Short-Answer Questions

1. (1 pt.) Explain in your own words why we need to refer to authorities to provide justifications for our beliefs. What are two problems that we may encounter when trying to justify our beliefs by referring to others. 

2. (2 pts.) The following argument, drawn from our textbook, is a type of ad hominem argument. Put it in a standard ad hominem argument form as illustrated on p. 132 (what’s the conclusion? What are the premises?). After you do this, use the critical questions listed at the top of p. 133 to analyze its quality. Is it poor or good? 

[NOTE: as you probably noticed, our textbook is written by a Brit. Fox hunting was restricted in 2005. The House of Commons is similar to the U.S. House of Representatives.]

“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 protesters 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 morally 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 House of Commons. And not one of those members who parade their tender conscience over fox-hunting has protested over the crime.”

3. (1 pt.) Ask a question to your classmates about either A) the reading itself or B) how to apply the ideas from the reading to everyday life.

Long-Answer Prompt

4. (2 pts.) Give an example from your personal experience (maybe from your friends/family or from a news article you heard/read) of either A) an argument from expert authority OR B) an argument from a position to know. Put it in standard form (using one of the argument forms presented in Sections 5.1 or 5.2) and evaluate it using the corresponding critical questions. 



Critical Thinking: The Basics is an accessible and engaging introduction to
the field of critical thinking, drawing on philosophy, communication and
psychology. Emphasising its relevance to decision-making (in personal,
professional and civic life), academic literacy and personal development,
this book supports the reader in understanding and developing the
knowledge and skills needed to avoid poor reasoning, to reconstruct and
evaluate arguments, and to 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 perseverance

• argument identification and reconstruction
• fallacies and argument evaluation.

With discussion questions and exercises and suggestions for further
reading at the end of the main chapters, 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 improving their thinking
skills and decision-making abilities.

Stuart Hanscomb is a Lecturer in Philosophy and Communication
at the School of Interdisciplinary Studies, University of Glasgow, UK.























Stuart Hanscomb

First published 2017
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada

by Routledge
711 Third Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 2017 Stuart Hanscomb

The right of Stuart Hanscomb to be identified as author of this work
has been asserted by him 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.

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: 1 [edition]. | New York : Routledge, 2016. | Series: The
basics | Includes bibliographical references and index.
Identifiers: LCCN 2016023190| ISBN 9781138826236 (hardback) | ISBN
9781138826243 (pbk.) | ISBN 9781315739465 (ebook)
Subjects: LCSH: Critical thinking.
Classification: LCC B105.T54 H363 2016 | DDC 160—dc23
LC record available at

ISBN: 978-1-138-82623-6 (hbk)
ISBN: 978-1-138-82624-3 (pbk)
ISBN: 978-1-315-73946-5 (eBook)

Typeset in Bembo Std and Scala Sans
by Book Now Ltd, London


List of boxes vi
Acknowledgements vii

Introduction: waking up to bad arguments 1

1 Rationality, cognitive biases and emotions 26

2 Critical thinking and dispositions 57

3 Arguments and argument reconstruction 78

4 Argument forms and fallacies 100

5 Arguments and social power: authority, threats
and other features of message source 115

6 Causal arguments, generalisations, arguments
from consequences and slippery slope arguments 167

7 Arguments from analogy 206

8 Further fallacies 218

Conclusion 228

Glossary 232
Select bibliography 243
Index 249


0.1 What is an argument? 4
0.2 Definition of critical thinking 24
4.1 Fundamental critical questions 107
4.2 Necessary and sufficient conditions 108
4.3 Fundamental critical questions and some examples

of sub-questions 109
5.1 French and Raven’s ‘bases of social power’ 118


My thanks go primarily to Benjamin Franks for his numerous helpful
suggestions and comments during the writing of this book.

Also to those students, GTAs and staff who contributed so much
to T&C, A-R-T, and CTC.

And to Tim Ewing (at for the
cover image.



My husband says I’m argumentative. He’s wrong though, and here are
three reasons why …

(Sacha T. Burnstorm, pers. comm.)

I wake up this morning to be told on the news that a culture that
forces people to get up early and start work at 9 is a form of ‘torture’.
I hit the snooze button and wonder if an appeal to human rights
could save me from having to give my 9 a.m. lecture. Unchangeable
bodily rhythms and the idea that we’re better suited to 10 or 11 o’clock
starts seem very important, but even if the hypothesis is correct, have
our lives thus far really been ‘torture’? Sleep deprivation is a well-
known method of cruel and unusual punishment, but anyone who
has endured it might be rightly dubious of classifying an early start in
this way. It’s good for headlines though.

I shuffle to the kitchen and put the kettle on. Soon my 3-year-
old son appears, yawning, looking for his breakfast. ‘Would you like
brown flakes or yellow flakes?’ I ask, knowing this is what’s called
a ‘false dichotomy’. There are several other cereal options but this
keeps things simple. As wilful as he can be, he seems content to have
his options framed in this narrow way first thing in the morning.
Hopefully this isn’t a form of torture.


Cycling to work, I take a route past a big field of cattle. There is
a hedge that I can see over, and the sight of my moving head seems
to spook one of the animals. He starts to run with me, parallel to
the road along the side of the hedge. As he passes other cattle, they
start galloping as well and before long I have caused a stampede.
After about 20 seconds the original runner slows to a halt and the
others do the same. They herd again, snorting and steaming, looking
agitated, and possibly slightly embarrassed.

I have picked up speed and become aware of my reluctance to use
the smallest and largest gear cogs (first and sixth) on my bike. Brief
reflection shows me that this is conditioned response caused by the
disintegration of the gear mechanism on my previous (ancient and
dilapidated) bike, which meant that the use of these gears ran a high
risk of the chain falling off. There is no reason to think that would
happen with my new bike, but the learning has transferred itself and
I limit myself for no reason.

Most animals and toddlers are not what we would call ‘critical
thinkers’, but nor are we much of the time. This series of events
is no exaggeration. Poor reasoning and the absence of reason-
ing on occasions where it would serve us well are everywhere.
Life is typically fast-paced and mistakes will happen, but even
when things are slowed down (for example, drafting a speech
rather than being interviewed on the radio), 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. our critical thinking capabilities are not what they might be, and
2. the particular forms of persuasive communication that make us

less likely to pay attention to, or even look for, poor reasoning,
and that are therefore more likely to win us round to their point
of view.

Arguments based on dubious, partial or irrelevant claims can be
remarkably effective if they target our cognitive and emotional



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 busi-
ness of life, three-fourths of the arguments for every disputed opinion consist
in dispelling the appearances which favour some opinion different from it.

(1962, p. 163)


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)

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)


[Arguments are] characterized by a particular structure, where one or
more statements … are given in support of a conclusion.

(Tindale, 2007, p. 1)

‘T o give an argument’ means to offer a set of reasons or evidence in support
of a conclusion.

(Morrow and Weston, 2011, p. xvii)


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

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 con-
clusion, 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 ‘having an
argument’ or a ‘having a row’ with someone, as in a bad-tempered
dialogue based on a disagreement. If we are ‘having an argument’
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, though, 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.
(I say ‘usually’ here because ‘preaching to the converted’ can involve


arguments, but in these cases they are used to re-enliven people’s
beliefs, often with the intention of encouraging action rather than
just verbal agreement.) Not all conversations involve arguments, but
many do. We talk to each other for a number 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.


Definitions of critical thinking tend to correspond with this defi-
nition of arguments. If an argument is a conclusion plus premises
(reasons given in support of the conclusion), then critical thinking
is the process of identifying what the argument is that is being put
forward, and determining whether or not the premises justify accept-
ing the conclusion (in other words, assessing whether it is a good
argument or not).

Robert Ennis, author of one of the most influential textbooks on
the subject, defines critical thinking as: ‘reasonable reflective think-
ing focused on deciding what to believe or do’ (1996b, p. 166).
Ennis’ reference to ‘doing’ as well as ‘believing’ indicates critical
thinking’s emphasis on deliberation and decision-making. It is very
much an applied area of study that intends to teach knowledge and
skills that have a clear practical application.

American philosopher John Dewey is often credited as being the
originator 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 deliberately 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’ve always believed, we actively con-
sider 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
critical thinkers (or 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. Practice – a lot of practice – in applying this learning to examples of
arguments and exchanges between people that involve arguments
and counter-arguments.

3. Self-re�ection on how we form our beliefs and how we interact
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 arguments, 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 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 own particu-
lar 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


deliberation and improved decision-making. Put another way, criti-
cal 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.


Critical thinking’s historical origins can be identified in two foun-
dational 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 wis-
dom and sound judgement (with self-knowledge among the most
prominent). In what is often taken to be the seminal work of mod-
ern philosophy, 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
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

(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 respon-
sibility for one’s destiny; to grow up. In An Answer to the Question:
What is Enlightenment? (originally published in 1784), Immanuel Kant
emphasised ‘release from … self-incurred tutelage. Self-incurred is this
tutelage when its cause lies not in lack of reason but in lack of resolu-
tion and courage to use it without direction from another’ (1963, p. 3).


These are examples of great philosophers and scientists challeng-
ing established knowledge, such as religious dogma and ancient
scientific 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. At bottom this is the importance of taking
responsibility for our beliefs; of thinking them through rigorously,
testing them out where possible and taking ‘ownership’ of them.
In this vein, the nineteenth-century philosopher Søren Kierkegaard
observed in The Present Age (originally published in 1846):

There are handbooks on everything, and generally speaking education will
soon consist of knowing letter-perfect a larger or smaller compendium of
observations from such handbooks, and one will excel in proportion to
his skill in pulling out the particular one, just as the typesetter picks out

(1979, p. 104)

His complaint concerned how, in an age proud to define itself as pro-
gressive in terms of social and scientific advances, individuals come
to see themselves as embodying this progress without themselves
needing to put the work in; without, as Cardinal Newman put it 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
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 prototypi-
cal critical thinker is the ancient Athenian, Socrates (469–399 bce).
He did not write anything, but his ideas, personality and philosophical


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 exam-
ple in the Apology, after engaging a politician with a reputation for
wisdom 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 talking 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 gener-
alisation which implies a lack of appropriate humility, or at least

For Socrates, this attitude was created and indulged by improper
education; in particular, an over-emphasis 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
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 palat-
able illusions of the speech makers. Not least of 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 painful or disquieting knowledge of our ignorance and a
commitment 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 in Athens. Famously he gave
the state little option but to kill him for his troubles, but in his own
defence, 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.

(1998, 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 fly. As we have seen, critical thinking
is a form of ‘rousing’, helping to improve our error-prone habits of
thought and action.

The method of the fly is dialogue: the exchange of views – of
explanations and arguments – between two (or more) people. (For
the sake of simplicity I will assume it to be a two-person discussion.)
A lot will be said about the nature and importance of dialogues in
Chapter 2, but aspects on its significance for Socrates will be briefly
discussed here. A dialogue can be two people with opposing views
trying to convince each other of the rightness of their particular
view; or it can be one person attempting to persuade an uncom-
mitted person. Socrates’ dialogues usually take the latter form, with


him questioning someone who believes they know something with
the aim of testing the firmness of the grounds of that belief. While
Descartes in his Meditations engaged in a kind of internal dialogue
to establish what he should and should not accept as true, Socrates’
method was 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, in a sense,
remind ourselves of why we know what we know. Alternatively
we discover that we do not know what we thought we knew, and
through reasoning come to appreciate our ignorance. It must be said
that reading many of Plato’s dialogues can leave us with the impres-
sion that Socrates did not always help people achieve this, but the
principle stands and remains basic to educational practices advocated
by philosophy. 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 person claiming to know something.
Teachers and peers can play this role, but very often dialogues are
symmetrical, with both parties having diverging convictions about
the matter at hand. Under these conditions the risk is that competi-
tion 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 promot-
ing 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.

• 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 hinder
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. While it is not my view 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 that can
be seen as basic to human fulfilment are typically associated with
an ideal critical thinker. 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 under-
stand the ways in which people are susceptible to persuasion (what
we now refer to as the psychology of persuasion or the psychol-
ogy of influence). Many of Aristotle’s insights in The Art of Rhetoric
are supported by modern research, including how emotions func-
tion to bias judgements, the persuasive power of familiar maxims
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 point of view, an appreciation of the ways in which
we can be unconsciously influenced matters because it allows us to
be on our guard against these processes, and because of how these
influences are able to explain some of the errors we make in our

As we shall learn about in Chapter 2, it is now very well known
in psychology and communication that arguments alone are rarely
enough to make us change our behaviours. Plato and Aristotle knew


this too. While Aristotle emphasised the value of rhetoric as com-
pensating for the limited intellect or attention span of audiences
(1991, pp. 75–6), Plato (through the voice of the teacher of rhetoric,
Gorgias) makes the point that:

It has often happened that I have gone with my brother and other doctors
to visit some sick person who refused to drink his medicine or to submit
to surgery … and when the doctors could not persuade him I have suc-
ceeded, simply by my use of the art of oratory.

(Gorgias, 2004, p. 18)

In the Phaedrus, Plato insisted that although knowledge of psychol-
ogy (‘the nature of the soul’) is relevant to the business of rhetoric,
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 prac-
tice, 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, for Plato, rhetoric is best taught and best practised
by philosophers.


There are two main places from which critical thinking emerges as a
discipline: philosophy and education. As we have seen, the spirit of
philosophy and the spirit of critical thinking are closely related, but
the analysis of arguments and reasoning is specific to the branch of
philosophy known as logic (the others are metaphysics, epistemol-
ogy and values (ethics and aesthetics)). Logic is primarily interested
in what is called ‘validity’ (see Chapter 4 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 conclusion on the basis
of certain premises. A simple example would be inferring that some-
one 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, then, logic in general is the study of arguments and reasoning,
informal logic is the study of arguments and reasoning as employed
in real-life contexts rather than in abstraction (which can be known
as ‘formal logic’). 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 represent these arguments in schematic
(abstract) forms, but a more complete understanding of them requires
contextual knowledge about the topic under discussion, the audi-
ence the argument is directed at, and certain characteristics of the
person presenting the argument. It is these considerations that infor-
mal logic seeks to make explicit, and in so doing it offers a systematic
approach to understanding and evaluating everyday arguments. This
involves skills concerning comprehension and interpretation (work-
ing 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 typical forms of reasoning and typical mistakes in
reasoning (fallacies); an understanding that is enriched by a range of
empirical considerations, 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 Anthony Blair and Ralph Johnson,
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.

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 condition 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.

The motivation of these philosophers is to make logic useful and
have a direct, practical application to daily life so that someone who
took a course in it would not only be better at logic, but better
deliberators among worldly affairs, and better academics and learn-
ers. Similar desires inspired the education-oriented ‘critical thinking
movement’ in the USA 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 criti-
cal thinking in American school curricula and sought, with some
success, to raise its profile and generate materials and techniques 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
passive 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
egocentricity – that have a deeper, more holistic effect on the
individual’s development. To use higher education scholar Ronald
Barnett’s expression (1997), students become ‘critical beings’
rather than just people able to argue. Dialogical context – arguments
encountered and analysed along with counter-arguments in
unfolding debates and discussions – Paul also sees as crucial for
nurturing this ‘strong’ conception of critical thinking (see Paul,
1984a). Having one’s beliefs subject to open-minded and respect-
ful questioning, along with a critical stance towards the positions
of others, encourages personal insight and responsibility. It is part
of what John 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 sine qua non [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) – Kuhn theorizes a progressively sophisticated way of under-
standing 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 these. Faced with this barrier,
they are liable to fall back on whatever first-hand experience they
have for explaining the phenomenon in question because, accord-
ing 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
theories. This perspective is correlated with critical thinking, and it
is clear why this is the case. If you think that knowledge is just there
to be discovered by those with the skills or experience, or that it is
just a matter of opinion and that multiple opinions have ‘equal legit-
imacy’ (ibid., p. 184), then argumentation is not likely to be seen as
valuable. So, a deepening appreciation of the complexity of branches
of knowledge along with the progress they have made helps establish
the value of argumentation. And as Kuhn points out, ‘people must
see the point of argument, if they are to engage in it’ (ibid., p. 201).


In this section I will address two issues associated with the ethical
consequences of critical thinking. The first is linked with its protective
function, and concerns critical thinking’s potential for empowering
individuals faced with the agendas of others, and the need to be
appropriately self-determining. Value is placed on knowing ourselves
and choosing the direction of our lives to the extent that this is pos-
sible; to be, as Harvey Siegel has it, rational and self-sufficient (1988,
pp. 55–61). Arguing for the virtues of critical thinking in this respect
can quickly elide with a generalised view of what constitutes an ideal
education in a modern liberal democracy.

Siegel offers a further reason for the ethical value of critical think-
ing in education: ‘democratic living’. He says:

Democracies rely for their health and well-being on the intelligence of their
citizens. … [S]uch 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

(Ibid., 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
effectively when engaging with the fast-moving, multi-faceted, multi-
opinioned, bias-laden debates that help define modern democratic
living. Although clearly informed by knowledge of a range of issues
and disciplines, 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 signifi-
cantly assisted by the knowledge, self-knowledge, and know-how of
critical thinking.

The second ethical consequence of critical thinking I wish to
discuss is less favourable, and concerns its place in the increasingly
instrumental approach to education that is now 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). But rather
perversely, some now see it as part of 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 trans-
ferable skills, be highly employable and ‘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 something else is striking. A well-respected
British university lists the following 16 attributes found in graduates:

• knowledgeable in their subject area;
• competent in applying their knowledge and skills;
• information literate;
• a skilled and ethical researcher;
• a critical, analytical and creative thinker;
• an entrepreneurial problem solver;
• someone who sees the big picture and understands the impor-

tance of context;
• experienced in working with clients, communities and partners

outside the university;
• an active citizen who respects diversity and has the cultural agility

to work in multinational settings;


• a flexible team worker;
• an independent learner;
• an efficient planner and time manager;
• an accomplished communicator;
• skilled in the use of IT;
• professional and adaptable;
• a well-rounded individual, reflective, self-aware and self-motivated.2

These entries, and the balance between the instrumental and
non-instrumental, are fairly typical across Anglophone universities.
‘Knowle dgeable in their subject area’ is at least top of the list, but is
quickly overwhelmed by the other 15 items, and there is no further
reference to characteristics such as enthusiasm for one’s subject, love
of learning, or students’ motivation to attain an ever-deeper under-
standing of the world around them.

The important point for our purposes is that critical thinking
appears on virtually all of these lists and is thus implicated 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 such
as critical thinking are certainly important, and of course many stu-
dents will want to know where their degree might lead them and
how it can take them there, but it is a matter of balance. We need
to avoid the implication that university is only for a certain type of
person; not all students will have or want a clear idea of future direc-
tions and are primarily at university to learn things about subjects
they care about. But perhaps more importantly, 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 empha-
sis 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 forgot-
ten. Understanding learning for the sake of a career and all that this
implies is easy for most people to understand and accommodate, 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 marketisa-
tion of higher education, it needs to be preserved.

It is not controversial to suggest that one facet of an emancipated
person from an educational perspective is their love of learning for
its own sake and being fascinated by the world (Peters, 1973). Certain
modes of critical thinking, however, conflict with this type of pas-
sivity. To the extent that critical thinking is the asking of questions
in order to further comprehension, then there is no tension between
it and an immersion in the world. But it also wants the learner to,
in a sense, take command of the situation through their scepticism.
They are encouraged to be active in the sense of being responsible
for making evaluations, but in an important, if unusual, sense of the
phrase, this stress on self-consciousness places the weight of the world
on their shoulders. A book teaching critical thinking is not going to
reject this perspective, but a balance is needed, and my belief is that
a full appreciation of critical thinking and what it is to be a critical
thinker accommodates this balance.

It is relatively straightforward to synthesise elements of the disci-
pline of critical thinking as a whole – especially if we incorporate
the more dialogical and non-adversarial approaches – to generate a
set of skills and attitudes that corrects corporate and other superfi-
cial appropriations. As Chapter 4 will demonstrate, of foundational
importance to the art of critical thinking is the ability to reach the
heart of the matter without over-simplification; the truth of what
the other person is trying to communicate. This is not the same as
speculating about symbolic and other underlying meanings, or writ-
ing exegeses of literary or arcane prose, but rather putting in order
and tidying up messages than are intended to have unambiguous, if
often complex, content. Critical thinking first requires open-minded
and generous reading or listening, and not until this is achieved can
we hope to offer an evaluative response.


In this book I seek to synthesise the most enlightening and use-
ful elements of these approaches. Chapter 1 seeks to explain rationality


with a particular emphasis on the psychology of judgement and
decision-making, and the role of emotions in argumentation.
Chapter 2 concentrates on critical thinking dispositions and
the nature of constructive dialogues. These opening chapters
involve, in part, critical thinking about critical thinking. Towards
explaining what I take to be its most valuable features, alternative
approaches to critical thinking’s subject matter and teaching are
considered and assessed. The conclusions I reach then shape the
remainder of the book where the focus is on the skills of argument

Chapter 3 looks at the nature of arguments and explains the art
of argument reconstruction – setting out the essential structure
and content of arguments expressed in natural language so that
they can be better understood and responded to. Chapters 4–7 are
concerned with a range of argument forms and their associated
fallacies. These include arguments from authority, ad hominem
arguments (those attacking the person rather than what the per-
son has to say), causal reasoning, generalisations, and arguments
from analogy. In each case my analysis will include discussions of
the persuasive (or rhetorical) potential of these argument forms
and fallacies; the psychological biases that predispose us to them;
their relationship with (the absence of) certain critical thinking dis-
positions, and their impact on attempts to generate constructive
dialogues. Chapter 8 explains some further fallacies that are impor-
tant to know about, and the book’s final chapter, the Conclusion,
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.

The book also contains a Glossary that provides definitions of key
words and phrases used in critical thinking. These key words and
phrases appear in bold in the text.


The examples used in critical thinking books usually involve gen-
eral subject matter – statements and arguments that do not require


specialist knowledge. Up to a point the strength of an argument is
determined by the form it takes (for instance, 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
purposes: first, their familiarity allows for a deeper understanding of
the kind of work that critical thinking does; and, second, they can
serve to illustrate how you should go about applying critical thinking
to the subject matter and situations (academic, professional, personal)
that you have significant knowledge of and responsibility for.

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 examples
and, where necessary, relevant background information will be
provided. These will come from a combination of academic and
non-academic sources, but in all cases I have tried to make them
as interesting 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
Harvard references with the full reference given in the bibliogra-
phy. Texts that serve only as examples of these ideas are fully ref-
erenced via the endnotes, but do not appear in the bibliography.

2. Since context (including dialogue) is so important for the proper
understanding and assessment of arguments, then students and
other readers will gain far more from applying their learning to
extended examples rather than brief and isolated ones. The 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 critical thinking, or
perhaps more pro�tably from the current a�airs of the place and
time you �nd yourself in. Students of critical thinking should get
into the habit of seeing patterns of argument and reasoning errors
in the world around them, and taking responsibility for �nding
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.

3. Extended arguments and argument-based narratives can be found
in various media, including documentaries, political speeches, and
works of �ction. As you will see, I have my favourites, including
Al Gore’s global warming documentary, An Inconvenient Truth
(Davis Guggenheim, 2006), Martin Luther King’s Letter from
Birmingham Jail (1963), a particular section of the �lm The Last King
of Scotland (Kevin MacDonald, 2007), and Naomi Klein’s treatise
on climate change and global capitalism This Changes Everything
(2014). But there are endless possibilities, and I can very happily
guarantee that the ideas and skills you will learn about in this book
will give you insight into, and the tools to analyse, any text you
choose that involves argumentation and rhetoric.

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
many references to it in the book, I will briefly explain the story. Set
in New York, a jury of 12 men have to reach a unanimous 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 punishment. 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 11 men to revisit the evidence, and as
they do, it starts to become apparent that the 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 the under-developed dialoguing skills of most of
the rest of the jury. From a critical thinking perspective the beauty of
the film lies not just with Juror 8’s ability to formulate and challenge
arguments, but with his capacity to stay calm, tolerate uncertainty,
not jump to conclusions, and remain independent from the pressure
of the majority (which often takes the form of personal attacks and
bullying). 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. Reasoning, rea-
sonableness and their opposites are put into sharp focus by the highly
recognisable behaviour of the characters. And the situation is not
trivial; life and death pivot on certain protagonists’ abilities to think
critically. It is old, it is in black and white, it is 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:


The aim of critical thinking is to make us better deliberators and
decision-makers through knowledge, techniques and a frame of mind
that do the following:

1. help us identify the sorts of questions we should be asking before
making significant decisions;

2. teach us about the pitfalls associated with reasoning in terms of:

i. weak arguments and their associated psychological biases;
ii. features of unconstructive dialogues;
iii. dispositions that make us prone to poor reasoning and uncon-

structive dialogues.



• Good primary sources for understanding the Socratic spirit and his
critique of rhetoric are Plato’s Apology and Gorgias (multiple edi-
tions are available). A good secondary source is David Melling’s
Understanding Plato (1987), Chapter 5.

• For a very good discussion of the wider contexts of argumentation,
including chapters on non-European and feminist perspectives,
see Berrill’s Perspectives on Written Argument (1996).


1 Known just as Informal Logic since the mid-1980s.
2 University of She�eld (2016) ‘The She�eld Graduate Attributes’. Available

at: www.she��eldgraduate/studentattributes



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)

I think I am smart unless I am really, really in love, and then I am
ridiculously stupid.

(Taylor Swift, The Guardian, 30 May 2015)

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 four
main headings: Rationality, Heuristics and biases, Emotions, and



In an investigation of critical thinking, two forms of rationality are
important to distinguish: ‘cognitive’ and ‘practical’ (or ‘functional’).
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. So, for instance,
Leonard’s denial that his wife might be having an affair despite her
acting out of character, displaying signs of guilt, ‘staying late at the
office’, coming home smelling of aftershave, her loss of interest in
sex (with him), and the reports of friends who’ve seen her out with
another man, is a likely case of cognitive irrationality. My Uncle
Vance’s claim to hate all pets, but loving my brother’s dog would also
appear to be cognitively irrational.

Practical rationality refers to the decisions we make and the actions
we take rather than to any beliefs that underpin them. A course of
action 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 Megan 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, makes the practically irrational deci-
sion to go out with friends.

Practical rationality requires self-understanding – knowing what
our wants are, and how these wants are prioritised – but it also requires
an accurate appreciation of how the world is in relevant respects
and of what needs to be done in order to achieve certain ends. In
this respect, it is reliant on cognitive rationality, but many practically
rational decisions are also made in the absence of cognitive rationality.


Cognitive psychologists will agree that decisions are influenced
by different forms of information processing, and from our own


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 instance, whether there are discrete systems of fast
and slow thinking, or a continuum running from automatic decision-
making through to decisions based on explicit and abstract reflection.

The more traditional 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 catego-
risation (or a continuum) could turn out to be more accurate. For
our purposes, the general distinction between relatively unreflective
and relatively 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)

We spend most of our lives operating in System 1 mode, 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 like 2 + 2. In contrast, System 2 is required
for more precise or unexpected driving manoeuvres, more complex
or unfamiliar sentences or problems to be solved. All critical thinking
goes on in System 2.


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 sit-
uations 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.


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. Many will not require such thought because they
are routine and predictable, but some demand our effortful con-
centration. 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 we make
choices about what we attend to, and this leaves a number of deci-
sions that would benefit from System 2 engagement in the hands
of System 1. System 1 will then get these right sometimes, and not
others, but whatever the case, 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 is
important for critical thinking:

1. Understanding the various heuristics we use can improve our self-
knowledge and motivate us to become less vulnerable to their
negative influence. It can, for example, contribute to a form of
self-awareness (referred to in Chapter 2 as ‘metacognition’) in
which we are more alert to their likely presence in certain situa-
tions and therefore able to employ strategies to avoid or mitigate
their bias.


2. They can themselves be educated by the practice of critical think-
ing. 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 author-
ity where it does not exist, can educate our quick as well as our
slow responses. It can, in other words, make our heuristics more

Since they systematically privilege some potentially relevant infor-
mation over other potentially relevant information, heuristics are, by
their nature, biased ways of thinking. However, their bias (meaning
‘unfair preference’) can be justified to the extent that it has the practi-
cal advantages mentioned. There are cognitive biases, however, which
do not have an obvious heuristic value. These can take the form of
predictable mistakes in fast and slow reasoning that are no more than
logical errors, or distortions of reality that are the product of deeper
motivations. We might say, then, that all heuristics are biases, but that
not all biases are heuristics. Unfortunately the language used to iden-
tify 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 (sometimes referred to as the ‘my side 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 indi-
viduals drop out of school significantly limited their interpretation
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 relation
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 evidence in
fact consists of.

