REASONING, ARGUMENTATION, AND

DELIBERATIVE DEMOCRACY

ln light of the latest research from cognitive and developmental

psychology, this key text explores reasoning, rationality, and
democracy, considering the unique nature of each and their
relationship to each other.
Broadening our understanding from the development of
reasoning and rationality in individuáis to encompass social
considerations of argumentation and democracy, the book
connects psychological literature to philosophy, law, political
science, and educational policy. Based on psychological
research, Moshman sets out a system of deliberative
democracy that promotes collaborative reasoning, rational
institutions such as science and law, education aimed at the
promotion of rationality, and intellectual freedom for all. Also
including the biological bases of logic, metacognition, and
collaborative reasoning, Moshman argües that, despite
systematic flaws in human reasoning, there are reasons for a
cautiously optimistic assessment of the potential for human
rationality and the prospects for democracy.
Reasoning, Argumentation, and Deliberative Democracy will
be essential reading for all researchers of thinking and
reasoning from psychology, philosophy, and education.

David Moshman is Professor Emeritus of Educational

Psychology, University of Nebraska-Lincoln, USA.
REASONING,
ARGUMENTATION, AND
DELIBERATIVE
DEMOCRACY

David Moshman

Routledge
Taylor&Franás Group

NEW VORK A N D LO N D O N
First published 2021
by Routledge
52 Vanderbilt Avenue, New York, NY 10017
and by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, 0X 14 4RN
Routledge is an imprint ofthe Taylor & Francis Group, an informa
business
© 2021 Taylor & Francis
The right of David Moshman 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 ñames may be trademarks or
registered trademarks, and are used only for identification and
explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
A catalog record for this title has been requested
ISBN: 978-0-367-31276-3 (hbk)
ISBN: 978-0-367-31277-0 (pbk)
ISBN: 978-0-429-31602-9 (ebk)
Typeset in Bembo
by Newgen Publishing UK
CONTENTS

Preface

1 Introduction
Overview
Reasoning and Rationality
Cognition and Development
Argumentation and Democracy
Conclusión

2 Development of Logical Reasoning

Early Logic
Progress in Logical Reasoning
Logical Reasoning in Adulthood
Mathematical Reasoning
Conclusión

3 Reasoning Beyond Logic

Scientific Reasoning
Moral Reasoning
Social Conventional Reasoning
Reasoning as Epistemologically Self-Regulated Thinking
Conclusión

4 Metacognition and Epistemic Cognition

Metacognition
Epistemic Development in Childhood
Epistemic Development Beyond Childhood
Complexities of Epistemic Cognition
Conclusión
5 Argumentation as Collaborative Reasoning
Collaborative Reasoning in a Logic Task
Additional Evidence for Collective Rationality
The Rational Social Ideal
Conclusión

6 Democracy as Collaborative Rationality

Deliberative Democracy
Is Democracy Psychologically Possible?
Rationality and Identity
Rational Institutions
Intellectual Freedom in Society
Conclusión

7 The Rational Construction of Rational Agency

“The Enigma of Reason”
Constructivist Accounts of Development
Development Across and Beyond Childhood
Development of Rational Self-Governance
Conclusión

8 Education for Rationality

Critical Thinking
Education as the Promotion of Development
Indoctrination in Curriculum and Instruction
Intellectual Freedom in Education
The Rational Ideal
Conclusión

9 Reasons and Persons

Becoming the Borg
“The Robot’s Rebellion”
Who Counts as a Person?
What Counts as a Reason?
Conclusión
Glossary
References
Author Index
Subject Index
PREFACE

Somewhere around 1980 I had the idea of writing what would

have been my first book: a small text on the development of
reasoning, called simply “The development of reasoning.” I had
been doing research on the development of logical and
scientific reasoning in adolescence since gradúate school in the
early 1970s, and was also very interested in the development
of moral reasoning. The book would have traced the
developmental course of several types of reasoning (covering
literaturas here addressed in Chapter 2 and part of Chapter 3)
and would also have addressed some of the associated
matters of metacognition, development, and education
discussed in Chapters 4, 7, and 8, respectively.
Over the past four decades I have continued to study and
write about reasoning. Portions of what I had in mind for my
reasoning book showed up in other publications, including my
books Adolescent Rationality and Development (Moshman,
2011a) and Epistemic Cognition and Development (Moshman,
2015). I continued to think about a book that would look directly
and comprehensively at the development of reasoning and
associated progress in rationality, but my conception of this
book gradually shifted in three major ways.
First, I increasingly found that there was much to say about
adult reasoning and rationality, without which the
developmental picture is incomplete. But it was not just
developmentalists who needed to see an account of human
reasoning and rationality that integrates the cognitive and
developmental literatures. I wanted to write a book for
cognitive, developmental, educational, and other psychologists
interested in a broad perspective on reasoning. Thus the term
“development” does not appear ¡n the title of this book and the
book is not primarily about children, though the perspective
throughout is substantially developmental.
Second, I have appreciated the efforts of psychologists and
philosophers working to extend the study of reasoning from
individual to group reasoning and to develop new conceptions
of reasoning as a social process of argumentation. I have also
appreciated the work of theorists of democracy to develop
models of deliberative democracy in which argumentation plays
a central role. I wanted to provide a comprehensive account
that construes reasoning and rationality as no less social than
individual, encompassing inter-individual processes of
argumentation, which can be seen as collaborative reasoning,
and rational social institutions, including deliberative forms of
democracy. I ultimately decided I could best communicate the
focus of the book with a title that proceeds from reasoning to
argumentation to deliberative democracy.
Finally, as I thought about the extended picture of reasoning,
and about the state of the world, I realized I also had an
argument to make about the role of reasoning in democracy
and the promotion of rationality at both individual and social
levels. I have tried to present the psychology of reasoning in a
way that not only extends to the social but also illuminates the
potential for enhancing democracy.
Reasoning, Argumentation, and Deliberative Democracy
provides an integrated treatment of the reasoning and
rationality of individuáis, as studied in the largely distinct
literatures of cognitive and developmental psychology. Going
beyond these psychological literatures, it also considers social
aspects of reasoning and rationality, drawing not only on social
psychology but also making connections to philosophy, law,
political science, and educational policy. The book has nine
chapters, beginning with an introduction that summarizes the
argument and introduces the main concepts. The second
chapter reviews research on the development of logical
reasoning. Chapter 3 expands the conception of reasoning
beyond logic to other forms of reasoning. Chapter 4 develops
the conception of rationality beyond various forms of reasoning
to include metacognitive knowledge about and control of
reasoning. Chapter 5 expands the conception of reasoning
beyond individual cognition, the focus of most of the cognitive
and developmental literature, to include collaborative
reasoning. Chapter 6 extends this social conception of
rationality beyond interindividual argumentation to societal
aspects of rationality. Chapter 7 goes beyond earlier
descriptions of developmental progress to look more directly at
the process of developmental change. Chapter 8 goes beyond
the explanatory focus of the previous chapters to consider the
deliberate promotion of reasoning, rationality, and democracy.
Finally, Chapter 9 turns to Science fiction to understand our
world of persons and reasons.
This book is aimed in part at researchers, theorists, and
students in cognitive, developmental, and educational
psychology concerned with reasoning, rationality,
metacognition, and/or advanced levels of development. I hope
cognitive psychologists will see the importance of a
developmental perspective in understanding human rationality
and that developmental psychologists will see the importance
of a serious look at adult cognition in understanding the
reasoning of children and the nature of developmental change.
I also hope all psychologists will see the importance of
conceptualizing reasoning socially while maintaining a focus on
individuáis.
The book is also intended for audiences beyond the field of
psychology. In particular, the focus on deliberative democracy
connects the psychological study of reasoning to questions of
law, philosophy, and political Science. For anyone aiming to
develop a psychologically plausible account of deliberative
democracy, or simply concerned about the state of democracy
in the 21st century, recent psychological claims of human
irrationality may be disheartening. I show in this book that there
is much evidence supporting rationalist assumptions and much
we can do to promote the sort of reasoning that makes
democracy possible and productive.
The book may serve as a supplementary text for students in
upper-level and gradúate courses, especially in psychology and
education, including courses in cognitive psychology, thinking
and reasoning, cognitive development, or lifespan
development. Because it is substantially interdisciplinary, it
does not assume readers will be expert in any particular
discipline. In the introductory chapter I provide an overview of
the book’s main argument and an introduction to 15 central
concepts from several relevant disciplines. I have also carefully
defined all terms as I proceed and included a glossary for
reference. I hope this book will help readers see the big picture
of reasoning and rationality in a way that connects theories,
concepts, and evidence from a variety of fields.
For early guidance in the study of reasoning, I am grateful to
Edith Neimark, my gradúate adviser at Rutgers University,
whose work spanned cognitive development (Neimark, 1970)
and adult thinking (Neimark, 1987; Neimark & Santa, 1975),
including Piaget’s conception of formal operations (Neimark,
1975) and related work on logical reasoning (Neimark &
Slotnick, 1970). She was hard to please but I always hoped for
the best when I gave her the latest draft of whatever I was
writing, and I was always delighted if she thought it was not too
bad.
I am grateful to Deanna Kuhn for detailed and helpful
feedback on the manuscript. I also thank Rick Bady, Austin
Dacey, Wim De Neys, Jeff Fraum, John Gibbs, Charlie Hauss,
Peter Levitov, Rick Lombardo, Henry Markovits, Hugo Mercier,
Adam Morfeld, Bill Overton, Harvey Siegel, Les Smith, Rich
Stamler, Keith Stanovich, Joe Starita, Truong Thi Nhu Ngoc
(Ruby), and Sam Walker for comments on various chapters.
As always, I am grateful to my wife Sara for everything and
more, even if that’s a logical impossibility.
1
INTRODUCTION

Research on reasoning has undermined traditional

assumptions of human rationality. The concept of democracy,
however, assumes that people are rational agents, competent
to govern themselves individually and collectively. If humans
are not rational agents, we must reconsider the political ideal of
democracy.
I argüe in this book that, despite systematic flaws in human
reasoning, there are three reasons for a cautiously optimistic
assessment of human rationality and the prospects for
democracy. First, research and theory in cognitive and
developmental psychology provide the basis for a more positive
assessment of individual rationality. Second, groups of people,
under the right circumstances, reason better than individuáis.
Third, developmental progress toward greater rationality can be
promoted through education. Psychological research provides
a basis for conceptualizing a system of deliberative democracy
that relies on and promotes collaborative reasoning, rational
institutions such as science and law, education aimed at the
promotion of rationality, and intellectual freedom for all.
I begin this chapter with a nontechnical overview of the book,
supporting and fleshing out the argument summarized in the
two paragraphs above. This is followed by three sections
providing a more technical introduction to 15 central concepts:
reasoning, rationality, reasons, persons, bounded rationality,
heuristics and biases, dual processing, metacognition,
epistemic cognition, developmental progress, developmental
process, argumentation, deliberative democracy, education,
and intellectual freedom.

