 179 Tversky, A. 27
Socrates 8–12, 28, 181, 247 Twelve Angry Men 22–3, 37, 49, 66, 70,
Sod’s law fallacy 195–6 75, 78, 84, 92, 148, 152, 160, 166–7,
sorites paradox 215–16 177–8, 201
sound argument 117
statements see propositions vagueness 104, 106, 248
staying focussed 82 virtues see dispositions
stereotypes 37, 172, 206
straw man argument 60, 99–100, Walton, D. 13, 63, 76, 79, 93, 117, 120,
112–13, 123 132, 165, 182, 202, 211, 220
sub-conclusions 108–11 ‘wholeheartedness’ 71–2, 248
superstition 198–9 witness testimony 135, 137, 147–9,
sweeping generalisation 203, 151, 177
205–6 Woods, J. 131, 164
System 1 and System 2 thinking
27–9, 42, 48, 79 Yalom, I. 77, 88, 191