(1991, p. 231)

In order to protect ourselves from the confirmation bias, we need
to develop the habit of looking for disconfirming evidence, but this
habit will be all the harder to acquire if we do not also become aware
of the range of other biases that contribute to it.


The confirmation bias is a variation on a pattern that is common to
many heuristics – seeing and interpreting events and ideas in a way
that preserves a coherent, but over-simplified, view of the world. And
since our world views are also partly the product of the other biases
and heuristics discussed in this section, its effect can be significantly
distorting. Richard Paul sees critical thinking as a corrective to ego-
centric 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

(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 over-estimate 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.

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
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 it 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 real reason makes us uncomfortable.
Often others will know that we are rationalising because the reason
given is not that plausible, or they have insight into our motivations.
In the moment, however (and perhaps for longer) the person ration-
alising 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 rationalisations;


and in stark contrast to his seemingly ultra-realist partner, the char-
acter 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.


In a classic study in social psychology (Lerner and Simmons, 1966),
participants who had been told they were taking part in an experi-
ment about ‘cues of emotional arousal’ found themselves observing
a peer receiving painful electric shocks in another lab experiment.
They were then asked to rate these people in terms of positive
and negative attributes. There were six separate conditions under
which this observation was made, and these varied in terms of how
unpleasant the situation was for the person receiving the shocks.
For example, in the ‘known reward’ condition, observers were
told the person was halfway through the experiment and that they
(the observers) could decide whether this subject should continue
receiving shocks, or should be subject to a neutral condition, or to
a reward condition (being paid 25 cents for each correct answer).
They voted (anonymously) and were told that the reward condition
had won. At the other end of the spectrum, observers had no control
over proceedings and were told that the person was only halfway
through the experiment, and so would be continuing to experience
painful shocks.

Counter-intuitively, the results showed that the better the cir-
cumstances were for the subjects receiving shocks, the more highly
rated they were in terms of attributes. Those suffering the worst
experimental conditions and over which observers could have no
influence, were rated the least favourably. This looks like kicking
someone when they are down, and in a sense it is, but Lerner and
Simmons hypothesised that this perception is motivated by a ‘just
world belief’. This is the seemingly quite widespread and often
unconscious idea that the world is fair and that people ‘get what
they deserve’. So, on observing participants’ suffering as a result


of nothing but bad luck, the observers’ just world view can only
be maintained by reaching the conclusion that somehow (because
they are not particularly good people compared to the others), the
participants deserve it. ‘Derogation of the victim’ is not uncommon –
consider some of the public opinion and media coverage of the
Syrian refugee crisis of 2015–16 – and ‘just world belief’ could be at
least part of the explanation.

For our purposes, this is an example of cognitive dissonance, a
term coined by psychologist Leon Festinger in 1957 referring to the
discomfort we feel when there is a contradiction between what the
evidence is showing us, and some deeply held belief about our self
or the world. Faced with this tension, we are prone to maintain the
belief at the cost of altering the reality we are faced with. In this case,
a just world belief is sustained by appraising the person as someone
who is deserving of their unfortunate fate.

Cognitive dissonance is an aspect of a broad set of theories, now
well established, that point to a profound need for consistency
between our various beliefs, our beliefs and our actions, and our
actions and our self-image. Once we commit to something (a cause,
a belief, a course of action) ‘we will encounter personal and inter-
personal 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 power-
ful techniques for ensuring 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 gas
and electricity?’ I am 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 5.


To avoid being trapped by the consistency heuristic, we need to be
careful what we commit to, both in terms of promises made and the
views we hold. In Chapter 6 (Section 6.2) we will consider how
absolutist statements and arguments can, through the need for con-
sistency, force us to defend unacceptable or implausible positions.


According to anthropologists, rules of reciprocation – a fair bal-
ance between what we receive and what we give – are found in all
cultures. Even if it is in the form of a ‘thank you’, a favour done mer-
its something 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

In Caveman Logic (2009), 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’ (ibid., p. 49).

Persuasion specialists know that an unsolicited gift in a charity
appeal envelope or as you step into 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
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. (Also at work here is the ‘contrast
effect’, which you will learn about below.)


In the world of negotiation, there is something known as a ‘recip-
rocal concession’. Early on in Twelve Angry Men Juror 8’s ‘not guilty’
vote means that a stalemate has been reached. On the one hand,
none of the other jurors seem willing to talk about the case, but, on
the other, 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’ll change his vote to ‘guilty’, but if there’s one more vote
for ‘not guilty’, then they must agree to discuss the case further. The
riskiness of this offer from Juror 8’s perspective 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.


The representative heuristic attempts to provide quick answers
to questions like ‘What’s the probability of individual X belonging
to group Y?’, or ‘What the probability of event P being the cause of
event Q?’. It does this by making reference to 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 it will pick.
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 representative 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 con-
clude 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,
a definition of which is an inaccurate generalisation about core fea-
tures of a particular category. Representativeness only has a chance of
providing the right answer if our idea of the typical group member
or causal factor is accurate in the first place. (More will be said about
stereotyping when we discuss generalisations in Chapter 6.)

The other kind of mistake caused by the representative heuristic
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
discrimination and social justice, and also participated in antinuclear
demonstrations. 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 informa-
tion 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 heuristic causes us to overlook
the conjunction rule (that the likelihood of X + Y is smaller than X
by itself), and thus to commit the conjunction fallacy. Tversky and
Kahneman also claim that representativeness is behind the common
fallacy of ‘base rate neglect’.


Base rate neglect (or sometimes ‘background’ rate neglect), is a
remarkably widespread fallacy. It refers to situations in which, when
making a judgement about a particular person or event, we over-
look relevant background information. 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). Asked if, for exam-
ple, Steve is more likely to be a librarian or a salesperson, most
of us will assume the former. However, this overlooks: (1) the fact
that there are many more salespeople than librarians; and (2) that
there is nothing in this description of Steve that precludes him from
being a salesperson. Our representative heuristic causes us to jump
to the conclusion that he is a librarian because his personality 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.

My parents used to live in a small rural town that was quite arty
and, because many of the younger people would leave to study and
find work in cities, it had a relatively old population. Sam, who lives
in this town, plays in a band, enjoys the occasional mosh pit, and is
partial to illegal stimulants. Is it more likely he is in his twenties or
his fifties? Our representative heuristic says twenties, but because of
base rates, it is considerably more likely he is in his fifties.

Ignoring or suppressing base rates can be used to persuasive effect.
A story that has made the news more than once in recent years is that
eating a small daily amount of processed meat increases our chances
of contracting colorectal cancer by 18 per cent. This sounds scary for
bacon roll eaters like me, and all the more so because news sources
are in no hurry to provide the absolute figures. The background rate
is that 6 in 100 people contract bowel cancer, and for those who
consume 50g or more of processed meat a day (about two slices of
bacon), this goes up to 7 in 100. This is less scary, and it will not
change my diet, whereas an 18 per cent rise from 30 in 100 to about
35 might.

Supporters of Scottish independence quite often use the argument
that Scotland has only had a representative government in the UK
Parliament for 34 out of the past 68 years. This sounds like quite a
serious problem with the union until you realise that it must be true
of many regions of England and Wales as well. Without a system of
proportional representation, it is an unavoidable outcome, but the
statistic presented in isolation can present a distorted perspective of
democratic norms.

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 50 per cent price rise appears dramatic,
and normally is. However, a 50 per cent rise in the price of lollipops
from 10p to 15p (assuming your entire diet is not based on them)
is probably not a cause for outrage because the base rate is so low.


Background rate neglect is quite a common error, and to an extent
we are used to arguers being called out on it in dialogues. A devel-
opment of this is the fallacy of suggesting false background rates in
order to minimise or normalise a situation that is in fact not trivial or
normal at all. In a speech relating to 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. It doesn’t happen in other places with this kind of frequency.2


The availability heuristic (also named and studied by Tversky and
Kahneman) judges the likelihood of an occurrence on the basis of
how readily it comes to mind. Availability is a fairly reliable meas-
ure of likelihood because common occurrences are often those that
come to mind with greater ease. I do not follow German league
football, but if asked whether Borussia Dortmund or Bayern Munich
has won their top division the most times in the past thirty years
Bayern Munich would immediately present itself as the answer,
and it would be the right one. Bayern’s presence in the Champions
League and the number of German internationals who have played
for them mean they make an impression on my memory that other
German club sides do not.

However, the availability heuristic can also predict certain kinds of
mistake. In a study where participants were asked to guess the rela-
tive likelihood of different causes of death, most would rate things
like accidents or tornadoes as significantly more likely than less ‘avail-
able’ causes such as asthma, strokes or diabetes.3 The reverse is true
(for example, at the time asthma caused 20 times more deaths than
tornadoes) but factors such as news worthiness, vividness, and emo-
tional impact mean that they come to mind more readily, skewing
our judgements. You would imagine that a contemporary study in
the UK (and possibly elsewhere) would reveal similarly exaggerated
ideas about the frequency of child abductions and terrorist attacks.

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 representative of their relative likelihood in comparison to more
mundane explanations.


The anchoring and contrast effects are examples of how the
contextualising of information can influence our perceptions and
interpretations. Anchoring happen when our judgement of a quan-
tity (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, while
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). A likely explana-
tion is that the initial figure is interpreted as a plausible (if extreme)
answer, and the individual’s subsequent deliberations keep this in
mind (are ‘anchored’ to it). Anchoring is known to be a pervasive
and powerful phenomenon in a range of real-life as well as labora-
tory settings. In negotiations there can be an advantage in making
the first move because even a demand expected to be excessive
will have a disproportionate influence over the range of prices (or
whatever is at stake) that is considered to be acceptable. When two
academics grade a student’s assignment, a more objective assessment
is made if the second marker is ignorant of the first marker’s grade.
Remaining ignorant is one way to counter the anchoring effect,
and but where this is not possible, a disciplined mind is required, for
example, undertaking prior research into acceptable price ranges in
a negotiation situation.

The contrast effect refers to the influence of nearby comparators
with contrasting qualities. Any parking ticket I’ve 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. £35 is still a lot of money to pay for
overrunning a meter by ten minutes, but it looks much more rea-
sonable when you’ve just been threatened with a £70 fine.


The contrast effect is commonly used as a sales and fund-raising
tactic (‘If you cannot afford £100, could you then donate £10?’),
but it has a psychologically protective function as well. In a study
looking at the coping mechanisms of women diagnosed with breast
cancer, a downward comparison was typically employed. For exam-
ple, 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.4 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 the people of
Newtown after the Sandy Hook shooting in 2012, President Obama
began with a piece of scripture emphasising the temporary nature of
our earthly lives in comparison to God’s eternal perspective. Under
the circumstances it is hard to see what else he could have done by
way of consolation.


When making quick judgements (in comparison to slow ones), we
are significantly more influenced by the characteristics of the people
presenting the 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, two features of the speaker’s (or
writer’s) character have been shown to be especially persuasive:
(1) the extent to which we perceive them as an authority; and
(2) how likeable we find them.

Forms of authority will be considered at length in Chapter 5,
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 par-
ticipants (by rewarding them with money for a prior task), and the
other half remained in a neutral state. Everyone was then asked to


listen 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
listened to a speech containing strong arguments, half to one con-
taining 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 influ-
ence on their assessment of the speeches. For the neutral group,
expert source had no significant 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, and
how much they like us. Of pronounced importance is how trust-
worthy (or reliable) they are, and their degree of perceived similarity
to us. Similarity can thus act as a heuristic. Short of time and infor-
mation about a product or course of action, its being promoted or
modelled by someone whom I identify with can serve as a short cut
to deciding in its favour; if it works for them, it might work for me
too. More will be said about the role of identity in argumentation
and persuasion under referent power in Chapter 5 (Section 5.5).


I used to hitchhike a lot in my twenties. Often I’d be the only person
standing at a service station or a roundabout thumbing a lift, but
when there were other hitchers waiting as well a phenomenon I
noticed was 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 conditions of uncertainty, such as what to order from
an unfamiliar curry house, what an acceptable volume of speech is in
a particular public library, or indeed the appropriateness and safety of
picking up hitchers in modern Britain, the behaviour of other peo-
ple becomes an important (if you will excuse the pun) rule of thumb.


Basing judgements on social proof can, of course, be unreliable.
For example, in the phenomenon known as ‘groupthink’ (Janis,
[1972] 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 intel-
ligent people to make very poor decisions. Irving Janis’ theory was
based upon mid-twentieth-century US foreign policy disasters such
as the Bay of Pigs fiasco in Cuba, and the escalation of the Vietnam
War, but aspects of his highly detailed analyses can readily be applied
to group or committee decision-making in a wide range 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 information and
points of view held by individual members only. It could be the case
that, put together, these individually held views amount to a strong
counter-argument, but a collective confirmation bias obscures this

Of particular relevance to social proof is the phenomenon of
pluralistic ignorance. Imagine you are sitting on a high-powered
committee that needs to decide on an important and urgent mat-
ter, and that you are feeling unsure about a piece of information,
an argument, or a decision about to be made. If you are reluc-
tant to admit your ignorance, or to give the appearance of lacking
conviction, then you will quite likely (while 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 majority 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 predictors of groupthink (such as the presence
of a dominant group leader intolerant of dissent, a strong sense of
identity or similarity among group members, and a crisis situation
in which emotions are running high) can be an inadequate evalu-
ation of alternative courses of action and more than likely a poor



Emotions are typically seen as being irrational and getting in the
way of critical thinking. There is some truth in this, but perhaps
not as much as we are inclined to think. In contemporary research
emotions are often viewed as forms of heuristic, assisting our more
immediate judgements, and they can also be viewed as ‘intelligent’
in a deeper sense as well. It is through these understandings that
the role of emotions in arguments will be illuminated, providing
us with the necessary tools for assessing what are called ‘appeals to


Emotions can be seen as cognitively rational when they are based on
accurate beliefs and judgements about the world. There are plenty of
occasions when we regard an emotional response as rationally justi-
fied, or where it would be irrational 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)

Anger is, generally speaking, justifiable when: (1) something frustrates
us – i.e. stops us from getting what we want; and (2) this is unfair or
unjust. So, if someone at work gets an undeserved promotion at our
expense; or when we witness or read about a legal or political injus-
tice, it is rational to be angry.

However, would we also not want to say that emotions can be
significantly irrational? For example, someone crying inconsolably
for a whole night after seeing Romeo and Juliet, the ‘red mist’ of
uncontrolled rage; phobias; obsessive jealousy, or the dizzy dispro-
portionate rollercoaster of romantic love. Our emotions appear to
make us think and do crazy things, but it is important to ask whether
it is the emotion per se that is irrational, or the belief that causes it that


is irrational. As philosopher Robert Solomon once put it: ‘Emotions
are not irrational; people are irrational’ (2003, p. 235).

Considering the biases and heuristics that have been the subject
matter of this chapter, this cannot be denied. For Solomon, the
emotions we feel when we are being cognitively irrational simply
correspond to these biases. So, when a small failure is pointed out
to him, it is the narcissist’s false belief that he is generally superior to
others that causes his excessively strong reaction of anger. However,
in a sense, the emotion – the strong reaction – is not the culprit, but
rather the distorted belief which underpins it.

This is fine as far as it goes, but it does not seem to explain a num-
ber of features of emotions, including the effects they have on our
perceptions and judgements once we are experiencing (or undergo-
ing or in the grip of) them. Very strong emotions can lead to what
is been called ‘cognitive incapacitation’: a situation in which we are
unable to think clearly. Strong emotions in particular are associated

• 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 focused 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, land-
owner George Middleton is throwing a New Year’s Eve party, and
he is a very happy and relieved man. He is deeply in love with
his fiancée Charlotte, and has recently recovered, against the odds,
from an operation to remove gall stones. In the kitchen the wait-
ing coachmen, 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 disputatious-
ness, 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!’

(Ibid., p. 299)5

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 is irrational? Presumably it is
the effects of the emotion itself; an example of what is called the
‘transferral of affect’ where the feeling expands beyond its rational
origins and attaches itself to irrelevant and inappropriate objects and

Strong emotions tend to distort and impede rational thought,
but mild emotions and mood 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 condi-
tion the judgements they make.6 This effect is sometimes referred to
as ‘emotional framing’, and is a valuable persuasion technique (for
more on framing in general, see Section 1.4). If a speech-maker finds
an angle on a topic that will predictably induce, say, anger in her
audience (e.g. terrorist threat), she can then use the resulting mood
to elicit the response she wants to promote other issues that would
be less persuasive were the audience not already in an angry frame of
mind. For example, in many countries, immigration is a controver-
sial topic that is associated with a range of emotions, including anger,
fear, pride, and compassion. If a politician is anti-immigration, then
he will want to play on anger and fear, and he will gain an advantage
if he can introduce the topic to an appropriately primed audience.
To put it another way, emotions and moods embody a confirmation
bias; when under their influence, we are, often quite unconsciously,
inclined to seek out features of our surroundings that confirm our



If emotions cause irrationality in the cognitive sense, they can also
assist more habitual forms of decision-making by serving as heu-
ristics (referred to sometimes as the ‘affect heuristic’). In recent
decades, Antonio Damasio’s (2000) ‘somatic marker’ theory has been
highly influential. The idea is that our initial response to many situ-
ations is an emotional one – broadly positive or negative – and this
‘feeling of what happened’ then serves to guide our attention 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, through helping to capture and then focus
our attention, emotion is a vital component of rational action. I’m in
the kitchen cooking and I hear my 3-year-old 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 (and that’s immediate as well), my compass quickly switches
and I see his younger brother walking on his sandcastle. Previous
experience has mellowed any potential irritation at my son’s extreme
reaction and instead empathy is triggered and suitably understanding
words and actions follow. The situation is sorted.

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 I’m encountering
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
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:

[M]any of the judgements we make are often too complex to review all
the relevant evidence. For instance, a comprehensive answer to the ques-
tion of how satisfied you are with your political leader might lead you to
think about current environmental policy, the state of health care, unem-
ployment 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 current
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 heuristics need
to be put aside and I need to be dealing with the evidence on its own
terms rather than on the basis of a system designed to find coherence
with existing schemas. By putting them aside I am able to make a less
biased judgement, and I’m also able to educate my heuristics. When
I return from my reflections back to the world, as it were, my ‘gut
reactions’ are better informed.

By itself then, what’s been called the ‘affect heuristic’

appears at once both wondrous and frightening: wondrous in its speed,
and subtlety, and sophistication, and its ability to ‘lubricate reason’; fright-
ening in its dependency upon context and experience, allowing us to be
led astray or manipulated – inadvertently or intentionally – silently and

(Slovic et al., 2007, p. 1349)


Emotions are intelligent; they are generated by what is currently
important to us, and they can also serve as reminders of what we
might otherwise overlook, or not look closely enough at. 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
at injustice, sadness at tragedy, fear at 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, feel-
ing pity motivates empathy, such that the pitiful image or vignette
engenders 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 pity is justified.

However, even if the pity is justified, true stories are rarely
straightforward, 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 Aylan
Kurdi featured on news media across the world, and had a gal-
vanizing effect on attitudes towards the Syrian refugee crisis in
some countries.

The image of the dead body – dressed like any other little
boy – on the beach and in the arms of a policeman, was unbearable,
particularly for anyone who has children of this age. 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.7

Indeed, so the emotion prompts a more empathetic engagement with
the severity of the 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 ‘availability’ it affords distorts the argument.


How, then, are we to counter this risk of imbalance? A couple of
suggestions are these:

1. Where the situation lends itself to this, a balance of emotion-led
reflections can help. In the Syrian case, several emotions will be
relevant to different facets of the situation; as well as compassion
for the refugees, there might be forms of personal and national
guilt, anger towards Assad and IS, and fear of the effects of mass
migration. This is a somewhat risky approach, however, since it
assumes that all important aspects will be emotionally compelling,
and that enough relevantly detailed stories will be accessible.

2. Part of the answer might be developing an intellectual apprecia-
tion of this kind of bias – as is the intention here. But knowing in
this abstract sense is rarely enough to influence behaviour, espe-
cially when up against the kind of engagement emotions bring. In
support of this, we need a range of what have been called ‘epis-
temic virtues’, and what are referred to in this context as ‘critical
thinking dispositions’. These include the ability to be flexible
in our thinking, to detach ourselves from what is currently com-
pelling (or our personal commitment) and assume a less biased
perspective motivated by the desire for truth and objectivity.
Chapter 2 will discuss various critical thinking dispositions, and
their relevance to different types of argument will be a theme of
this book.


There is a category of arguments known as ‘arguments from emo-
tion’ or ‘appeals to emotion’. In theory, this could include any
emotion you care to name, but analysis is weighted towards those
which are more common or ‘basic’, and those which have tended to
be associated with rhetorical techniques: appeals to fear, pity, guilt,
pride, anger (or indignation), and humour (or amusement).

Unlike other argument forms that will be covered in this book,
appeals to emotion tend not to have a clear structure, but instead
piggy-back on other types of argument. For example, slippery
slope arguments (see Chapter 6) gain much of their persuasive
traction from inducing fear in audiences. Also, the subject matter of


most arguments will involve emotions to some extent; usually we
stand to lose or gain something depending on whose conclusion is
right or accepted, and what we lose or gain can cause us to feel fear,
guilt, anger, pity, joy, pride, and so on.

For this reason, the preceding discussion of emotions 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 cloud our judgement. 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
the key distinction we need to make when assessing the emotional
effects of arguments is between those arguments that aim to gener-
ate emotion in order to persuade, 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 who 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 Jim
Sturman QC, defending Ms Clarence, said 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’. What is implied here, however, is not that being moved by
the events will bias our judgement, but that being moved is an appro-
priate response to the circumstances. The pity 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 understood legally (and

Put another way, when the defence lawyer explains the circum-
stances of the killing, he is offering premises that are relevant to the
conclusion (that Tania Clarence is not responsible for her actions).
These premises arouse pity in us, but the lawyer’s aim is not to use
this pity to get away with a poor argument, it is a natural consequence
of the subject matter under discussion. The lawyer’s reference to being
‘moved’ becomes a short-hand for a rightful judgement of the case.

As another kind of example, consider the difference between
the student who asks for an extension for their assignment deadline


because of tragic family circumstances, and the student who burst
into tears in my office before telling me that he cannot submit his
essay because his computer broke, losing all his files. Of course,
this second example might not be designed to be manipulative,
but for the sake of argument let’s assume it is. Let’s assume that he
knows that this is a poor excuse (whether or not it is true), but that
if he is able to generate pity in me from the start, I will be suitably
softened up. ‘Softened up’ here means my pity for him makes me
more receptive to the tragic facets of his situation, and therefore
more likely to be biased in his favour when making a judgement
on an extension.

In order to better protect ourselves from the problematic influ-
ence of emotions in arguments, we need to ask these questions:

Q1: Is this an issue that has generated strong emotions in me?
Q2: Is the arguer attempting to cloud my judgement on the issue by

invoking strong emotions instead of presenting strong arguments?
Q3: Do the emotions arise as a natural consequence of the subject

matter of the argument?
Q4: If so, are the emotions generated representative of a balanced

view of the issue?
Q5: If so, is it the strength of the argument (including insights assisted

by the emotions evoked), rather than the feelings themselves
(i.e. the affect heuristic) that leads me to accept or reject the


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 empha-
sising some aspects of the matter at the expense of others.

Whereas some might say that my Uncle Vance is an alcoholic,
others would say he’s a ‘party animal’. In his Autumn Statement
in 2015, the UK Chancellor of the Exchequer, George Osborne,
backed down on tax credits cuts after pressure from the Lords and


from the general public. The opposition said this was a ‘U-turn’,
his supporters said it demonstrated how he is flexible and prepared
to listen.

All heuristics and biases frame situations because they are system-
atically selective in what they lead us to take from them. Another
form of framing is the use of figurative language. Metaphors, analo-
gies and allegories try to explain or illuminate some aspect of the
world by comparing it with something recognisable to an audience.
Arguments from analogy work in a similar way, and this will be the
subject matter of Chapter 7.

Broadly speaking, there are two purposes 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 audience
is currently familiar with; a basic educational process. In Naomi
Klein’s book, This Changes Everything (2015), analogies with the
anti-slavery, women’s and civil rights movements form a central
argument for why rapid and radical change to the economic sys-
tem that sustains climate change is in fact achievable. Alternatively,
framing can provide a way of coping with a situation. For exam-
ple, I have always felt that it is healthier for students to see exams
as a ‘challenge’ rather than something to be scared of or just to
‘get out of the way’. (See also 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 ideologies,’ 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.8

Things become more interesting when both sets of needs are relevant.
In his Letter from Birmingham Jail, Martin Luther King 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 law-breakers from history and the
Bible – such as Jesus, Paul, Lincoln, and Jefferson – who are founda-
tional to American values. He then reframes the issue: ‘the question
is not whether we will be extremist, but what kind of extremists we
will be’. 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 provide a potentially helpful angle for an educated audience to
engage with his argument. Unlike propaganda, it is part of a respectful


Rhetoric, as explained in the Introduction, is the art of persuasive
communication. Arguments also aim to persuade, but a good way to
distinguish between the two is, whereas arguments appeal to System 2
thinking, rhetoric appeals to System 1 thinking. 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) do involve arguments. However, the qual-
ity 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 rea-
son that it is important to learn about fallacies (weak arguments that
have the appearance of being strong), and to understand our cogni-
tive frailties as we have been doing in this chapter. These two sets of
information will help protect us from forms of persuasive communi-
cation that benefit from keeping System 2 asleep in the back.

It should also be recognised that rarely (if ever) can argumentation
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. We have already seen
two ways in which this occurs: one in the section on arguments
from emotion (for example, the unavoidability of pity in the Tania
Clarence case), and the other in the section on framing regarding
Martin Luther King’s careful choice of analogies. We should also
bear in mind that all arguments have a source and a medium and


these will rarely be persuasively neutral (think of authority and like-
ability, and see as well the ‘halo’ and ‘horn’ effects to be discussed in
Chapter 5).

When we are practising argumentation rather than rhetoric, it
would be wrong to use these features of arguments to gain an advan-
tage, but awareness of them can be used more legitimately if it means
the difference between someone listening to your argument and
someone switching off. In their book on ‘principled negotiation’,
Fisher and Ury make a helpful suggestion along these lines:

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
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)

This is not about obscuring or twisting the truth, but rather a sugges-
tion for how we compose our arguments in certain situations so that
they are heard.


From what you have read in this chapter, draw up a list of things
we could do to protect ourselves from the dangers of heuristics
and biases. This will include ways to prevent poor decision-making
in groups; protection from those who will deliberately deploy
persuasion techniques in order to influence us, and ways to assist
the management of our emotional responses to arguments. 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 on the basis of the chapter’s content.


• 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) who explores how heuristics are exploited by professional

• For methods of debiasing, see Larrick (2004), and for an evalua-
tion of dual process theories of reasoning, see Osman (2004).

• Hank Davis’ Caveman Logic (2009) is primarily an argument for
the dangers of heuristic reasoning. In contrast, Gerd Gigerenzer,
Peter Todd, et al.’s Simple Heuristics that Make Us Smart (1999),
argues that they have more value than we have tended to assume.

• For a review of the anchoring effect, see Furnham and Boo

• For a recent analysis of groupthink, see Baron (2005).
• A very good philosophy book on emotion and rationality is

Brady (2013).
• And a good psychology source is the chapter ‘Emotions and

Cognition’ in Keltner, Oatley, and Jenkins (2013).


1 Roads to Freedom (London: Routledge, 1996), p. 116.
2 For a transcript, see

3 Slovic et al., as cited in Kahneman (2012), p. 138.
4 Discussed in Brown (1986), pp. 165–7.
5 Reproduced by permission of The Random House Group Ltd.
6 For an experiment along these lines, see DeSteno et al. (2004).
7 A. Withnall, (2015). Available at:


8 Wisdom, Information and Wonder (London: Routledge, 1991), p. 48.



It is not enough to have a good mind, rather the main thing is to apply
it well.