Overview
The remaining eight chapters of the book begin with the
biological bases of logic and proceed from there to reasoning of
several sorts, metacognitive aspects of rationality, rational
argumentation, the ideal functioning of democratic societies,
the nature of development at individual and social levels, the
role of educational institutions in the promotion of rationality,
and defending against the Borg.
Chapter 2 (Development of Logical Reasoning) considers
the development of logical reasoning, beginning with Piaget’s
conception of logic as intrinsic to biological self-regulation.
Psychological research reveáis a sensorimotor logic of action
already developing in infancy and shows that even preschool
children routinely make logically correct inferences. The
development of logic is in large part the development of
(metalogical) understanding about the nature of logic and
associated regulation of one’s inferences. Research shows
systematic and universal progress in logical reasoning over the
first 12 years of life, leading to the ability (beginning about age
11 or 12 years) to determine the logical implications of ideas
one deems hypothetical or even false. Development beyond
that is much less universal and related more to experience and
education than to age. There is extensive evidence that even
adults fall far short of logical perfection, with substantial
individual differences. Mathematical reasoning, which also
involves matters of logical necessity, shows a similar
developmental pattern of universal childhood progress over the
first 12 years, with substantial individual differences in further
development and systematic errors at all ages.
One might think the development of logical reasoning is the
basic story of reasoning and rationality. Perhaps all we need to
complete the picture is more detail about logical functioning at
various ages across the lifespan.
 Chapter 3 (Reasoning Beyond Logic) argües to the contrary
that, even if we interpret logic broadly to go beyond deduction,
reasoning is more than logic. Causal reasoning is discussed as
a form of reasoning central to science that cannot be reduced
to logic because it involves matters of empirical fact and
theoretical interpretation that go beyond logical necessity.
Reasoning beyond logic also includes principled reasoning,
which is central to morality, and reasoning on the basis of
precedent, which is central to social tradition. Logical, causal,
principled, and precedent-based reasoning are all concerned
with matters of truth and justification, though they conceive of
such matters differently. Reasoning is defined as
epistemologically self-regulated thinking, meaning it involves a
focus on the justification and truth of one’s conclusions. This
involves the generation and use of reasons, including but not
limited to logical reasons. As with logical reasoning, research
on causal, principled, and precedent-based reasoning shows
systematic and universal progress in the first 12 years of life.
Here too, however, adults show substantial individual
differences in the extent of development beyond that, and all
fall far short of rational perfection.
One might think this expanded conception of reasoning, in
which logical reasoning is one of several forms, provides a
comprehensive framework. Perhaps all that remains is to flesh
out the picture with detailed research on the development and
functioning of various sorts of reasoning.
Chapter 4 (Metacognition and Epistemic Cognition) argües
to the contrary. Taking into account the psychological reality of
logical inferences in very young children, it makes the case that
the development of reasoning is in large part the development
of metacognition, expanding on the discussion in Chapter 2
about the development and role of metalogical understanding.
Of particular importance to reasoning is the aspect of
metacognition concerned with the justification and truth of
beliefs— epistemic cognition. The chapter traces the
development of epistemic cognition in childhood and beyond.
Drawing on a variety of theorists, it suggests there is a
theoretical consensus that epistemic development is largely a
process of recognizing subjectivity and then re-establishing a
basis for objectivity (at least as a meaningful ideal) in múltiple
domains and at múltiple levels of abstraction. Looking across
Chapters 2-4, cognitive development is presented as progress
toward rationality, thus distinguishing it from other cognitive
changes. A distinction can be made between basic
development, referring to the universal progress in rationality
across the first 12 years of life, and advanced development,
referring to further progress that is highly variable across
individuáis.
Again one might think we now have the basic framework of
reasoning and rationality with only specifics left to be filled in
about our understanding and regulation of our various
inferential processes across development. What more could
there be?
Chapter 5 (Argumentation as Collaborative Reasoning)
notes that the picture so far presented, consistent with the
literatures of cognitive and developmental psychology,
construes reasoning as a process that takes place in an
individual mind. An emerging literature, however, suggests that
argumentation among people can be construed as a process of
collaborative reasoning. The chapter reviews evidence that
when small groups engage in reasoning they are sometimes
more rational than any of their individual members. Of course it
is also true that groups can be less rational than their members.
Examination of the conditions that foster group rationality
suggests the importance of respect for reasons and persons. It
appears that rationality can be enhanced through
argumentation, but the possibility and quality of argumentation
depend on the size, circumstances, and functioning of the
group, including the respect of group members for each other’s
intellectual freedom.
Having supplemented the standard individualistic picture of
reasoning and rationality with a social picture, it might appear
that the account is now comprehensive. Again, what more
could there be?
Chapter 6 (Democracy as Collaborative Rationality) adds
another whole dimensión to the discussion by distinguishing
two meanings of social (following Piaget and many other
theorists): (1) interindividual and (2) societal. Argumentation
among individuáis in small groups (the topic of Chapter 5)
addresses the social aspect of reasoning only in the first sense.
What remains is the nature and role of society. Many theorists
in philosophy, law, and political science (especially neo-
Kantians such as John Rawls) have construed democracy as a
project of collaborative rationality based on respect for reasons
and respect for persons, and some in this tradition have
specifically advocated “deliberative democracy” as the ideal
form of democracy. This chapter connects democratic theory to
research on reasoning and rationality. Along the way, it
examines identity as a theory of oneself that coordinates the
personal and the social, notes the dangers of identity for
rationality and democracy, and proposes a concept of
rationalist identity consistent with deliberative democracy.
Argumentation is crucial to social institutions, such as
legislatures, law, and science, that play important roles in the
rationality of democratic social systems. Intellectual freedom,
seen in Chapter 5 as crucial to argumentation, is
correspondingly crucial to democracy, especially deliberative
democracy.
We now have an account of reasoning and rationality
extending from the biological bases of logic early in Chapter 2
to the ideal functioning of democratic societies and rational
institutions in Chapter 6. Now, one might think, we must surely
be done, unless we want to talk about rationality across the
galaxy or in nonbiological forms. But the picture remains
incomplete even if we remain on Earth and stick to human
rationality.
Chapter 7 (The Rational Construction of Rational Agency)
turns our attention to processes and patterns of development.
Previous chapters described the course of developmental
progress in logical reasoning (Chapter 2); in scientific, moral,
and social conventional reasoning (Chapter 3); and in the
metacognitive basis for reasoning (Chapter 4). Chapter 7,
however, looks more directly at fundamental processes of
developmental change and general patterns of progress in
rational agency. Drawing on Chapters 5 and 6, moreover, it
considers the role of argumentation and democratic institutions
in the development of individuáis and the possibility that groups
and societies might themselves show developmental change. I
argüe that rationality is neither inherent in the human genome
ñor learned from human environments, though genes and
environments are obviously important. Rather, rationality is
constructed over time by increasingly rational agents through
rational and social processes of reflection and coordination.
Subjectivity is inevitable but reflection on subjectivity enables
progress to metasubjective forms of objectivity and increasingly
rational self-governance at individual, social, and societal
levels.
And what remains beyond this seemingly comprehensive
account of the development of reasoning and rationality? What
more is there to be said?
Chapters 2-7, even as each went beyond the previous one,
remained focused on describing and explaining reasoning,
rationality, and development. What remains for Chapter 8
(Education for Rationality) is the question of how we can
promote the development of reasoning and rationality. The
chapter begins with the concept of critical thinking, which has
long been popular in education and, although variously defined,
is closely connected to rationality in its concern for reasons.
Peer argumentation is proposed as an educational approach
that promotes the development of rationality. The chapter also
notes the prevalence of indoctrination in curriculum and
instruction, especially in elementary and secondary education.
Extending previous discussions about the role of intellectual
freedom in collaborative reasoning (Chapter 5), deliberative
democracy (Chapter 6), and the development of rational
agency (Chapter 7), Chapter 8 addresses the central role of
intellectual freedom in any educational program aimed at the
promotion of rationality. The chapter concludes with a brief
sketch of the ideal rational agent.
And finally, I do turn to questions of rationality across the
galaxy and in nonbiological forms. But my focus remains on
human rationality on the planet Earth, which I now reconsider
from the perspective of science fiction.
Chapter 9 (Reasons and Persons) begins with the Borg, a
collective entity introduced in the televisión series Star Trek:
The Next Generation. The Borg is composed of biomechanical
life forms with limited individual agency. Within the Star Trek
universe, the Borg is a fearsome threat because it assimilates
all societies and civilizations it encounters. In the real world, I
suggest, the more immediate threat is not that the Borg will
arrive from elsewhere and assimilate us, but rather that we
ourselves will turn into the Borg. Resisting that tendency
requires active reflection on rationality. It also requires efforts to
support reasons and persons, neither of which can exist without
the other. I ¡Ilústrate the importance of personhood with two
legal cases in which a court faced specific questions as to
whether a particular individual was legally a person. In 1879,
the Ponca chief Standing Bear was found in a federal court in
Omaha, Nebraska to be a person within the meaning of the law
and thus entitled to determine where he would live. Nearly five
centuries later, the android Data is likewise found, in an
episode of Star Trek: The Next Generation, to be a person with
the right to make his own choices. Turning finally to an episode
from the original Star Trek, I ¡Ilústrate the centrality of truth and
justification. The alternative to the Borg is a world of persons
and reasons.

Reasoning and Rationality

In the remaining three sections I introduce the major concepts
of the book. First, I consider the nature of reasoning, especially
in relation to rationality. I then tum to cognitive and
developmental psychology, which have generally focused on
the reasoning, rationality, and development of individuáis. This
individualist account is then supplemented by social
considerations of argumentation and democracy that extend
and enrich standard cognitive and developmental accounts of
reasoning and rationality.
I begin with reasoning, a central concern of philosophy at
least since Aristotle and a major topic of psychological research
since the 1930s. The study of reasoning raises questions of
rationality, leading to further discussion concerning the nature
and role of reasons, agency, and personhood.

Reasoning as Epistemologically Self-Regulated

Thinking
I use the term “reasoning” in much the way it is used both in
ordinary discourse and in the scholarly literatures of philosophy
and psychology. Although reasoning has been defined in a
variety of ways, I attempt here to capture what I believe to be
something cióse to a consensus. Reasoning is generally seen
to be (1) inferential; (2) under the control of an agent; and (3)
aimed at reaching conclusions that are true or at least justified.
With respect to the first, reasoning is generally seen as a
process by which we infer something that goes beyond what
we have perceived. With respect to the second, automatic
inferences play a crucial role in cognition but are usually
distinguished from reasoning, which is deliberate. And with
respect to the third, reasoning provides reasons for belief or
action, which is why it makes good sense to cali it reasoning.
With that in mind, I define reasoning as epistemologically
self-regulated thinking. I define thinking as the deliberate
application and coordination of one’s inferences to serve one’s
purposes. I define inference as going beyond the data. And I
use the term “epistemológica!” to refer to matters of truth and
justification, based on a definition of epistemology as the study
of knowledge, especially with regard to normative matters of
justification and truth (Moshman, 2011a, 2015).
These definitions recognize that inferences can be entirely
automatic. People routinely make inferences that they did not
mean to make without any awareness that they have made
them. You are making a variety of grammatical and semantic
inferences right now as you extract meaning from the sequence
of words in this sentence (Graesser, Singer, & Trabasso,
1994). Even young children routinely make a variety of
inferences, including logical inferences (more on this in Chapter
2).
We do not generally regard our ongoing inferential
processes as thinking, however. Rather, we reserve the term
“thinking” for cognitive acts aimed at serving some purpose,
involving some degree of awareness and control of our
inferences. If you stop reading to study a particular sentence
that confused you, focusing on specific words and grammatical
relations, we might say you are thinking about that sentence.
But it would seem odd to say that you have thought about every
sentence you have read. Reading always involves the
Processing of information and is always inferential, but it seems
useful to distinguish (a) a deliberate effort to make sense of
something confusing from (b) the ongoing and largely
automatic processing of text. More generally, it is important to
distinguish acts of deliberation from automatic processes. Thus,
following common (but not universal) practice, I reserve the
term thinking for deliberate actions that serve the purposes of
an agent.
Reasoning, in tum, is a type or aspect of thinking, but not all
thinking is reasoning. Thinking may be aimed at practical goals,
such as solving some problem or making some decisión, and
may be satisfied with pragmatic solutions. But sometimes we
think in order to determine the truth of some matter, or at least
to reach the most justified conclusión. Thus, again following
common (but not universal) practice, I have defined reasoning
as a subset of thinking in which we regúlate our inferential
processes for epistemic purposes— that is, to reach
conclusions that are true, or at least justified.
Given this definition, múltiple forms of reasoning can be
distinguished. I consider logical reasoning in Chapter 2. I then
proceed in Chapter 3 with causal, principled, and precedent-
based reasoning, which are central, respectively, to the
domains of science, morality, and social convention (Moshman,
2015). Later, in Chapter 5, I turn to argumentation, which I
construe as a collaborative form of reasoning.

,
Good Reasoning Bad Reasoning, and Not Reasoning
Reasoning of all sorts serves epistemic purposes. Thus we can
evalúate it with respect to how well it serves those purposes.
This suggests a distinction between good reasoning, which
reaches justifiable conclusions, and bad reasoning, which
reaches conclusions that cannot be justified. Such a distinction
is indeed meaningful and important, but conceptual and
empirical considerations complícate the picture. Sometimes
what looks like good reasoning or bad reasoning is actually not
reasoning at all. Thus we need an additional distinction: We
must distinguish reasoning from not reasoning. Only if
someone is actually reasoning is there a legitímate question of
whether the reasoning is good or bad.
Consider a mathematical example: What is 4 * 3? An
answer of 12 is correct, but if this answer comes to you
immediately you are not reasoning. Right answers do not
necessarily show good reasoning. This is not only because the
reasoner may have been reasoning poorly and got lucky but
also because the person in question may not be reasoning at
all. Suppose, however, the question were asked of a child of
about 8 years who had not memorized basic multiplication facts
but understood the logic of multiplication. If the child
determined that 4 x 3 means 4 + 4 + 4, added 4 + 4 to get 8,
and then added another 4 to get 12, this would indeed be good
reasoning. And if the child made a mistake along the way and
ended with an answer different from 12, that would be less than
good reasoning, maybe bad reasoning, but still reasoning.
Just as right answers do not necessarily show good
reasoning, wrong answers do not necessarily show bad
reasoning. One outcome of research in cognitive psychology is
evidence that much of what might seem at first to be bad
reasoning is actually not reasoning at all. Much of our cognition
consists of automatic processes, including a variety of
heuristics and biases, that are often helpful but may also lead
us astray (Kahneman, 2011).
These considerations about reasoning raise questions of
rationality, which can be seen in two ways. In the next
subsection I present an objectivist conception of rationality as a
matter of good reasoning rather than bad reasoning. The
subsection after that supplements this with a subjectivist
conception of rationality as a matter of rational agency, which is
what makes reasoning (good or bad) possible at all. As
discussed above, we must distinguish good reasoning from bad
reasoning, and must also distinguish reasoning (whether good
or bad) from automatic cognitive processes. The objectivist
aspect of rationality enables us to do the first. The subjectivist
aspect enables us to do the second.

Rationality as Good Reasoning

Perhaps the most common conception of rationality, at least
among psychologists, is that it is a matter of good reasoning.
This conception of rationality is objectivist in that it assumes
reasoning can be evaluated objectively against logical,
statistical, or other norms. There is sometimes dispute about
what norms apply in particular cases, but the concern with
norms connects empirical psychology to the normative
considerations of philosophy. To what extent, asks the
objectivist conception of rationality, do people reason the way
logicians and other philosophers say they should?
The answer is we often do and often don’t. Sometimes when
we don’t conform to logical or other norms it is unclear whether
we should have done so. But in many cases there is a
consensus among experts that the failure to conform to
normative standards cannot be justified. Both good reasoning
and bad reasoning are common. The objectivist conception of
rationality highlights the rationality of good reasoning.

Rationality as Rational Agency

The objectivist conception of rationality, as just seen, evaluates
reasoning by applying standards proposed by logicians and
other philosophers. But why should we listen to logicians and
philosophers? The short answer is that they are operating at a
higher level of understanding with regard to logical and
epistemological matters (more on this in Chapters 2, 4, and 7).
But they are people, nonetheless. Without people there could
be no reasoning or rationality at all. Rationality does involve
objective norms but it is not a set of pre-existing rules.
Rationality, in its subjectivist sense, is a characteristic of
people, who, even when they reason badly, are rational agents.
To be a person, then, is to be a rational agent. To be a
rational agent is to act on the basis of reasons. To act on the
basis of reasons is to form, maintain, revise, and act on one’s
own beliefs rather than to have one’s beliefs and actions
caused by physical, biological, social, or other forces beyond
one’s control. Working through what we mean by reasoning
and rationality takes us, it seems, beyond the causal universe
to a world of reasons and persons.

Reasons and Persons

The human body is a physical object subject to causal forces.
But we rarely explain the movements of human bodies in
causal terms. Instead, we generally think and act on the
presumption that we and others are persons who engage in
actions on the basis of reasons.
If a tornado were to blow me across a field, for example, no
one would ask why I crossed the field. The motion of my body
was caused by strong winds. But in most cases where I cross a
field, there would indeed be an expectation that I have and
could provide an explanation. Maybe I was playing a game.
Maybe I was going someplace. Maybe I was looking for
something. Maybe I was avoiding someone. Maybe I was
taking a walk. In each case I had a reason for Crossing the field.
Crossing the field was an action, not just a movement of my
body, and is explained by my reasons, not by a physical or
other forcé causing something to happen. This is what it means
to be a rational agent.
This picture of persons as rational agents is not entirely true,
but it is not simply a fiction. The presumption of rational agency
is perhaps best described as an idealization. We are part of the
causal universe, even when we think we are determining our
own beliefs and behavior, but we are also, to some extent,
rational agents. To the extent that we recognize and valué our
rational agency, we respect both reasons and persons.
Respect for reasons and persons enhances our rational agency
and, in the ideal, creates a world of reasons and persons in
which reasoning and rationality flourish.
The opposite of this is the world of the Borg. In Chapter 9,
using examples from Star Trek, I conclude the book with an
idealized visión of a world of persons and reasons.

Cognition and Development

Reasoning has long been a major topic of study in both
cognitive and developmental psychology, with substantial
attention given to issues of rationality in both fields. In this
section I discuss theoretical conceptions from cognitive and
developmental psychology that play important roles in the
present analysis of reasoning and rationality. Drawing first on
cognitive psychology, I consider the bounded nature of human
rationality, the role of heuristics and biases, and the plural
nature of cognitive processes. Drawing on developmental and
educational psychology, I consider metacognitive, and
especially epistemic, aspects of reasoning and rationality.
Finally, I consider what it means to make developmental
progress, and what processes account for such progress.