(René Descartes, Discourse on Method [1641] 1968)

My husband says I’m overly sensitive to criticism. BUT WHAT DOES

(Sacha T. Burnstorm, pers. comm.)

Learning about heuristics and biases is not the same thing as learning
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 do not just ‘do’ critical thinking,
but ‘become’ a critical thinker. ‘Becoming’ a critical thinker, how-
ever, does not mean some cultish, full-blooded transformation in
your personality, as seems to be suggested by some theorists. Harvey
Siegel, 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’ (1988, p. 41). 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. Even if creating ‘a certain
sort of person’ is possible, this will not be a desirable profile for many
of us, but this should not stop us attempting to develop and enhance
critical thinking dispositions. For each individual, these tendencies
will merge 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

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 moti-
vating 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 moved by this omission; disappointed in ourselves,
perhaps angry.

The second point concerns the distinction between encouraging
dispositions that will tend to make us better critical thinkers, and
seeing critical thinking as a discipline that will foster these disposi-
tions. Writings in this area tend to be framed in terms of the former,
but the latter is implied as well. Therefore, answers to Robert Ennis’
question, ‘What dispositions does an ideal critical thinker possess?’
(1996b) tell us: (1) which characteristics we need to develop in order
to have a readiness for, or be predisposed towards, thinking critically;
and (2) which characteristics can be acquired as a result of learning
critical thinking knowledge, values and skills. These will, of course,
be mutually reinforcing, but from an educational point of view, the

Justin Kitchen
Disposition approach
Justin Kitchen
Second introductory remark
Justin Kitchen
First introductory remark


second should be the primary aim. Education will develop disposi-
tions 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 in order to be a better critical thinker
seems to put the emphasis in the wrong place. Instead, the knowl-
edge, skills and values of critical thinking are taught, and we then
expect certain dispositions to develop out of this culture.

There have been a number of attempts to formulate lists of criti-
cal thinking dispositions, including those by Richard Paul (1995,
Chapter 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 I
believe have the most direct influence: love of truth, open-mindedness,
flexibility, modesty, self-knowledge, meta-cognition, and what I’m
calling ‘dialogical dispositions’. Where appropriate, however, I will
also indicate related or complementary dispositions.


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 on
the basis of rational enquiry – the use of reason and evidence. It is
important to recognise that we can have a love of truth, but not a
commitment to rational enquiry as a means of attaining it. Instead
we could regard notions like faith, feeling or intuition 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 pro-
vide an argument for why these domains require a different type of

Justin Kitchen
critical thinking dispositions
Justin Kitchen
Love of truth


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 love of truth, and it
will leave us especially vulnerable to the confirmation bias and
other ego-defensive biases that serve the self rather than objective
knowledge. 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 it is arguably this need
that dampens his inquisitiveness and makes him unwilling to go to the
lengths of Juror 8 in scrutinising the arguments put forward by the


Since seeking truth requires us to listen to the views and reason-
ing of others, and an appreciation of the fallibility of our own
beliefs and convictions, then open-mindedness must be a fun-
damental disposition 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 what premises present themselves
and what conclusion these should then 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 her indi-
vidual perspectives and the value of other perspectives (and thus the
value of dialogue) as a way of opening her mind. Often other peo-
ple will have positions and arguments that we had not thought of
before listening to what they are saying. Open-mindedness entails a
willingness to change one’s mind, either in the direction of another’s

Justin Kitchen
Desire to be right vs. Love of truth
Justin Kitchen


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 main-
tain and defend, regardless of whether or not they are true or coherent.

(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 accompany 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 par-
ticipant 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’. 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 philosophers
writing in this field – such as Valerie Tiberius (2008, Chapter 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 – to function well in the
world in spite of – uncertainty.

Endurance is also required in circumstances where open-minded
decisions need to be made. As we know, System 2 thinking is

Justin Kitchen


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 psychologist Irvin
Yalom (1980, p. 312) calls ‘simultaneous ambivalence’. This is the
result of remaining clearly focused on the for and against of both (or
all) the options available, rather than letting one of them dominate
our attention and, therefore, that way incline us to a less trouble-
some, but biased, decision.

As well as the need for conviction, we must also recognise that a
vital feature of a fulfilled life is substantial time spent being ‘unre-
flectively absorbed by what we value’ (Tiberius, 2008, p. 67). This
is also a different psychological mode to critical thinking, and one
that is similarly at odds with it. If both this type of absorption and
reflective thinking are so important, then we can see how being
‘ready’ (to use Bohm’s expression) to move between the two is also


Modesty (or humility) is primarily understood as possessing accurate
perceptions of ourselves and the status of our beliefs in relation to
others. The exaggerate pride in being right we identified in 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 generalised sense of equality. They will thus bring
an absence of egocentricity to a discussion that allows for more open

Also, however, modesty is important for critical thinking for a
slightly different reason. Critical thinking is empowering in terms
of the insights it provides, and how it improves our ability to suc-
cessfully interrogate the arguments of others. These are no mean
abilities, and have the potential to create a sense of superiority in the
learner. Modesty is a corrective to this. It will incline us, for exam-
ple, to realise that critical thinking is just one among many important
practices; that virtually anyone can become a better critical thinker if
they commit themselves to learning it; and, perhaps most important
of all, to recognise that if people are not schooled in this way, it
does not follow that: (1) they are unintelligent; (2) they do not have

Justin Kitchen
Modesty 1
Justin Kitchen
Two benefits of modesty


worthwhile beliefs; or (3) they are not worthy or respect. (For more
on this kind of issue, see the section, ‘Don’t be a smart arse’, below.)


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 actual
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 they
will not always be something that critical thinking believes we
should be prepared to bracket in the context of a discussion. That is
a choice for an individual. However, what is important for critical
thinking is that we are aware of what these deep-seated commit-
ments are. One reason for this is that it allows us a clear choice about
what we do and do not want to put out there for critical appraisal
by others. A similar reason is that it permits us to assess which of
our assumptions we want to question and which we do not. Also,
foundational beliefs will permeate many of our other beliefs, so rec-
ognising what Walton refers to as ‘dark side commitments’ (1992,
p. 255) will provide important premises in arguments concerning a
very wide range of topics.


Closely related to self-knowledge is the disposition to be meta-
cognitive. The most specific meaning of this term refers to an
awareness of our thought processes, and is exemplified by ideas like
‘alertness to loss of control of one’s thinking’, and ‘the impulse to
stand back and take stock’ (Perkins et al., 1993, p. 8). It is not about
formal knowledge of our cognitive biases, but rather the disposition
to monitor and assess the quality or mode of our thinking in differ-
ent situations. An important aspect of this concerns knowing when
to think critically and when not to. For example, Perkins et al. refer to
the importance of the ‘detection of complex thinking situations’


(ibid., p. 8), which will include knowing when to switch from System 1
to System 2 thinking.

David Bohm discusses the ‘blocks’ we have that make us unaware
of some of the contradictions in our beliefs, and therefore less open-
minded. To assist in our understanding of these blocks, he suggests
we become sensitive to mild emotional responses to ideas that we
encounter. ‘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)

With reference to ideas discussed in Chapter 1, 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 meta-cognition is simply not welcome
or necessary: painting for pleasure, unself-conscious dancing, and
an evening with a Breaking Bad 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
meta-cognition’s snooze button.


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 in particular 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 present arguments to convince others of our position; arguments
provoke questions and counter-arguments, and via this process the
open-minded, flexible, self-aware person should be able to edge
closer to the truth about the issue under discussion. Good quality
dialogues are thus profoundly important, and so the ability to con-
duct them constructively has equivalent importance. In part, this is a
matter of knowing and applying certain rules (see below), but there
are also dispositions 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 pos-
sible. 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, and not being a ‘smart arse’ –
before providing some basic rules for constructive dialogues.


One reason we might be reluctant to be clear about our reasons
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 significant social
pressure to conform. Twelve Angry Men and the Milgram experiments
(which will be discussed in Chapter 5) are dramatic examples of
pressures associated with groupthink and forms of authority, but as
we saw in Chapter 1, the underlying processes are common. Social
media is a prime example; one academic researcher writes about how
she came to the 2015 UK general election as a floating voter who
raised questions about the policies on all sides, but who was con-
fronted ‘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 a number of reasons, including ‘rocking
the boat’ (destabilising an established, possibly hard-won, equilib-
rium). In certain circumstances or at certain times, critical thinking
(or at least its expression) is not appropriate, and, depending on the
exact context, ‘boat rocking’ could be one of them. However, as
the previous discussion of the characteristics of groupthink demon-
strates, this is often not a judgement call that is easy to make. It takes

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 pre-
ordained order to our lives (no final set of truths or essential self to
be discovered). At a cultural level, Kant meant something like this
when he described the Enlightenment as an emergence from a ‘tute-
lage’ that is ‘self-imposed’, not by an inability to reason, but by a lack
of ‘resolve and courage’. Critical thinking gains much of its signifi-
cance from a profound freedom that comes with the understanding
that existence has no ultimate answer or purpose. Words like ‘active’
and ‘judgement’ serve as reminders that in a very important sense
our lives are what we choose to make of them. There is a freedom
and 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.


In dialogues it is very easy to become side-tracked, so the critical
thinker always tries to stay focused on the overall point of the discus-
sion. Losing focus can happen by accident, but it can also be the aim 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 Chapter 5), red herrings
(see Chapter 8) and straw man arguments (see Chapter 3).

Part of the art of staying on track is asking the right questions at
the right times. Ennis (1996a, pp. 373–5) identifies various types of


• ‘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 focused’ might be persistence or persever-
ance, and it is noteworthy that John Dewey includes the former in
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.


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 orientations,
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

(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 words, as epistemic dispositions. Respect and concern for
others – qualities that seems 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 though that, while recognising it is not a defining
characteristic, Ennis sees the need to include ‘care about the dignity
and worth of every person’ among his list of critical thinking dispo-
sitions. His reason is it would serve as a ‘corrective’ against critical
thinkings misuse, implying that it is a powerful ability that has the
potential to hurt (humiliate, disempower, oppress) others. However,
I will argue: (1) that this is not enough of a reason to include it as
a core disposition; but also (2) that 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 disposi-
tions, such as those relevant to being a good sports person or a good
business person. The good sports person is not the same things as a
good person more generally conceived, but if we want them not
to use their talents and dispositions for ill-intent, then they need to
also have virtues such as respect and compassion for others. So, in a
sense, we would add this basic ethical disposition to 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 profes-
sions 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 criti-
cal 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 core, then as having greater
importance to critical thinking than it has to other practices (such as
sport). The reasons in question concern the functioning of construc-
tive 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
dialogue with.

Conclusion: Therefore concern for the welfare of others is a
disposition of an ideal critical thinker.

The initial premise has already been explained, but the second is in
need of some elaboration. One reason constructive dialogue is less
likely if we are not respectful towards other participants is that it
could be an incentive for them to disengage. A person detecting signs
of disrespect might leave the discussion entirely, or be reluctant to
give it their full energy, or to be entirely open about their position
and reasons 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 ourselves from
our commitments in order to objectively assess alternatives, 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, we need to be
willing to devote time and energy, and a kind of selflessness, 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 … [T]hat part of the truth which
turns the scale, and decides the judgment of a completely informed mind,
they are strangers to; nor is it ever really known, but to those who have
attended equally and impartially to both sides, and endeavoured to see the
reasons of both in the strongest light.


To be willing to do this, I would argue, 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 inde-
pendently of believers.


‘Argument, on this model,’ says Michael Gilbert about his theory of
‘coalescent argumentation’, ‘is among persons, not between theo-
ries’ (1994, p. 112). Because many of our beliefs – and certainly
many of those worth debating – are personal, then to enter into
argumentation dialogues can be to run a significant risk. It is more
than a matter of the possibility of finding out you are wrong in a
way that is analogous to getting an answer on a test wrong; it is find-
ing out that a belief that is central to your values and commitments
is wrong. 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.

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, runs the risk of
creating ‘not just able arguers, but argumentative arguers: proficient,
pedantic and petty …’ (ibid., pp. 180–1). Taking pleasure in argu-
ment 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.


In place of the ‘war’ metaphor, Cohen suggests a number of alter-
natives, including collaborative frames such as ‘brainstorming’ and
the nineteenth-century American tradition of ‘barn raising’ (which
is still practised by communities like the Amish). In place of listen-
ing 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 intimacy
that, especially if reciprocated, is profoundly rewarding. The care-
ful, respectful listening and thus opening up of the other’s world is
one reason for this. Another is that the sharing of ideas in a dialogue
towards new, mutually generated, important insights is an excellent
basis for bonding. Philosopher Bertrand Russell describes his meet-
ing with novelist Joseph Conrad in a way that demonstrates this

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 pas-
sionate love, and at the same time all-embracing, I came away 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 focused 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


maintain concentration in some deliberative situations, wholeheart-
edness is an intellectual absorption in a subject where ‘the material
holds and buoys his mind up and gives an onward impetus to think-
ing’. It is complex, reflective thinking with its own momentum, in
which questions and ideas arise ‘spontaneously’.

Overall, it can be seen that 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 encour-
aged in the right ways, can serve as a gateway to understanding and
compassion. Critical thinking can thus be motivated not just by
truth-seeking, but by connection and intimacy.


Becoming a critical thinking can change a person in a couple of
respects. On the one hand, they have, or have honed, the disposi-
tions so far discussed, and, on the other, they have a vocabulary that
is distinctive. In terms of exercising one’s ability, one has to be aware
that this can, as Ennis (1996a) puts it, ‘intimidate and confuse’ others
who have not had this training, or who are not otherwise inclined
to think in this way.

However, this does not mean that we should not engage people
who are less inclined to think in this way in dialogue, 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 must not
believe that we are in some sense superior.

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 smart arse. On the whole, we need to be very care-
ful when using comedy routines as examples of arguments, because
for the most part their intention is to entertain rather than to seek
the truth. (It should be recognised, however, that there are genuine
hybrids, such as Comedy Central’s The Daily Show, and the work
of people like Michael Moore, Mark Thomas and Dave Gorman.)
However, I’m going to break this rule here and consider a story told
by comedian Stewart Lee. It is an example that I believe is justified
because the comedian’s own published reflections 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, fraternising mainly with cab drivers … where that
was just accepted as a point.2 … after a moment he said ‘Well, because
homosexuality 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.’3

As abhorrent as the cabbie’s stated view is, Lee is being a smart arse
because the argument he presents is going to derail him. 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 any-
thing 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.’4

The virtue that has been called ‘deliberative friendliness’ captures
the type of constructive approach that would mitigate smart-arsery.
It has been defined as: ‘the willingness to entertain discussion in
a manner that does not unnecessarily offend or alienate interlocu-
tors’ (Aikin 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 encouraging, ‘sporting’


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, the features of arguments and argumentation that
arise from this aspect of the discussion of dispositions are these:

• Dialogues can be valuable for reaching mutually satisfying and
edifying answers, rather than a win-lose framework.

• A recognition of the complexity and hard-to-get-at nature of the
basis of many of the beliefs that we hold. This means that dia-
logues involving these issues need to be conducted with respect,
sensitivity, and tolerance (including for apparent 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 of
argumentation, and can be an additional aim.

• Careful and sustained listening is privileged as a skill, and as a
disposition (the desire to discover another’s worldview).

• Sensitivity is required, not just to the complexity, subtlety and
distinctiveness of the positions others hold, but to their style of
thinking, vocabulary and conversational norms. These are not
necessarily those of the critical thinker, but this is not the same as
being unintelligent or uninformed. And even if someone is these
things, that does not mean they cannot be engaged in some level
of argumentation.


In addition to this analysis of dialogical dispositions, it might be help-
ful to provide a summary of rules of conduct (influenced by Ennis,
1996a; Walton, 2006; and the work of pragma-dialecticians such as
Eemeren and Grootendorst, 2004) that should be followed in order
to give dialogue the best chance of success. The person embodying
critical thinking dispositions will be inclined to these ways of behav-
ing, in which case, these guidelines can serve as a kind of summary
of how the ideal critical thinker comes across when engaging in


argumentation. It can also, however, function as a stand-alone checklist
that has value regardless of underlying dispositions to conduct oneself
in these ways.

1. The discussants should be allowed to speak freely – both in terms
of expressing their view, and in terms of being critical of the
views of other discussants.

2. If a discussant is asked to explain their viewpoint (for example,
provide more clarity, or provide reasons why they hold that
view), they must be prepared to do so. This is especially impor-
tant since premises are often implicit. (See Ennis’ list of ‘clarifica-
tion questions’, above.)

3. Discussants have a duty to listen carefully to, and avoid misrepre-
senting, each other’s views.

4. Unless the issue is about the person (or persons) involved, per-
sonal attacks (ad hominem arguments – see Chapter 5) should
be avoided where possible. Often these are fallacious, but even
when 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

7. The discussion should usually only end when all the 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 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).


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 of the characters in
the play/film. The quality of the content of their arguments, and


in particular 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.

2. A good way to loosen up our biases is through what is known
as ‘counter-attitudinal advocacy’. This means writing or speak-
ing in favour of a position that you do not hold and/or to argue
against a position that you do hold. Its effectiveness with respect
to changing minds is well established in psychology and com-
munication research (see Petty and Cacioppo, 1996, Chapter 8),
but its weakness is that you can only get people to act in this
way under particular circumstances. A critical thinking class is just
such a circumstance, however. For example, class members’ views
on contentious contemporary issues can be sought and a debate
organised in which the participants argue counter-attitudinally.

3. As an exercise in personal reflection (rather than class discussion),
you might want to attend to the presence and development of
your own critical thinking dispositions and behaviours as you
work your way through this book and/or your critical thinking
class. (For a discussion of the relationship between academic work
and dispositions, see Hanscomb, 2015.)


• Aikin, S. F. and Clanton, J.C. (2010) Developing group-deliberative
virtues. Journal of Applied Philosophy, 27(4), 409–24. This has particu-
lar relevance to avoiding the problems associated with groupthink
that were discussed in Chapter 1.

• A very useful source of advice for constructive dialogues can
again be found in Robert Ennis’ Critical Thinking (1996a) (see
‘Thinking Critically When Discussing Things with Others’, 1996a,
pp. 371–5).

• For a more in-depth look at the nature of dialogues (including
‘Rules for a Critical Discussion’), see Walton (2006), Chapter 5.

• For an interesting discussion of flexibility, see Tiberius (2008)
Chapter 3.



1 Diana Beech, ‘Attitude is everything’, Times Higher Education, 21 May 2015.
2 He accepts in a footnote that he is stereotyping cab drivers.
3 How I Escaped My Certain Fate: The Life and Deaths of a Stand-up Comedian

(London: Faber & Faber, 2011), pp. 80–2.
4 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 astonished/angered him here is the cab driver’s prejudicial



A physician cannot treat a disease … properly without diagnosing it
correctly. 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 biases
and dispositions and towards the nature of arguments themselves.
The majority of it is devoted to explaining argument recon-
struction; 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 con-
cepts and techniques such as ambiguity and vagueness, straw man
arguments, implicit premises, and the principle of charity. Prior
to this, though, I will provide a reminder of what an argument
is, and explain how the sentences which comprise the prem-
ises and conclusions must be what are called ‘propositions’ or



In the Introduction, it was explained how the identification and
analysis of arguments are at the heart of critical thinking. An argu-
ment, 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; i.e. to try to 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 traditions
of formal and informal logic, they are usually set out in this form:

Premise 1:
Premise 2:

For the sake of brevity, a premise is abbreviated to ‘P’ and a conclusion
to ‘C’, so that we have:


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 punish-
ment 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.


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 looked at shortly.


Premises and conclusions must be what are called propositions (or
statements). A proposition is a sentence that makes a claim about
something that can be (in theory) 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’ to ‘The running of FIFA
is not unlike an organised crime syndicate’ to ‘I’m feeling hungry’.

Propositions are contrasted with other kinds of sentence, most

Questions: (‘Where’s my drink?’)
Directives: (‘Well, go and get me another one then!’)

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 eval-
uated in this way; in other words, they must be propositions.

When reconstructing arguments, part of the challenge can be
converting rhetorical questions into statements. If my wife asks me


if I would like to watch an episode of Game of Thrones, there is no
statement implied in this; she is simply asking me a question to
which I need to supply an answer. If, however, as another char-
acter is dispatched in a graphically drawn-out, gory and disturbing
manner, 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 like ‘Game of Thrones would be just as
good without quite so much gruesomeness’, or ‘Game of Thrones
would be even more engaging if the writers didn’t feel obliged to
shoehorn in plotlines that allow for horror rather than advancing
the overall story.’


The skill of argument reconstruction and evaluation is one way in
which many of the things learned in critical thinking are applied
and practised. In this section you will be learning about the recon-
struction part (which argument (or arguments) is (or are) being put
forward), and the rest of the book will be relevant to the evaluation
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

1. In each case you will be applying it to new arguments, and the
forms these take across different types of communication vary

2. Quite often what people are saying 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 (that are set out below), but master-
ing 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 apparent, as will be the way in which this ability will transfer


itself to improved questioning and comprehension in any aspect of
life of which arguments are a part.

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 minimising
errors in our reasoning by improving our ability to generate strong
arguments and evaluate the arguments of others. Since formulating
our own arguments usually involves responding to the arguments of
others, then the quality of these evaluations is of fundamental impor-
tance. But assessment can only be effective if the argument has been
properly understood in the first place.

The truth of this is perhaps fairly obvious, but worth highlighting
because its intellectual obviousness can mask how difficult it is to
achieve in practice. Many of the reasons behind this difficulty were
discussed in Chapter 1, and the interpersonal, ethical and epistemo-
logical significances of careful reading and listening were addressed
in Chapter 2. 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 other-
wise bring down) a straw effigy than a real person. The metaphor
represents the difference between, on the one hand, X’s real position
and, on the other, a weaker position ascribed to them by Y that is
easier to argue against.

For example, in recent years, an argument by the right-of-centre
UK Tory government against the left-of-centre Labour Party has
been made along these lines: ‘Labour are against our austerity mea-
sures, and say that the government should not be making cuts, so
clearly they do not think that the budget deficit is a serious issue.’
They do think it is serious, but they also believe that reviving the
economy by avoiding cuts is the best way to reduce the deficit. It is
easier to argue against a party who deny the importance of reducing
the deficit than it is to argue against a party who not only have a
different approach to reducing it, but one that is liable to be popular
with a significant proportion of the electorate.

Sometimes, as in this case, a distorted version of an opponent’s
position is employed deliberately, but the straw man label also applies
to instances where it is a mistake. From a dialectical point of view,


being misrepresented 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 con-
firmation bias is implicated (see Chapter 1), and the discussion in
Chapter 2 about open-mindedness and paying careful attention to
both the words and the circumstances of the other is important
for avoiding unintentional straw man arguments. In particular, we
need to avoid, where possible, hearing arguments second hand. As
John Stuart Mill says, we should

hear them from persons who actually believe them; who defend them in
earnest, and do their very utmost for them. [We] must know them in their
most plausible and persuasive form; … must feel the whole force of the
difficulty which the true view of the subject has to encounter and dispose
of; else he will never really possess himself of the portion of truth which
meets and removes that difficulty.

(1962, p. 163)

Critical thinking scholars are fairly consistent in listing the tasks
that need to be performed towards reconstructing arguments,1 and
in this tradition the rest of this chapter will help you develop and
refine this skill under the following headings: identifying premises
and conclusions; argument-friendly rewording; implicit premises and
conclusions; and argument structure (sub-conclusions, re-ordering of


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, if the overall point is
not clear, it is crucial to get into the habit of asking people to clarify
what they mean (which can be easier said than done if you think


back to the discussion of groupthink in Chapter 1). With written
arguments the responsibility lies with us to read, and if necessary re-
read, 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 recon-
struction 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
indicators. 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:

in which case.

And premises can often be identified by terms 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. One reason is that reading and
re-reading a passage help us to understand it more deeply, and not
be tempted by the short-cuts 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 argu-
ments. 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 sometimes closer to the sun than Neptune because it
has a more elliptical orbit than its neighbour. Notice that the word
‘because’ is used here as well, and with both arguments and explana-
tions 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.


Part of the skill of argument reconstruction is to make what is pre-
sented both argument-friendly and as concise as possible without
losing its meaning. Great care must be taken with this because
premises and conclusions can contain subtleties, so that inattentive
rewordings can lead to important nuances being lost in the edit. The
rule for avoiding this is: if in doubt, retain the original words.

The three main aims of argument-friendly rewording are:

1. Translating information in the form of non-statements (such as
rhetorical questions) into statements.

2. Condensing the argument where necessary through:

i. using more efficient expressions;
ii. leaving out material that does not contribute to the premises or


3. Clarifying ambiguous and vague sentences.

The passage below is from Martin Luther King’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 sometimes 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
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.

It is immediately evident here that King’s argument employs a series of
rhetorical questions. Each, however, contains a relevant point, and so
they need to be preserved in the reconstruction in the form of state-
ments. Some of his sentences are quite wordy and can, arguably, 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 condemning some-
one (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 pro-
test because it precipitates violence.


This presents the essential structure of the argument (which is an
argument from analogy; see Chapter 7) and an appeal to various
authorities (see Chapter 5), 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 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, however,
the detail King offers is important to include, resulting in the following

P1: To condemn our actions because they precipitate violence is like
condemning the robbed man because his possession of money pre-
cipitated the evil act of robbery.

P2: This is like condemning Socrates because his unswerving commitment
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 precipitated 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 anal-
ogy with the robbed man but be more satisfied with the Socrates and
Jesus analogies.

In summary then: if they are to form part of the reconstruction,
then questions and other non-statements must be re-worded; but if
in any doubt as to whether significant re-wording or condensing of
sentences might lose important content, then leave them in their
original form.


Ambiguity and vagueness

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

Did you hear that Jason was fired from his job at the orange juice factory?
Apparently he was squeezed out.

Ambiguity can also cause serious problems. A 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.2

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 children as
a total population are poor …’ (This has since been corrected, by the

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 the 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 genuine
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.3


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). In light of 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 who has caused the mix-up. Let’s hope it is Fraser,
but either way, what we are presented with is ambiguous.

When teasing out the premises and conclusion of a person’s argu-
ment, ambiguous sentences have to 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 prin-
ciple of charity (see below), or

• reconstruct more than one version of the argument, applying
alternative meanings of ambiguous sentences in each case.

Sometimes, though, you might come across, not just words and terms
with ambiguous meanings, but concepts which are indeterminate
(Freeden, 2005). ‘Democracy’ is an example; no amount of disam-
biguation can overcome its complex and contested nature, and all a
reconstruction can do is either stay with the original wording, or use
the context of the argument to generate implicit premises indicating
which aspects of this idea seem to be most relevant to the discussion
(such as forms of equality, or human rights). With this latter strategy,
however, we need to be highly sensitive to the provisional nature
of the reconstruction and recognise that the way the term is used is
something that is likely to have to be negotiated by the discussants.

If an ambiguous sentence is one which can mean more than one
thing, a vague sentence lacks precision. In many contexts, an absence
of precision is entirely appropriate; ‘Give me a moment’, ‘I’ll be there
soon’, ‘Just give me a few potatoes.’ My 4-year-old does not like
vagueness. He recently asked me how long a ‘moment’ is, and he is
also not too wild about his astronomy books saying that Jupiter has
‘more than 60 moons’. I’m kind of with him on this one; it could
mean about 62, or around 70, or over 100. Or it could mean they do
not have a clue. Which is it?


The Intergovernmental Panel on Climate Change (IPCC) has
been urged to change its style of reporting probabilities to the media
because of a mismatch between the vague terms it was using and
what these are taken to mean by the general public. For example,
in statements like ‘Anthropogenic influences have very likely con-
tributed to Arctic sea-ice loss since 1979’,3 they use ‘very likely’ to
mean >90 per cent certainty, but it has been found that the public’s
understanding of ‘very likely’ is >70 per cent certainty.4 This rep-
resents a significant discrepancy 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 is “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; ‘I cannot respond to this until I know exactly what is
meant by …’

Unlike ambiguous sentences, vague statements can form premises
and conclusions in arguments where (1) and (2) above do not repre-
sent a problem (‘I’ve only got a couple more things to do, therefore
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 precise;
either through an educated guess or through further research. As
with the re-wording of ambiguous sentences, the ‘principle of charity’
should also be applied (see below).


A lot 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 visu-
ally, such as advertisements with minimal written text, or gestures


like the two-handed diving motion used to footballers to make the
point to the referee that a player took a dive and was not in fact
fouled.5 Partial arguments are most common though; a simple exam-
ple might be:

Only extroverts are energised by social situations, so Susan is an

The whole argument would run:

P1: Only extroverts are energised by social situations.
P2: Susan is energised by social situations.
C: So Susan is an extrovert.

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 when
recommending their use, Aristotle notes in The Art of Rhetoric that
‘obscurity is produced … from length of reasoning, and … it is a
waste of time as one is stating the obvious’ (1991, p. 195). Being
explicit can be unnecessarily wordy and have the effect of being bor-
ing, patronising, and in some cases harder to follow.

There is a balance to be struck, however. A couple of years ago,
cycling along the same stretch of road where I caused the cattle
stampede (see Introduction), a woman driving towards me slowed
down her car, lowered the window and shouted: ‘They’ve cut the
hedges!’ I was very confused as to why she was telling me this. I said:
‘Er, yes, it looks great …’ She then said something about hawthorns,
and the penny dropped – thorny hedge debris would be littering the
edge of the road, causing a puncture risk. It was very thoughtful of
her, but her initial argument was so condensed that the communica-
tion nearly failed.

Reconstructed, her argument was, presumably:

P1: If the hedges have been cut, there will be hawthorns on the road.
P2: If there are hawthorns on the road, you risk getting a puncture.


P3: You don’t want to risk getting a puncture.
C: You should be careful (or, you should cycle away from the edge of the


This illustrates how conclusions as well as premises can be implicit. In
this case the inclusion of the conclusion would have been helpful, but in
other cases it is unnecessary for the same reasons that it is often unneces-
sary to include all the relevant premises. For example, towards the end
of his speech following the Sandy Hook shooting in 2012, President
Obama said, ‘We can’t tolerate this any more. These tragedies must end.
And to end them we must change.’ The conclusion is ‘therefore we
must change’, but by this point in the speech, it is clear what he means.

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 that
is implied by a sarcastic question: ‘Prime Ministers have regularly
used war abroad to distract from constitutional matters or problems
at home, as history shows. But it couldn’t happen today, could it?’6 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.

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 in order 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 argu-
ment, 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 functions
of reconstructing arguments. For example, should we want to respond
to the letter above, it is important to make P2 and C1 explicit.




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)
Overall Conclusion

Arguments containing sub-conclusions are usually what are called
chained arguments. In a chained argument each sub-conclusion
forms a premise for a further argument, and in these circumstances
it is critical that we put the premises in the right order in our recon-
struction. 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: So 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 should not 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: So 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, ‘Gulf oil spill: a hole in
the world’, about the Deepwater Horizon disaster in the Gulf of
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
contact 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 (77 km). This is the most astonishing claim of all. In a gulf that
often sees winds of more than 70 km 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 77 km trip. None
of this sloppiness would have been possible, however, had BP not been
making its predictions to a political class eager to believe that nature
had indeed been mastered. 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 environ-
mentally sound way,’ she told the Senate energy committee just seven
months ago.6


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, proclaimed 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 environ-
mentally sound way.’

P2: (implicit) A statement like this 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

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.

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, such that the basic structure is simply:



A simple example might be:

Because it is 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.


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 like ‘Another
consideration that supports this point is …’; 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.


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 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.’

This kind of language is recommended. It indicates that you can-
not always be sure that you have grasped what the other person is
intending to say, and that the true meaning of the argument and
its intention rest with them. Such language connotes respect and
modesty, and, as discussed in Chapter 2, better serves the aim of a
constructive dialogue.

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
statements; (2) the inclusion of implicit premises; (3) disambigua-
tion; and (4) making precision out of vagueness, we should aim to
put the argument 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 to 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 arguments. It should be noted, however,
that over-eagerness with respect to the principle of charity can lead
to the opposite of a straw man – an ‘iron man’ argument. These are
cases where we misrepresent someone else’s (or our own) position as
being stronger than it is. This could be motivated by a desire to make
it easier to defend, but in the context of the principle of charity, it is
primarily an error of interpretation.7


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).


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


1. It’s okay to lie to children about the existence of Santa Claus until
they’re older, isn’t it? So it’s okay not to tell my fiancé that I used to
be a man until after we’re married.

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 intro-
duced benefit cuts. Some Tories say that food banks are part of the
welfare state, but food banks are charities and separate from the wel-
fare state. They are a sign that the welfare state is failing. (Adapted
from MP Mhairi Black’s maiden speech in the House of Commons
(UK), 14 July 2015.)

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, 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 ille-
gal 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 April 2011).


• For exercises on visual arguments, see Morrow and Weston
(2011) A Workbook for Arguments; and Lunsford and Ruszkiewicz
(2009) Everything’s an Argument.

• For an instructive article on straw man and iron man arguments,
see Aikin and Casey (2016).


1 Some good examples are Johnson and Blair (2006); Scriven (1976); Bowell
and Kemp (2010), and Morrow and Weston (2011).


2 L. Berk (2010) Infants, Children, and Adolescents, 7th edn (Harlow: Pearson
International Edition), p. 74.

3 IPCC (2014) Climate Change 2014: Synthesis Report: Summary for Policy Makers,
p. 5. Available at:

4 J. Painter (2013) Climate Change and the Media (London: I.B. Tauris).
5 I am wary of using visual arguments for reconstruction and evaluation

because the skill required is more one of interpretation than reconstruction,
but in the Further Reading you will find some suggestions for good critical
thinking books that do go down this path.

6 N. Klein (2010) ‘Gulf oil spill: a hole in the world’. Available at: www.

7 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.



[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 fallacy (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 arguments, and argument forms (or ‘schemes’).



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 of


formal logic, but because I do believe they have some value in terms
of the effective communication of arguments, and because you will
hear these terms being mentioned in some circles, I will pay some
attention 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 it 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 deductive argument, then they must
also accept the conclusion because the truth of the conclusion is
entirely contained within the truths presented in the premises.

The argument above claims that there are only two options that
could be true. If one option is then rejected, the remaining one must
be the case.

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 it has changed form due to tectonic shifts.

While it is entirely reasonable to reach this conclusion (science is
based on this kind of inference), it does not necessarily follow from the
premises. In other words, it would not be a contradiction to reject the
conclusion while 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 deductive
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 with-
out 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.

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.

We find, then, that most arguments can be presented in deductive
or inductive forms without losing ‘the point of the argument’, so the
primary trajectory of critical thinking has been towards more helpful
ways to categorise arguments.



In recent decades Douglas Walton (1995, 2006) has provided a char-
acterisation of everyday argumentation in terms of what are called
plausible arguments (also known as ‘presumptive’ or ‘defeasible’
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, particularly in
the context of dialogues. 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 con-
tradicts 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

Plausible arguments are highly pragmatic. They operate in situations
where decisions have to be made fairly quickly, often in the absence
of precise information, but in circumstances that are generally famil-
iar. In other words, they operate in many of the situations in which
we find ourselves in daily life.

Working with plausible arguments is not the same as System 1
thinking, but it is significantly vulnerable to biases. Although result-
ing from a degree of reflection, plausible reasoning is nevertheless a
form of generalising that occurs in the thick of life, and it will inevi-
tably rely on heuristics. However, it also embodies an awareness of
its own limitations (and thus the provisional nature of its conclusions)
and it is this that places it on a different level to System 1 thinking.
Plausible reasoning can perhaps be characterised as System 2 think-
ing in a hurry.

Clearly enough, plausible arguments are useful in familiar situa-
tions and when time is scarce, but we can also view them as the start
of a dialogue in which they provide the catalyst for more careful and
nuanced reasoning. Thinking slows down as ideas are passed around.


Definitions and classifications become more accurate, and probabili-
ties and alternatives become as well considered and researched as is
reasonable under the circumstances.


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 the majority of the
rest of this book too.

Examples of argument forms (sometimes referred to as ‘argu-
ment schemes’) include ‘arguments from authority’, ‘arguments
from analogy’, ‘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,
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 author-
ity, analogy, popular opinion, and so on. Establishing criteria 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 (1988, p. 177) as ‘a mistake
in reasoning … which occurs with some frequency in real arguments
and which is characteristically deceptive’.


A fallacious argument is one that is not only bad, but bad in a way

1. Conforms to a typical type of error that we are prone to make
when constructing arguments.

2. Tends to create the (superficial) impression of being a good

Fallacies are, in short, poor arguments of certain types 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 important. 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 course or
textbook must make choices about what to include and what to
prioritise. A sensible criterion for inclusion is a fallacy being ‘charac-
teristically deceptive’ since these will be the more disabling ones in
processes of decision-making.

This book has defined critical thinking in terms of the attempt to
avoid reasoning errors, therefore identifying fallacious arguments is
clearly central to its purpose. As the previous discussion has indicated,
however, 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 text-
books 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,


1. A particularly poor argument conforming to an argument form
is often called a ‘fallacious ad hominem argument’, ‘fallacious
causal argument’, and so on. Alternatively, they can be called
‘weak’ or ‘poor’, the terminology does not need to be that

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. Straw man arguments are potentially one exam-
ple (see Chapter 3), as are circular arguments (see Chapter 8).
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’ is a fallacy that can be committed when constructing a
causal argument.


‘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) 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

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 can inter-
rogate 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.



Perhaps the best starting place for critical questioning is to ask yourself:

• What kinds of questions would you ask about a claim in deciding
whether or not to believe it (and therefore act on it)?


• What kinds of questions would you ask about this claim in decid-
ing whether or not to believe it (and therefore act on it)?

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.)

Ralph H. Johnson and J. Anthony Blair identify three fundamental
critical questions, otherwise known as the ‘Acceptability, Relevance,
Sufficiency’ criteria (generating the slightly unfortunate acronym,
ARS).1 ARS criteria are briefly explained in Box 4.1.


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

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 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.


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 eutha-
nasia is undoubtedly a form of killing, sufficient for establishing
the conclusion as well. However, many people 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 on the basis of the premise
being unacceptable. Someone else might argue that euthanasia is
wrong because it contravenes a religiously derived prohibition
on suicide. In response to this, it could be argued that religious
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 for legalising euthana-
sia. Possible abuses of the law, and old and terminally ill people
feeling 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


In logic, the concepts necessary and sufficient conditions are impor-
tant. 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 conditions means, simply, that
everything needed for something being 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 suf-
ficient to keep me happily occupied for an hour, but many other things
could fulfil this role as well.

Box 4.3 shows how each of these fundamental questions can be con-
nected to a range of possible sub-questions (many of which we will
look at in detail in subsequent chapters).



1. Are the premises acceptable, i.e. likely to be true, or likely to be acce-
pted as true or plausible by the people the arguer is attempting to

• 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 represent 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?

• 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 seeking
to establish?

• Has the appropriate kind of evidence been presented? For
example, what the general public thinks about who should gov-
ern them is relevant to who should govern them, but what they
think the effects 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?

• If the conclusion is a generalisation, is this one that can be sup-
ported on the basis of the evidence provided?

• Has relevant information or have perspectives been over-looked?
• What alternative conclusions can be drawn from the evidence?


On a British political analysis TV programme in 2015, the possi-
ble repeal of the fox hunting ban in Britain was being discussed.2


Rock guitarist, astrophysics PhD, 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 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 could not;
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 analogy.
It is no coincidence that we have come across this type of argument
previously in this book because they are extremely common. They
are also often quite weak, but even when they are weak, they can be
significantly persuasive. Arguments from analogy are, then, impor-
tant to analyse and will be a central focus of Chapter 7, 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

Q4. Can convincing counter-analogies be found?

Applied to Neil’s argument, the likely answers (i.e. the basis for fur-
ther 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.

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

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 factors like the degree of
cruelty, the degree of human skill and courage involved, and the
victim having a sporting chance, are among several that differen-
tiate the two practices.

4. Portillo mentioned the counter-analogy of bullfighting, but this
is not as strong as it might be since there is also a lot of support
for banning this. Other possibilities are other traditional forms of
hunting (deer, grouse, etc.) that are less controversial.

The first two critical questions tend to be where our basic ration-
ality will take us when encountering arguments from analogy,
but I think that the third one is slightly less intuitive (it is also the
hardest one to answer since it involves a lot of imaginative work).
However, even if all three 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 on 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, note that critical question Q1 for
arguments from analogy is a form of the acceptability criterion; Q2
concerns relevance, and Q3 and Q4 question the premises’ sufficiency.


When we looked at the process of 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
reconstructions is also about making implicit assumptions explicit.

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 slippery
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 6).
In the process of identifying something as a slippery slope argument
it has become apparent that 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 consequences will be significantly negative. (These in turn
generate critical questions that examine the likelihood of the sequence
of events; whether it really will be unstoppable once it has started, and
whether or not the final consequences are 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 5), 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. In light of the guidance
on reconstructing enthymemes offered in Chapter 3, 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.


The argument forms and associated fallacies that will be discussed in
subsequent chapters have been selected because they are particularly
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, disposi-
tions, and constructive dialogues. From a study of the argument forms
we commonly use and the ways in which they can go wrong, we
can learn a lot about, not only argument reconstruction and evalua-
tion, but also 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 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 5–7 will be investigated in terms of 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.


My view is that there is limited value in applying the ARS questions
to relatively unfamiliar and de-contextualised arguments, so instead
of providing some of these (as I did at the end of the last chapter),
my recommendation is that you should 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 the ones recommended in this book;
• essays and other assignments that you have written.


1. 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).

2. Two sources in which the Acceptability, Relevance, Sufficiency
criteria are employed explicitly and with clarity are Johnson and
Blair, Logical Self-Defense (2006); and Hughes, Lavery and Doran,
Critical Thinking (2010).

3. If you would like to know more about deductive and inductive
arguments, some good sources are Bowell and Kemp (2010) and
Ennis (1996a).

4. Ennis (1996a) also has a helpful chapter on the distinctive nature
of value-based arguments.


1 Originally presented in 1977 in their book, Logical Self Defense.
2 This Week, 2 July 2015.




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

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

7 October 2005)

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 Introduction, 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

In the context of argumentation, one of the most effective ways of
deflecting requests to justify one’s beliefs is to appeal to an authority.
By doing so we are saying things like: ‘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 part of a larger system with which the responsibility lies.
The implication of the second is that there are reasons for thinking
that 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 con-
sideration for critical thinking.

Since we cannot become experts in all of the fields that are rel-
evant to the decisions we make in our personal and professional
lives, reliance on authority is unavoidable. There are also plenty
of circumstances (and not just in the military) where an unques-
tioning approach to authority is the best course of action or the
best policy. 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, and
so on) lends itself to a policy of not questioning orders. Modern
analyses of arguments from authority recognise the limitations of
individual powers, and the critical questions they generate embody
this understanding. In Johnson and Blair’s words, ‘We are urged to
be filters of opinions rather than sponges who soak them up indis-
criminately’ (2006, p. 167).

These concerns are central to understanding our place in a repre-
sentative democracy. On the one hand, we need to be appropriately
informed on the issues we are voting on (and we need to be willing
to exercise critical thinking), and, on the other, we need to choose
candidates who are knowledgeable and trustworthy. What consti-
tutes trustworthiness in our elected representatives partly depends on
the preferences of the voter in question, but many of us will be look-
ing for someone who shares our values, who is willing and able to be
a critical thinker, who has a wide set of critical thinking dispositions
(including those, like courage and empathy, that are important for
generating constructive dialogues and avoiding groupthink), and
someone who is in other respects a good person (e.g. honest and

The authority of the political representative derives from some
combination of her legitimacy, her expertise, and her character.
Character matters because shared values matter (they are in an
important sense ‘like us’), and because we must trust them to act


appropriately in circumstances that are not entirely predictable
in terms of previous experiences. They will, for example, need
to vote on matters that their current expertise does not prepare
them for, and they will need to conduct themselves in pressured
situations that will test their courage, persistence, decency, self-
control and humility. To a great extent, this is because failings in
these ways will impede their critical thinking, but it can also be
because we want them to represent what is best in us and to act
as role models.

The flipside of appealing to authority in support of our positions
is attempting to discredit the views of those who oppose our beliefs.
This can be on the basis of their lack of expertise; their biased view
on the subject; inconsistency between what they advocate and what
they do, or simply because of their undesirable character (or links
to other undesirable characters). All of these forms of argument are
known as ad hominem arguments.

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 a number of 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 collec-
tion of people). By influence is meant a change in behaviour and/or
beliefs and attitudes.

In their widely employed analysis of the subject, social psy-
chologists John French and Bertram Raven (1959) identified five
types of social power: (1) legitimate power; (2) expert power;
(3) reward power; (4) coercive power; and (5) referent power.
A sixth type – information power – was later added by Raven
(1965). Other theorists (from a range of disciplines including
argumentation, rhetoric and leadership) have suggested fur-
ther categories and sub-categories, including witness testimony
(which I will treat as a sub-type of information power) and eth-
otic power.1 These six types and one sub-type are explained in
Box 5.1, and it is these categories that will form the bones of this



• 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 pronounce-
ments on ancient Egypt.

• Information power: There are plenty of situations in which, although
a person is not an expert, they do nevertheless possess knowledge
that is pertinent and valuable in a certain situation. The black-
mailer trades on information power, as does my wife when she
hides the channel changer when the cricket is on TV. An important
subdivision of information power is witness testimony, referring
to someone’s being present at and observing a one-off event. In
court cases the information power of a witness is of course very
important in 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 indi-
vidual, by virtue of which the external agent can assert his power’
(French and Raven, 1959, p. 265). Thus, it is important to remem-
ber that legitimate authority covers both official positions/jobs
AND authority associated moral codes and social norms.

• 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 hair style). Anyone able to influence the behaviour
of others through rewards has ‘reward power’, which extends this
category beyond the positional and normative reach of legitimate

• Coercive power: The power of someone who is able to punish some-
one 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 with
whom you personally connect. They demonstrate values and other
qualities that are (or you would like to be) an important part of


your identity, and so, without the need for requests, orders, or
threats, you desire to conform to group or person’s norms, and
those norms have power over you.

An additional basis 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
someone 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 less general
concept that tends to refer to an official or otherwise socially recog-
nised sanctioning of someone’s power. The phrase ‘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 com-
ment about the source or location of social power. Put briefly, a person’s
power does not simply reside in them independently, but is actualised
by being recognised or accepted by other people. My mother might
be an expert on parenting but her advice will not be heeded by me
unless I acknowledge this. This dependency explains the importance
of signalling power via, for example, uniforms and badges.

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 commands X, we should consider believing or doing X.
These include appeals to expert, legitimate and ethotic authority;


arguments based on information power (including witness
testimony) arguments based on referent power (a species of
which is known as the ad populum argument), 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 person who is presenting the argu-
ment (the explicit or implicit message that you should take them
seriously because of who they are), or they can be referring to a
third person.

2. Negative arguments – otherwise known as ad hominem
arguments – which claim the opposite; that because certain
beliefs or actions are those of, or those associated with, a particu-
lar 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 infected (by
lack of expertise, biased views, poor character, and so on) such
that anything that springs from it is also infected.


A vast amount of the information we rely upon to make decisions
comes from other people. Since it is not possible to find out every-
thing for ourselves in a direct manner, we have no choice but to
include second-hand information in many of our rational delib-
erations. The question of which information to trust becomes the
question of who to trust.

In the context of public decision-making, Charles Willard (1990,
p. 20) makes this sobering observation:

Action needs facts; decision-makers are dependent on the custodians of
facts, the [academic] disciplines. As it is impossible for public actors to
acquire expertise in the range of subject matters that confront them, we
need to rethink the very ideas of public knowledge and competence.

Part of this re-think concerns the ways in which group decision-
making is approached, but of more direct relevance to this chapter
is the need for competence in our assessment, not so much in the
ideas experts put forward, but in the trustworthiness of the experts


in question. By ‘expert’ we typically mean someone who has
undergone substantial education or training. They have significant
experience of working in the field of their expertise, and the level
of their knowledge is often verified by qualifications, influence, 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? [Acceptability]
Q2: If so, is this field relevant to the issue under discussion?

Q3: If so, is the authority cited to be trusted? [Acceptability]
Q4: If so, is the strength of the conclusion reached appropriate?


Each of these, though, is associated with a number of important sub-
questions, and some of these will be discussed below.


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
psychological condition, then we need to be more cautious. This
much is obvious, but if we’re 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, but if
someone commenting on the Battle of Culloden is only introduced


as a ‘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 recog-
nise that generalist and interdisciplinary writing and research have a
couple of vitally important functions:

• Most of the big issues affecting us (such as climate change, global
inequality, terrorism) require multiple disciplinary perspectives to
make sense of 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 pay-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 (including any
influence they might have on policy-makers). The books of journal-
ist and political activist Naomi Klein are good examples. Her 2015
work This Changes Everything combines climate science with ethics,
politics and economics. Few people, including academics, are experts
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 con-
tinual 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 sensi-
tive 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 urgent but
complex problems. Our evaluation of such efforts needs to be differ-
ently oriented: to be more generous and tolerant, and to recognise


that they belong to an applied agenda rather than the pursuit of
knowledge for its own sake.


Trustworthiness refers to our confidence that the individual’s exper-
tise will be appropriately applied to the current situation. Indicators
of trustworthiness include the extent to which the expert’s views on
the issue are shared by her peers (consensus), the ways in which she
appears to be applying her knowledge to the problem (judgement),
and whether some form of bias might be compromising her objec-
tivity. 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.


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 audience 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
explain and defend 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 sim-
ply pointed out that a reply to Wilson’s paper elicited a damning
response in the form of a letter to the journal Nature (the journal
that published Wilson’s article) from 140 eminent evolutionary biol-
ogists. His argument came down to numbers: yes, Wilson is a leading


authority in evolutionary biology, but on this issue he is vastly out-
numbered by other authorities. Who are you going to trust? The first
part of Wilson’s response to this was to draw an analogy with a 1921
paper entitled ‘One-hundred physicists against Einstein’, which erro-
neously 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 disci-
plines (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 consensus 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
to Nature 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 angle on 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 not to 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 critical
thinker. What to look out for in this respect is broadly covered by the
critical thinking dispositions (see Chapter 2), and the relevant critical
question supplementing the broad reliability question is:


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; or being open-minded rather than dogmatic
when considering options?)

In some situations an additional relevant question(Goodwin, 2011)
might be:

What do they have to lose if they turn out to be wrong?

Of course, a learned practitioner might have developed some habitual
weaknesses in this respect, in which case the direct (or abusive) ad
hominem argument can legitimately be employed against him. This
will be discussed further under ‘Arguments appealing to character’.6


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 per cent of scientists rather
than 97 per cent), 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 signifi-
cant cause of global warming, and that the consequences of global
warming will be very serious indeed.

P2: A clear majority of scientists believe that immediate reductions in car-
bon emissions will slow down global warming, leading to less severe

C: Therefore we need to act now to reduce carbon emissions.


This example also highlights the value of an extra critical ques-
tion 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 severe, so we ignore them at our
peril. For most of us at least, the same cannot be said for authori-
tative views on the surface temperature of Venus, or the history
of cricket.


It is unsurprising that psychologists have identified an ‘expertise heu-
ristic’ which prompts us to jump to conclusions when expertise – or
the appearance of expertise – is associated with an idea or product
(Smith et al., 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 example,
Carol Vorderman, an engineering graduate and TV presenter (ini-
tially doing the maths on the game show Countdown) is 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 is discussed in
more detail below, under ethotic arguments.)

Rapid message delivery, steady eye contact, and even face shape
(Smith et al., 2014) are among the many non-verbal and paralin-
guistic signs of expertise and trustworthiness. Owen Hargie and
David Dickson (2004, p. 344) identify the ‘Three Ts of expert
power’: (1) titles (Professor, Doctor); (2) threads (clothes, uniforms);
and (3) trappings (other signs, such as framed degree certificates,
shelves of technical books and journals, or specialist equipment).
During the UK Labour Party leadership election in the summer of
2015, a letter was published in The Guardian in support of one of
the candidates. It was signed in this way:


Richard Wilkinson Emeritus Professor, University of Nottingham, Kate
Pickett Professor, University of York, Steve Keen, Professor, Kingston
University, Elizabeth Dore, Emeritus Professor, University of Southampton
and 23 others.7

The full list is online and the nearly everyone there is a professor.
Very few though are professors of politics, so presumably their titles
are meant to add to the persuasiveness of the argument presented in
the letter by virtue of the expertise heuristic.


Kant pointed to ‘laziness and cowardice’ as

the reasons why so great a proportion of men … gladly remain in lifelong
tutelage, and why it is so easy for others to establish themselves as their
guardians. It is so easy to be immature. If I have a book to serve as my
understanding, a pastor to serve as my conscience, a physician to deter-
mine my diet for me, and so on, I need not exert myself at all. I need not
think, if only I can pay: others will readily undertake the irksome work
for me.

([1784] 1963, p. 35)

If ‘laziness and cowardice’ 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. In both cases, the fault lies with over-
reliance on limited sources – on oneself in the case of the immodest
person, and on too few others in the case of the dogmatic tradition-
alist. 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 like courage, open-mindedness and

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 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 argument, 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 her own authorities in support of her 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 situ-
ation in a number of respects: they are responsible for finding the
relevant authorities; for understanding them as far as they can, and,
importantly, 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 (so long as
groupthink is avoided) could itself lead to a softening of biases.


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 et al., 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 are located.
Examples of position to know range from our local geographical and
cultural knowledge (where the local library is; where the best place
to look for job ads is; 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,8 and altogether juicier revelations found in autobiogra-
phies or information given to us in confidence (or obtained through
snooping or computer hacking) on things like love affairs and criminal

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 contrast between expertise and experience. Consider, for exam-
ple, the difference between knowing all about cancer from a scientific
perspective and living through cancer; or between knowing about
heartbreak from a literary and psychological point of view and hav-
ing 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 experi-
ences. 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 understanding. For a number of


reasons – including contextual and ‘felt’ qualities – it offers a qualita-
tively distinct perspective on an issue, rather than one that is simply
waiting to be reduced to and replaced by a theoretical account.

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

Q1: Is appealing to position to know appropriate in these circum-
stances? [Relevance]

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


Q1 is worthy of some elaboration, mostly in the form of sub-ques-
tions. First, we should ask whether this is information we are able
to find out for ourselves, and, if so, question our motives for relying
on others (which might of course 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 posi-
tion-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?)


For certain kinds of news stories (such as extreme weather events,
or 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 sta-
bility in the information’s availability so that it is there for anyone
who happens to be in a ‘position to know’. The eye witness, in
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, eye-witness 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 Brian’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 where 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 subconscious level as well (see rationalisation and cognitive
dissonance, 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’, 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 follow-
ing critical questions can be posed (based on Walton et al., 2008, p. 91):

Q1: Is what the witness says internally consistent? [Acceptability]
Q2: Is what they say consistent with what is otherwise known about

the event, including what others have said? [Acceptability]


Q3: Is the witness liable to be biased? (In Twelve Angry Men, for exam-
ple, it is suggested that one of the witnesses was motivated by the
chance to testify in court and feel important, rather than his genu-
ine knowledge of events surrounding the murder.) [Relevance]

Q4: Is what they have said plausible? (In the case of miracles and
UFOs, it is presumably not.) [Acceptability]

Q5: If so, is the strength of the conclusion reached appropriate?


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 his 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.

In a recent example, it was proving so difficult to find someone
not linked to possible suspects to chair a UK inquiry into child
sex abuse – one that includes government-level suspects like for-
mer Prime Minister Ted Heath and stretches back at least until the
1970s – it was necessary 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 doc-
umentary The Great Global Warming Swindle (Martin Durkin, 2007)
that the scientists who have found evidence for anthropogenic 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 voluntary and unpaid.)

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
circumstances true? [Acceptability]

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 out-
right on the basis of an ad hominem attack whereas at most it
should be called into doubt.) [Sufficiency]

Towards the end of Martin Luther King’s Letter from Birmingham Jail
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 support 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 realized
that few members of a race that has oppressed another race can under-
stand or appreciate the deep groans and passionate yearnings of those
that have been oppressed, and still fewer have the vision to see that injus-
tice must be rooted out by strong, persistent, and determined action.

(King, 1963)

The ad hominem argument in this passage, reconstructed in a
summarised 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 psychological condi-

tion 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 is such that
they are unable or unwilling to put themselves in the shoes of the


people that King 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’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 lay-person’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 on an ad homi-
nem argument of this type, then it would be fair to regard his, quite
firm, conclusion as inappropriate (Q3).