Bounded Rationality
Our concern in this book is with human rationality, not all
possible forms of rationality. Human reasoning is limited by the
human information processing system. In evaluating human
reasoning against rational ideáis, we must recognize that not all
ideáis are humanly achievable. There may be some valué in
evaluating human reasoning against standards of rational
perfection, but it is also important to evalúate human reasoning
against what a human mind could reasonably be expected to
achieve in a reasonable period of time. In fact, it is rational to
do the best one can with the mental resources at hand and to
reach a conclusión while there is still time to act on it.
Sometimes people properly settle for a satisfactory solution
rather than continuing to search for an ideal one.
One well-established constraint on human cognition, for
example, is working memory (Barrouillet & Gaillard, 2011). Is it
rational to consider only a small number of items at once rather
than taking everything into account simultaneously? For a
human being, there is often no choice but to simplify or
approximate. Even if we should consider everything at once,
we can’t.
Limitations, however, can often be overcome. If you ask me
to multiply 8,756 by 3,472 in my head, the best I can do is
provide an approximate result, not because I don’t understand
multiplication but because the information processing demands
of the task overload my working memory. But I can provide an
exact result using pencil and paper, or more easily with a
calculator. Rationality is not just doing the best possible under
the circumstances but creating better circumstances. Thus, the
limitations of the human information processing system are not
the last word, but they are a factor to be considered.
Heuristics and Biases
One might hope that, even if our information processing
systems must function within strict constraints, such as working
memory limits, our inferential processes are all rational at least
in the sense of conforming to logical or other norms. Alas, as
we will see in Chapter 2, research and theory since the 1930s
have suggested that our inferential processes often lead us
astray. Serious doubt about human rationality became
increasingly prevalent and influential beginning with research
by Peter Wason in the 1960s (Wason & Johnson-Laird, 1972).
Beginning in the early 1970s, Kahneman and Tversky
(Kahneman, 2011) were particularly important in providing
evidence that inference often involves heuristic processes that
are subject to a variety of biases. To refer to these processes
as “heuristic” suggests that they are pragmatically useful in
providing satisfactory answers to questions we are likely to
face. To refer to “bias” suggests that our inferences generate
conclusions that deviate systematically from what rational
norms require. Much research in the latter third of the 20th
century suggested that human beings, far from being rational
agents, are largely driven by nonlogical processes beyond our
control.

Dual Processing
Dual process theories construe cognitive functioning as an
interaction between distinct systems or levels of inferential
Processing. Such theories aróse in cognitive and social
psychology in the 1970s and have increasingly dominated the
psychological study of reasoning since the 1990s (Barrouillet &
Gauffroy, 2013; Evans, 2007; Evans & Stanovich, 2013;
Kahneman, 2011; Klaczynski, 2000; Markovits, Brunet,
Thompson, & Brisson, 2013; Ricco, 2015; Ricco & Overton,
2011; Stanovich, 2004, 2011). Current philosophical accounts
of reasoning are also generally consistent with dual process
conceptions and sometimes refer to them explicitly (Richard,
2019; Siegel, 2019; Staffel, 2019).
Dual process theories generally distinguish two systems or
levels of processing. Despite differences in the precise
characterization of these systems or levels, there appears to be
a widely accepted distinction between a system of automatic
and intuitive processes (often just called System 1) and a
system of deliberate and reflective processes (often just called
System 2). System 1 involves heuristics and biases that are
often highly adapted to the cognitive issues we face but may
also lead us astray. System 2, relying on deliberation and
reflection, may overcome the limitations of System 1 but is itself
subject to systematic biases due to strong commitments to
protect and maintain our theoretical beliefs and personal
identities.
With regard to rationality, then, System 1 is often rational in
the objectivist sense but is not rational in the subjectivist sense.
There is no agent responsible for its operations. System 2 is
rational in the subjectivist sense of agency and respect for
reasons but can be systematically biased by self-serving
motivations, such as the human tendency to apply critical
thinking selectively to disfavored ideas (Klaczynski, 2000).
Human rationality relies on the interplay of both systems
(Stanovich, 2004).
But human rationality, as already noted, is bounded by
human constraints. Even if System 2 functioned perfectly, it
could not replace System 1. Given the stringent limits of
working memory and the effort required for self-regulation,
ongoing application of System 2 to everything is not
psychologically realistic for human beings. To function
adaptively we rely largely on System 1, which involves
automatic and intuitive processes not subject to the
metacognitive demands of self-regulation and the strict
constraints of working memory. The deliberate and reflective
processes of System 2 are profoundly important but are just the
tip of the cognitive iceberg, most of which operates below the
surface of consciousness.
 Further examination of System 1 shows it to be a collection
of systems rather than a single system. There are many
systems of automatic inference and associated intuitions.
Inferring basic emotions from people’s facial expressions, for
example, is a complex cognitive task that nearly all of us
perform automatically and effortlessly. It appears that evolution
and early development provide us with a special system of
inferential processes highly adapted to the specific task of
inferring emotions from facial expressions. Similarly, speaking a
human language requires complex linguistic transformations
that are routine among 4-year-old children, though even adults
are mostly unable to explain or justify their grammatical
intuitions. The diverse subsystems of System 1 operate
autonomously, independent of each other and beyond the
reach of System 2 (Stanovich, 2004).
There is reason to believe that System 2 is also internally
complex (Ricco, 2015; Ricco & Overton, 2011; Stanovich,
2011). System 2 is generally seen as both deliberate and
reflective. Thinking, as defined earlier in this chapter, is
deliberate, and thus associated with System 2. Reasoning,
defined as a form of thinking, is also deliberate and thus
associated with System 2. But reasoning, as defined earlier,
involves epistemological reflection on one’s inferences,
whereas thinking need not be reflective. Most versions of dual
Processing confíate two distinctions: (1) the distinction between
automaticity and deliberation and (2) the distinction between
intuition and reflection. This leads to a conception of just two
systems, one automatic and intuitive and the other both
deliberate and reflective. A distinction of reasoning from
thinking, however, reminds us that thinking can be deliberate
without being reflective. Thus, we need to distinguish (1)
múltiple systems of automatic inference and associated
intuitions, (2) deliberate but pragmatic processes of thinking,
and (3) reflective processes of reasoning.
Research on dual Processing has important implications for
understanding the development of reasoning and rationality. In
demonstrating the major role of automatic, intuitive processes
in human cognition at all ages, dual process conceptions
remind us that deliberation supplements automatic inferences
rather than replacing them, and that reflection supplements
intuition rather than supplanting it.

Metacognition and Epistemic Cognition

Metacognition is cognition about cognition, including (1)
conceptual knowledge about knowledge and inference and (2)
procedural knowledge implicit in cognitive self-regulation
(Moshman, 2011a, 2015; Schraw & Moshman, 1995). Thus
defined, it is a very broad concept, including much that is
discussed using related terms, such as self-monitoring, self-
knowledge, introspection, reflection, cognizance,
consciousness, metarepresentation, attribution, perspective
taking, theory of mind, executive function, and executive
control. Under varying terminology, metacognition has been a
central topic of developmental and educational psychology
since the 1970s. The human mind’s understanding of itself and
its ability to regúlate itself show dramatic and universal age-
related progress over the first 12 years of life. Metacognitive
development often continúes well beyond that, but there are
substantial individual differences in the nature and extent of
further progress (Demetriou & Spanoudis, 2018; Moshman,
2011a, 2015).
Thinking is metacognitive in that it involves the deliberate
control of inferences to serve the purposes of the thinker. In the
case of reasoning, this self-regulation is epistemological,
meaning that it is particularly concerned with truth and
justification. This requires knowledge about matters of
epistemology— that is, epistemic cognition (Greene, Sandoval,
& Bráten, 2016; Moshman, 2015). Thus epistemic cognition is
the subcategory of metacognition most directly relevant to
reasoning. Of course, few people other than philosophers read
the literature of epistemology. But beginning in early childhood,
we all think about matters of justification and truth and we all
make dramatic progress in our understanding of these matters.
Early development of epistemic cognition can be seen in
children’s increasing recognition of their own subjectivity in a
variety of situations, their progress in recognizing the legitimacy
of múltiple perspectives, and their later progress in
understanding that, even if there isn’t always one true answer,
some ideas may be better than others. Adolescents and adults
go beyond this in recognizing more abstractly that knowledge is
frequently subjective, perhaps inherently subjective, raising
general questions about the status of objectivity. Some come to
understand the challenge of subjectivity more fully than others
and some, to varying extents, respond to that challenge by
constructing rationalist epistemologies that motívate and
support better reasoning (Chandler, Hallett, & Sokol, 2002;
Moshman, 2015).
Thus, there is great potential for development in epistemic
cognition beyond childhood and associated progress in
reasoning, but great variability in the extent to which that
potential is realized (Greene et al., 2016; King & Kitchener,
1994; Kuhn, 2005; Moshman, 2015). Metacognition and
epistemic cognition are the topic of Chapter 4, with further
discussion about the development and promotion of epistemic
cognition in Chapters 7 and 8.

Developmental Progress
Development is change, but not all change is development. The
prototypical example of developmental change is the case of an
immature organism over some portion of the lifespan making
anatomical and physiological progress toward maturity, as in
the attainment of sexual maturity over the course of puberty.
Four aspects of such change are particularly notable, providing
the basis for a definition of development that distinguishes
developmental changes from other sorts of change. First, the
change takes place over an extended period of time. Second,
the change is regulated by internal organismic processes rather
than imposed directly by the environment. Third, the change
involves a succession of qualitatively distinct structures or
modes of functioning rather than simply a process of growth in
the quantitative sense of getting larger or faster. And finally, the
change is Progressive, rather than arbitrary, neutral, regressive,
pathological, or specific to particular environments or cultures.
Development, then, is a type or pattern of change that is
extended over time, self-regulated, qualitative, and Progressive
(Moshman, 2011a).
Developmental change extends far beyond anatomy and
physiology to psychological realms of behavior and cognition
(Piaget, 1967/1971). As we will see in Chapters 2-4, children
show systematic age-related changes in logical, scientific,
principled, and precedent-based reasoning and in associated
conceptions of knowledge. By the age of 12 or 13 years, nearly
all individuáis have achieved a level of rational agency virtually
unseen in children under the age of 10 years, with strong
potential in most cases for substantial further development in
adolescence and adulthood. Development beyond age 12,
however, is far from universal and is less related to age than to
experience and education.

Developmental Process
The accounts of developmental change provided in Chapters
2 -4 are largely descriptive, consisting mostly of a succession of
stages or levels through which the individual passes in the
course of development. But what explains such development?
To explain development requires an account of the process
itself, not just descriptions of a series of outcomes.
Research looking closely at developmental transitions
suggests that progress in reasoning and rationality comes
about through rational processes of coordination and reflection
that transform earlier structures into better ones and make
implicit knowledge explicit. The full discussion of developmental
process is deferred until Chapter 7, however, because it needs
to incorpórate considerations mostly absent from Chapters 2 -4
because they are mostly (though not entirely) absent from the
literaturas of cognitive and developmental psychology. The
missing considerations are social considerations, including both
interpersonal rational exchanges (discussed in Chapter 5) and
the rational institutions of society (Chapter 6).

Argumentation and Democracy

Argumentation is both cognitive and social. It is closely
associated with reasoning but not always seen as a form of
reasoning. I suggest that it be seen as collaborative reasoning,
and that such reasoning can be seen as the foundation of
democracy or, specifically, of an ideal form of democracy
known as deliberative democracy.

Argumentation as Collaborative Reasoning

Argumentation is often seen as a matter of dispute. To argüe is
to disagree, which is to be disagreeable. But disagreement is a
fact of life, and verbal dispute is far from the worst response to
it.
Argumentation may thus be seen, somewhat more
positively, as a matter of debate. Given two positions held by
two different people, we argüe to determine which, and thus
who, is right. Even if neither person changes position, the
debate may help others choose the better alternative. Thus,
argumentation potentially serves a rational purpose.
But sometimes both positions have some merit, and neither
debater is entirely correct. This may render the argumentation
inconclusive and perhaps seemingly pointless. But if the
process of argumentation is seen as collaborative reasoning,
the arguing parties may find a way to coordínate their
perspectives to create a new position better than either of the
original positions, and thus may both profit from the exchange.
As discussed in Chapter 5, research shows that, under the right
circumstances, groups may use argumentation to operate more
rationally than any of the individuáis composing them.
Deliberative Democracy
Democracy is commonly seen as majority rule. But majorities
may vote to ban minority religious and political views or to act
against minority groups of all sorts. If rule of the majority is all
there is to democracy, then democracy is a danger to
minorities.
Liberal democracy may be defined as a social system based
on equal respect for all persons. Equal respect for all persons
entails a right of each to particípate in democratic decision-
making, including a right to have your vote count equally to
others, which is the basis for the democratic principie of
majority rule. But equal respect for all persons also entails
respect for the fundamental rights of each individual, which
constrains what democratic majorities may legitimately do
(Rawls, 1971, 2001). These fundamental rights are typically
specified in democratic constitutions protecting freedoms of
belief and expression, legal equality, and rights to privacy and
due process.
Many theorists of democracy have argued for some versión
of what has been called deliberative democracy (Báchtiger,
Dryzek, Mansbridge, & Warren, 2018b; Niño, 1996). A
deliberative democracy is a form of liberal democracy that
takes as its ideal the achievement of consensus through
rational processes of argumentation. If we think of democracy
as rooted in respect for persons, deliberative democracy is
equally rooted in respect for reasons. Respect for reasons, and
associated processes of argumentation, serve to generate
better decisions and are thus in the best interest of the social
system. By recognizing persons as rational agents, moreover,
respect for reasons furthers respect for persons. To respect a
person is not just to allow that person a vote but to seriously
consider his or her beliefs and reasons.
Ideally, deliberation leads to consensus. When it fails to do
so, majority rule remains the basis for making decisions that
must be made. Even when a decisión must be made, it ideally
remains clear that the majority could be wrong. Ideally a
consensus remains that the process of argumentation followed,
¡f necessary, by a vote, is the rational ideal, and that we are all
best off in the long run ¡f we respect that process even when we
disagree with particular outcomes.
Deliberative democracy rests on a conception of persons as
rational agents. The conception of persons as rational agents,
like the corresponding conception of societies as deliberative
democracies, is an idealization. People and societies routinely
fall far short of rational norms. Nonetheless, I argüe in Chapter
6 that people are sufficiently rational for deliberative democracy
to be the ideal form of collective governance. Acknowledging
systematic and pervasive human irrationality, I suggest three
reasons for cautious optimism. First, as seen in Chapters 2-4,
people are not as utterly irrational as some have claimed. On
the contrary, adults at their best are rational in ways beyond the
competence of children, and even children show increasing
rationality over the course of childhood. Second, as seen in
Chapter 5, argumentation under the right circumstances
enhances reasoning. Deliberative democracies draw in múltiple
ways on the rational powers of argumentation to operate more
rationally than the individuáis who compose them. Finally, as
discussed in Chapter 8, democratic societies can promote
rationality, and thus democracy, through education.

Education for Rationality and Democracy

Education potentially serves many purposes. Two widely
accepted purposes for education in democratic societies are
the promotion of rationality and democracy. We can foster both
through education aimed at the promotion of (1) the rational
agency of individual students; (2) the social competencies that
enable collaborative reasoning; and (3) the civic
understandings and valúes that enable deliberative democracy.
Education may also undermine rationality and democracy,
and often does, through indoctrination and censorship. In
Chapter 8 I discuss both the problems of actual schools at all
levels of education and the ideal of an educational system
aimed at the promotion of rationality and democracy.

Intellectual Freedom
A recurring theme of the book is that rationality requires
intellectual freedom. This includes freedoms of belief,
expression, discussion, inquiry, and access to information and
ideas. Intellectual freedom is crucial to rational argumentation
(Chapter 5), deliberative democracy and rational institutions
(Chapter 6), developmental progress (Chapter 7), and
academic work at all levels of education (Chapter 8).
Intellectual freedom can be justified on at least three
grounds. First, intellectual freedom is a fundamental right of
individual persons as it respects their rational agency, which is
what defines them as persons. Second, intellectual freedom
serves the pursuit of truth by enabling collaborative reasoning.
Third, intellectual freedom is necessary for democracy,
especially deliberative democracy. But intellectual freedom is
not just the absence of government censorship. Intellectual
freedom, broadly defined as freedom to engage the intellect, is
presented throughout much of the book as the basis for
reasoning, rationality, argumentation, democracy, development,
and education.