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 reason
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 memorable. We
find the details of particular human experiences engaging in a way that
abstract reasoning tends not to be, which is why documentaries and
books with serious messages trying to reach wide audiences are either
told via stories, or mix personal experience with more technical argu-
ments. (The climate change films, An Inconvenient Truth (2006) and The
Age of Stupid (2009) are quite typical in this respect.)



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 meta-error of over-confidence. However, like
most dispositions, there is a balance to be sought, and due regard for


power and authority, in all of 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 intuition 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 should not expect others (who have not expe-
rienced the same thing) to believe us. The critical thinker would
know this limitation, 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.

A further type of situation where personal conviction can be jus-
tified 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 or all others do not share or
understand, 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 where ‘HANDLE WITH
CARE’ is stamped in large letters. The line between admirable
integrity and a 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 criti-
cal questions that we will come across in this book, there are subtle
forms of justification which help establish the credibility of lonely
decisions like these. Ethotic authority is one, and another is sim-
ply the person’s willingness to acknowledge how their situation can
seem peculiar to others; that by normal standards (i.e. standards that
they otherwise respect), what they are choosing would be regarded
as wrong or foolish.




Stanley Milgram’s experiments in the 1960s measuring and analys-
ing obedience to authority remain influential and important. They
are quite well known these days (beyond those studying psychol-
ogy), 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 attention to the processes of
power and authority that explain the results. These are discussed at
length in Milgram’s book Obedience to Authority ([1974] 2005), and
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 persuasion that
is associated with appeals to authority.

Motivated by the high obedience levels among soldiers and
Nazi officials in the World War II death camps, and by the trial of
Adolf Eichmann in 1961,11 Milgram wanted to examine obedience
behaviour in the laboratory. Members of the public were asked to
volunteer for a memory experiment at Yale University. If selected,
they would turn up at the lab, have the procedure explained to
them, and meet their partner, another (apparent) volunteer called
Mr Wallace.

As one of these participants you would be (seemingly randomly)
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 the he would try to choose
the correct pairing. The twist in this procedure is that every time
Mr Wallace gets one of these wrong, you are required to deliver
an electric shock (the supposed purpose of the experiment being to
test the effects of pain in motivating 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 is not
actually receiving shocks) starts to make mistakes and the shock level
duly increases. As the shocks become more severe, at various stages
he starts, for example, groaning (135V), demanding to be allowed
to leave (150–65V) (the cry of ‘Get me out of here!’ becomes con-
tinuous 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 predictable reluctance, the experimenter has a series
of carefully scripted prods to help you continue (‘Please continue’;
‘The experiment requires that you continue’; ‘It is absolutely essen-
tial 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 per cent of subjects went as far as 300V, and 65 per cent
went to 450V. Variations of the experiment carried out by Milgram
and by other psychologists in a wide variety of countries and cultures
across several decades show similar results. 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 stranger under circumstance in which the punishment for diso-
bedience is no more than the experimenter’s disapproval. Milgram

[I]f 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, multi-layered
explanation that includes an evolved disposition towards obedience
(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 have entered it. These ‘binding factors’ include:
(1) their voluntary agreement to participate (commitment and
consistency); (2) receiving payment for participating (recipro-
cation); and (3) the forms of expert and legitimate authority
represented by science and its methods, Yale University, the lab-
oratory 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 not. They all contribute to a ‘barrier’ of anxiety
(Milgram, 2005, p. 154) that the majority of subjects were unable to
surmount, leading Milgram to conclude:

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 (e.g. Blass, 2000) is that it
is the combination of legitimate and expert power that makes disobe-
dience so difficult. 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 invariably comes with forms of reward and coercive power.

Numerous factors in the experiment ensure the experimenter’s
legitimacy: his management of the procedure; his control of the
space (the lab); his relationship to the place (the university) in which
the experiment takes place; and the epistemic and moral authority of
science (which is signified by all of the above), plus his demeanour
and dress. The wording of his ‘prods’ makes no direct reference to
any punishments for not continuing, 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.

(Ibid., 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 subjects were unable to overcome was the result of fear
of the experimenter’s disapproval, which itself – Milgram plausibly
speculates – is in large part caused by a generalising of an internal-
ised association between authority, rewards and punishment which
begins in childhood.


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? [Acceptability]
Q2: Is this person currently in a position to exercise this power?

[Relevance] (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 (Jay Roach, 2000).)

Q3: Are there other considerations that might override the person’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.


In argumentation, coercive power is usually analysed in terms of what
are called ad baculum arguments. Ad baculum means in Latin ‘to the
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’; their basic form being:

P1: If you do not bring about X, then I will commit myself to seeing to it
that 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 general linguistic usage, these terms are fairly interchangeable,
but one important difference is relevant to our discussion. The man-
ner in which we relate 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 legitimate
power (coming from the law courts, one’s teacher, spouse, and so on),
whereas a threat implies someone ‘taking the law into their own
hands’. That said, the IRA issued plenty of bomb warnings, and the
Mafia will also ‘warn’ people, but arguably these organisations use this
language precisely in order to create or sustain an air of legitimacy.
‘Threat’ sounds thuggish, ‘warning’ sounds principled.

Important for us though is that we recognise this distinction
between legitimate and non-legitimate appeals to force. In the fields
of applied ethics and politics this is foundational for discussions on
whether and under what circumstances direct action (for example,
blockading, occupying, boycotting), war and terrorism are accept-
able. These are profoundly important questions for those who regard
the physical punishment of others as extreme and uncivilised, and
very much a last resort.


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 her power? [Relevance]

Q2: Where not legitimate, is the person making the threat able to
use the force they have threatened? [Acceptability]

Q3: Is the person being threatened able to comply? [Acceptability]
Q4: Is the person 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.

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 criteria
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 situation 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 unmet targets is grounds for dismissal, it
is only these grounds that could be the basis of 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 sat-
isfactorily 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 (per-
haps through manufacturing 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 baculum arguments 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 unavoidable (in trying to convert a racist, for
example), and is perhaps implied in quite a number of discussions.
Open-mindedness is limited by what is clearly false, implausible and
morally unacceptable, and so argumentation has the power to begin
or end relationships. Also, however, a consequence of argumenta-
tion is that we find out who we do and do not agree with, and
sharing views is a determinant of liking and respect.


Ad baculum arguments are a subcategory 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 conclu-
sion. 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 1, emotions tend
to cloud and bias our thinking. Once the fearful possibility is raised,


critical questions can be overlooked or our capacity to investigate
them accurately is to some degree impaired. An extreme case is ter-
rorism, the potency of which relies on ‘an irrational tendency in
human nature … to magnify unexpected and “mysterious” evils out
of their true proportion’ (Narveson, 1993, p. 154). The antidote,
Narveson says, is: ‘Rationality, plus a modicum of courage. The life
expectancy of the average citizen is very little reduced by terrorism,
whereas the expected evil if we accede to the terrorist’s demands is
great’ (1993, p. 154).

The critical questions to apply to the fear element of ad bacu-
lum arguments (i.e. the general questions applicable to all appeals to
emotion) can be found in Chapter 1).



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.

In the case of threats or warnings issued legitimately, further dis-
cussion can ensue around the nature of this legitimacy, but if it is
simply someone forcing you to do or believe something or else suffer
the consequences, then they are effectively declaring the discussion
over. This is what happens when peace talks collapse, and typically
this is what war is – an unwillingness or inability to live with a disa-
greement that has proved unsolvable through dialogue.

Implicit ad baculums can occur via the disapproval we often
signal through non-verbal cues such as facial expressions and by
paralinguistic cues such as 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. (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 at some level, but do not want to license
this feeling, or communicate it, precisely because it will prematurely


close down the dialogue. In cases like this, the dispositions of
meta-cognition and self-control become valuable for maintaining
a constructive dialogue.

A final point worth considering is that there is one type of dialogue
in which threats are accepted as legitimate – the negotiation.12 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.13


In his discussion of individual responses to the experimental condi-
tions, Milgram distinguishes between a ‘Professor of Old Testament’
(2005, pp. 49–50) who disobeyed orders on the basis of a differ-
ent (‘good’) authority (God); and a woman (referred to as Gretchen
Brandt), whom he sees as having firmly integrated ethical values
which are able to empower 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
courage – similar to that needed to prevent groupthink – is required.
Not only is it a matter of disobeying orders from an authority whose
position is still upheld by a situation or by the perceptions of the
majority as legitimate, but of transgressing 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 embar-
rass us). Milgram makes the point by suggesting that readers think of
someone they have respect for:

preferably someone older that yourself at least by a generation, and who
represents an authority in an important life domain [teacher, priest, possibly


a parent] … a person you would refer to with some title [Professor, Dr,
Father, etc.] … a person who represents to you the distance and solemnity
of a genuine authority. To understand what it means to breach the eti-
quette of relations with authority, you need merely present yourself to the
person, and in place of using his title … address him using his first name,
or perhaps even an appropriate nickname.

(Ibid., p. 152)

Chances are we will not be able to do it, and even if we could,
we would experience profound anxiety beforehand. Milgram’s
point is that hierarchical etiquette of this kind is so deeply embed-
ded in our relations with others that it will significantly contribute to
our inability to disobey what are clearly inappropriate and unethical

Not only do we need courage to stand up to de-legitimise a pow-
erful authority but, as we saw in Chapter 1, the anxiety that such
situations provoke can severely disable our ability to think critically.
In a state of anxiety, 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 rationalisations for
our continued obedient behaviour, such as seeing the experimenter
as responsible for the harm rather than ourselves, 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
have 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 situation. In this case the experiment is saturated from the begin-
ning with the ideology of science so that subsequent requests and
commands are liable to be interpreted through the lens of scientific


legitimacy and expertise. For the subject, what is going on could be
seen as wrong, but the behaviour of those who continue to represent
this frame of reference suggests otherwise. In most instances, this
contributes to obedience, and only in a minority of cases is the com-
peting frame of independent ethical thinking able to gain enough of
a foothold to help leverage disobedience.

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.


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
giving ourselves ‘treats’ – a session on Angry Birds, an episode of
The Walking Dead, 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? [Acceptability]

Q2: Are you able to do what is required to receive the reward
offered? [Acceptability]

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. Perhaps most crucial is Q1 if Milgram is right to suggest
that the approval we seek from the experimenter is conditioned by
a lifetime of associating approval with obedience. System 1 think-
ing will happily, but erroneously, generalise early family and school
experiences to our relationship with the source of authority in this


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).

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
character of the person upholding a position to help establish the
plausibility 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

In this section I am going to use ‘ethotic authority’ to refer to
arguments appealing in a positive way to moral character, and ‘abu-
sive ad hominem’ for arguments that appeal to it in a negative way.


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 it begins to make sense. Perhaps 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 argues for campaigning against further cuts to low income ben-
efits, 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 contribution
of another colleague to a project we are working on, this would be
down to the modest person’s relatively unbiased insight into these
types of situation. In another kind of example, if a person’s good
character is typical of Christian ethics, then the appeal is less to them
as a person and more to them as someone who instantiates the legiti-
mate authority (for many) of Christianity.

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 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 seems to be this for many
Americans, Sir David Attenborough for many British people, and
Nelson Mandela globally. In cases where these people are cited in
support of a claim, or as making that claim themselves, we will, as a
matter of fact, be more likely to agree with them. But the circum-
stances 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 arguments are most
clearly defined by situations in which, with no further expertise
implied, a culturally conditioned notion of a good person 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 discuss-
ing the case even though they are all convinced of the defendant’s
guilt. Juror 9 is quite explicit when he argues that he is willing to
accept the request because he is impressed by certain characteristics of
Juror 8 – such as his willingness to stand up to the majority – which
are not relevant to the case itself.

In his unusual and very readable social psychology text book,
Roger Brown (1986) endorses the practical necessity of this type of


reasoning under circumstances of unavoidable ignorance. He describes
the hypothetical situation of being a member of a committee that
has to make some technical decisions on an issue where he is clue-
less (concerning bringing cable TV to Cambridge, Massachusetts, in
1983). Ideally he should absent himself, but not wanting to admit his
ignorance, his next best bet is to base his decision on the choices made
by those people on the committee he most admires in ethotic terms:

In a good expressive discussion … information is transmitted that is not
part of any relevant argument. Each participant in some degree expresses
his … assertiveness, intelligence, education, fair-mindedness, compas-
sion, prejudice, cowardice, and so on.

(Ibid., p. 234)

The critical questions relating to ethotic arguments are these:

Q1. Is the person someone who can reasonably be regarded as hav-
ing good character? [Acceptability]

Q2. Is the good character of the person of the right kind to be rel-
evant to the situation in question? (Counteracting the ‘halo
effect’ (see below).) [Relevance]

Q3. Is the weight placed on their character proportionate for the
situation in question? [Sufficiency]

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 or practice-specific, but the
preceding discussion indicates that in many cases it will encompass
someone who exhibits some of the core critical thinking dispositions.

Q2 highlights the situation-specific nature of good character;
Juror 8 is the right beacon in a jury room, but not necessarily on
the battlefield.

Q3 serves as a reminder that placing significant emphasis on eth-
otic arguments is often most appropriate in situations of uncertainty.
As is explained in relation to 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 respon-
sibility for our actions in situations where we should be.



Someone possessing ethotic power will tend to function as a role
model. A role model is someone we imitate, and this notion is help-
ful 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 recent example
being the fashion among young Christians for wristbands 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. Making reference 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 Ripley (from the Alien films) do?’;
‘What would Grandma do?’

A common approach in marketing is to employ the ethotic qualities
of celebrities to enhance the appeal of products. The idea is that the
intelligence, sincerity, or good taste of people like George Clooney or
Stephen Fry 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
psychologist Edward Thorndike – refers to our tendency to gener-
alise from a person’s known positive qualities to an assumption that
they possess other positive qualities. An intelligent person is more
likely (without any evidence to support this) to be judged a mor-
ally good person than a less intelligent person; someone attractive is


assumed to be above average in intelligence, and so on. This effect is
explained by Daniel Kahneman, who calls it ‘exaggerated emotional
coherence’, as a best guess made on the basis of our feeling about
someone. We meet Joan at a party (to use his example) and find her
‘personable and easy to talk to’. In a later conversation the question
is raised as to whether she is likely to contribute to charity.

What do you know about Joan’s generosity? The correct answer is that
you know virtually nothing because there is little reason to believe that
people who are agreeable in social situations are also generous contribu-
tors to charities. But you like Joan and you will retrieve the feeling of liking
when you think of her. You also like generosity and generous people. By
association, you are now disposed to believe that Joan is generous … Real
evidence of generosity is missing in the story of Joan, and the gap is filled
by a guess that fits one’s emotional response to her.

(2012, p. 82)

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.


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 characteristics
rather than more behavioural features (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 critical
thinking dispositions. In certain circumstances, a person’s views can
rightfully be taken less seriously if they are dishonest, dogmatic, arro-
gant, inflexible, cowardly, disrespectful, 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 moves that we tend to resort to automatically when
under pressure and when emotions are strong.



One form of direct ad hominem concerns the questioning of some-
one’s motivations for appealing to authority in their arguments.
This is worth paying some attention to here because it focuses on
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
recognition 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 epigraph 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 refers 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
personality that inspires love and makes people want to follow them
(see Weber, 1978, pp. 242–3; Burns, 1978, Chapter 9).14

In all of these cases there is a comforting sense of security and
certainty that comes with handing over responsibility to someone
else. In some situations (such as small child to parent) there is noth-
ing wrong with this, but in many others it can be a dangerous
illusion. With people in positions of power considered as charis-
matic, there can often be a willingness to assume this responsibility,
and history attests that the two motivations combined can have
tragic consequences.

Determining when an appeal to authority is wrongly motivated
is significantly context dependent, but the questions we need to ask
are clear enough. First:

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 there being 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 in an important sense
indirect, then they can carry all manner of prejudiced assumptions
relating to people’s credibility.15 Racial and other stereotypes are
fallacious direct ad hominem arguments. When hearing or making
direct ad hominem arguments, we should pause to consider the critical


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).
To mitigate prejudice, dispositions associated with listening and
respect are fundamental. Also important is a certain form of meta-
cognition 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).


Wife: I really wish you’d be nicer to my family …
Husband: Oh, and I suppose you’re a model of tolerance towards mine

all the time, are you?

This kind of interchange is remarkably common; instead of respond-
ing directly to a criticism, we hit back with a version of the same
criticism (tu quoque translates from Latin as ‘you too’). Tu quoque is a
peculiar piece of argumentation because, on the one hand, it’s typi-
cally 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 are 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 cannot 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 serious
way because inconsistency, or hypocrisy, is seen as a significant char-
acter 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, but not
if his film is not making an impact on publics and policy-makers in
the first place. Arguably it was his presence in these places (plus the
commitment to the cause that his touring demonstrated) that made
people pay attention.16

On the other hand, the Church of England criticising payday
loan companies, while at the same time investing in one of them
(Wonga) looks like a stronger tu quoque argument.17 As does a recent
attack of the Speaker of the House of Commons John Bercow for
(among other things) spending £367 ‘taking a car to Luton [about
35 miles from central London] to deliver a speech on how MPs were
restoring their reputation after the expenses scandal’.18

In the wife and husband dialogue in the epigraph, things are
more complicated. It is possible that the husband’s response is
entirely defensive and that his counter-accusation is simply a
means of deflecting the discussion away from the fact that he does
indeed need to be nicer to his in-laws. It is also possible that, even
if true of the husband, the wife’s accusation is indeed mirrored
by her own behaviour. 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 conclu-
sion 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 reciprocity in so many social
encounters (see Chapter 1). Since we cannot expect perfection
in our moral and social behaviour, our standards must, to a great
extent, 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 reciprocity that can cause a group
(or couple) conspiracy of silence. In order not the 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 our-
selves 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 premises or arguments will go unchallenged, and the
‘illusion of unanimity’ will be strengthened.


Guilt by association is a variation of ad hominem arguments in which
an arguer’s association with a person or organisation considered dis-
reputable is transferred onto them. These arguments can be strong if
the association in question has an actual and relevant bearing on the
credibility of a person’s position. This might be circumstantial (for
example a football official who is on friendly terms with a number
of club owners); direct (such as a male politician’s membership of a
club known to be misogynistic), or tu quoque guilt by association (a
rebellious rocker who frequents royal garden parties and has friends
in the Establishment).

An example of a fallacious guilt by association argument was the
Daily Mail’s smear campaign against Ed Miliband (then leader of the
UK Labour Party) based on his father’s Marxism. Journalist Tim
Stanley pointed out the absurdity of this, saying: ‘The Daily Mail
published a silly thesis (Ralph Miliband was a commie, so maybe
Ed is too) under a daft headline (‘The Man Who Hated Britain’).19


More recently I’ve seen a placard in a demonstration against
‘Obamacare’ which read:


The argument is flimsy and vague, but the negative association will
be powerful for many Americans. In the 2015 UK General Election,
Nigel Farage – leader of the right-wing UK Independence Party – was
constantly being let down by his racially insensitive party members
and supporters. The lowest point for Farage was when UKIP turned
out to be the party of choice for one of the country’s most hated
individuals – child serial killer, Ian Brady.

During his election campaign in 2015, the candidate for the
Labour leadership Jeremy Corbyn was accused of ‘sharing a plat-
form’ with two Arab leaders who are anti-Semitic.20 It is clear that
Corbyn is not a racist, but whatever the intentions of this attack,
what could stick to someone perceived by many as being quite far to
the left like Corbyn is an aura of the more dangerous or unpleasant
aspects of radical politics.


The influence on our beliefs or actions originating from the com-
mitments 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


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).

A variation that appeals to referent power is:

P1: Everyone (or a large majority of people) belonging to my referent
group Y believes X about subject Z.

P2: Whenever something is generally accepted by my referent group
Y about subject Z as true, there’s a strong likelihood that it is

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? [Acceptability]
Q2: If so, is X a domain of knowledge where popular belief is rel-

evant to its truth? [Relevance]
Q3: If so, is the population whose beliefs are referred to the one with

the appropriate knowledge of X? [Acceptability]
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?) [Acceptability]

Q5: If so, is the conclusion reached on the basis of general belief
appropriate (e.g. suitably cautious). [Sufficiency]

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 indeed worse than
its bite; that Morton’s fish and chips are the best in town. Another
is practice-based knowledge: gardeners knowing that grass benefits
from being cut whereas weeds do not; comedians knowing that
jokes about sex will go down well with most audiences, or football
managers knowing that making a substitution as you are about to
defend a corner kick is risky.

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. Religious belief is an out-
standing example; it could be true, but millions believing that it is
does not count as a reason for a neutral 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 America Christian right is a specific and serious example,
and the exposing of commonly held beliefs as myths is common in
popular science forums.21

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 cul-
ture, but some beliefs remain common because they are comforting
or otherwise useful.

Q3 is covered by previous analyses of position to know arguments.
Q4 is directed at ideas like groupthink, but also covers dogmatic,
lazy, fearful, antagonistic, or competitive group mind-sets that create
dangerous biases and unreliable communication. 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 situations 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 mili-
tary practices, but they are not going to be the last word on the matter.

Twelve Angry Men is also helpful for illustrating the important
difference between ad populum arguments that refer to actions and
those that refer to beliefs. As discussed in Chapter 1, 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 where going along with
the majority – despite not agreeing with them – can be the best or
right thing to do. Needless to say, these kinds of deliberation come
with their own critical questions (for example, 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.

Some recent cases of social media-based ‘mob pressure’ or ‘witch
hunts’ do, however, raise some intriguing questions concerning the
power of ad populums. One high profile example is footballer and (at
the time) convicted rapist Ched Evans, in whom several clubs were
interested on his release from prison in 2014, but in each case the
vigorous petitioning of fans led to the clubs deciding against this.
Evans has always denied the charges, and at the time of writing he is
yet to resume his professional career.

Another is the forced resignation of Nobel Laureate Sir Tim Hunt
from his academic post at University College London, after making
sexist comments (in the form of a joke) at a conference. UCL’s deci-
sion is 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, sometimes …


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.22

The influence of social media in the Ched Evans case is clearer cut
than with UCL and Tim Hunt, but if we assume that this medium
is able to amplify what appears to be popular moral views in a way
that has, or can have, this kind of impact, then the helpfulness of the
critical questions in guiding our analysis of such situations is plain.
The answer to each of them is not obvious 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.


As the discussion of social proof in Chapter 1 shows, appeals to
popular beliefs and behaviours are known to be extremely persua-
sive; having an immediate and powerful effect across a wide range
of situations. Schultz et al. (2007) measured the effects of social
proof – and in particular referent power – on energy consumption
by adding to the electricity bills of the residents of a particular street
information which demonstrated their household consumption in
comparison with their neighbours. This had an effect on electricity
usage referred to as a ‘rush to the middle’; those with higher than
average bill consumed less, and those with lower consumed more.

This is further evidence for the power of social proof, but the aim
of the study was to find a way to persuade people to use less energy,
not bring everyone closer to the average. The solution was to add
to this purely descriptive norm a moral norm, originally in the form
of a smiley face for lower than average consumption and a sad face
for above average. The effectiveness of the moral norm can seem-
ingly be attributed to the broad legitimacy of environmental values
(or at least of not being wasteful), but the adoption of this practice
by a progressive company called Opower serves as an example of
referent power. The moral (or ‘injunctive’) norm method works,
but not always for customers with a non-progressive political lean-
ing (Republican voters), some of whom increased their consumption.
Unlike liberal customers, they will not identify with the overtly


progressive approach of Opower and the kind of people who might
aim to reduce consumption. But rather than this simply not affecting
how much electricity they use, it could lead to a slight increase as an
act of defiance.23

The general understanding among social psychologists is that we
tend to rely on referent power in particular when we are unsure how
to behave in a particular situation (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 obe-
dience rates for the remaining (real) subject dropping dramatically
(2005, pp. 117–22). Exactly what constitutes an acceptable level of
energy to consume is a difficult question, and it makes sense that it
falls under the kind of situation in which we turn to those who are
‘like us’ for a benchmark.


The way we handle arguments from popular opinion or that carry
referent power is, in part, a function of our courage and of our will-
ingness to take responsibility for our commitments. Zachary Seech
rightly emphasises the pressure that can come from ad populum argu-
ments, not so much towards changing our minds, but to silencing us
through fear of appearing stupid or out of touch:

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 this can be more fearful than the consequences referred
to in ad baculum arguments. One of Christopher Tindale’s ad popu-
lum 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 you do have something to prove. Like
Socrates, you are regarded as an irritant.

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 and
improve ourselves, then it is important to be receptive to the, often
painful, feedback that they can provide. While it can sometimes be
right to be sceptical of isolated comments 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, colleagues, and so on tends to indicate a failing of disposi-
tions such as courage, modesty or open-mindedness.


As with other chapters, use the in-text references as a guide to further
reading on specific aspects of this chapter. A couple of further rec-
ommendations are Douglas Walton’s book Ad Hominem Arguments
(2009), and for a concise and illuminating discussion of ad baculum
arguments, see J. Woods (1995) ‘Appeal to force’.


1. Assess the following tu quoque argument (from Roger Scruton,
On Hunting (1999), pp. 139–40) by applying the ad hominem criti-
cal questions to it.

It puzzles me that [those who oppose fox-hunting] should have sin-
gled 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 morally 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. Annie is a line manager in an organisation with direct responsibil-
ity for about 20 staff. Although her team works relatively well, she
feels that there is more she could do to improve its performance.
She feels that the management training she has so far received is
not enough to enable this improvement, so she is looking to learn
more. Her own line manager is in the same situation and regards
herself as largely self-taught. There are endless popular books on
how to be a good (or ‘great’ or ‘extraordinary’) manager, but these
tend to advocate a particular approach, she suspects that many are
quite shallow, and how is she to know which of these she should
be consulting? There are also a vast range of academic textbooks
on the subject, but which should she choose? In her experience,
different leadership styles and techniques will be effective in dif-
ferent situations (different types of organisation, different types of
employees, and different types of leaders/managers etc.) so any
answers she finds will need to involve decisions made by her on
what methods to use in her particular context. This then suggests
that she will need to find out as much as possible to allow her to
make these decisions, but how does she go about this? With work
and family commitments she does not have time to study for a
part-time management degree, or to read immense amounts of
academic literature, but without this depth of learning, she realises
that she will be quite heavily reliant upon the authority of a range
of authors and colleagues to steer her in the right direction.

Using some of the ideas and critical questions explored in this
chapter, make some recommendations about how Annie should
approach this problem.



1 A modern advocate of this, drawing on Aristotle’s and Seneca’s ideas on virtues
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) has seven.

3 The Life Scientific, BBC Radio 4, 28 July 2015.
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 E. Pennisi (2011) ‘Researchers challenge E.O. Wilson over evolutionary
theory’. Available at:

6 My analysis of these issues has been significantly influenced by Heather
Battaly’s article ‘Attacking character’ (2010).

7 The Guardian, 15 August 2015, p. 32, and available at:

8 You know who you are!
9 See R. Mason (2014) ‘Fiona Woolf resigns as chair of government’s child

abuse inquiry’. Available at:
fiona-woolf-resigns-chairman-child-abuse-inquiry. (For other reasons, Lowell
resigned (or was sacked) in 2016.)

10 Adapted from Walton (2006, p. 123) and Tindale (2007, p. 89).
11 Described and analysed in Hannah Arendt’s book Eichmann in Jerusalem: A

Report into the Banality of Evil (Harmondsworth: Penguin, 1965).
12 For an insightful discussion of this, and ad baculum arguments in general, see

Woods (1995).
13 It should also be noted though that an alternative to this bargaining approach

exists – ‘principled negotiation’ (Fisher and Ury, 1991). In this model the attempt
to understand the other side’s core commitments in order to find creative solu-
tions is promoted in place of a more superficial haggling over positions.

14 For two excellent example of charismatic leadership (as well as a range of
rhetorical techniques), see the character Andreas Wolf in Jonathan Franzen’s
novel Purity (London: Fourth Estate, 2015, p. 262), and Forest Whitaker’s
portrayal of Idi Amin in the film The Last King of Scotland (Kevin MacDonald,
2007), in particular the section where Amin is persuading the young Scottish
doctor to be his personal physician (between approx. 24 and 36 minutes).


15 A very good book devoted to this subject is Miranda Fricker’s Epistemic
Injustice (Oxford: Oxford University Press, 2007).

16 See A. Hoffman (2011) ‘Talking past each other? Cultural framing of skepti-
cal and convinced logics in the climate change debate’, Organization & Envi-
ronment, 24(1), 3–33.

17 See

18 The Guardian, 25 July 2015.
19 T. Stanley (2013). Blog. Available at:


20 H. Rifkind (2015) ‘Jeremy Corbyn is not an anti-Semite but he is reaping what
he sowed’. Blog. Available at:

21 For example, Discovery Channel’s Mythbusters, and the Fortean Times column

22 The Guardian view on the Tim Hunt affair: an explosive combination of sci-
ence, sexism and social media (2015). Editorial. Available at: www.theguardian.

23 See O. Payne, Inspiring Sustainable Behaviour (London: Earthscan, 2012),
pp. 97–9.




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

(William James, ‘The will to believe’, 1967)

Generalisations … never right, always fun.
(Henry Rollins, Talk is Cheap, vol. 3)

When decisions are made, they are often justified in term of the
consequences they will 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 goal-line technology in football will not impede the
flow of the game, eating less processed meat will make us healthier,
and so on. But we of course – 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

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
consequences, 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 usually problematic,
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, are fundamental to the students’ and practitioners’ ability
to think critically in these areas. Critical thinking as a subject will
explore the basic forms of this kind of reasoning, and sometimes dis-
cuss its association with the philosophy of science (see the end of this
chapter for recommended 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, 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 kinds 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.


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 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
technology I assumed that (typically) my MP3 player was a bit slow.
It otherwise worked fine though, so this was not much of a prob-
lem, 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 while going through this ritual
and let go after a couple of seconds. When I looked back at the
screen a few seconds 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 nothing actually appears on the screen
until after 6 or 7 seconds.

Apart from a need to reassess my relationship with technol-
ogy, 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 theoretical work of the
natural and social scientific disciplines and to the work of histori-
ans. Each discipline will have distinctive 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
reasons for acting in certain ways different in kind from the mechan-
ical 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 (even if not always the distance and direction I want
it to go), but 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 neces-
sary conditions, including causality, but that the actual basis of our
experience of causation is itself unknowable.

It is this limitation on our knowledge, combined with the com-
plexity of the world, and combined with what Kahneman (2012,
pp. 74–8) suggests 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 criminal behaviour; or of particular events like the
Reformation, the American Civil War, and the fall of apartheid in
South Africa are complex, intertwined and hard to establish with
certainty. As we have seen, we have a tendency to jump to con-
clusions about straightforward cases of cause and effect, and with
complex phenomena an added problem is of over-simplifying our

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, edu-
cation) will work together to implement solutions for child abuse
prevention, 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 Brussels? Was it all of these
things combined, or some of them, or something else entirely? This
is clearly not a matter for a research project or a team of practition-
ers, 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, however, important to sur-
vey some general principles of causal reasoning. The aim here is to
seek to clarify these through the description and analyses of causal
reasoning errors. After this, I will explore the dispositions and psy-
chological 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 recreate conditions in which
the cause and effect underlying the correlation can be adequately

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 particular instances like these, we can normally
still find historical precedents or comparisons with subsequent
events (for example, the causes of the 2008 financial crisis shared
some features with the 1929 crash, and the causes and conse-
quences of the 13 November 2015 shootings in Paris are not
entirely dissimilar to other terrorist attacks), and thus we find
that inductive arguments from analogy (see Chapter 7) 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 confidence 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.



When attempting to establish the cause of a happening, we can be
presented with two broad kinds of relationship between it and its
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 the event’s 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 relation 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’m being too quick or
too lazy in my hypothesising? If I reflect, then perhaps it was Uncle
Vance’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
arrived home from work.

The main fallacies associated with causal reasoning fall under the
headings of coincidence, direction of cause, shared cause, multiple
cause, placebo effect, and mistaking cause for correlation (the Sod’s
law fallacy). These apply to either one or both of the types of cor-
relation outlined in the previous two paragraphs.


If two events regularly occur together but are not causally related, the
chances are that they have a shared cause (see below). With singu-
lar or unusual events, there is a strong possibility of mistaking mere


coincidence 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 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 ‘abstain-
ing 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 beliefs.

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 long-livers, these
recommendations are largely anecdotal; no one knows.

One other example is more serious because it identifies a particu-
lar type of mistake made in formal scientific reasoning. Below we’ll
look at the placebo effect as the basis of a discrete category of causal
reasoning error, but it’s also possible to infer a placebo effect where
none exists. The phenomenon known as regression to the mean
predicts that an extreme measurement for any phenomenon will
typically be followed by a less extreme one. As hard-to-please cricket
pundit Geoffrey Boycott was keen to point out, getting too excited
about young England batsman Ben Stokes’ record-breaking 258 runs
in the Second Test against South Africa in January 2016 is premature;
he has to consistently score high to really prove his self. In terms
of everyday causal reasoning, regression to the mean is particularly
problematic when trying to understand pain and other symptoms.
These tend to occur in cycles – severe  less severe  severe, and
so on – and there’s a reasonable chance that people will seek medi-
cal 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,
but for regression to the mean as well.


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 plau-
sible a causal connection seems, it is worth reversing it to see if there
is another possibility. In fact, the more plausible the direction of cause
seems, the more revelatory a reversal can be. Group psychotherapist
Irvin Yalom (1980, p. 394) 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 confronted with
the negative effect this was having on the others, and Yalom encour-
aged her to take more of a risk by making more honest and direct
comments. At this point she revealed that she was an alcoholic, and
said that the shame that this led her to feel was the reason for keep-
ing herself hidden. Yalom, however, proceeded to reverse her causal
argument, suggesting, ‘She did not hide herself because she drank,
she drank because she hid herself.’ 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 isolation, rather than her isolation resulting
from the shame associated with her drinking.


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 aver-
age than non-religious people, this could be because religion leads to
greater happiness, but it could also be that a third variable – for example,
religious families being more stable – results in both the happiness and
the religious commitment.

In a contrasting example some recent research 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 direction of cause is usually mentioned in press coverage,
and the article itself, though very cautious about causation, suggests
that identifying with the goth sub-culture does increase vulnera-
bility 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 conditions 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 over-
looking them.


System 1’s fondness for jumping to conclusions is abetted by a
dislike of complexity. What has been called the ‘narrative fallacy’
(see below) is explained by a desire for coherence in our under-
standing 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 over-simplification of causes. For example, there is a widely
held belief that it was the Great Fire of London in 1666 that ended
the bubonic plague epidemic in England that had started the previ-
ous 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 black rats that carried the plague fleas, the plague was


on its way out anyway. Theories suggest 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 availability of the fire and the plague, and the
rough plausibility of one destroying the other, is no doubt at least
partly responsible for the myth’s perpetuation. But whether or not
that is the case, the attraction of simple causal stories is again dem-
onstrated. Further examples include the cause of the American
Civil War (which was not just about slavery), and it is interesting
to note that many researchers into environmental communication
and behaviour change cite the inherent complexity of the causes
and effects of climate change as a reason why the public often fail to
engage with the issue.3


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 coincidence, nor a mistake in the direction of cause, nor an
overlooking of a shared cause, and nor does it have to be part of
a multiple cause. Instead the cause is very much connected with
the mistaken cause, but in a way that is peculiarly indirect. I say
‘peculiarly’ because 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 altogether that has its own begin-
ning 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 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

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 is predicted (by a fictitious test) to
make an intellectual ‘spurt’; tell the other teacher 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 stu-
dents 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 particular contexts. As a culture we are reasonably
familiar with these phenomena now, but we can nevertheless be sur-
prised by the potency of unconscious processes in affecting physical
and social change. For this reason, and perhaps because of their rela-
tive invisibility, they are easy to overlook or dismiss.



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 30 mph
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 inconvenient time,
and is essentially about bad luck. It’s Sod’s law that my boiler breaks
down the very month that my 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 driv-
ing example, it’s not Sod’s law that you ran someone’s dog down, it
was because you were driving too fast and could not 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.

A somewhat different set of motivations lie behind this next
example. In October 2015, the singer and activist Charlotte Church
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 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 indi-
cated a plausible connection between these things. It is known as the
‘food shock’ argument and connects a food shock in 2011 (extreme
weather such as droughts leading to globally significant food short-
ages which have a disproportionate effect on poorer places, especially
those reliant on imports), 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-one-hundred-year event to a one-in-thirty-year phenom-
enon, and potentially a seven-in-ten-year 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


The critical questions that can be applied to causal arguments
correspond to the headings above:

Q1: Has a correlation been identified? [Acceptability]
Q2: If so, how convincing is the causal reasoning that attempts to

explain the phenomenon in question? [Sufficiency]

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 rela-

tionship (rather than a mere correlation) has been dismissed too
quickly or for the wrong reasons?


Errors in causal reasoning are exacerbated by several features of our

• 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 all the more likely that we will
jump to roughly coherent, but incorrect conclusions.

When this year’s Halloween pumpkin began to rot, the first thing that
happened was the jaw collapsed, closing up the mouth and making
it appear more round again. My 4-year-old reasoned that it ‘wants to
be a pumpkin again’, i.e. rejecting the grotesque face I’d carved. And
like many kids of his age he has surmised that the clouds rain in order
to water the plants. The need for causal explanations is evidenced
by kids’ constant ‘why’ questions, but abstract scientific reasoning is
not available to their cognition, in which case they attempt to make
sense of the world in accordance with what they do understand –
conscious intentions.

Two areas of research that make particularly interesting con-
tributions to our understanding of causal reasoning errors are the
psychology of superstition, and what’s been termed the narrative


In his investigations in to 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
example. In his studies of superstition, however, this causal con-
nection 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 repeti-
tion 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.

In the pigeon’s case, it cannot fathom the underlying causal fea-
tures of the situation, but as a species humans have, as we know,
become increasingly 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 will not make System 1’s anxiety at the prospect of wearing blue
rather than black go away without a fight.

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.
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 result-
ing 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 coherence of a causal explanation – for example, that it is con-
sistent with known truths – is a necessary but far from sufficient
condition 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). It refers to our attraction to, and
confidence in, coherent but simple explanations at the expense of an
appreciation of complex and unknown factors. Our desire to make
sense of the world rather than acknowledge the limitations of our
understanding leads us to invent (or readily accept), and then act
upon, straightforward, but inaccurate, causal stories.

One of Taleb’s examples is especially revealing of the entrenched
and automatic nature of this way of thinking. After hearing Taleb
speak at a conference, a fellow academic congratulates him on his
theory (as presented in an earlier book) but also reveals his frustra-
tion because he had been planning to write a book on the same
subject. He then proceeds to hypothesise that Taleb would not have
had the idea had he been brought up in a Protestant country where
outcomes are so firmly linked to individual efforts rather than luck
and circumstance. According to Taleb’s theory, of course, there will
be a multiplicity of causes of how he came by this theory and of its
subsequent success, many of them unknown. This learned admirer,
however, manages to reduce this to a simple narrative, and therefore
commit the very fallacy he is discussing. If this is not ironic enough,
for a short while Taleb finds himself actually agreeing with him
(ibid., p. 63).



The narrative fallacy provides psychological insight into all of the
causal fallacies explored in this chapter, which share two essential
features: seeing causes where they do not exist, and overlooking
complexity. 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 overlook complexity, and you can see how vulnerable
we are to false claims about how buying this product, performing
this behaviour, or voting for this party, will make us happier. To
protect ourselves, we need to do the following:


• work towards greater clarity about what we can and cannot
know, control or predict;

• develop meta-cognitive awareness of the conflict between how
things are in terms of what we can control and predict, and how
we’d like them to be;

• develop meta-cognitive awareness of our proneness to supersti-
tion and the narrative fallacy.

We need to learn to tolerate uncertainty and be more like Juror 8
in Twelve Angry Men (see Chapter 2). The illusion of control, like
other self-serving biases, is in opposition to modesty, understood as
a realistic perception of our attributes and abilities. From a dialogical
perspective, encountering over-confidence in others can be demo-
tivating and lead to antagonism. Moreover, jumping to conclusions
has an unfortunate self-fulfilling tendency. The principle of com-
mitment and consistency (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, implying as it does dispositions such as wisdom,
trustworthiness and not being unduly influenced by peer pressure. If
a conclusion is prematurely declared, therefore, retracting it this 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.’


Causal arguments are often generalisations, but not all generali-
sations are causal arguments. Another type involves arguing from


qualities that characterise a particular class of things (aircraft, poli-
ticians, pandas, scientific theories, vegetables, and so on) to the
expected qualities of individual instances of that class (the 737 I’m
currently sitting on, Hillary Clinton, Ling Ling, string theory, run-
ner beans). Perhaps the most recognisable version of a generalisation
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

In a similar fashion to causal arguments, generalisations are fun-
damental to scientific reasoning. A significant proportion of the
methodological 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 rec-
ommendations for critical thinking textbooks which go into detail
about generalisations, often with a philosophy of science perspec-
tive, but you should also bear in mind that texts on natural and
social scientific research methods are also an important (and often
overlooked) critical thinking resource.

Generalisations are most commonly classified into two kinds:
absolute (also known as ‘universal’, ‘strict’, or ‘hard’), and non-absolute
(also known as ‘non-universal’, ‘inductive’ or ‘soft’). Absolute gener-
alisations are those that claim that all of a certain type of thing are
a certain way (like ‘all countries have flags’), and non-absolute gen-
eralisations claim that ‘most’ or the majority of a type of thing are a
certain way (like ‘most countries have land borders with other coun-
tries’). Non-absolute generalisations can be subdivided into statistical
and non-statistical forms. The latter will use vague terms like ‘most’,
while 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 4). 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 typically involve qualifiers like ‘all’ or ‘most’, and they
certainly never involve statistics. As with all plausible arguments,
their most important feature is their practical and contextualised
nature. They are saying that prior personal experience and common
knowledge have established the right to presume X about situation
Y in circumstances where more rigorous investigation is not practi-
cal; ‘meat that is before its sell-by date and which has been stored in
the fridge is edible’, for example.

Like all defeasible arguments, there is a background assumption
that there are very likely to be exceptions to the rule, but that the
burden of proof falls to those who want to claim an exception
in this instance. Defeasible generalisations, however, are vulnerable
to stereotyping; to being derived from ‘common knowledge’ that
is in fact inaccurate, and personal experience that is skewed by
the confirmation bias. This will be addressed under ‘sweeping
generalisations’, 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.


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, 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.

(1991, p. 74)

This is a case of generalising from too few examples to a conclusion
about the whole, and as in many cases of this (we’re all guilty), the
‘too few examples’ come from our own particular experiences. It is
understandable why we do this; our experiences are vivid and con-
crete and thus collude with the availability heuristic. They exaggerate
the relevance of the pattern in question, prompting us to jump to a

Hasty generalisations are also known as the ‘fallacy of insuffi-
cient 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 market research and opinion polls) are not fal-
lacious when appropriately derived and when their limitations are
known and accounted for. Where this is not the case, hasty generali-
sations 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 would
typically imagine – 1000–2000 people); and it is highly unlikely that
they would be representative in terms of the demographics they rep-
resent (age, sex, race, region, social class, and so on).

A variation of 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 particularly
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 difference between an unrepresentative sample and
cherry picking is that with cherry picking a more complete set of
information is available but not shared, whereas with an unrepre-
sentative sample only a partial data set exists.

Superficial and one-sided communications such as advertisements,
company annual reports, CVs, or social media personal profiles will
cherry pick like there is no tomorrow. This might be undesirable,
but is at least something that we expect to happen. In other contexts
it is more troubling. It’s a common tactic of organised corporate and
political denial of the health risks from smoking and global envi-
ronmental issues, and, as Ben Goldacre argues, is commonly used
in research carried out by advocates of alternative therapies and, in
particular, nutritionism.6 All reporting of research findings is at risk
from cherry picking, though, but in theory at least avoiding it is rela-
tively straightforward. It involves what is called a ‘systematic review’,
which puts in place rules for unbiased selection and questions to
guide assessment of previous research on a subject. The ‘traditional’
method was open to 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 always peer-reviewed in aca-
demic journals, the review part of it (i.e. a summary of the research
findings to date) was not. 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.7


The other topic Deana Kuhn used to study people’s everyday rea-
soning skills was criminals re-offending. In one example a participant
accounts for re-offending 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 commit-
ting a crime is probably habit.’ 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’ (1991, p. 60).


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 re-offending 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 re-offending – but
should this be the only cause suggested, then the arguer runs the risk
of presenting a sweeping generalisation.

An example of 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’s the work of System 1
and is in need of qualifying. (There are many examples of killing
that can be argued to be morally acceptable, such as combatants in
the theatre of war, euthanasia, capital punishment, social revolution,
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
graduates tend to be better than average critical thinkers does not
mean an employer should assume this is the case for 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
over-simplifications 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 dialogical
handling of plausible arguments), 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 over-simplification 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 she


belongs; and over-simplifying and therefore 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.


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?

Q2: If so, is the generalisation made based on a representative sample?

For defeasible generalisations

Q3: Is the generalisation plausible? [Acceptability]

For all generalisations

Q4: Is this a context in which applying a generalised truth is appro-
priate? [Relevance]


It seems that we are fond of generalising; 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 his Rhetoric, Aristotle observed how generalisa-
tions appeal to prejudicial tendencies in audiences. We like to
make quick leaps from particulars to the general, and in persuasive
communication he identifies the maxim as especially effective for
providing generalised legitimacy to the audience’s limited experiences.
Maxims are broad truths communicated in a single sentence, and,
if spoken by the right person, they carry a kind 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 familiarity 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’).

As cognitive-behavioural therapy (CBT) demonstrates, gener-
alisations do not necessarily lose their appeal even when they work
against the individual. For CBT, unhappiness and mental disorders
are caused and maintained by people’s inaccurate (or ‘inefficient’)
beliefs. Originator Aaron Beck identified ‘basic errors’ of thought in
patients such as over-generalisations and black-and-white thinking,
and among the ‘irrational beliefs’ listed by 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

Because something once affected one’s life, it will indefinitely affect it.

Because I had bad luck once, I will always have it.9

CBT seeks out unrealistic and inflexible patterns of thought that have
become automatic responses to situations and guides to behaviour.
In a therapeutic context the aim is to help people replace these with
healthier (for the most part, more accurate) ways of thinking, and the
relative success of this form of intervention is evidence of 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 erroneous
beliefs about the nature of luck, and, on the whole, people suffer-
ing depression tend to wrongly attribute negative events to features
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 are, then, strangely satisfying, and when not creating
the kind of problems CBT confronts, they are also quite entertaining.


Spoken word artist and comedian Henry Rollins observes how they
are ‘never right, always fun’;10 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 necessarily 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 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.

Towards protecting ourselves against this tendency much of the
discussion in the equivalent section under ‘Causal arguments and
causal fallacies’ is pertinent. In addition, however, over-generalisations
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 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, tempting 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 problems (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’, ‘usu-
ally’, and so on – allow for flexibility in the positions we can 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 in order to avoid an absolute gen-
eralisation 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 particular problem rather than 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,11 Matthew Dentith makes
the point that sometimes they are true, but that they have come
to be defined as something that is not true. He is arguing against
this ‘no true Scotsman’ attitude, and is concerned that we will miss
uncovering real conspiracies 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 the-
ory, is it?’ … We have been told that conspiracy theories are bunk, and
so we treat them as such.

(2015, p. 39)


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

Q1: Is this a situation in which the consequences of a decision are
the appropriate standard of evaluation? [Relevance]

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? [Acceptability]

Q4: If so, is it better/worse than consequences arising from alterna-
tive decisions? [Acceptability]

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? [Sufficiency]

On the subject of the use of drones in attacks on terrorist suspects, the
following 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 cri-
teria 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 Q2. However, because there are a number of other arguments for
and against drone strikes and their relationship to just war theory (e.g.
greater precision leading to fewer civilian casualties, but also less like-
lihood of surrender since the enemy do not feel that they are fighting
a real enemy), then Q4 is complicated 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.


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; Bowell
and Kemp, 2010; Lunsford and Ruszkiewicz, 2010), but as Douglas
Walton (2015) and others have argued, slippery slope arguments are
not necessarily weak.

They share the general structure of arguments from negative
consequences, but with these features:

• They claim that by bringing about an initial, relatively incon-
sequential (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 significant 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 catastrophic) negative consequence.

Slippery slope arguments are quite common in deliberative reason-
ing, and perhaps most conspicuously in debates about drug legislation
and euthanasia. In both of these 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 perceived 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 the public acceptance of non-voluntary

Communicating the seriousness of climate change is impeded by
its complexity and the fact that many of its impacts will not affect
us for a number of decades. It also has the form of a slippery slope
argument. Under the heading ‘Increasing magnitudes of warming
increase the likelihood of severe, pervasive, and irreversible impacts’,
the IPCC’s 5th Assessment (2014, p. 15) states:

Some risks of climate change are considerable at 1 or 2°C above preindus-
trial levels (as shown in Assessment Box SPM.1). Global climate change
risks are high to very high with global mean temperature increase of 4°C
or more above preindustrial levels in all reasons for concern (Assessment
Box SPM.1), and include severe and widespread impacts on unique and
threatened systems, substantial species extinction, large risks to global
and regional food security, and the combination of high temperature and
humidity compromising normal human activities, including growing food
or working outdoors in some areas for parts of the year (high confidence).
The precise levels of climate change sufficient to trigger tipping points
(thresholds for abrupt and irreversible change) remain uncertain, but
the risk associated with crossing multiple tipping points in the earth
system or in interlinked human and natural systems increases with rising
temperature (medium confidence).13

At the moment, carrying on as we are has relatively minimal direct
impact on the quality of life of western industrialised nations, and it
of course avoids the ‘inconvenience’ that mitigating measures will
bring for many. However, as the passage claims, such inaction will
likely result in tipping points, the timing of which is hard to predict
precisely, but which will be irreversible and lead to a range of very
serious final 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
consequences. Also, in order 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? [Acceptability]

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 policy-makers, 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 2°C rise will be very
serious indeed. The argument thus satisfies Q3, and it also satisfies Q1.
Q4 can be in some doubt in so far as sceptics have argued that miti-
gation policies are harmful to developing economies, but since the
consequences of climate change will almost certainly be worse eco-
nomically speaking, and since the concessions made by developed
countries at the Paris conference in 2015, this does not seem to
pose a threat to the argument. 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 very important that the
passage quoted makes reference to the details of the research which
confirms these predictions. With respect to Q2a, we are, ominously,
told that 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 continu-
ing to emit greenhouse gasses to the extent that we are is running
a significant risk. In light of the devastating consequences brought
about by these tipping points, the cautionary principle seems the
only logical one to apply (Q5), and thus the continued lack of coor-
dinated international action is the wrong policy.

The following example, from an article about the Scottish
Independence referendum of 2014 appears to be less successful. The
passage below is from a section subtitled ‘What good will devo
max do?’ ‘Devo max’ is short for ‘devolution maximum’, which


here refers to more powers (such as setting income tax rates) being
awarded to the Scottish government by the UK government in

If the Scots decide to vote against independence, David Cameron is already
promising that more powers will be devolved to the Scottish parliament.
Many have interpreted these additional powers as equating to devo max.
But what would be the likely outcome of the Scots being granted devo max
as a concession following a no vote? … Does anyone believe for a split sec-
ond that a Scottish government run by the Scottish National Party devoted
to extricating the Scots from the British state would be placated with devo
max? Once the Scots have it, what’s to stop them, just like any good nego-
tiator, from continually asking for additional powers and threatening to
separate if they don’t get them? Wouldn’t Scotland and England continue
to grow further apart within the UK until all that would be left to say is that
they are the two largest national components of one excessively decen-
tralised state? What good does this do for England, Wales and Northern
Ireland? The English must know that in the long term, offering devo max is
a disastrous policy fraught with dire consequences for the union.14

The author is describing how, once a short-term gain is made (pre-
venting independence), a sequence of cause and effect will occur
over which Westminster would have little control, and which would
end badly for them. This does appear to be a slippery slope argument,
in which case we can apply the relevant critical questions.

The consequences of a decision like this are certainly relevant to
its evaluation (Q1); and the loss of the union’s integrity (and the
eventual Scottish independence possibly implied by this) is not a
good outcome as far as a Unionist UK Government is concerned (Q3).
Q4 is tougher to answer because if devo max had not been on the
table, then the UK could have lost Scotland in 2014. More criti-
cal for the strength of the argument, however, is the plausibility of
the series of consequences that the author suggests (Q2 and Q2a).
A Nationalist government in Scotland might well negotiate in the
way claimed, but since this is a situation in which the no vote has
(currently) won the day, how confident can we be that the Scottish
people who are not wanting independence will not see the potential
for this slope themselves? If they do, then any party with this agenda


(most obviously the Scottish National Party) risks their majority
being weakened or being voted out of the Scottish Parliament.

So, even though you could argue that offering devo max places
the UK government in a weak position in any future negotiations
(and thus the unfolding sequence is out of their control) (Q2a),
whether the necessary forces will be at work in Scotland that would
lead to such negotiations is debatable. We can conclude then that
this is a relatively weak slippery slope argument. For the arguer there
seems to be no doubt that this move will end in disaster, whereas at
best it is a possible outcome that a Unionist UK government might
be prepared to risk (Q5).


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 not 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 policy-makers 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 set off alarm bells,
and although in some cases this might be the aim, to misapply the
label can lead to misplaced fear or panic.

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 most common example is legalising
gay marriage; a move whose detractors often claim will open the
legal door to practices like marrying animals, cars and siblings. This,


however, is more obviously a form of argument from analogy
(see Chapter 7) rather than a slippery slope argument. In the case
of gay marriage, the analogy is weak, in which case there is no
reason to suppose that social and legal systems will have to allow
these undesirable consequences to happen. If it were a strong anal-
ogy though, the detractor presumably sees something inherently
troubling in the initial step, in which case it is some distance from
the basic structure of a slippery slope argument (where the conse-
quences of the initial step are unproblematic or positive). From the
perspective of some of those opposing gay marriage, it really is as
wrong as incest.

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 intui-
tively distinguish between a heap of sand and a non-heap (perhaps
a ‘pile’ or a ‘mound’), but the process of adding single grains of
sand to the non-heap 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

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 a difference between bald and non-bald (adding or subtracting
single hairs creates the same kind of problem). We make this dis-
tinction 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 (‘please, just five more minutes’), saplings and trees, or,


more seriously (for versions of the abortion debate), the point at
which a zygote becomes an embryo, or an embryo a foetus.

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 the basis of roughly where the
line should be drawn, and, very importantly, on the basis that a line
has to 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
55 mg or 155 mg.

Stronger sorites-type slippery slope arguments are found in situa-
tions where rules are not enforceable. My wife and I are always up
early because of work and the children, but when the latest series of
The Walking Dead or Game of Thrones is released on DVD, we face
a problem. We say we will limit ourselves to one or two episodes
a night, but often end up watching three or four. ‘Just one more?’
when this is 40 minutes less sleep is hard to resist, but the cumula-
tive effect is the problem. This is without doubt a slippery slope, and
because it is so addictive, we sometimes just have to not step onto it
and leave the box set until Friday night.

Insensitivity to small, incremental changes kills frogs (if put in boil-
ing water, they jump out, but when placed in very gradually boiling
water, or so some claim, they do not notice and fail to escape) and
causes sleep deprivation in humans like me. But it also has its ben-
efits. In what behavioural therapists call ‘systematic desensitisation’,
phobias are cured through incremental exposure to the fearful object.
Someone with arachnophobia cannot get close to spiders, but if they
can tolerate being on the other side of a large room as a contained
spider, this can be the start of a process of counter-conditioning. The
person is encouraged to move closer and closer to it – one step at a
time if necessary. The lid comes off the container and they get closer
still, until finally the beast is running across their hands. At each stage
the small step must be accompanied by a manageable level of fear,
and it is maintaining this level that is made possible by the relatively
painless incremental exposure.


I recently gave up taking sugar in coffee, not quite one grain at a
time, but pretty close. In a lengthy but painless process, avoiding the
cold turkey of total abstinence, I succeeded in turning a heap into a
non-heap and a non-heap into nothing.