Conclusión
The remaining eight chapters are meant to be read in order but
don’t have to be. This first chapter provided an overview of the
argument and an introduction to the central concepts of the
book. That should suffice so readers with particular interests
may sample from the remaining chapters as they please.
Readers interested in individual cognition and development
may wish to read Chapters 2, 3, and 4 most closely, and then
proceed as their interests direct. Readers most interested in
social cognition, democracy, or education may want to skim or
skip Chapters 2 -4 to focus on later chapters. Star Trek fans are
free to go right to Chapter 9.
2
DEVELOPMENT OF LOGICAL
REASONING

There is a long history of psychological controversy regarding

the origins, nature, and development of logic, and its relation to
thinking and reasoning (Braine & O’Brien, 1998; Evans, 2007,
2017; Henle, 1962; Inhelder & Piaget, 1955/1958, 1959/1964;
Johnson-Laird, 2006; Johnson-Laird & Byrne, 1991; Langer,
1980, 1986; Mercier & Sperber, 2017, 2019; Markovits, 2014,
2020; Moshman, 1990, 2004b; 2015; 2018; Noveck, 2019;
Overton, 1990; Piaget, 1928/1972, 1967/1971; Ricco, 2015;
Ricco & Overton, 2011; Rips, 1994; Smith, 1993; Wason &
Johnson-Laird, 1972). In this chapter I present theories and
research concerning the early emergence of logic,
developmental progress in logical reasoning, and associated
developmental progress in knowledge about the nature of logic.
I then turn to the extensive literature on logical reasoning in
adults, showing that it is best understood in a developmental
context. Finally, I raise questions about the scope of logic in
relation to reasoning, leading to Chapter 3, which considers
reasoning that goes beyond logic.

Early Logic
Logic, one might think, is the culmination of the development of
reasoning. But one would be wrong about that. Logic, it
appears, is where it all begins.
Consider this deduction: If an object is here or there, and it is
not here, it is there. A logician might cali that a disjunctive
inference, but you don’t need to be a logician to act on a
sensorimotor understanding that if something is here or there,
and not here, look there. In fact, you don’t even need language.
Disjunctive logic can be observed in the behavior of infants still
in the first year of life (Cesana-Arlotti et al., 2018; Piaget,
1937/1954).

The Origins of Logic

The most systematic and convincing answer to the question of
the origin of logic, in my view, comes from Jean Piaget’s
(1967/1971) classic Biology and Knowledge, which addressed
the fundamental question of how a living organism could have
knowledge, especially what he called logico-mathematical
knowledge, which concerns matters of logical necessity. Living
things, Piaget argued, are, most fundamentally, self-regulating
systems, and logic is intrinsic to self-regulation at physiological,
behavioral, and cognitive levels. Among their many logical
aspects, self-regulating systems include structures of inclusión
(classification) and structures of ordering (seriation).
With regard to inclusión, after detailed analysis of múltiple
examples, Piaget (1967/1971) concluded:

Thus, the classificatory function seems to be found in

every organization structure, and this fact constitutes a
remarkable structural isomorphism between biological
and cognitive organizations. Of course, we are not talking
of the same kinds of classification; sometimes the
inclusions of the subclasses or substructures within
classes or structures are, as it were, incorporated in a
material organization . . ., sometimes they are an
immanent part of a functioning . . ., [and] sometimes they
arise out of it.
(pp. 163-164)

Similarly, concluding a section on structures of ordering, Piaget

(1967/1971) wrote:
 To sum up, order structures do seem at the outset (from
the DNA stage) to be inherent in every biological
organizaron and in its functioning. At the other end, order
structures are produced by thought, but a thought which
also is ordered in its functioning. Between these two
extremes are found all the intermediary stages, and the
parallelism between these and inclusión structures could
be described in greater detail. Here again we are
confronted with a fundamental isomorphism between
biological and cognitive structures.
(p. 166)

Transformaron of biological (including behavioral and

cognitive) systems tends to maintain or enhance their
equilibrium. Logic is implicit in biological systems, including
their psychological functioning, and becomes increasingly
explicit at more advanced levels of equilibrium. In Piaget’s
words, “thought starts in structures that are immanent to the
living organization, but by reconstructing them at its own level it
extends and enriches them in endless ways” (p. 164).
McCulloch and Pitts (1943) published a classic analysis of
the logic inherent in neural networks that was favorably cited by
Piaget (1967/1971) as an example, at the level of nervous
functioning, of his thesis that logic can be found in biological
structures at all levels. McCulloch and Pitts (1943) wrote:

Because of the “all-or-none” character of nervous activity,

neural events and the relations among them can be
treated by means of propositional logic. It is found that the
behavior of every net can be described in these terms,
with the addition of more complicated logical means for
nets containing circles; and that for any logical expression
satisfying certain conditions, one can find a net behaving
in the fashion it describes.
(p. 115)

But Piaget (1967/1971) cautioned:

 The first question to ask is whether this is indeed a case
of “logic” or of a cognitive mechanism. One can certainly
speak of some kind of logic, but only on condition that a
careful distinction be made . . . between the structures
inherent in a function, which thus intervene as factors in
its internal mechanism, and the structures produced by
the function, which inaugúrate functioning of a higher
order, that is, “behavior.”
(p. 222)

In other words, even if there is a logic intrinsic to the

functioning of the nervous system, this does not mean an
organism is making logical inferences, much less engaging in
logical reasoning. To understand the transition from the logic
implicit in all biological functioning to the increasingly explicit
logic of human childhood, we must consider the sensorimotor
logic of infancy.

Sensorimotor Logic
Piaget (1936/1963) conducted longitudinal observations of
each of his three children over the course of infancy, resulting
in a detailed account of the development of what he called
“sensorimotor” intelligence, beginning with the elementary
reflexes of newborn infants and culminating in the emergence
of what he called “representational” intelligence in the latter part
of the second year of life. Much of the analysis focused on how
infants devise means to serve their ends. “Deduction,” Piaget
wrote, “appears at its beginnings as the direct extensión of
earlier mechanisms of assimilation and accommodation, but on
a plañe which begins to become differentiated from direct
perception and action” (p. 349). Thus “sensorimotor deduction”
(p. 347), involving observable trial and error, is transformed in
the second year of life into “mental deduction” (p. 348), which
requires representational intelligence. In related work, Piaget
(1937/1954) looked at how infants, through the coordination of
their increasingly intelligent sensorimotor action-schemes,
construct increasingly objective sensorimotor conceptions of
space, time, objects, and causality. Such conceptions include
matters of implicit logic, such as the reversibility of
displacements in space.
The most thorough and systematic investigation of infant
logic can be found in a series of two volumes by Langer (1980,
1986). The study involved 12 infants at each of the ages 6, 8,
10, and 12 months (reported in the first volume) and 12 older
infants/toddlers at each of 15, 18, 21, and 24 months (reported
in the second volume). The sample was mostly cross-sectional
but included a longitudinal aspect in that some of the 10-month-
olds had previously been assessed at 8 months. Over múltiple
triáis, each participant was presented with a series of carefully
devised combinations of a diverse set of objects, varying in
arrangement and subject to a variety of manipulations, with
interventions by the experimenter on some triáis. Children’s
videotaped actions were analyzed with respect to logical
considerations, such as classification, ordering, mapping,
equivalence, replacement, substitution, commutativity,
combinativity, correlation, negation, reversibility, and exchange.
The results showed ongoing progress in logic throughout the
period of 6-24 months.
Langer’s interpretation of his results is generally consistent
with Piaget’s account of sensorimotor development but more
directly focused on specific matters of logic. On the
fundamental issue of the origin of logic in relation to language,
the substantial logic already seen in children of 12-15 months
strongly supports Piaget’s (1967/1971) claim that “logico-
mathematical structures are extracted from the general
coordination of actions long before they make any use of
language, either natural or artificial” (p. 181).
Critics of Piaget’s account of infant development have
argued that he underestimated the extent of innate knowledge
and/or the rate of early development. Such claims generally
refer to new evidence that competencies observed by Piaget
appear to be present even earlier than he acknowledged
(Gopnik, 2009). But virtually no one has argued that Piaget
overestimated infant competence. The consensus among
developmentalists is that infants of any age are at least as
logical as Piaget thought.

Preschool Inference
Though logic can be seen prior to language, there is no doubt
that the representational intelligence and emerging language of
2-year-olds have major implications for the nature and scope of
logic, and for our ability to study it. Scholnick and Wing (1991)
analyzed uses of the word “if in conversations involving 4-year-
olds and adults (mostly their parents and teachers). They found
that the 39 4-year-olds (as well as the adults) used “if
commonly, in most cases in sentences that “resembled the first
premise of a conditional syllogism” (1991, p. 249). The children
made inferences that were classified as modus ponens (of the
form Ifp then q; p; therefore q) and modus tollens (of the form If
p then q; not-q; therefore not-p), both of which are consistent
with the logic of conditional relations. They also made
biconditional inferences, which involve interpreting If p then q
symmetrically to entail Ifq then p. There is room for considering
how much logic should be read into such isolated inferences,
but it seems clear that preschool children make logical
inferences at least in the sense that they often make inferences
consistent with rules of logic.
In closely related research, longitudinal analyses following
three preschool children from ages 2 to 5 years (Scholnick &
Wing, 1992) showed early progress from (1) comprehending
adult use of if to (2) making inferences from adult if sentences
to (3) child production of if. “Within 6 months of speaking their
first if," Scholnick and Wing (1992) concluded, “children
produced ifs at the same rate and in the same forms as adults”
(p. 1). Further longitudinal analyses including two additional
children (for a total of five) showed, “By the time the children
produced their own ifs, their inferences resembled adults.”
(Scholnick & Wing, 1995, p. 319).
 Another line of research on preschool logical inference
concerns transitivity. Transitive inference involves the logic of
ordering (seriation). If A is larger than B, and B is larger than C,
we can conclude that A is larger than C. But even without
making a transitive inference, a child who is told that A is larger
than B and B is larger than C might conclude A is larger than C
simply because she has heard about A being large (in the first
premise) and C not-large (in the second premise). Bryant and
Trabasso (1971), in a study of 4-year-olds, provided a way to
address this problem. In what has become a standard
methodology, children learn (for example) that A is larger than
B, B is larger than C, C is larger than D, and D is larger than E.
They are then asked the relative size of B and D, which have
not been compared directly. In this expanded format, the child
has learned about B being both not-large (first premise) and
large (second premise) and about D being both not-large (third
premise) and large (fourth premise), so a conclusión that B is
larger than D cannot be based on a categorical distinction
between things that are or are not large. The children’s success
on this task was attributed to genuine transitive inference.
Going beyond human cognition, researchers using
nonverbal variations on Bryant and Trabasso’s (1971)
methodology have found what many deem evidence for
transitive inference in a variety of mammals and birds, and
even in fish and insects (Tibbetts, Agudelo, Pandit, & Riojas,
2019), though skeptics have proposed and provided evidence
for alternative interpretations (Zentall, Peng, & Miles, 2019).
Regardless of what inferences we attribute to various species,
however, it seems clear that preschool children make logical
inferences, including transitive inferences, by the age of 4
years.
Braine & O’Brien (1998; see also Rips, 1994) developed a
theory of “mental logic” that helps explain both preschool logical
competence and subsequent development. The theory has
three components. The first is a set of elementary “inference
schemas” such as If p then q; p; therefore q (modus ponens)
and p or q; not-p; therefore q (disjunction) that are assumed to
be available to everyone beginning in early childhood. The
second component is a “reasoning program” that applies the
schemas in lines of reasoning. This includes a relatively
effortless “direct-reasoning routine” that is also assumed to be
universal beginning in early childhood. The reasoning program
also includes some more effortful “indirect-reasoning strategies”
that develop later, if at all. Finally, the third component of
mental-logic theory involves pragmatic considerations of
meaning, comprehension, and conversational expectations.
Thus in Braine and O’Brien’s (1998) mental-logic theory,
which is supported by substantial research, the inference
schemas and direct-reasoning routine together comprise a
basic mental logic. Inferences that can be made on the basis of
this logic are routinely made beginning in the preschool years,
whereas other inferences, which require more effort and
control, are seen less routinely and only later in development, if
at all.

Progress in Logical Reasoning

As we have seen, human development is rooted in the implicit
logic of self-regulatory systems and proceeds through the
sensorimotor logic of infancy to the verbal inferences of
preschool children. What, one may ask, remains to develop?

Knowledge of Inference
Perhaps the most important initial development marking
progress beyond the preschool years is emerging knowledge of
inference. Although preschool children routinely make
inferences, it appears they do not know they are making
inferences, do not attribute inferences to others, and fail to
recognize inference as a source of knowledge for themselves
and others. Beginning at age 6 years, however, children are
aware that they and others make inferences and show
increasing knowledge about the nature and role of inference
(Miller, 2012; Moshman, 2015; Pillow, 2012; Sodian & Wimmer,
1987).
Knowledge about inference includes knowledge about the
logical properties of inferences, a matter of metalogical
understanding. Metalogical understanding concerns the
necessary truth of logical knowledge and the special kind of
justification provided by logical proofs. We will consider
metalogical understanding in Chapter 4 as part of epistemic
cognition, people’s knowledge about the justification and truth
of beliefs and claims. Metacognition, including epistemic
cognition, is no less important than logic for rational agency and
thus democracy.

Classification and Seríation

In a series of studies with múltiple collaborators, Inhelder and
Piaget (1959/1964) interviewed over 2,000 children ranging in
age from 3 to 9 years as they examined and manipulated
assorted objects that could be classified and ordered in a
variety of ways. The objects, presentaron, and questions were
designed to elicit behavior and explanations from the children
relevant to class inclusión, hierarchical classification,
complementary classes, multiplicative classification (matrices),
seriation, and múltiple seriation.
The studies showed a regular progression from preschool
“graphic collections” and isolated inferences to behavior and
explanations that show children understand and apply the logic
of classification and seriation beginning at age 7 or 8 years.
Understanding the logic of hierarchical classification, for
example, includes understanding that a class may be divided
into subclasses and logically must have at least as many
members as any subclass within it. Inhelder and Piaget
interpreted their results as part of Piaget’s general theory of the
emergence of what he called “concrete operations,”
emphasizing the role of logical structures in providing for a level
of logical reversibility and equilibrium that is typical by age 7 or
8 years but not seen in preschool children.
 Critics of Piaget’s account have questioned his structural
analysis of logical reasoning and devised tasks to show that
logical competencies associated by Piaget with concrete
operations are already available in the preschool years. In a
classic early example already noted, Bryant and Trabasso
(1971) interpreted their evidence for transitive inference in 4-
year-olds as a challenge to Piaget’s theory that the logic of
seriation is not understood until age 7 or 8 years. Of course
there is more to the logic of seriation than making a transitive
inference, which raises complex theoretical issues. For present
purposes, however, what is notable is that in childhood, as in
infancy, critiques of Piaget have overwhelmingly focused on
showing logical competence earlier than he acknowledged.
Thus there is a long-standing consensus that 8-year-old
children are at least as competent as Piaget’s evidence
indicated.