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
provokes, and from the anxiety caused by the loss of control that
results in this end point. (For more on fear, see ad baculum argu-
ments in Chapter 5.) 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; Scotland remaining part of the UK)
is far more likely to be successful than one requiring quite radical
changes that are perceived as undesirable. Despite the evidence for
anthropogenic climate change and its negative consequences being
overwhelming for two decades, the burden of proof in terms of
the need for social change has remained with the climate scientists
and their supporters. Under these circumstances the complexity of
a slippery slope assists our deep-seated desire to ignore the problem,
or rationalise it away.

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 they are 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.



Because of their complexity, slippery slope arguments are often pre-
sented in a shortened (or ‘compressed’) form. We might hear, for


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 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 terrorist threats. Before 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 construc-
tive dialogue and argument reconstruction are required to fill in the
various steps. Only once this effort is made, can the argument be
properly evaluated.

We have seen that although many slippery slope arguments do not
stand up to scrutiny, some do, and because they deal in severe nega-
tive consequences, then we need to be careful about what we dismiss.
Notwithstanding the preciousness of our time and our developing
ability to read signs of unreliability 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.


1. The passage below is from Martin Luther King’s Letter from
Birmingham Jail (see Chapter 1 for context and background).
Many of the argument forms discussed in this chapter can be
found in it. Attempt to identify and reconstruct some of these
and, using the appropriate critical questions, assess its strength as
best you can. Taking into consideration King’s audiences (white
Southern clergy, politicians and the American public in general),
you might also want to evaluate its rhetorical power.


If this philosophy [of nonviolent direct action] had not emerged, I am
convinced that by now many streets of the South would be flowing
with floods of blood. And I am further convinced that if our white
brothers dismiss as ‘rabble-rousers’ and ‘outside agitators’ those of
us who are working through the channels of nonviolent direct action
and refuse to support our nonviolent efforts, millions of Negroes,
out of frustration and despair, will seek solace and security in black
nationalist ideologies, a development that will lead inevitably to a
frightening racial nightmare.

Oppressed people cannot remain oppressed forever. The urge
for freedom will eventually come. This is what has happened to the
American Negro. Something within has reminded him of his birth-
right of freedom; something without has reminded him that he can
gain it. Consciously and unconsciously, he has been swept in by what
the Germans call the Zeitgeist, and with his black brothers of Africa
and his brown and yellow brothers of Asia, South America, and the
Caribbean, he is moving with a sense of cosmic urgency toward the
promised land of racial justice. Recognizing this vital urge that has
engulfed the Negro community, one should readily understand pub-
lic demonstrations. The Negro has many pent-up resentments and
latent frustrations. He has to get them out. So let him march some-
time; let him have his prayer pilgrimages to the city hall; understand
why he must have sitins and freedom rides. If his repressed emotions
do not come out in these nonviolent ways, they will come out in omi-
nous expressions of violence. This is not a threat; it is a fact of history.
So I have not said to my people, ‘Get rid of your discontent.’ But I have
tried to say that this normal and healthy discontent can be channelled
through the creative outlet of nonviolent direct action.

2. Over the next few days pay particular attention to the use of
causal fallacies, hasty and sweeping generalisations, and slippery
slope arguments in your own thoughts and conversations, and in
other sources (notably the media). Make a note of some of them,
including such things as the contexts in which they arise, their
degree of emotionality and automaticity, and the effect they had
on you and other people.




• Deanna Kuhn’s book The Skills of Argument (1991) is a very inter-
esting investigation into how people formulate causal arguments
and associated generalisations. In it, she theorises four stages of
sophistication 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.


• A very readable and insightful discussion of generalisations
(including 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 relation-
ship to scientific method is Chapter 10 of Robert Ennis’ book
Critical Thinking (1996a).



• For a good recent overview of slippery slope arguments, see
Anneli Jefferson (2014) ‘Slippery slope arguments’, Philosophy
Compass, 9/10, 672–80. And for a more advanced history and
analysis, see Douglas Walton (2015) ‘The basic slippery slope
argument’, Informal Logic, 35(3), 273–311.


1 Fortean Times, Sept. 2013.
2 L. Bowes et al. (2015) ‘Risk of depression and self-harm in teenagers identify-

ing with goth subculture: a longitudinal cohort study’, The Lancet. Available
at:, Vol. 2.

3 See, for example, Kollmuss and Agyeman (2002) ‘Mind the gap: why do
people act environmentally and what are the barriers to pro-environmental


behaviour?’ Environmental Education Research, 8(3), 239–60. See, in particular,
p. 254.

4 M. McGrath (2015) ‘Global warming increases food shocks threat’. Available
at:, and R. Bailey
et al. (2015) ‘Extreme weather and resilience of the global food system’. Avail-
able at:

5 B.F. Skinner (1948) ‘“Superstition” in the pigeon’, Journal of Experimental
Psychology, 38, 168–72.

6 See B. Goldacre (2008) Bad Science (London: Fourth Estate, pp. 97–9, 165–70).
7 For a good overview of systematic reviews in medicine, see P. Hemingway

(2009) ‘What is a systematic review?’ Available at:

8 ‘The will to believe’, in Selected Papers on Philosophy (New York: Dutton,
1967), p. 110.

9 M. Mahoney and A. Freeman (eds) Cognition and Psychotherapy (New York:
Plenum Press, 1985).

10 Talk is Cheap, Volume 3 (audio recording, 2004).
11 Fortean Times, February 2015.
12 D. Brunstetter and M. Braun (2011) ‘The implications of drones on the just

war tradition’, Ethics and International Affairs, 25(3), 337–58.
13 IPCC (2014) Climate Change 2014 Impacts, Adaptation and Vulnerability: Sum-

mary for Policymakers. Available at:

14 B. Glass (2014) ‘Wise up England, you’d be better off without Scotland’.
Available at:



Argument by analogy is often the most powerful and compelling type of
argument we can use.

(Michael Scriven, 1976, p. 210)

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 (situation,
event, etc.). Analogies are extremely common in communication.

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

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 (GWTW) did not need to. Regardless of how strong


this comparison is (which is the kind of question that will be dealt with
in the rest of this section), the argument stands a chance of being per-
suasive to the degree that GWTW will be known to, and liked by, many
readers of the paper. 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).

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 grabs the audience’s attention
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 argu-
ments, 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.


An argument from analogy is one which claims that, because a certain
state of affairs is true about idea or situation Y, so it is likely to be true of
idea or situation 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.
P3: Z is true of Y.
P4: Z is linked with features a, b, c, etc.
C: Therefore Z is also true of X.


The film The Last King of Scotland (Keven MacDonald, 2007), set in
the 1970s, is about a young Scottish doctor (Nicholas) who, when
volunteering in Uganda, ends up becoming the President (Ide
Amin’s) personal physician. Initially he is reluctant to accept Amin’s
offer because his motivation for coming to Uganda is to help ordi-
nary people, but Amin persuades him by arguing, initially, that by
helping the President he will be helping the people (for example,
by having a role in setting up a national health service). Nicholas
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 something 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 called the target feature (the decision being based
on duty rather than personal preference).1

Just how arguments from analogy should be categorised has
been the subject of a lot of debate among philosophers and other
academics in recent decades. One important and generally accepted
distinction (first identified by Aristotle (1991, Chapter 2.20) is
between these two types:


• Real-life (or historical) analogies, where the comparison is with
empirical truths or historical events 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 of the primary case.

Extended inductive analogies can sometimes form the entire struc-
ture of lengthy books and documentaries. For example, Joel Bakan’s
The Corporation (2004) argues that the problem with large, publicly
traded companies is that by law they have no option but to behave
like psychopaths, whose symptoms and influence are systematically
analysed and applied to corporate behaviour. 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’ Omertà! Sepp Blatter’s FIFA Organised Crime Family (2014)
argues that the corruption in football’s governing body is best under-
stood in terms of the Mafia, and thus the appropriate alarm, moral
disgust and responses we have towards such organisations should
rightfully be directed towards Blatter’s FIFA (as it subsequently has

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
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.2

Another hypothetical analogy, well known in philosophy and
theology, is William Paley’s ‘watchmaker’ argument for intelligent
design (published in 1800). In it he states that on encountering a
watch for the first time (in comparison to a stone), its complexity
and evident function would give us every reason for believing that it
has been designed and constructed by an intelligent being for a par-
ticular purpose. 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. I mean that the contrivances of nature surpass
the contrivances of art, in the complexity, subtlety, and curiosity, of the
mechanism; and still more, if possible, do they go beyond them in number
and variety: yet, in a multitude of cases, are not less evidently mechani-
cal, not less evidently contrivances, not less evidently accommodated to
their end, or suited to their office, than are the most perfect productions
of human ingenuity.

(Paley, [1800] 2009, p. 19)

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)? [Acceptability]
Q2: Are there relevant similarities between X and Y? [Relevance]
Q3: Are there dissimilarities between X and Y that undermine the

similarities? [Sufficiency]
Q4: Can convincing counter-analogies be found (or in the case of

hypothetical analogies, be hypothesised?) [Sufficiency]
Q5: Is the conclusion drawn from the comparison of the appropriate

strength? [Sufficiency]

Prior to these, a question it is wise to ask ourselves is whether the
analogy is actually intended to serve as an argument, rather than to
explain or illuminate the primary case.

In his Dialogues Concerning Natural Religion (published in 1779,
two decades before Paley’s book), David Hume employed these
questions to good effect against design arguments. We can grant that
someone 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 similari-
ties between the watch and the workings of nature (Q2), but Q3
begins to reveal weaknesses in the argument. Despite their complex-
ity and functionality, living organisms and their component parts are
noticeably different from man-made artefacts, and in many respects
(irregularities, shapes, textures, and so on) share more with the con-
trasting 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 analogy for the creation of nature is the creation of living
organisms as a result of procreation. A designer is not an observable
or necessary feature of the continuation of animal and vegetable spe-
cies, 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
vegetation. The cause, therefore, of the world, we may infer to be some-
thing similar analogous to generation or vegetation.

([1779] 1990, Part VII, 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 8)). A
further argument of Hume’s against the design argument, therefore,
is a variation of Q5. Applying this to Paley, we can say that he con-
cludes too much from his analogy.


Analogies 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


features of the hypothetical scenario, to engage more fully with the
topic. In this respect, analogies have an educational function. Less
benignly though, they can be distracting, play to our prejudices, and
lead us to accept conclusions that we should not accept. The clever
persuader will employ analogies and other forms of figurative lan-
guage 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 appropri-
ately memorable and moving events would aid the US audience (in
2006 at least) to engage with his message.

Even where figurative language is not directly used as part of an
argument, its presence can still influence audiences. Aristotle (in The
Art of Rhetoric) and modern research (see Sopory and Dillard, 2002,
for a review) provide us with the following two criteria to evaluate
the persuasive power of analogies and metaphors.



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 immedi-
ately reveal their affinity’ (1991, p. 220). This is a real art; effectively
knowing what is on the tip of your audience’s tongue, and then
finding the right word 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 crystalize the ‘affinity’ between the conditions
of the working classes and slavery. Even if this connection is under-
stood abstractly, the power of this imagery can be the difference
between inaction and revolutionary action.

Metaphors are of course basic to poetry because they have the
potential to be beautiful (or moving in some other respect). Part
of the art of persuasion through metaphor is to know what will be
pleasing to the audience in question. This can be in a poetic sense,


but also something which has the right kind of emotional resonance
which, when associated with the primary case, adds to its appeal. For
many years now the Automobile Association in the UK (the AA) has
used the strapline ‘the fourth emergency service’, something which
is not literally true, but which borrows connotations of security,
reliability and selfless deeds that many associate with responses to
999 calls.



It is important to get the positioning as well as the content of meta-
phors right. The effect of using a single metaphor rather than multiple
ones, and of employing the metaphor early on in the message can help to
provide a clear frame for engaging with the overall argument. This
process is explained by the ‘superior organisation’ theory which sees
metaphors as facilitating our understanding and memorising of the
arguments presented by providing a ‘structure’ to link 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 than if this stylistic feature were absent
(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 recent and well-received political speeches:
President Obama’s moving and sympathetic address in Newtown,
Connecticut, shortly after the Sandy Hook school shooting in
December 2012; and ex-UK Prime Minister Gordon Brown’s fiery
pro-union speech in Glasgow just prior to the Scottish independ-
ence referendum in September 2014.3

The only metaphor Obama used in his speech 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 equivalent


of having your heart outside of your body all the time, walking
around’. This is a vivid metaphor for vulnerability; something we
care about more than anything is continually exposed to circum-
stances 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, among other things, to gun laws).

In Brown’s speech, his only metaphor referred to the risks of an
independent Scotland in terms of an ‘economic minefield’, and an
‘economic trapdoor down which we go 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.

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 con-
dolences rather than arguments, and in the Brown case other arguments
(concerning, for example, national (British) pride) were more appropri-
ate as openers. If a metaphor functions as a joker, it makes sense that
Brown would choose to play his on the economic theme.


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 these 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
context and to use our imaginations’ (2007, p. 198).

Imagination is important both for formulating appropriate analo-
gies for our own arguments and for formulating counter-analogies
when questioning other people’s arguments (see Q3). This imaginative
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 come up
with convincing analogies in the heat of the moment. Producing
counter-analogies is, however, less tricky since we can usually fol-
low 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 counter-analogies in other unsuccessful and
less well-motivated wars such as Vietnam (wasteful, futile, a national
embarrassment to the USA) and the Crusades (self-righteous, milita-
rist and imperialist folly).

As the design argument example demonstrates, arguments from
analogy tend not to be particularly strong. Where hypothetical analo-
gies have creative qualities, however, they are effective for generating
discussion and catalysing stronger arguments and deeper understand-
ing. There is a cognitive reason for this, concerning stimulation of
the imagination, and a motivational reason as well. A number of
literary philosophers (such as Søren Kierkegaard, Jean-Paul Sartre,
Iris Murdoch and Martha Nussbaum) have written about the impor-
tance of stories (including allegories) for de-personalising topics;
their aesthetic distance provides an engaging focus while helping
to avoid some of the emotional 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, is

effective because it starts with a case sufficiently different from our every-
day 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.


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’, Informal
Logic, 21(3), 199–218.

• Christopher Tindale (2007) Fallacies and Argument Appraisal,
Chapter 10.

• Trudy Govier (2013) A Practical Study of Argument, Chapter 11.


1. There are a number of examples of brief and extended analogies
in this chapter (such as the film The Corporation, and Thomson’s
‘famous violinist’) 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. You can of course, find your
own examples as well – this is easy to do once you start tuning in
to them.

2. Apply the critical questions to Dale Jamieson’s intriguing argu-
ment from analogy found in his book Reason in a Dark Time:4

A revealing example of elite scientific ignorance was on display in the
Supreme Court during the oral arguments in Massachusetts versus EPA.
After being gently corrected for confusing the troposphere with the
stratosphere, Justice Scalia replied, ‘Troposphere. Whatever. I told you
before I’m not a scientist.’ As laughter swept the courtroom, Scalia
added, ‘That’s why I don’t want to have to deal with global warming,


to tell you the truth.’ Justice Scalia’s scientific ignorance is sad but not
surprising. What is truly disturbing was his indifference to his igno-
rance, and his stated desire to ignore an important problem because it
has a scientific dimension. Even worse, Scalia was correct in surmising
that his attitudes would be widely and sympathetically shared among
his audience of lawyers, students, journalists, and others. Imagine
a Supreme Court justice expressing comparable ignorance and atti-
tudes toward problems centering on religion, politics, or economics:
‘Supply and Demand. Whatever. I told you before, I’m not an econ-
omist. That’s why I don’t want to have to deal with monopolies, to
tell you the truth.’ Were a justice to say that, I doubt that it would be
viewed as a charming eccentricity.


1 The choice of terminology is closest to Trudy Govier’s (2013), but ‘target
feature’ is Hughes, Lavery and Doran’s phrase (2010, p. 215).

2 J.J. Thomson (1971) ‘A defence of abortion’, Philosophy and Public Affairs, 1(1),

3 Both of these speeches are easily located for viewing online.
4 Oxford: Oxford University Press, 2014, pp. 62–3.



The European Union is something you’re either for or against. Personally
I’m somewhere in between.

(Sacha T. Burnstorm, pers. comm.)

The previous three chapters have been about argument forms (and their
associated fallacies) that warrant extended analysis and exploration.
The focus of this chapter is a selection of fallacies and rhetorical
techniques that could not be located among these analyses, but
that 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


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.


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 4) 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.


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


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 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 – like walking in a mountainous landscape, or contemplating
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

P1: If X, then Y.
P2: Not Y.
C: Therefore 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 scat.
P2: You don’t like scat.
C: Therefore you don’t like all forms of music.

The fallacious version of modus tollens is known as ‘denying the

P1: If you like all forms of music, then you must like scat.
P2: You don’t like all forms of music.
C: Therefore you don’t like scat.

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.


A circular argument (otherwise known as ‘begging the question’) is
when the conclusion of an argument is required in order to establish


one or more of the premises. A circular argument is expressed in this

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.

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 encounter them,
but articulating them can sometimes be difficult. Setting them out
in a formal structure is slightly tricky as well, but Bowell and Kemp
(2010) achieve this by joining two separate arguments together like

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’s 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’s probably


X: How can we rely on the word of the Bible?
Y: Because the Bible is the word of God.
X: But how do we know God exists?
Y: 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 justification
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: ‘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 reasoned 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.


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 health
care 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’ll 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 has not been presented as an


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 (possibly
implicit) conclusion:

P2: Of the options available, these beliefs/courses of action should be

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?) [Acceptability]

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
further alternatives, but to think creatively in terms of how apparently
incompatible options might be synthesised into a previously
unconsidered theory or solution.

False dichotomies play squarely into our tendency to see situations
in black and white, especially if strong emotions are involved. We hear
quite often in the TV talent show The X Factor how it’s ‘all or nothing’
or ‘Now or never’ for those auditioning, whereas in many cases it
does not actually come down to a choice between superstardom and
working at McDonald’s. News media will typically simplify stories in
this way as well, making us all the more vulnerable to thinking that
issues present us with more clear-cut choices than they usually do.
In his handbook, How to Win Campaigns (2010, pp. 181–2), former
Greenpeace Strategic Adviser Chris Rose recommends that campaign
messages are framed in binary (‘either/or’) terms because of the clarity
and emotional engagement this brings. While this makes sense from
the point of view of rhetoric and the practicalities of initiating social
change, as critical thinkers our aim is both engagement and truth.


We should be aware though that 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 Changes Everything Naomi Klein
argues that we really do have a clear choice between the continuation
of unregulated capitalism and catastrophic climate change (a ‘battle
between capitalism and the planet’). She is critical of the ‘fetish of
centrism – of reasonableness, seriousness, splitting the difference’
(2015, p. 22); 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 seems plausible and delays actions
towards radical change. Klein might not be right, but the point stands
that many situations 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.


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 perfectionist 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, that is because, in the vast majority of cases,
it is. For example, after mass shootings in the USA, we usually hear a
variation on this argument:

P1: If the USA 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

C: Therefore there’s no point in the USA changing its gun control laws.


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’s 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
supporter 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.’2 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’s no point in lowering the speed
limit in the UK to 50 mph 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.

On a less life-and-death note, a now-retired pundit on BBC’s
football highlights programme Match of the Day would regularly
argue that ‘Even if we introduce video technology for penalties and
goal line decisions, referees will still make mistakes with off-sides, late
tackles and so on, so what’s the point?’ The point is that fewer errors
are better than more errors.

Life is such a thing that perfect solutions are rarely feasible, and
this 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 insignificant
difference can be dismissed at once. If I can rescue a single person from
death or misery, the fact that there are many others I cannot rescue is
irrelevant to the moral worth of doing this. Huge problems sometimes
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.

(1986, p. 126)



If you hear the term ‘red herring’ in relation to someone’s argument
(or your own) it means: (1) it is not relevant to the main point being
debated, but (2) has the appearance of being relevant such that it is
liable to shift the focus of the debate. In a way that is analogous to
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 2010, the cap on university fees was raised from £3,000 to
£9,000 per year for English and Northern Irish students studying
in the UK. The government minister responsible for universities at
the time was asked a question in a radio interview along the lines
of ‘What do you say to students who start university next year and
who have to pay three times the fees of students this year?’ His
response was ‘But you must remember that the monthly amount
they will be paying back will in fact be less than students have to
pay now.’ Somewhere along the line this point might have some
relevance to the general debate on higher education fees, but it is
some distance from the matter at hand. The interviewer’s question
is perhaps unfair, but nevertheless the minister’s response seems to
be a deliberate attempt to move the discussion to a place where he
is more comfortable.

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, where 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.

After a bad run of games in 2013, the then Chelsea manager José
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.’3 By employing
the contrast effect and the impression of an unexpectedly worldly
answer from a football manager, Mourinho, true to form, managed
to deflect the question, but it is a clear case of equivocation. 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 is not the case; it simply has a different meaning to
‘crisis’ as referring to fatally serious, urgent, on-going and large-scale
humanitarian situations.


1. The ‘perfectionist fallacy’ assumes that a perfectionist argument
will always be weak, but can you think of situations in which
reasoning in this way constitutes a strong argument? If so, in what
ways are these situations different from fallacious examples?

2. Choose a topical issue and construct two versions of an argument
in relation to it; one that employs a false dilemma, and one that
employs a true dilemma. A good model for this is Gore’s An
Inconvenient Truth documentary. Towards the end he dispels
some false choices (such as environment vs. the economy), but of
course the entire debate around climate change hinges on some
very real choices.


Most textbooks on informal logic and critical thinking will discuss
these fallacies. Some good examples are Bowell and Kemp (2010),
Johnson and Blair (2006), and Hughes, Lavery and Doran (2010).


1 Today, BBC Radio 4, 28 September 2015.
2 Classic Rock, December 2015, p. 22.
3 B. Jefferson (2013) ‘José Mourinho: Chelsea losing to Basel and Everton isn’t a

crisis, Syria is a crisis.’ Available at:


I want people who can think, who can paint pictures and communicate …,
and be prepared to have discussion and debate and dialogue and

(Construction Sector, Departmental Manager)1

What would Davis do?
(Sacha T. Burnstorm, pers. comm.)

As stated in the Introduction, 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

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;
iii. dispositions that make us prone to fallacious reasoning and

unconstructive dialogues.


The aims of assessing the arguments of others and improving our own
arguments are closely related. This is partly because a significant part
of putting forward a position is to demonstrate its strength in com-
parison to rival positions, but it is also because we should assess our
own arguments by the same standards that 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 transcend or manage them take us so far,
but the examples and chapter exercises provided are also designed
to encourage individual 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 particular contexts, we can
understand all the better their power and prevalence. On the other
hand, this reflection is valuable for the individual in and of itself.
Self-understanding and the role it plays in improving our lives are
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
nevertheless are personal because, as stated, these skills and this
knowledge apply equally as much 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 (such as
the one asking you to reconstruct and evaluate arguments from your
own essays or other assignments in Chapter 4).

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, will I set out, in the form of a list, ways in which you
can apply this learning to the practices of academic and professional

1. When constructing arguments, the guidelines for argument recon-
struction and evaluation can be adapted so that we ask ourselves
questions like: What is my conclusion? What are my reasons for
supporting it? Are these acceptable, relevant and sufficient?

2. Be careful to make your language as clear and precise as possible;
avoid ambiguity, vagueness and equivocation.

3. Be watchful of the tendency to allow absolutist premises and
conclusions to appear in our 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 6, but also from the discus-
sion of false dilemmas and the perfectionist fallacy in Chapter 8.
System 1 likes to jump to conclusions of a somewhat black-and-
white nature, but where this is undesirable we might try asking
ourselves ‘What would Davis do?’ (Davis being the name of
Juror 8 in Twelve Angry Men, something we only learn right at
the end of the film.)

4. If you are going to use authorities to support your conclusions,
apply the critical questions to them and make sure you are using
them for the right reasons. In the end you must take responsibil-
ity 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 principle
of charity) are not peripheral but central to critical thinking. 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 hearing (what
Dewey calls ‘wholeheartedness’). Since the confirmation 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 par-
ticular 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 over-confidence 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).
But we are 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 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 smart-
arsery. Where permitted by the subject matter, choose authori-
ties and analogies that will be recognisable and appealing to your

10. If we listen carefully to the views of others, we might find that
we agree about 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 argumentative, be a
critical thinker.


1 Quoted in G.W. Hinchliffe and A. Jolly (2011) ‘Graduate identity and
employability’, British Educational Research Journal, 37, 563–84.


Abusive ad hominem argument: See ad hominem argument.
Ad baculum argument (appeal to force): An argument that involves

a threat, such as: ‘Accept X, or I will ensure that negative conse-
quence Y will occur.’

Ad hominem argument: An argument that attacks the arguer
rather than their argument. A negative feature 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 asso-
ciation with people or institutions of questionable character, bias,
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
reflective consideration of relevant factors.

Affirming the consequent: A fallacious version of a valid deduc-
tive argument known as modus ponens, which runs: ‘P1: If X, then
Y; P2: X; C: Therefore Y.’ The invalid version makes the mistake


of affirming the consequent (Y) rather than the antecedent (X)
of the first premise, leaving us with ‘P1: If X, then Y; P2: Y; C:
Therefore X.’

Ambiguity and vagueness: An ambiguous sentence is one that has
two or more possible meanings. In argument reconstruction, one
meaning must be settled before a sentence can constitute a premise
or conclusion. A vague sentence is one that is imprecise. In argu-
ment reconstruction if this vagueness is problematic when inter-
preting the meaning of a sentence, then it should be made more

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 (situation,
event). Where a comparison is being made with a real-life event
or other state of affairs, it is known as an inductive analogy, and
where the comparison is with an invented situation, it is known as
a hypothetical analogy.

Anchoring: This is the biasing effect of being exposed to prior
relevant information when making a judgement. For example, if
I am asked to guess the age of an actor I recognise but know little
biographical detail about, such as Daniel Craig, my answer will be
affected by the framing of a previous age-related question, such as
‘Is he older or younger than 30?’ or ‘Is he older or younger than
40?’ I am liable to guess lower after being presented with the first
question, than if presented with the second one, because my esti-
mate is anchored by this prior number.

Appeal to emotion: A form of argument in which a particular
emotion (such as fear, pity or guilt) is instrumental in establishing
the conclusion. In strong forms of these arguments the emotion is
usually an appropriate response to the argument’s subject matter.
In weak forms, the emotion has a distorting effect on the audi-
ence’s understanding and assessment of the argument.

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 argu-
ments, argument from consequences, and so on) that share a


general structure and are often associated with particular 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 of (or likely
to be true of) X.

Argument from authority: An argument in which we are encour-
aged to accept the conclusion on the basis that a relevant author-
ity (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 appropriately ordered
premises and conclusions.

Argumentation: The exchange of arguments in dialogues.
ARS criteria: The assessment of arguments in terms of the accuracy

of their premises; the relevance of their premises to the conclusion,
and whether the premises are sufficient for establishing the conclu-
sion. See also critical questions.

Assumptions: An assumption, in the context of an argument, is a
belief that has relevance to the argument, but which has not been

Availability heuristic: The quick rule by which we judge the like-
lihood of an occurrence on the basis of how readily it comes to

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. So, if judging whether a shy person is
more likely to be a librarian or a salesperson, the relative number of
people employed in these jobs needs to be taken into consideration.
See also representative heuristic.

Burden of proof: If a belief is taken to be generally established, the norm
among a group of people, or an agreed default position, then this tends
not to be a position that requires defending if challenged. Instead, the


burden of proof falls to the challenger, and should they fail to convince
those holding the established view, then the matter is closed.

Causal arguments: Arguments that claim that correlated events (X
and 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, while 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 is established.

Circumstantial ad hominem argument: See ad hominem

Coercive power: The social influence someone has resulting from
their ability to punish the actions of others (through violence,
insults, fines, detentions, and so on). Note that one does not need
to have legitimate power in order to have coercive power.

Cognitive dissonance: The discomfort experienced as a result of
an (often unconscious) 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.
It is especially powerful when commitments are publicly declared.

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 domi-
nating presence of critical thinking. Examples include participants
being willing to defend any claims that they make, and the dialogue
not being considered complete until all parties have been able to
express their views, and these views are given due consideration.

Contrast effect: The effect that nearby comparators with contrast-
ing qualities have on our perception and interpretation of infor-
mation. For example, if you are trying to impress someone with
your singer-song writer abilities, do not do it after that someone
has just been listening to Bob Dylan.

Critical questions: The questions which are important to ask in
order to establish the strength of an argument.