Hypothetico-Deductive Reasoning
The evidence seems clear that children are capable of logical
reasoning by the age of 7 or 8 years (Inhelder & Piaget,
1959/1964)— if not even earlier, as proposed by many critics of
Piaget’s stage theory. What, then, if anything, remains to
develop? Obviously, we may leam various new facts and skills
useful in particular environments, and we may get more
efficient at processing some kinds of information. But do we
continué to make progress toward more advanced forms of
reasoning or higher levels of rationality?
Philosophers since Plato and Aristotle have theorized about
the nature of advanced reasoning and logic. In the early days of
psychology, James Mark Baldwin (1895) postulated a “hyper-
logical” stage of mental development marked by the emergence
of puré thought, form without content. Piaget (1928/1972) took
up the question of formal reasoning in some of his earliest
work. By formal reasoning, Piaget (1928/1972) meant “formal
deduction,” which
 consists in drawing conclusions, not from a fact given in
¡mmediate observation, ñor from a judgment which one
holds to be true without any qualifications (and thus
incorporates into reality such as one conceives it), but in a
judgment which one simply assumes, i.e. which one
admits without believing it, just to see what it will lead to.
(P■69)

Piaget’s (1928/1972) research showed that formal or

“hypothetico-deductive” (p. 69) reasoning begins to be seen at
age 11 or 12 years. The deductions of younger children may be
logical, but such deduction “bears only upon the beliefs which
the child has adopted himself (p. 67). Formal reasoning, in
Piaget’s conception, enables the adolescent to reason strictly
about hypotheses in a constructed realm of possibility that is
explicitly distinguished from empirical reality. “To be formal,” he
proposed, “deduction must detach itself from reality and take up
its stand upon the plañe of the purely possible, which is by
definition the domain of hypothesis” (p. 71).
As Piaget subsequently developed his structural theory of
developmental stages, his conception of formal reasoning
became his theory of formal operations. Inhelder and Piaget
(1955/1958) provided the definitive statement of that theory
along with supporting evidence from a series of 15 studies in
which over 1,500 children and adolescents, ranging in age from
4 through 16 years, were interviewed about their responses to
a variety of scientific tasks, involving apparatus such as a
pendulum or balance beam, that required logical reasoning
about possibilities and hypotheses. The results showed isolated
inferences in the preschool years and concrete operational
reasoning reflecting the logic of classification and seriation
beginning about age 7 or 8 years, consistent with earlier and
later work (Inhelder and Piaget, 1959/1964). Also consistent
with earlier work (Piaget, 1928/1972), formal reasoning began
to be seen at about age 11 or 12 years.
Inhelder and Piaget (1955/1958) continued to define formal
reasoning as hypothetico-deductive reasoning—that is, as
reasoning that begins with at least one premise that is purely
hypothetical, or even known to be false, and proceeds from
there through strictly logical deductions to a conclusión that
follows necessarily from the premises, with full awareness that
if a premise is false the conclusión may be false as well.
Children can deduce what follows from premises they accept.
Hypothetico-deductive reasoning extends the range of
deductive reasoning to the realm of possibility, enabling
exploration of the logical consequences of premises one does
not accept. Such ability is crucial to scientific hypothesis
testing, which reserves judgment on the truth of theories even
as it deduces their predictions. Hypothetico-deductive
reasoning continúes to be widely studied not only in the
transition to adolescence (Markovits, 2014, 2020; Moshman
2011a) but also in the reasoning of adults (Evans, 2007).
More broadly, formal reasoning construes reality as a subset
of possibilities. Hypothetico-deductive reasoning enables the
rigorous and systematic elaboration of logical possibilities,
which in turn enables one to reconsider reality within a larger
hypothetical context. Even young children consider possibilities,
as can be seen in the imaginative play of preschoolers. But
children’s possibilities are relatively direct extensions of reality.
Formal operations transcend concrete operations in that
possibilities are now formulated in such a way that reality can
be understood and evaluated as the realization of a particular
possibility. This enabled better hypothesis testing on the
scientific tasks of Inhelder and Piaget (1955/1958), but the
implications, they noted, extend far beyond that. With the
attainment of formal operations, all sorts of accepted ideologies
and conventions may be seen as open to critique and
reconsideration on the basis of newly constructed possibilities.
Finally, formal operations are second-order operations in
that they operate on concrete (first-order) operations, such as
classification and seriation, to create logics that recognize
classes of classes and relations of relations. Formal operations,
in Piaget’s theory, involve the “complete combinatorial system”
of propositional logic, which represents a more advanced and
stable form of equilibrium than the múltiple concrete logics of
classes and relations. Thus there is a metacognitive aspect to
Piaget’s conception of formal operations as second-order
operations that derives from his metacognitive conception of
development as a reflective process of constructing more
equilibrated structures. We will return to these matters in
addressing metalogical understanding as an aspect of
epistemic cognition in Chapter 4 and reflection as a
developmental process in Chapter 7.
Psychologists have been skeptical of the more technical
aspects of Piaget’s theoretical account of formal operations,
which are widely seen as unclear and difficult to test.
Nevertheless, developmental research has strongly supported
Inhelder and Piaget’s (1955/1958) empirical observation that
hypothetico-deductive reasoning emerges at about age 11 or
12 years (Moshman, 1990, 2011a; Moshman & Franks, 1986).
Recent research has been aimed at providing a more detailed
picture of the transition from concrete to formal reasoning by
investigating progress across the preadolescent years in the
elaboration and manipulation of mental models of possible
states of affairs (Markovits, 2014, 2020).
The major challenge to Piaget’s theory of formal operations
has come from research on adult reasoning, which seems to
show much less rationality than does Piaget’s research. The
study of child development since the 1960s has consisted in
large part of efforts to show that infants and children are more
rational than Piaget claimed. The study of adult reasoning since
the 1960s, in contrast, has focused largely on demonstrating
systematic deviations from logical, statistical, and other norms.
Such deviations have been interpreted by many to show that
human irrationality is not just a matter of the occasional mistake
but a matter of deep flaws in how we process information.

Logical Reasoning in Adulthood

I now consider the literature of adult logical reasoning, first
historically and then in the context of the developmental
literature on logical reasoning. Does the theory of formal
operations overestimate human rationality? Does the literature
of adult reasoning underestimate it? As we will see, a
combination of dual process theory and a developmental
perspective enables us to reconcile the seemingly divergent
findings. I conclude that logical reasoning does indeed develop,
but the adult reasoning literature helps put this into context by
addressing the nature and limitations of advanced logical
functioning.

The View from Cognitive Psychology

Psychological research on logical reasoning can be traced to a
series of studies published in Germán by Stórring beginning in
1908 (cited in Woodworth & Sells, 1935) and one in English by
Wilkins (1928). Drawing on Aristotelian logic, these studies
presented abstract categorical syllogisms, each consisting of
two premises and a conclusión. Each premise and each
conclusión was stated with letters in one of four possible forms:
All X are Y, No X are Y, Some X are Y, and Some X are not Y.
Some of the syllogisms were valid and others were not. For
example, the syllogism All A are C; All B are C; therefore,
Some B are A is not valid because the conclusión, although
possible, does not follow necessarily from the premises.
Results showed high rates of error in evaluating the validity of
abstract syllogisms, including systematic errors associated with
particular patterns.
Woodworth and Sells (1935) provided a theoretical analysis
of results from these early studies, suggesting that, despite
some evidence of logic, participants’ identification of valid
inferences was compromised by three factors. First, there was
a tendency to interpret “some” as “some but not all” (which,
depending on context, is often a reasonable interpretation in
ordinary language) rather than as “at least some and perhaps
all” (which is how the term is used in formal logic). Second,
there is a tendency toward “caution” that rendered participants
wary of conclusions involving “AN” statements. Third,
participants’ judgments of validity were influenced by whether
the conclusión matched the premises in “atmosphere,” which
involved being affirmative (“AN”) vs. negative (“No”) and/or
being universal (“AN” or “No”) vs. particular (“Some are” or
“Some are not”). Four decades before the rise of the heuristics
and biases literature (introduced in Chapter 1), the tendency
toward “caution” may be regarded as a bias, and the
“atmosphere effect” may be considered an early example of
what would come to be called a heuristic. Woodworth and Sells
(1935) also provided new evidence on syllogistic reasoning
(later published in greater detail by Sells, 1936) in support of
their analysis.
Chapman and Chapman (1959) questioned the evidence for
the postulated “atmosphere effect” and provided new evidence
from research in which they presented sets of premises; in
each case, they asked participants to choose their own
conclusión from a set of choices (rather than providing
complete arguments, as in all the earlier research). They found
that neither “atmosphere” ñor a combined consideration of
“atmosphere” and “caution” (proposed by Sells, 1936) could
account for their pattern of results. They proposed that many or
most of the seemingly erroneous responses could be
accounted for on the basis of reasonable interpretations of
language and reasonable construals of task expectations,
consistent with the first explanation from Woodworth and Sells
(1935).
Mary Henle (1962) provided a classic philosophical and
empirical defense of the role of logic in thinking. After reviewing
philosophical analyses and psychological research, she used
examples from her own research to ¡Ilústrate cases where
apparent errors in logical reasoning were due not to errors in
the reasoning process itself but to other factors. In some cases,
the problem was “failure to grasp or accept the logical task” (p.
370) by focusing on the truth of conclusions rather than
whether they follow from the premises. Other cases involved
“restatement of a premise or conclusión so that the intended
meaning is changed” (p. 371). Some involved omitting a
premise or slipping in additional premises. Henle concluded
that, if one considers the premises that participants were
actually starting from and the task as they understood it, there
was no evidence for faulty reasoning. On the contrary, “the
laws of logic . . . are widely discernible in the thinking process”
(p. 373). She noted that logical inference “need not be explicit”
(p. 373) but may involve “an implicit logical process” (p. 374).
Henle provided some examples of rapid automatic inferences in
daily life, such as “I do not want to wear the same dress two
days in succession. I wore this dress yesterday; so I do not
want to wear this dress today” (p. 374). She added, “if people
were unable to reason logically, so that each arrived at different
conclusions from the same premises, it is difficult to see how
they could understand each other, follow one another’s
thinking, reach common decisions, and work together” (p. 374).
The modern era of research on adult reasoning is commonly
traced to the early work of Peter Wason in the 1960s (reviewed
by Wason & Johnson-Laird, 1972). Wason expanded the focus
of research beyond the categorical syllogisms of Aristotle to
modern propositional logic, focusing particularly on conditional
reasoning, which involves reasoning from a premise of the form
“If X then Y.” His “selection task” is likely the most famous task
in the extensive literature of logical reasoning. In the standard
versión, you are presented with a set of four cards, each with a
letter on one side and a number on the other, and a hypothesis
about those cards. The hypothesis is to be tested by turning
one or more of the four cards, but no more than necessary. For
example, the hypothesis might be, “If a card has a vowel on
one side, then it has an even number on the other side” and the
corresponding cards would show E, K, 4, and 7 on the visible
sides. Which of these cards must be turned to determine
conclusively whether that hypothesis is true or false? The two
most common responses are to choose just the E card or the E
and the 4. The logically correct response, however, is E and 7.
The E card, which has a vowel on the visible side, would falsify
the hypothesis if it had an odd number on the other side. The 7
card, which has an odd number on the visible side, would falsify
the hypothesis if it had a vowel on the other side. Both of these
cards must be turned because either could falsify the
hypothesis. No other card needs to be turned because no other
card could falsify the hypothesis, regardless of what is on the
other side. If the hypothesis is not falsified by what appears on
the other side of the E and 7 cards, then it is true.
Research on logical tasks has continued actively ever since
and has generated a variety of theories of logical reasoning
aimed at explaining both successes and failures. Two major
sorts of theories provide general accounts of basic logical
processes: mental logic theories and mental model theories.
Mental logic theories (Braine & O’Brien, 1998; Rips, 1994),
which postúlate rules of logic, were discussed earlier in relation
to logical inference in preschool children. Mental model theories
(Johnson-Laird, 1983, 2006; Johnson-Laird & Byrne, 1991)
propose that logical reasoning involves the elaboration and
manipulation of mental models of possible states of affairs,
rather than the application of rules, a conception that (as noted
earlier) has been applied to developmental research by
Markovits (2014, 2020). There is substantial evidence for both
sorts of theories, suggesting that both logical rules and mental
models play a role in logical reasoning.
Beginning in the 1970s, the literature on errors in logical
reasoning was increasingly supplemented by research on what
carne to be called “heuristics and biases” (Kahneman, 2011),
as discussed in Chapter 1. This was initially a rather distinct
literature that was seen by many as demonstrating that people
do not just make logical mistakes but rely on a variety of
heuristics and biases rather than engaging in logical reasoning
at all. The trend in the last three decades of the 20th century
was toward conceptions of human beings as deeply and
fundamentally illogical and, more generally, irrational.
 But the evidence for human logical abilities, including the
developmental research reviewed earlier, is multifaceted and
difficult to deny. The major resolution to this problem has been
within the context of dual processing. Dual process theories, as
discussed in Chapter 1, have been the dominant approach
since the turn of the century in explaining human reasoning and
rationality. The standard dual process view is that we can be
logical and often are, but that we fall short so often because our
logical processes are not our only processes. The usual
assumption here is that logic is part of System 2 deliberation
and reflection, which does not always override System 1
automatic processes and intuitions. But that assumption is
contrary to the views of both Piaget and Henle, as discussed
above, and has been seriously questioned within 21st century
cognitive psychology.

Is Logic a System 2 Ability?

In recent years, dual process researchers have questioned the
usual assumption that logic is a System 2 ability. Research
methodologies that impose strict time limits or extraneous
burdens on working memory, which have been shown to
dampen System 2 responses, have been found not to have the
expected dampening effect on logic. Results indícate instead
that people commonly make System 1 logical inferences. A
major challenge for current dual process theorizing is to explain
the origin, nature, and role of what dual process researchers
cali “implicit logic” (Trippas & Handley, 2018), “fast logic” (De
Neys, 2018), and “logical intuitions” (De Neys, 2012;
Pennycook, 2018; Thompson & Newman, 2018).
As seen earlier in this chapter, developmental research
shows automatic and intuitive forms of logic in the early years
of life. This is fully consistent with the independent conclusions
in 21 st century cognitive psychology that logic is a System 1
ability. What distinguishes System 2 is not logic but reasoning,
which is relatively general and requires intensive use of working
memory. But development is not the replacement of System 1
with System 2. Rather, combining developmental and dual
process insights, we should see System 2 as supplementing
System 1. The development of reasoning is, at least in part, a
matter of comprehending and gaining control over systems of
logic that can be traced back to early childhood and, even
before that, to basic modes of biological functioning.

From a Developmental Perspective

The study of adult reasoning has long been quite distinct from
the study of the development of reasoning. Adult reasoning is
studied by cognitive psychologists, whereas the development of
reasoning is studied by developmental psychologists, who are
generally most concerned with childhood and adolescence. The
study of the development of reasoning is generally seen as part
of the field of cognitive development, which could be seen as
the overlap of cognitive psychology and developmental
psychology but is more commonly understood as a subdivisión
of developmental psychology. Research on adult reasoning is
generally published in journals of cognitive psychology, such as
Thinking & Reasoning, where clear evidence of systematic
deviation from logical, statistical, or other norms is deemed
theoretically interesting. Research on the development of
reasoning is generally published in developmental journals,
where the focus is often on showing clear evidence of some
competence at an earlier age than anyone has shown before.
Research on adult reasoning generally ignores development
(for a recent example, see Evans, 2017), and researchers who
study the development of reasoning often fail to address the
adult reasoning literature (with notable counterexamples in the
study of adolescence, such as Klaczynski, 2000; Markovits,
2013). The result of the communication gap between cognitive
and developmental psychology is an apparent counter-
developmental trend from preschool logic, as demonstrated in
the developmental literature since the 1970s, to adult
irrationality, as demonstrated in the cognitive literature since
the 1960s (Moshman, 2004b).
 Considering the literatura on adult logical reasoning from a
developmental perspective yields at least four important
insights. First, as already discussed, there is a strong basis in
early childhood research and biological theorizing (Piaget,
1967/1971) for the sorts of implicit logic recently “discovered” in
dual process research. Second, a developmental perspective
shows the importance of logic, which has been doubted by
theorists such as Mercier and Sperber (2017; for discussion,
see Mercier & Sperber, 2019; Moshman, 2018; Noveck, 2019).
Some of the logic tasks in the literature of adult reasoning may
be artificial or peculiar, but logic is fundamental across the
lifespan.
The third developmental insight involves standards for
comparison. Lacking a developmental perspective, it is natural
to look at adult rationality in comparison to normative ideáis
from philosophy or by comparing adults to each other, leading
to the conclusión that adults fall far short of logical and other
ideáis and some are even more irrational than others. This is
an important part of the picture, but a developmental
perspective complements the view from cognitive psychology.
Adult shortcomings and individual differences in rationality must
be seen in the context of the developmental reality that all
normal humans beyond the age of 12 or 13 years show forms
of reasoning and levels of rationality virtually never seen in 8-
year-olds, who in turn show forms of reasoning and levels of
rationality never seen in 4-year-olds, who in turn show rational
competencies far beyond the sensorimotor logic already seen
in infancy.
Finally, even among adults, it is important to distinguish
developmental differences from other individual differences.
Not all individual differences are developmental, but some can
only be understood by recognizing that some individuáis are
operating at a higher developmental level than others, and that
developmental progress remains possible, though not
inevitable, long beyond childhood. But development is not
simply a matter of avoiding mistakes or learning particular rules
or strategies. A clear understanding of what (potentially)
develops in adolescence and adulthood and how such
development takes place (the topic of Chapter 7) provides the
basis for educational efforts to promote such development
(Chapter 8).

Mathematical Reasoning
Mathematical reasoning is closely related to logical reasoning
but forms a major category of its own. It ranges from basic
conceptions of numbers to complex statistical reasoning.

The Concept of Number

We generally take numbers for granted and think of learning to
count as a simple matter of repeating and memorizing a
sequence of number ñames in whatever language is spoken by
the adults around you. Epistemologically, however, there are
complex questions to be asked about what a number actually is
and how it is possible for the human mind to know whatever it
is we know about numbers (McCulloch, 1961).
Piaget (1941/1965a, 1967/1971) proposed that a number is,
both epistemologically and developmentally, a synthesis of
classification and seriation. The number 5, for example, is the
(abstract) class of all sets of 5 things. If I have 5 horses, 5
dogs, and 4 cats, the first two sets fall in the same numerical
category (sets of 5) and the third does not, for reasons having
nothing to do with the characteristics of horses, dogs, or cats.
But 5 is not just a different category from 4 or 6. The number 5
is defined in part by its being simultaneously larger than 4 and
smaller than 6 in the series of integers, with all that logically
entails. Mathematical knowledge, like logical knowledge,
involves necessary truths and a conception of justification as
proof. Logico-mathematical knowledge thus differs from
empirical knowledge of the physical world, which is contingent
on what the world is like and subject to revisión on the basis of
new data. Only a synthesis of classification and seriation can
account for the logical properties of number and the associated
logic of elementary arithmetic.
Piaget (1941/1965a) conducted a detailed investigaron of
“the construction of number in relation to logical operations,”
hypothesizing “that the construction of number goes hand-in-
hand with the development of logic” (p. viii). Over a series of
ten studies, children ranging in age from 3 to 8 years (mostly 4
to 7) were interviewed about tasks assessing their
comprehension of the logic of elementary mathematics,
including their understanding of conservation of quantity, one-
to-one correspondence, part-whole relations, cardinal and
ordinal properties of number, multiplication of classes and
numbers, equivalence, and reversibility. Piaget concluded:

Our results do, in fact, show that number is organized,

stage after stage, in cióse connection with the gradual
elaboration of systems of inclusions (hierarchy of logical
classes) and systems of asymmetrical relations
(qualitative seriations), the sequence of numbers thus
resulting from an operational synthesis of classification
and seriation. In our view, logical and arithmetical
operations therefore constitute a single system that is
psychologically natural.
(p. viii)

A major alternative to Piaget’s account was provided by

Gelman and Gallistel (1978), who proposed that understanding
of number involves the coordinated application of five principies
of counting: the one-one principie (tag each item with one and
only one unique number ñame); the stable-order principie (don’t
change the order of the numbers); the cardinal principie (the
last number in the count is the total number of items; the
abstraction principie (any set of items, however diverse, can be
counted); and the order-irrelevance principie (the order in which
the items are tagged does not matter). Their research and that
of many others provides a detailed picture of how children
construct the logic of elementary mathematics in the preschool
years, including evidence often interpreted as showing earlier
understanding than Piaget indicated. There appears to be a
consensus, however, that the logic of numbers is increasingly
understood over the course of early childhood, and that
children of any given age are at least as competent
mathematically as Piaget indicated (Feiman, Hartshorne, &
Barner, 2019).

Probabilistic and Statistical Reasoning

Mathematical reasoning of course extends far beyond numbers
and counting (Cauley, 1992; Smith, 2002). Probabilistic
reasoning, for example, enables us to be rigorous about our
uncertainty by coordinating deduction and probability. Piaget
and Inhelder (1951/1975) found developmental progress in
conceptions of chance, with concrete forms of probabilistic
reasoning emerging about age 7 or 8 years and formal
elaboration of possibilities at age 11 or 12 years.
Not surprisingly, developmental researchers have sought
and found evidence for statistical inferences at increasingly
early ages, even in infancy (Sobel & Kirkham, 2007), whereas
the adult cognition literature has shown systematic deviations
from basic probabilistic and statistical norms in adults
(Kahneman, 2011). Once again, we see Piaget’s
developmental account challenged by developmentalists as
underestimating the rationality of young children and by
researchers on adult reasoning as failing to acknowledge the
irrationality of adults. Here too, a developmental dual process
approach resolves the apparent developmental reversal by
showing how statistical inference is increasingly supplemented
over the course of development by statistical reasoning, but
never replaced.
Even in matters of probability, mathematical reasoning is
strictly deductive and thus remains within the logico-
mathematic domain of necessary inference. Can logic go
beyond that? Does reasoning go beyond logic? We will expand
the account of reasoning in the next chapter.
Conclusión
Let us conclude with four questions about logic. Was logic
invented by logicians? Is it a property or outcome of language?
Is it how we think? Is it how we ought to think? The answers, I
suggest, are No, No, Mostly Not, and Sometimes.
Starting with the first question, logicians at least since
Aristotle have been constructing and reconstructing systems of
logic, but even if we give a logician credit for inventing some
system of logic, the invention is not a free creation. It is an
explicit systematization of a logic that was already implicit in an
earlier system. As we have seen, logic existed long before
logicians discovered it.
With regard to the second question, logic is most visible in
language and is commonly seen as a feature or property of
language or as an ability that language makes possible. As we
have seen, however, developmental research shows clear
evidence of logic in prelinguistic children. Research with
nonhuman species, moreover, shows what many take to be
evidence of logical inference in a variety of mammals and birds,
perhaps even in wasps. Language greatly facilitates the
development and use of logic but is not the source of logic.
With regard to the third question, we sometimes think in
accord with rules of logic, but our thinking does not inherently
follow such rules and often deviates systematically from them.
There is an important role for logic in thinking, and especially in
reasoning, but neither thinking ñor reasoning is simply a matter
of following rules of logic.
Turning to the fourth question, maybe we ought to think in
accord with rules of logic, at least when we are reasoning. We
should indeed make logical inferences when they are relevant
and avoid inferences contrary to relevant rules of logic.
Reasoning, however, is more than logic. Beyond logical
reasoning, as we will see in the next chapter, there are other
types of reasoning that cannot be reduced to logic or to each
other. At the very least, in addition to (1) logical reasoning, we
must consider (2) scientific reasoning, which is crucial to
understanding the world; (3) principled reasoning, which is
crucial to morality; and (4) precedent-based reasoning, which is
crucial to social systems.
3
REASONING BEYOND LOGIC

The prototypical example of logic is deduction. If we define

logic strictly as deduction, then we can say a rational agent
engages in logical reasoning when (or to the extent that) the
agent makes a deliberate inference in accord with strict rules
that are understood by the agent to generate a conclusión that
follows necessarily from the premises. Such inferences are
deductive; their deliberate and reflective application, as
discussed in Chapter 2, is deductive reasoning.
Philosophers often distinguish induction from deduction,
some distinguish abduction from induction, and some discuss
informal logic. This raises questions about the scope of logic.
Are there nondeductive systems of logic? What exactly is logic?
Induction is generally defined as reaching a conclusión that
goes beyond what the premises require. Is this logic? Are there
rules of induction, even if the results are not logically
necessary? Abduction is defined as inference to the best
explanation. Is this logic? Are there rules of abduction? Informal
logic may not follow formal rules. Are there other kinds of rules?
As we will see in this chapter, scientific interpretation and
explanation require much more than following rules to reach
necessary conclusions and, thus, go beyond anything like
deductive logic. The same is true for various sorts of moral and
social reasoning. For present purposes, I use the term logic to
refer to strict norms of good reasoning, especially rules of
deductive necessity. This definition acknowledges that logic in
its broader sense may be more than strict deduction.
Even in this broader sense of logic, however, reasoning
includes more than logic. I will propose in this chapter that
scientific, principled, and precedent-based reasoning are
epistemologically and psychologically distinct from logical
reasoning and from each other. This suggests a much richer
picture of human reasoning than the account of logical
reasoning presented in Chapter 2, with a variety of ways people
may achieve or fall short of rational norms.

Scientific Reasoning
We look first at scientific reasoning. As we will see, scientific
reasoning is epistemologically distinct from logical reasoning in
ways that children increasingly come to understand. The
development of scientific reasoning is correspondingly distinct
from the development of logical reasoning but shows a similar
pattern of progress.

Scientific Reasoning as Empírica! and Causal

Reasoning
Outside of logical reasoning, perhaps the most-studied form of
reasoning is scientific reasoning. But what exactly is scientific
reasoning?
For a start, scientific reasoning has long been seen as
including an ability to design unconfounded experiments
(Inhelder & Piaget, 1955/1958). If I want to know which of three
variables is responsible for some effect, I must manipúlate the
variables independently of each other. If I flip three switches
and a light goes on, I don’t know which switch (or switches)
caused the light to go on. If I just wanted the light to go on, no
problem. But if my goal is to understand what causes the light
to go on, I must flip one switch at a time, holding everything
else constant. The quest to understand, rather than just create
some effect, is empirical in its concern for what the world is
really like and causal in its concern for how things work. If this
is all there is to scientific reasoning, however, scientific
reasoning does not appear to be anything more than the
application of logic to understanding the world. Even if it turns
out that the effect of a given switch depends on the position of
another, which complicates the causal logic, this remains a
matter of logic, involving strict rules of inference and a single
correct conclusión.
Somewhat more broadly, scientific reasoning is often
presented as fundamentally a matter of testing hypotheses and
evaluating evidence. This includes designing unconfounded
experiments to test causal predictions, but it goes beyond that.
The four-card “selection task” (Wason & Johnson-Laird, 1972)
presented in the previous chapter can be seen as a study of
hypothesis testing. Suppose we take the conditional If X then Y
to represent a hypothesis that X causes Y. We can test this by
getting further evidence about (1) a case where X is present;
(2) a case where X is not present; (3) a case where Y is
present; and (4) a case where Y is not present. The task is to
determine which of these four sorts of research must be done
to reach a definitive conclusión about the truth or falsity of the
hypothesis. The solution requires (a) the insight that we must
do any research that could turn up a case of X without Y,
because any such case would disconfirm the hypothesis that X
causes Y; and (b) the further insight that we need not do any
other research, because no other case could disconfirm the
hypothesis. Thus, we must do studies 1 and 4 because either
of those could disconfirm the hypothesis. If either of these
studies turns up a case of cause without effect, the hypothesis
is false. If no such disconfirming case shows up, then the
hypothesis is true regardless of what we might have found in
studies 2 or 3, which need not be done because they could not
disconfirm the hypothesis. As we saw in Chapter 2, this is a
notoriously difficult task. However difficult such hypothesis
testing may be psychologically, however, the task remains a
matter of logic with clear right and wrong answers distinguished
on the basis of formal rules without regard for what we mean by
X or Y. Hypothesis testing with a simple hypothesis in a limited
domain of potential data may be mostly or entirely a matter of
logic.
Scientific reasoning, however, is not just controlling
variables, testing hypotheses, and evaluating evidence in
limited environments involving a few switches or a set of four
cards. Science aims to explain the world, which requires an
ongoing process of formulating, testing, and revising theories.
Our theories influence what data we collect and guide our
interpretations of the data. Scientific data do not “speak for
themselves,” and their relevance and import for various
scientific theories may be a matter of reasonable disagreement.
We coordínate theories and data in an ongoing effort to
improve our knowledge (Kuhn, Amsel, & O’Loughlin, 1988).
This includes broad use of relevant background information— in
contrast to logic, which relies on isolating premises from the
world to determine what follows from those premises alone.
Scientific reasoning is rational in its coordination of theories and
data to make progress in causal understanding, but
psychologists and philosophers of Science generally agree that
such progress is not simply a matter of following rules of logic
to a necessary conclusión (Carey & Smith, 1993; Koslowski,
1996, 2013; Koslowski, Marasia, Chelenza, & Dublin, 2008;
Rosen berg & Mclntyre, 2020).
Scientific reasoning is thus empirical and causal (Koslowski,
1996; 2013; Koslowski et al., 2008; Rosenberg & Mclntyre,
2020; Sodian, 2018). It is empirical in that it posits a world
outside itself and seeks to explain that world. This requires
maintaining a distinction between theory and evidence and
seeking evidence about what the world is really like. In addition,
scientific reasoning is generally causal in that its explanations
generally posit causal relations within the world. Science is not
the accumulation of information. It is the formulation of causal
theories and the seeking of evidence to test those theories as
part of an effort to explain some aspect of reality. As Koslowski
et al. (2008) put it in the title of an article: “Information becomes
evidence when an explanation can incorpórate it into a causal
framework.”
 Science, then, may be defined as the empirical study and
theoretical explanation of the world. Correspondingly, scientific
reasoning is reasoning aimed at empirical justification and truth.
This includes the reflective coordination of theories and data for
the purpose of making progress in our causal understanding of
the world.