Critical thinking disposition: See disposition.
Deductive argument: An argument in which the conclusion fol-

lows logically (or necessarily) from the premises, so that if the
premises are true, the conclusion must also be true.

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 typically do not employ qualifiers
such as ‘all’ or ‘most’, but 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 simul-

taneously 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.

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.

Ego defences: Ways in which we distort reality in order to keep pain-
ful 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 well aware of what has been omitted.

Equivocation: This is a fallacy linked to ambiguity, where 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. See also red herring.

Ethotic power (ethotic authority): The social influence of some-
one who is regarded as a broadly ‘good’ or ‘virtuous’ person (like
Barack Obama or Oprah Winfrey, for some).

Expert power (expert authority): The social influence a person has
by virtue of their expertise (or perceived expertise) in a certain area.

Explanation: In contrast to arguments, with explanations it is
agreed that something is true, and a person is then offering rea-
sons to explain why it is true (for example, what has caused it).
With an argument, it is not agreed that something is true, and the
arguer presents reasons in order to convince the other person of
what they believe to be true.

Fallacy: Fallacies are weak arguments of certain types that are used
frequently and are liable to be convincing to those not thinking

False dilemma/false dichotomy/false choice: A set of falla-
cies in which the options available in a particular situation have
been misleadingly represented. ‘False dilemma’ indicates that the
options offered and the implied pay-off between them are not fully
accurate; ‘false dichotomy’ means that only two possibilities have
been presented but that there are in fact more than this, and ‘false
choice’ is a more general term that covers two or more options
and does not imply a pay-off between them (as a ‘dilemma’ does).

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. The purpose can be
to enable an audience’s understanding of the thing in question,
or to put it in a more favourable light, and to encourage certain
forms of action in relation to it. For example, in response to calls
for alcohol to be banned on planes after a series of ‘air rage’ inci-
dents, comedian Doug Stanhope (on the Deadbeat Hero CD) sug-
gests we consider all the air rage that has been prevented as a result
of people being mellowed out by alcohol.

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 a thing have


a certain quality (e.g. ‘all planets have a gravitational pull’), and
non-absolute (also known as ‘non-universal’, ‘inductive’ or ‘soft’)
generalisations are those that claim that only ‘most’ or ‘the majority’
of instances have this quality (e.g. ‘most planets in our solar system
are bigger than Mars’).

Groupthink: A set of processes of social influence (such as collective
over-confidence and pluralistic ignorance) that causes a group
of otherwise rational and intelligent people to make very poor

Guilt by association: See ad hominem argument.
Halo effect: Our pronounced tendency to generalise known positive

features of a person to other, unknown, aspects of them. If there is
someone we find likeable at work, we tend to wrongly assume we
would like them in other contexts as well; or if someone is kind to
non-human animals, we might jump to the conclusion that they are
also compassionate towards human beings. 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 tool for quick decision-making; a rule of thumb that is
applied to certain types of situation that, although lacking preci-
sion, makes better than chance judgements. See also System 1
and System 2 thinking.

Horn effect: See halo effect.
Hypothetical analogy: See analogy.
Inductive analogy: See analogy.
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 critical thinking.

Information power: The social influence a person has by virtue
of the information they have 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 experience
of an unusual event.

Legitimate power: The social influence a person has resulting from
the position they hold, or moral and legal principles their deci-
sions and actions uphold.

Meta-cognition: This can sometimes simply mean self-awareness,
but a more specific usage refers to our awareness of the mental
processes (e.g. thoughts and emotions) we are currently experienc-
ing with a view to regulating these or the judgements, decisions or
actions that might arise from them.

Mistaking cause for correlation: Otherwise known as ‘Sod’s law
fallacy’, in which a person overlooks a causal relationship between
two events and instead declares them to be a coincidence. 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’.

Narrative fallacy: This can be understood as a psychological basis
of the over-simplification of causes, and refers to our attrac-
tion to, and confidence in, coherent but simple explanations at
the expense of an appreciation of complex and unknown factors.

Necessary and sufficient conditions: A necessary condition is
one that must be fulfilled in order for something to be the case
(belong to a particular category, cause something to happen, and
so on); 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 does not have to be.

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.

Over-simplification of causes: The fallacy of assuming a more
straightforward causal explanation for a phenomenon than is actu-
ally the case. In many cases multiple causal factors need to be
taken into account. 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 lowering speed limits).

Placebo effect: Beneficial medical outcomes resulting from a
patient’s belief that the treatment they are receiving will improve


their health, rather than from any direct physical effects of the
treatment itself.

Plausible argument: A style of argument employed in deliberations
and other dialogues in which generally accepted assumptions
act as the basis for reaching conclusions. They are recognised
as provisional in nature, and can function as opening moves in
what becomes a more systematic process of reasoning about
the matter in hand. See also burden of proof, and defeasible

Pluralistic ignorance: A socially based form of ignorance in which
indecision or inaction results from the mutual misinterpretation of
(often) non-verbal communication. One person’s non-urgent or
calm demeanour is seen by another as a sign of insight that a situa-
tion is under control, but in fact this demeanour is either the result
of believing that similar body language in others has this same mean-
ing, or that they do not want to appear to be anxious or out of their

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 and conclusions. ‘The reasons
for this are’, and ‘this is so because’ are examples of premise indi-
cators; and ‘thus’, ‘therefore’ and ‘in which case’ are examples of
conclusion indicators.

Principle of charity: When reconstructing arguments, the prin-
ciple 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 (orders). All
premises and conclusions in an argument must be propositions,
and sometimes argument reconstructions require the rewording
of information originally presented in non-propositional form
(such as rhetorical questions).

Psychology of persuasion: The study of the psychological prin-
ciples underlying persuasive communication.


Rationalisation: To rationalise is to provide a reason why some-
thing happened, but one that is not the real reason – a fact that
is only unconsciously known to us. Rationalisation is a form of
ego defence in that it functions to protect us, at least temporarily,
from an aspect of reality (such as unsavoury motivations) that we
do not want to acknowledge.

Reciprocation: A heuristic linked to the norm of equitable
exchange of favours, goods, and so on. If someone gives some-
thing to us, we feel strongly obliged to give them something
in return.

Red herring: A fallacious argument with the potential to divert
other arguers away from the issue being discussed because of its
apparent relevance and often emotive content. It can be employed
deliberately or accidentally. See also equivocation.

Referent power: The form of social influence held by a person or a
group that we identify with. The sense of, and desire for, belong-
ing make us likely to conform to the attitudes and behaviours of
that person or group.

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 score.

Representative heuristic: A simple rule whereby quick answers to
questions like ‘What’s the probability of individual X belonging
to group Y’, or ‘What is the probability of event P being the cause
of event Q?’ are provided by making reference to what a typical
member of group Y, or a typical cause of Q is like.

Reward power: The social influence someone has resulting from
their ability to reward the actions of others (through gifts, treats,
praise, bonuses, and so on). Note that one does not need to have
legitimate power in order to have reward power.

Rhetoric: The art of persuasive communication.
Self-serving bias: An automatic and systematic tendency to over-

estimate those features of ourselves and the world that are core to
our sense of self-esteem and our motivation to succeed.

Shared causes: See overlooking shared cause.
Slippery slope arguments: Arguments that share the general

structure of arguments from negative consequences, but 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.

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. This misrepresentation can be deliberate or accidental.

Sub-conclusions: Conclusions reached on the way to an overall
conclusion in a larger argument. The conclusions of component
arguments function as further premises in the larger argument.

Sweeping generalisation: See hasty generalisation.
System 1 and System 2 thinking: System 1 is the ‘fast thinking’

that is employed in situations we are familiar with (or think we
are familiar with) and which is reliant on heuristics. System 2 is
the sort of ‘slow thinking’ that we associate with focused attention,
problem solving and critical thinking.

System 1 candy: A term to characterise certain features of arguments
or other persuasive messages that encourage System 1 (heuristics-
based) decision-making in situations where System 2 ought to be

Tu quoque: See ad hominem argument.
Vagueness: See ambiguity and vagueness.
Witness testimony: See information power.


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, online: 1–10.

Aikin, S. and Clanton, J. (2010) Developing Group-Deliberative Virtues. Journal
of Applied Philosophy, 27(4), 409–24.

Aristotle (1991) The Art of Rhetoric. London: Penguin.
Barnett, R. (1997) Higher Education: A Critical Business. Buckingham: 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.

Battaly, H. (2010) Attacking Character: Ad Hominem Argument and Virtue
Epistemology. Informal Logic, 30(4): 361–90.

Berrill, D. (ed.) (1996) Perspectives on Written Argument. Cresskill, NJ: Hampton

Blass, T. (2000) Obedience to Authority: Current Perspectives on the Milgram
Paradigm. Englewood Cliffs, NJ: Erlbaum.

Bohm, D. (2004) On Dialogue. London: Routledge.
Bowell, T. and Kemp, G. (2010) Critical Thinking: A Concise Guide, 3rd edn.

London: Routledge.
Brady, M. (2013) Emotional Insight. Oxford: Oxford University Press.

Select BiBliography244

Brinton, A. (1986) Ethotic Argument. History of Philosophy Quarterly, 3(3),

Brown, R. (1986) Social Psychology. New York: Free Press.
Burns, J.M. (1978) Leadership. New York: Harper & Row.
Cialdini, R. (2007) Influence: The Psychology of Persuasion. New York: Harper Collins.
Cohen, D.H. (1995) Argument is War … and War is Hell: Philosophy, Education,

and Metaphors for Argumentation. Informal Logic, 17(2), 177–88.
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. London: Rider.
Damasio, A. (2000) The Feeling of What Happens: Body, Emotion and the Making

of Consciousness. London: Vintage.
Davis, H. (2009) Caveman Logic. Amherst, NY: Prometheus Books.
Descartes, R. ([1641] 1968) Discourse on Method and the Meditations. London:

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. New York: D.C. Heath & Co.
Eemeren, F. van and Grootendorst, R. (2004) A Systematic Theory of

Argumentation: The Pragma-Dialectical Approach. Cambridge: Cambridge
University Press.

Ennis, R. (1996a) Critical Thinking. Upper Saddle River, NJ: Prentice Hall.
Ennis, R. (1996b) Critical Thinking Dispositions: Their Nature and Accessibility.

Informal Logic, 18(2–3), 165–82.
Facione, P. (2006) Critical Thinking: What It is and Why It Counts. Hermosa

Beach, CA: Measured Reasons LLC.
Facione, P., Sanchez, C., Falione, N. and Gainen, J. (1995) The Disposition

towards Critical Thinking. The Journal of General Education, 44(1), 1–25.
Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford, CA: Stanford

University Press.
Fisher, R. and Ury, W. (1991) Getting to Yes: Negotiating an Agreement Without

Giving In. London: Random House.
Fiske, S. (2005) What’s in a Category? in A. Miller (ed.) The Social Psychology of

Good and Evil. New York: Guilford Press.
Flew, A. (1985) Thinking About Thinking. London: Flamingo.
Freeden, M. (2005) What Should the ‘Political’ in Political Theory Explore?

Journal of Political Philosophy, 13(2), 113–34.
French, J. and Raven, B. (1959) The Bases of Social Power, in D. Cartwright

(ed.) Studies in Social Power. Ann Arbor, MI: Institute for Social Research.

Select BiBliography 245

Furnham, A. and Boo, H.C. (2011) A Literature Review of the Anchoring
Effect. The Journal of Socio-Economics, 40(1), 35–42.

Gigerenzer, G., Todd, P. and the ABC Research Group (1999) Simple Heuristics
that Make Us Smart. Oxford: Oxford University Press.

Gilbert, M. (1994) Feminism, Argumentation and Coalescence. Informal Logic,
XVI(2), 95–113.

Gilbert, M. (1997) Coalescent Argumentation. Englewood Cliffs, NJ: Erlbaum.
Glover, J. (1986) It Makes No Difference Whether or Not I Do It, in P. Singer

(ed.) Applied Ethics. Oxford: Oxford University Press.
Goldacre, B. (2008) Bad Science. London: Fourth Estate.
Goodwin, J. (2011) Accounting for the Appeal to the Authority of Experts.

Argumentation, 25, 285–96.
Govier, T. (1988) Problems in Argument: Analysis and Evaluation. Dordrecht:

Foris Publications.
Govier, T. (2013) A Practical Guide to Argument, 7th edn. Belmont, CA: Wadsworth.
Hamblin, C. (1970) Fallacies. London: Methuen.
Hanscomb, S. (2015) Assessments and the Self: Academic Practice and Character

Attributes. Journal of Learning Development in Higher Education, 8.
Hargie, O. and Dickson, D. (2004) Skilled Interpersonal Communication, 4th edn.

London: Routledge.
Hughes, W., Lavery, J. and Doran, K. (2010) Critical Thinking, 6th edn.

Peterborough, ON: Broadview Press.
Hume, D. ([1779] 1990) Dialogues Concerning Natural Religion. London: Penguin.
Hume, D. ([1748] 2008) An Enquiry Concerning Human Understanding. Oxford:

Oxford University Press.
Janis, I. ([1972] 1982) Groupthink, 2nd edn. Boston: Houghton Mifflin Co.
Jefferson, A. (2014) Slippery Slope Arguments. Philosophy Compass, 9/10,

Johnson, R. and Blair, A. (2006) Logical Self Defense. New York: IDEA.
Kahneman, D. (2012) Thinking Fast and Slow. London: Penguin.
Kant, I. ([1781] 1929) Critique of Pure Reason, trans. N. Kemp Smith. Basingstoke:

Kant, I. ([1784] 1963) What is Enlightenment? in On History, ed. and trans.

L. White Beck. New York: Macmillan.
Keltner, D., Oatley, K. and Jenkins, J. (2013) Understanding Emotions, 3rd edn.

New York: Wiley Global Education.
Kierkegaard, S. ([1846] 1979) Two Ages, ed. and trans. H.V. Hong and

E.H. Hong. Princeton, NJ: Princeton University Press.
King, M.L. (1963) Letter from Birmingham Jail. Available at: http://kingencyclo

Select BiBliography246

Klein, N. (2015) This Changes Everything. London: Penguin.
Kuhn, D. (1991) The Skills of Argument. Cambridge: Cambridge University

Larrick, R. (2004) Debiasing. In D.J. Koehler and N. Harvey (eds) Blackwell

Handbook of Judgment and Decision Making. Oxford: Blackwell.
Lerner, M. and Simmons, C. (1966) Observer’s Reaction to the ‘Innocent

Victim’: Compassion or Rejection? Journal of Personality and Social Psychology,
4(2), 203–10.

Lunsford, A. and Ruszkiewicz, J. (2010) Everything’s an Argument, 5th edn.
London: St Martin’s Press.

Makan, J.M. and Marty, D.L. (2001) Cooperative Argumentation: A Model for
Deliberative Community. Long Grove, IL: 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.

Melling, D. (1987) Understanding Plato. Oxford: Oxford University Press.
Milgram, S. ([1974] 2005) Obedience to Authority. London: Pinter & Martin.
Mill, J.S. ([1859] 1962) Utilitarianism, On Liberty: Essay on Bentham. Glasgow: Collins.
Morrow, D. and Weston, A. (2011) A Workbook for Arguments. Indianapolis:

Narveson, J. (1993) Morality and Violence, in T. Regan (ed.) Matters of Life and

Death, 3rd edn. New York: McGraw-Hill.
Newman, J.H. ([1852] 1982) The Idea of a University. Notre Dame, IN:

University of Notre Dame Press.
Nussbaum, M. (1990) Love’s Knowledge. Cambridge: Cambridge University

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 edn. Oxford: Blackwell.
Osman, M. (2004) An Evaluation of Dual-Process Theories of Reasoning.

Psychonomic Bulletin and Review, 11(6), 988–1010.
Paley, W. ([1800] 2009) Natural Theology. Cambridge: Cambridge University Press.
Paul, R. (1984a) 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.

Paul, R. (1984b) Critical Thinking: Fundamental to Education for a Free
Society, Educational Leadership, Sept., 4–14.

Paul, R. (1995) Critical Thinking. Santa Rosa, CA: Foundation for Critical Thinking.
Perelman, C. and Olbrechts-Tyteca, L. (1969) The New Rhetoric: A Treatise on

Argumentation. Notre Dame, IN: University of Notre Dame Press.

Select BiBliography 247

Perkins, D., Jay, E. and Tishman, S. (1993) Beyond Abilities: A Dispositional
Theory of Thinking. Merrill-Palmer Quarterly, 39(1), 1–21.

Peters, R. (1973) The Aims of Education, in R. Peters (ed.) The Philosophy of
Education. Harmondsworth: Penguin.

Peterson, C. and Seligman, M. (2004) Character Strengths and Virtues. Oxford:
Oxford University Press.

Petty, R. and Cacioppo, J. (1996) Attitudes and Persuasion: Classic and
Contemporary Approaches. Boulder, CO: Westview Press.

Plato (1993) The Last Days of Socrates, trans. H. Tarrant and H. Tredennick.
London: Penguin.

Plato (2004) Gorgias, trans. C. Emlyn-Jones and W. Hamilton. London: Penguin.
Raven, B. (1965) Social Influence and Power, in I.D. Steiner and M. Fishbein

(eds) Current Studies in Social Psychology. New York: Holt, Rinehart, &

Rose, C. (2010) How to Win Campaigns, 2nd edn. London: Earthscan.
Rosenthal, R. and Jacobson, L. ([1968] 2003) Pygmalion in the Classroom.

Carmarthen: 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. New York: McGraw-Hill.
Scruton, R. (1999) On Hunting. London: Yellow Jersey.
Seech, Z. (1993) Open Minds and Everyday Reasoning. Belmont, CA: Wadsworth.
Siegel, H. (1988) Educating Reason. New York: Routledge.
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 edn.

Philadelphia, PA: Psychology Press.
Solomon, R. (2003) Emotions and Choice, in R. Solomon (ed.) What is an

Emotion?, 2nd edn. Oxford: 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.
Sutherland, S. (2007) Irrationality. London: Pinter & Martin.
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. Chicago: Nelson-Hall, pp. 7–24.

Taleb, N. (2007) The Black Swan. London: Penguin.
Taylor, S.E. and Brown, J.D. (1988) Illusion and Well-Being: A Social

Psychological Perspective on Mental Health. Psychological Bulletin, 103(2),

Tiberius, V. (2008) The Reflective Life. Oxford: Oxford University Press.

Select BiBliography248

Tindale, C. (2007) Fallacies and Argument Appraisal. Cambridge: Cambridge
University Press.

Tomić , T. (2013) False Dilemma: A Systematic Exposition. Argumentation,
27(4), 347–68.

Toulmin, S. (1958) The Uses of Argument. Cambridge: Cambridge University

Tremain, R. (2000) Music and Silence. London: Vintage.
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),

Waller, B.N. (2001) Classifying and Analysing Analogies. Informal Logic, 21(3),

Walton, D. (1989) Informal Logic. Cambridge: Cambridge University Press.
Walton, D. (1992) The Place of Emotion in Argument. Philadelphia, PA: Penn

State University Press.
Walton, D. (1995) A Pragmatic Theory of Fallacy. Tuscaloosa, AL: University of

Alabama Press.
Walton, D. (1998) The New Dialectic: Conversational Contexts of Argument.

Toronto: University of Toronto Press.
Walton, D. (2006) Fundamentals of Critical Argumentation. Cambridge: Cambridge

University Press.
Walton, D. (2009) Ad Hominem Arguments. Tuscaloosa, AL: University of

Alabama Press.
Walton, D. (2015) The Basic Slippery Slope Argument. Informal Logic, 35(3),

Walton, D., Reed, C. and Macagno, F. (2008) Argumentation Schemes.

Cambridge: Cambridge University Press.
Weber, M. (1978) Economy and Society. Berkeley, CA: University of California

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.

Philadelphia, PA: Penn State University Press.
Woods, J. (2013) Errors of Reasoning: Naturalizing the Logic of Inference. London:

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. New York: Basic Books.


ad baculum arguments 120, 140–4, 162
ad hominem arguments 117, 120, 231
ad hominem arguments,

circumstantial 132–4
ad hominem arguments, direct (or

abusive) 125, 151–2
ad populum arguments 120, 157–60
affect heuristic 47, 48, 52, 64
affirming the consequent 218–220
ambiguity 88–9, 230
analogy, arguments from 110, 171,

199, 206–17
analogy, hypothetical 209–10, 215–16
analogy, inductive 209
anchoring 40–1
argument reconstruction 81–97, 230
argumentation 14–15, 17, 23, 54–5,

70, 74, 115, 128, 142, 185, 226, 231
argument-friendly rewording 85–90
arguments, definition of 4
arguments, structure of 79
Aristotle 12, 41, 54, 91, 208, 212
ARS criteria see fundamental critical

assumption 63, 92, 112, 169, 180, 225

authority 115, 119, 230
authority (as a heuristic) 41–2, 55
authority, arguments from 112, 119
availability heuristic 39–40, 134, 186

background rate neglect see base rate

base rate neglect 37–9
biases see heuristics
Bohm, D. 61, 64
burden of proof 103, 128, 163, 185, 201

causal arguments 168–71
causal fallacies 172–82
chained arguments 93
charisma 152
cherry picking 186–7
circular argument 211, 220–2
coercive power 117–18, 136, 140
cognitive dissonance 33–5, 131
cognitive rationality see rationality
cognitive-behavioural therapy 190
coincidence 172–4
commitment and consistency 33–5,

138, 183


conclusion 4, 79
confirmation bias 30–1, 43, 46, 60,

83, 185, 187, 230
conjunction fallacy 37
consequences, arguments from 192–4
constructive dialogue, guidelines for

a 74–5
continuum fallacy 199–200
contrast effect 35, 40–1, 53, 226
convergent arguments 95–6
‘counter-attitudinal advocacy’ 76
courage 65–6, 143, 144–5, 162, 163
critical questions 104, 106–13, 116,

121, 130–2, 134–5, 139, 141, 143,
146, 149, 158, 161–3, 164, 179,
189, 193, 195, 197, 210, 216,
223, 230

critical thinking, definition 5–7,
24, 228

critical thinking dispositions see
dispositions, ‘critical thinking

Damasio, A. 47
deductive arguments 100–3, 114, 219
defeasible generalisation 184–5, 188
‘deliberative friendliness’ 73
democracy 17–18
denying the antecedent 218, 220
Dewey, J. 5, 67, 71–2, 230
dialogical dispositions 64
dialogue 10–11, 14–15, 39, 54, 64–76,

83, 89–90, 96, 103, 127–8, 134–5,
143–4, 152–4, 162–3, 182–3,
189–92, 201–2, 214–15, 228–9

direction of cause 174
dispositions 6, 8, 16, 21, 24, 50, 57–76,

113–14, 116, 124–5, 127, 134,
143–4, 149, 151–2, 154, 162–3,
171, 182–3, 190, 201, 214, 228;
‘critical thinking movement’ 15

ego defences see rationalisation
emotion, appeals to see emotion,

arguments from

emotion, arguments from 50–2
‘emotional framing’ 46
emotions 6, 12, 21, 41, 43, 44–52,

142, 152, 223, 231
emotions and rationality see

emotions as heuristics 47–8
Ennis, R. 5, 66–7, 72, 74, 76
enthymeme 91, 113, 231
equivocation see red herring
ethics 7, 13, 17, 140, 148, 188
ethotic arguments 147, 149, 152
ethotic power (authority) 119,

expert authority 120–8, 138
expert power 117–18
explanations 84–5

fallacies 14, 54, 66, 104–6, 113,

false choice see false dilemma
false dichotomy see false dilemma
false dilemma, 222–4
familiarity 190, 207, 211, 216
flexibility 59, 61–2, 76
‘flow’ 71
framing 52–4, 145, 207, 223
fundamental critical questions (ARS

criteria) 107–8

generalisations 183–92, 230
Gilbert, M. 70
‘graduate attributes’ 18–20
groupthink 43, 65, 84, 116, 128, 144,

156, 158, 159
guilt by association 150, 156–7

halo effect 127, 150
hasty generalisation 185–7
heuristics 29–30, 31, 41, 42, 44, 45,

47–8, 53, 55, 103, 113, 148, 181
higher education see ‘graduate

horn effect 151
Hume, D. 131, 135, 169, 210–11

Index 251

identifying premises and conclusions

imagination 214–15, 225
implicit assumptions 112
implicit premises and conclusions

Inconvenient Truth, An 23, 134, 155,

212, 227
inductive arguments 101–3, 114
informal logic 14
information power 117–18, 128–30
inquisitiveness 60, 71, 202
‘iron man’ argument 97

Kahneman, D. 28, 37, 39, 56, 151, 170
Kant, I. 7, 10, 66, 127, 170
King, M.L. 53–4, 85–7, 202–3
Kuhn, D. 16–17, 185–7, 204

legitimate power 117–18, 136, 138
Letter from Birmingham Jail see King,

likeability 41–2, 55
listen 61, 69, 75
listening 71, 74, 82, 154
love of truth 59–60

meta-cognition 29, 63–4, 144,
154, 183

metaphors 212–13
Milgram experiments 136–40, 144–6,

152, 162
Mill, J.S. 3, 69, 83
mistaking cause for correlation 178–9
modesty 62–3, 152, 163, 183
modus ponens see affirming the

modus tollens see denying the

multiple causes 175–6

narrative fallacy 175, 180, 181–3
necessary and sufficient conditions

108, 181
negotiation 36, 40, 55, 144, 165

‘no true Scotsman’ fallacy 191–2
Nussbaum, M. 44, 135, 215

open-mindedness 60–1, 83, 163, 187,
202, 230

over-confidence 134, 183, 231
over-simplification of causes 175–6

perfectionist fallacy 224–5
placebo effect 173–4, 176–7
Plato 8–13, 25, 29
plausible arguments 103–4
pluralistic ignorance 43
position to know, argument from see

information power
post hoc ergo propter hoc 172
practical rationality see rationality
pragma-dialectics 14, 74
precedent, argument from 198–9
premise 4, 79
premise and conclusion indicators 84–5
principle of charity 90, 96–7, 230
propositions (statements) 80–1
psychology of persuasion 12

rationalisation 32–3, 131
rationality 27
reciprocation 35–6, 138
red herring 226–7
referent power 117–18, 157–8,

regression to the mean 173–4
representative heuristic 36–7
respect for others 67–8
reward power 117–18, 146–7
rhetoric 9, 12–13, 15, 23, 29, 54–5,

117, 126, 134, 142, 150, 161, 179,
189, 201, 211, 223, 229

rhetorical questions 80–1, 85, 86
Russell, B. 30–1, 71

self-fulfilling prophecy 177
self-knowledge 6, 7, 9, 18, 29, 59, 63,

74, 229
self-reflection 6, 229–31


self-serving bias 31–3, 231
shared causes 174–5
slippery slope arguments 50, 112,

194–202, 231
smart arse 72–3, 231
social power 117–19
social proof 42–3, 161
Socrates 8–12, 29, 163, 229
statements see propositions
staying focussed 66–7
stereotypes 36, 153, 175, 188
straw man argument 82–3
sub-conclusions 93–6
sweeping generalisation 187–9
System 1 and System 2 thinking

28–30, 64, 103, 181, 224
‘System 1 candy’ 54, 175

System 1 thinking 41, 54, 57, 147,
180–1, 188, 230

‘systematic desensitisation’ 200

trustworthiness 123, 183
tu quoque arguments 154–6, 163
Twelve Angry Men 23–4, 36, 60, 65,

68, 75, 131, 132, 135, 143, 148,
159–60, 183, 230

vagueness 89–90, 230

Walton, D. 61, 63, 76, 101, 103, 114,
131, 147, 163, 184, 194, 204

‘wholeheartedness’ 71–2, 230
witness testimony 118, 130–2
Woods, J. 113, 165

  • Cover
  • Title
  • Copyright
  • List of boxes
  • Acknowledgements
  • Introduction: waking up to bad arguments
  • 1 Rationality, cognitive biases and emotions
  • 2 Critical thinking and dispositions
  • 3 Arguments and argument reconstruction
  • 4 Argument forms and fallacies
  • 5 Arguments and social power: authority, threats and other features of message source
  • 6 Causal arguments, generalisations, arguments from consequences and slippery slope arguments
  • 7 Arguments from analogy
  • 8 Further fallacies
  • Conclusion
  • Glossary
  • Select bibliography
  • Index