Epistemologies of Science and Logic

The philosophy of Science is in large part the epistemology of
Science in that its central concern is understanding the nature
of scientific truth and the justification of scientific claims and
methodologies. Philosophers of Science generally agree that
scientific knowledge is not a direct copy of the world, a logically
necessary consequence of available data, or a collection of
information. Nevertheless, Science aims to understand a world
outside itself and is in that sense fundamentally empirical. Data
are not sufficient for science, but without data there is no
Science.
Logic, in contrast, makes no use of data except for premises
accepted without question. It reaches conclusions that must be
so regardless of any data beyond the accepted premises. If you
claim there are more dogs than cats in a room, you should be
able to justify that with evidence about the number of dogs and
cats, whereas if you claim that there are more dogs than
animals, you are clearly wrong regardless of evidence. There
must be at least as many animals as dogs in any room. This
conclusión is not based on long experience counting animals in
a variety of rooms. It is based on understanding the necessary
logic of hierarchical classification. More generally, logical
necessities require no empirical evidence and cannot be
contradicted by evidence because they are not about what the
world happens to be like (Moshman, 1990, 2015; Smith, 1993).
Thus, scientific reasoning is distinct from logical reasoning in its
reliance on data.
In addition to being empirical, scientific reasoning is causal
reasoning. Scientific theories are generally causal theories.
This does not mean everything is reducible to physics. Our
conceptions of causality must reflect the complexity of what we
seek to understand. Biological systems show self-regulatory
forms of causation not seen in physics. Psychological
functioning and social organization further complícate causal
explanation. Dynamic systems manifest complex and circular
forms of causality in which the systems themselves are causal
agents (Witherington, 2011). Whatever the complications of
causality, however— and ignoring, for present purposes, the
subatomic mysteries of quantum mechanics— Science is
fundamentally causal (Moshman, 2015).
Logic, in contrast, is implicative, not causal. In logic, the
conditional If p then q can be understood as telling us that p
implies q. In science, however, the major concern would be
whether p causes q. Thus, scientific reasoning is distinct from
logical reasoning in its postulation of a real world that can be
understood through causal theorizing. Causality plays no role in
logical reasoning but is crucial to scientific reasoning.
The epistemology of science is not just a concern of
philosophers. On the contrary, we all have “intuitive
epistemologies of science,” which “include ideas about the
sources and justification of scientific knowledge, its constructive
and tentative character, its embeddedness into cultural and
historical contexts, the role of theories, observation and
experiments, and the role of rational argument in the
construction of scientific knowledge” (Sodian, 2018, p. 234).
Research in developmental psychology, educational
psychology, and science education shows universal progress in
understanding the “nature of science” over the course of
childhood (Sodian, 2018), with progress possible but much less
universal in adolescence and beyond (Moshman, 2015).
Understanding the nature of science is in large part a matter of
understanding its epistemology, which is crucial to engaging in
scientific reasoning. We will look more generally at epistemic
development in the next chapter.
Development of Scientific Reasoning
Piaget (1936/1963) discussed sensorimotor experimentation in
connection with what he called “tertiary circular reactions.” In
primary circular reactions, first seen about the age of 1 month,
the infant attempts to reproduce a favorable outcome centered
on its own body, such as getting its thumb back in its mouth to
recreate the experience of sucking on it. In secondary circular
reactions, beginning about age 4 months, the infant attempts to
reproduce an outcome involving an external object, but the
object is so tied to the infant’s actions that it is not yet seen as
external. In tertiary circular reactions, in contrast, beginning at
about 12 months, the focus shifts to the object itself and
associated matters of causality and space, resulting in
intentional variations in action to see what happens. In an
example of the transition from secondary to tertiary circular
reactions, Piaget (1936/1963) described how his son, Laurent
(at 10 months, 2 days), discovered the possibility of throwing a
case of soap and letting it fall. As expected in a secondary
circular reaction, however, he was content simply to reproduce
this result because what interested him “was not the objective
phenomenon of the fall— that is to say, the object’s trajectory—
but the very act of letting go” (p. 268). But eight days later,
Piaget saw evidence of a tertiary circular reaction when Laurent
repeatedly dropped a small piece of bread, even breaking off
fragments and letting them drop. In this case,

he pays no attention to the act of letting go whereas he

watches with great interest the body in motion; in
particular, he looks at it for a long time when it has fallen,
and picks it up when he can.
(pp. 268-269)

The following day:

Laurent is lying on his back but nevertheless resumes his

experiments of the day before. He grasps in succession a
celluloid swan, a box, etc., stretches out his arm and lets
 them fall. He distinctly varíes the positions of the fall.
Sometimes he stretches out his arm vertically, sometimes
he holds it obliquely, in front of or behind his eyes, etc.
When the object falls in a new position (for example on
his pillow), he lets it fall two or three times more on the
same place, as though to study the spatial relation; then
he modifies the situation.
(p. 269)

Sensorimotor causality shows a closely related

developmental trend toward objectivity. Over the first year of
life, infants construct causality in relation to the causal impact of
their own actions (Piaget, 1937/1954). As they move into their
second year, “causality will be really objectified and spatialized,
thus becoming detached from the action itself to be
externalized in the universe of perception, free to be applied in
retum to the visible aspects of the action itself (Piaget,
1937/1954, p. 271). Thus, children in the second year of life
engage what they have come to see as an objective world of
causal interactions among objects in space and they actively
conduct experiments on that world. In Piaget’s (1936/1963)
words: “The objectification of causality is the source of
experimentation” (p. 278).
All of this, however, is a matter of practical intelligence at
what Piaget called the sensorimotor level. Over the second
year of life, infants construct a representational intelligence that
enables them to mentally represent the world and act mentally
on those representations. This enables them to represent and
think about causal relations, with ongoing development in their
causal reasoning over the course of childhood as children
reconstruct their early sensorimotor insights at the level of
representational intelligence. Research with children beyond
infancy showed systematic progress in causal understanding
corresponding to the more general construction of concrete
operational structures at about age 7 or 8 years and formal
structures at about 11 or 12 years (Piaget, 1971/1974), with
related progress in scientific reasoning leading at the formal
operational level to reflective hypothesis testing, including the
ability to isolate variables in order to determine their effects
(Inhelder & Piaget, 1955/1958).
Developmental research on children’s scientific reasoning
has focused mostly on providing evidence showing earlier
competence than Piaget indicated with respect to causal
inference and hypothesis testing (Gopnik, 2012; Sodian, 2018).
As in the case of logical reasoning, there appears to be a
consensus that children of any given age are at least as
competent as Piaget claimed, if not more. Research with
scientific reasoning in adults, however, shows major deviations
from Piaget’s conception of mature scientific reasoning,
including strong tendencies to test hypotheses by seeking out
supporting data rather than by seeking data that could falsify
the hypothesis (Moshman, 2011a; Wason & Johnson-Laird,
1972). As in the case of logical reasoning, the scientific
reasoning of adults not only falls far short of rational norms but
even seems to fall short of the reasoning many
developmentalists attribute to young children.
A solution to this apparent developmental paradox has been
provided by Deanna Kuhn (1989, 2005; Kuhn et al., 1988), who
highlighted the metacognitive nature of advanced scientific
reasoning. Children engage in active exploration of the world,
make causal inferences, and construct intuitive theories, but do
not reflectively differentiate and coordínate theories and data.
Progress in scientific reasoning consists of progress in “thinking
about theories, rather than merely with them, and thinking
about evidence, rather than merely being influenced by it. This
development is thus metacognitive” (Kuhn, 1989, p. 688). In her
later work, Kuhn (2005) especially emphasized the epistemic
basis for advanced reasoning, noting that adolescents and
adults show substantial individual differences in epistemic
cognition and scientific reasoning and that we all fall short of
scientific ideáis. We will address metacognition and epistemic
cognition in more detail in the next chapter.
This metacognitive approach is fully consistent with a dual
process model. People of all ages have many reasons to
notice, seek, retain, and recall information supporting their
views. There is no reason to expect this tendency to disappear
at advanced levels of development. But we can complement,
and sometimes override, our natural heuristics and biases with
deliberate efforts to seek data systematically and to identify
falsifying evidence, and with reflective coordination of theories
with emerging data. In scientific reasoning, as we already saw
in logical reasoning, a developmental dual process conception
helps reconcile developmental research showing the rationality
of young children with cognitive research documenting the
limited rationality of adults.

Moral Reasoning
Moral reasoning is reasoning about matters of morality, which
includes respect and concern for the rights and welfare of
others. Some people construe morality more broadly to include
respect for a variety of societal norms that vary across cultures.
As we will see, however, there is good reason to distinguish
morality from social convention and strong evidence that this
distinction is understood by young children and increasingly
respected over the course of development. I focus in this
section on moral reasoning, which is rooted in principie, and
then turn in the following section to social conventional
reasoning, which is rooted in precedent.

Moral Reasoning as Principled Reasoning

Consider this simple case. You hire a group of three 8-year-
olds to clear snow around your house for $15. They are equally
competent, work equally hard, and contribute equally to getting
the job done well. How much should each get paid? No, this is
not a trick question. Each should get paid $5. Solving this moral
problem is no more difficult than dividing 15 by 3, an objective
matter of following mathematical rules. The children themselves
would understand that. All three would likely object to
differential pay unless it could be justified on the basis of some
children working harder and/or accomplishing more.
Situations like this are common, and the basic principie of
equality that applies in such a situation is elementary and
universal. But the principie is not just mathematical. If we had
15 flowers and three vases, we would see the mathematical
logic of putting 5 flowers in each vase, but we could decide,
say, to put 7 in one and 4 in each of the others, or put 6 in each
of two vases and just 3 in the other. Some people might deem
some distributions esthetically objectionable, but there would
be no question of any moral objection. Giving some children
more or less than others for no good reason, however, is a
different matter. People are not vases. People have rights of
fair treatment that vases do not have. People have moral
obligations to treat each other on a basis of mutual respect,
which requires fair treatment for all. In the case of people,
unlike the case of vases, the matter of distribution across three
options is not just mathematical; it raises issues of individual
rights and corresponding moral obligations.
But what if one of the children worked much harder than
either of the others and accomplished much more? What if they
worked equally hard but one was much more competent and
ended up doing most of the work? All other things being equal,
we should treat all persons equally, but at least some other
things are often unequal. Situational factors, such as differential
effort or ability, may or may not justify deviations from strict
equality of treatment. Deviations from a mathematical rule of
strict equality may be permitted or even required in some cases
by the underlying moral principie of equal respect for all
persons.
What is a principie, and what makes it morally obligatory?
We may define a principie as an abstract norm whose
application to cases requires rational judgment rather than strict
imposition of a concrete rule. Moral principies apply fully and
equally to all persons, consistent with the defining moral
requirement of respect for persons (Dworkin, 1985). But a
principie is not just a rule to be followed. Principies serve as
guidelines and constraints, ruling out some possibilities and
showing how some of the remaining possibilities are better
justified than others, but there may not be a single objectively
correct answer, even if some answers are better justified than
others (Moshman, 1995, 2011a, 2015). In the case of the three
snow shovelers, if one child is more competent than the others
and shoveled more snow, a case might be made for that child
being paid more, but a case might also be made for paying the
children equally. All other things being equal, however, it would
be clearly unfair for that child to be paid less. There does not
appear to be any rule to tell us exactly what the three payments
should be, but moral principies limit the range of legitímate
possibilities and provide guidance in choosing among the
remaining possibilities.
Perspective taking is intrinsic to principled reasoning. Moral
principies arise from the coordination of múltiple perspectives
and ideally take all possible perspectives into account. This
includes not only individual perspectives but the perspectives of
social groups. Perspective taking, moreover, is not just
cognitive but also emotional, including empathic responses that
play crucial roles in moral judgment and action (Gibbs, 2019).
Nevertheless, moral justifications must be principled (Bloom,
2016).

Moral Epistemology
“Everyone,” proclaimed Piaget (1932/1965b), “is aware of the
kinship between logical and ethical norms. Logic is the morality
of thought just as morality is the logic of action” (p. 398).
If we construe logic narrowly as a matter of following strict
rules that yield necessary conclusions, this conception misses
the principled nature of morality. Principies are not rules;
principled reasoning requires a more subtle epistemology
allowing for rational diversity in interpretations and judgments.
Some judgments are morally more justifiable than others, but in
many cases there is no single demonstrably correct moral
answer and little reason to believe that logic alone could supply
one (Moshman, 2015). Moral reasoning is not simply the social
application of logical reasoning.
On a somewhat broader interpretation, however, Piaget’s
statement expresses the view, central to his work on moral
judgment, that there is a moral rationality intrinsic to human
social interaction. This is a broadly Kantian view that is widely,
but not universally, shared by moral philosophers. Although the
rational basis for morality has been doubted in recent decades
by many psychologists, there is strong psychological evidence
for development toward moral rationality (Gibbs, 2019;
Moshman, 1995, 2011a; more on this in the next subsection).
Moral reasoning thus resembles scientific reasoning in
having a rational basis that cannot be reduced to logic (Bloom,
2016). But moral reasoning is not scientific reasoning. Scientific
reasoning aims to determine what the world is really like and to
provide causal explanations of what goes on within it. Moral
reasoning, in contrast, aims to evalúate the world and to tell us
what to do. Unlike science, morality (like logic) is normative
rather than descriptive.
Moral justification and truth, then, are distinct from the
justification and truth of both logical necessities and scientific
explanations. Morality has an epistemology of its own, distinct
from the epistemologies of both logic and science (Moshman,
2015). Moral reasoning, on the basis of moral epistemology,
incorporates principled reasoning and perspective taking in
ways that distinguish it from both logical and scientific
reasoning.
The distinction is not just that morality is a social matter.
Morality, which involves respect or concern for the rights or
welfare of others, is also epistemologically distinct from matters
of social convention and personal identity (Moshman, 2015;
Nucci, 2001, 2014; Smetana, 2011; Smetana, Jambón, & Ball,
2014; Turiel, 1983, 2008; see below for more on social
convention).
Development of Moral Reasoning
Extensive research since the 1960s has shown dramatic
progress over the course of childhood in children’s ability to see
themselves from the perspective of other people. Selman
(1980) proposed that development beyond childhood involves
the emerging ability to engage in third-party perspective taking,
which involves seeing one’s reciprocal relation to another
person from the (meta)perspective of a real or hypothetical third
party. He also proposed a still more advanced, and perhaps
less universal, level of perspective taking that involves taking
the perspective of the social order.
Perspective taking is not itself moral. Understanding the
perspectives of others may serve the immoral purpose of
exploiting them more effectively by anticipating their reactions.
Nevertheless, Kohlberg’s (1981, 1984) Piagetian theory of
moral development proposed that moral development beyond
childhood involves recognizing the moral valué of relationships,
which requires third-party perspective taking. He also proposed
that development may proceed beyond that to a
reconsideration of human relations from the perspective of the
social system, which requires Selman’s societal level of
perspective taking.
John Gibbs (2019) proposed a neo-Kohlbergian theory that
posits four stages of moral development, drawing on
increasingly advanced levels of perspective taking, which is
construed as both cognitive and emotional. Children make
progress in understanding and respecting reciprocity, but the
reciprocity of Stage 2, typical of older children, is a pragmatic
reciprocity of self-interested deais. In adolescence, developing
individuáis construct the ideal reciprocity and genuine mutuality
of Stage 3, which requires third-party perspective taking. Some
individuáis in at least some societies proceed from there to the
social system perspective of Stage 4, rooted in societal
perspective taking.
Research on the development of moral reasoning has
looked specifically at principled reasoning about matters
including free speech, religious liberty, democratic decision-
making, forms of government, and student rights (Helwig, Ruck,
& Peterson-Badali, 2014). Participants across a range of ages
and cultural contexts have been presented with a variety of
moral and social dilemmas and have been asked to explain and
justify their responses. Helwig (1995), for example, asked 12-
year-olds, 16-year-olds, and college students to evalúate
potential laws restricting freedoms of speech and religión.
Participants were also asked to evalúate the exercise of
intellectual and religious freedom in relatively straightforward
cases and in more challenging cases involving, for example,
harmful religious practices or the use of racial slurs. Regardless
of age, students generally showed substantial support for
freedoms of speech and religión, but also recognized limits on
these rights in cases where they potentially conflicted with other
rights or valúes. Students of all ages disagreed with each other
about how best to resolve complex cases, but their justifications
showed clear understanding of relevant civil liberties principies.
Related research (reviewed in Helwig et al., 2014) supports
the conclusión that principled moral reasoning is common in
adolescents and adults, who often show explicit understanding
of principies implicit in the moral judgments of children. The
development of principled reasoning is not so much the
emergence of new principies as the increasing metacognitive
awareness, understanding, and control of one’s principies.
Unlike children, adolescents and adults understand that
principies, although morally obligatory, are not rigid rules.
A broader look at developmental and cognitive research on
human morality shows the same pattern already observed in
the study of logical and scientific reasoning. Developmental
research on children has focused largely on showing empathy,
perspective taking, and moral action earlier than expected by
Piaget (1932/1965b) and by Piagetian theorists such as
Selman, Kohlberg, and Gibbs (Bloom, 2013). Research with
adults, in contrast, has focused on showing systematic
deviations from rational moral norms. Moral judgment has
come to be seen as mostly the result of a variety of cognitive
and moral heuristics and biases. Many researchers have
concluded that our moral and immoral judgments and actions
are largely driven by emotions and intuitions, with only a limited
role for reasoning, not only in childhood but throughout our lives
(Haidt, 2001).
In recent years, however, many researchers have provided
new evidence for the important role of reasoning in moral
judgment. Landy and Royzman (2018) reviewed much of the
relevant literature and proposed a developmental dual process
account of moral reasoning that they cali the moral myopia
model. By “moral myopia,” Landy and Royzman (2018) mean
“singularly attending to one salient aspect of a moral problem
rather than thinking about múltiple moral considerations . . . in a
more complex, integrative way” (p. 71). Although children make
developmental progress in their ability to decenter from one
salient aspect of a problem and to coordínate múltiple
considerations, the moral myopia model acknowledges that
adults still show moral myopia. Depending on circumstances,
however, adults may show reasoning beyond the capacity of
children, with some more likely than others to do so. Landy and
Royzman (2018) argüe that their model accounts for evidence
that “reasoning can be applied to resolve situations in which
moral concerns clash and . . . can promote differentiating
between moral and conventional transgressions” (p. 71).
Integrating developmental research with dual process
theory, it appears that principled moral reasoning and third-
party perspective taking emerge at the transition to
adolescence but that they supplement and sometimes shape
emotional and intuitive reactions, rather than replacing them. A
developmental dual process approach (Landy & Royzman,
2018) provides an integrative picture of moral reasoning that
accounts for both developmental progress (Gibbs, 2019) and
human limitations (Haidt, 2001).

Social Conventional Reasoning

Reasoning about social conventions, like reasoning about
morality, is a matter of reasoning about social norms. As we will
see, however, social conventional reasoning is precedent-
based rather than principled and, thus, rests on a distinct
epistemology.

Conventional Reasoning as Precedent-Based

Reasoning
Consider the three young snow shovelers a couple of years
later. Suppose they now have a history of getting jobs they can
do together. Early on, one of them was arranging these jobs (in
addition to doing a third of the work) and it was agreed that the
arranger should get 40% of the pay and the others 30% each.
Later, a second one took over the role of lining up jobs and got
40% of the pay. Now they have agreed that the third should
take over this role. How should their income be distributed?
Once again this is not a trick question. The new arranger
should receive 40%. But why? Perhaps the job of arranger is so
challenging and time-consuming that the arranger should
receive 50%. Perhaps it is so minimal, due to friendly
neighbors, that the pay should be divided equally. Even with
more information there may be no single correct answer to the
question of how the income should be distributed. But having
decided as a group to pay the arranger 40%, and especially
after maintaining that división for a second arranger, the group
established a social convention and a precedent for future pay
allocations, leading to legitímate expectations on the part of
individuáis. It would not have been morally wrong at the start
for the three children to agree that they would share job income
equally regardless of who arranged the job. But having settled
on a different división of income, to change the established
practice requires justification. Even when múltiple choices are
legitímate, a group choice or practice sets up a precedent that
may be a legitímate social constraint on future choices.
Social groups establish social conventions that are not
required by general moral principies but nevertheless have
normative forcé within that group. Ignoring precedent within a
group undermines the mutual expectations between individuáis
that make the group a group. In the case of larger groups, up to
and including societies and nations, there may be social
conventions that define the group across generations, even as
individual members are replaced, and historical precedents that
serve as a basis for reasoning within the group. Precedents,
like principies, require application and interpretation in
particular cases and may not yield single correct results, but
are nevertheless binding in that they must be considered and
apparent deviations must be explained and justified. The
difference is that moral principies are universally binding on all
rational social agents, whereas precedents are only binding in
specific social contexts.
Reasoning about matters of social convention routinely
involves logical, empirical, and moral considerations. In the
case of the young jobs group, the total of their individual
earnings cannot be greater than the group’s overall income (a
logical necessity), it may be useful to consider evidence about
their relative contributions (a matter of empirical truth), and they
should be fair to each other (a matter of moral principie).
Precedent, however, provides a basis for reasoning in specific
social contexts that goes beyond logic, science, or morality.

Epistemology of Social Conventions

Social conventions are norms of behavior specific to particular
social groups or systems that indícate what one should or
should not do within those social contexts. Unlike moral norms,
they do not have a universal rational basis that makes them
obligatory for everyone. Every social system has its own history
and, thus, its own conventions that are relative to that history.
But social conventions are not just empirical generalizations
concerning typical behavior in particular social systems. They
are normative, like moral norms, but lack the universality of
moral norms.
We must therefore consider an epistemology of social
convention distinct from the epistemologies of logic, Science,
and morality already discussed (Moshman, 2015). Social
conventions may sometimes consist of rules that can be
rigorously applied, but they generally require judgment and
interpretation that go beyond logic. Social conventions may be
based on empirical knowledge of social phenomena, but they
are neither empirical generalizations ñor scientific theories.
Finally, social conventions are normative, but their application
lacks the universality of moral norms. Social convention is thus
an epistemic domain with its own conceptions of justification
and truth and its own prototypical form of reasoning: precedent-
based reasoning.

Development of Precedent-Based Reasoning

Gibbs (2019), drawing on Kohlberg’s (1984) stages, proposed a
social system morality as the fourth and final stage of his theory
of moral development. Evidence across several decades has
shown that children generally construct a Stage 3 morality of
mutual expectations, reciprocity, and third-party perspective
taking around the transition to adolescence, and that Stage 4
social system morality is constructed after that and is less
universal. Meanwhile, social cognitive domain theorists (Turiel,
1983, 2008; Smetana, Jambón, & Ball, 2014) have provided
extensive evidence that children distinguish morality from social
convention on epistemological grounds by the age of 4 or 5
years. Young children already understand, for example, that
hitting is wrong because it hurts others regardless of whether
there is a rule against it, whereas lining up in certain ways at
certain times is only obligatory in particular social contexts
where this is expected. Thus, there is evidence for both early
competence and extended development in conceptions of
social convention.
Central to the development of social conventional reasoning
is the development of its prototypical form: precedent-based
reasoning. Such reasoning has been much less studied than
logical, scientific, or moral reasoning, but relevant evidence
does exist. Paul Klaczynski (2011; see also Klaczynski &
Felmban, 2019) interviewed individuáis of ages 8, 11, 14, and
17-18 years about vignettes in which a rule was violated under
conditions that did or did not present mitigating circumstances.
For example, a principal must decide whether to enforce a rule
against teasing. In one condition the violator was retaliating
after a long history of being teased (mitigating circumstance). In
another condition the violator was simply jealous of a student
who got higher grades. Analysis of the responses showed
systematic differences between the 14-year-olds (who
responded much like the older teens) and the 11-year-olds
(who responded much like the 8-year-olds). Participants of all
ages recognized the legitimacy of rule enforcement by
authorities and understood that a general or arbitrary failure to
enforce rules tends to undermine both their deterrence valué
and the perceived legitimacy of the authority failing to enforce
them. Unlike the preteens, however, the teens understood that
declining to enforce a rule because of mitigating circumstances
does not set the same sort of precedent as routine or
unjustified failures to enforce. Failure to enforce a rule sets a
precedent for future nonenforcement, but the general precedent
for nonenforcement is limited if a special exception is justified
on the basis of specific mitigating circumstances, thus creating
a narrower precedent.
As with other forms of reasoning, a developmental dual
process model best accommodates the data of developmental
and cognitive psychology. Human inferential processes rely
heavily on perceived analogies beginning in infancy and
continuing throughout our lives (Goswami, 1992). The
development of precedent-based reasoning does not supplant
our ongoing reliance on automatic and intuitive analogical
inferences, but supplements and regulates such inferences.
Social conventional reasoning is more than precedent-based
reasoning and routinely makes use of logic, evidence, and
moral principies. It appears, however, that conventional
reasoning is epistemologically distinct from logical, scientific,
and moral reasoning in its reliance on precedent-based
reasoning. We will retum to precedent-based reasoning in the
discussion of constitutional reasoning and legal tradition in
Chapter 6.

Reasoning as Epistemologically Self-Regulated

Thinking
In Chapter 1 I defined reasoning as epistemologically self-
regulated thinking. This definition was not crucial in Chapter 2
because logical reasoning is widely recognized as reasoning. In
the present chapter, however, the question of what counts as
reasoning arises. What does it mean for thinking to be
epistemologically self-regulated? Having considered three
additional forms of reasoning, we now look at reasoning more
generally.

The Epistemic Basis of Reasoning

Scientific, moral, and social conventional reasoning differ from
each other, as we have seen, but share one characteristic that
distinguishes them all from logic. They are all subjective in a
way that logic, arguably, is not. Of course, all reasoning, even
logical reasoning, takes place from the (subjective) viewpoint of
the reasoner. But logical reasoning conforms to rules so strict
that they leave no room for any answer other than the correct
one. Scientific, moral, and social conventional reasoning, in
contrast, are subjective in a deeper sense. There are often
rationally justifiable differences of opinion about such matters.
But this doesn’t mean that all opinions are equally justified.
Some scientific, moral, and conventional claims or conclusions
are better justified than others, and some are just plain false.
Research in cognitive and developmental psychology shows
that people distinguish logic from other domains on
epistemological grounds, with increasingly explicit
understanding of this distinction as individuáis construct
increasingly reflective levels of metalogical understanding. In
scientific, moral, and social conventional domains, extensive
research has converged on an account of epistemic
development as progress from (1) objectivist conceptions of
knowledge in which everything is a matter of fact or logic, to (2)
subjectivist conceptions in which everything is relative, to (3)
rationalist conceptions that find a basis for objectivity despite
inherent subjectivity and, thus, restore a role for justification
and a meaningful conception of truth.
I have defined thinking as the deliberate application and
coordination of one’s inferences to serve one’s purposes.
Thinking is thus intrinsically metacognitive in that it involves
knowledge and control of cognition. Reasoning is thinking
regulated specifically for purposes of truth and justification and
thus involves epistemological aspects of metacognition. In the
next chapter we will look in much more detail at matters of
metacognition, and especially epistemic cognition.

Reasoning as the Generation and Use of Reasons

Reasoning provides us with reasons for what we believe and
what we do. Automatic inferences routinely modify our beliefs
and cause us to act, but only reasoning, because it is deliberate
and reflective, provides reasons. When we believe or act on the
basis of reasoning we can say why we believe what we do or
acted as we did. Consistent with the definition of reasoning as
epistemologically self-regulated thinking, we may also define it
as the generation and use of reasons.
But it is not just individuáis who generate and use reasons.
Groups of two or more individuáis engaged in argumentation
generate and use reasons and thus engage in collaborative
reasoning. Collaborative reasoning, however, is not simply a
collection of reasons. In collaborative reasoning, each of the
collaborators is a rational agent— that is, an agent who believes
or acts on the basis of reasons and who engages with other
rational agents on this basis. To be a rational agent, moreover,
is to be a person— that is, the sort of agent who can particípate
in argumentation and, more broadly, in deliberative democracy.
We will retum to these matters of reasoning in argumentation
and deliberative democracy in Chapters 5 and 6, respectively.

Conclusión
We are faced throughout life with questions of what to believe
and what to do. We should address such matters logically, to
the extent that logic is relevant, but we should not expect logic
to answer all our questions. As we have seen in this chapter,
we supplement our logical reasoning with scientific, moral, and
social conventional reasoning.
It might seem the account of múltiple forms of reasoning in
Chapters 2 and 3 provides the outline of a comprehensive
theory of reasoning. What remains, it might seem, is to fill in the
details concerning the development and functioning of the
various forms of logical, scientific, moral, and social
conventional reasoning, and perhaps consider whether there
are any additional sorts of reasoning not reducible to any of
these. As already noted, however, all such reasoning is rooted
in metacognition, and especially in epistemic cognition. We
cannot understand reasoning without understanding how
people control their inferences and what people know about
justification and truth. We now tum to these matters.
4
METACOGNITION AND EPISTEMIC
COGNITION

In the previous chapters we defined (1) thinking as the

deliberate application and coordination of one’s inferences to
serve one’s purposes and (2) reasoning as epistemologically
self-regulated thinking. Thinking (including reasoning) is thus
metacognitive in its cognitive self-regulation of cognition.
Reasoning, specifically, is epistemological in its concern with
truth and justification.
In this chapter we look more directly and systematically at
metacognition, focusing especially on epistemic cognition. Our
conceptions of justification and truth are crucial to our
reasoning. As we will see in Chapter 6, moreover, epistemic
cognition is also crucial to democratic self-governance, which
requires not only rational deliberation among peers but also
appropriate respect for expertise. Thus, democracy requires
widespread knowledge about the nature of knowledge. We will
see in Chapter 8 that epistemic cognition is a crucial
consideration in education aimed at the promotion of rationality
and democracy.
The development of metacognition begins early and
proceeds rapidly across childhood. Epistemic cognition shows
a universal course of development across childhood with
substantial potential for further development beyond that in our
knowledge about justification and truth. A developmental
sequence seen in childhood, beyond childhood, and in múltiple
epistemic domains is a transition from (1) a conception of
knowledge as objective truth, to (2) a conception of knowledge
as subjective opinion, to (3) a conception of knowledge as
justified belief.

Metacognition
I begin in this section with metacognition, which is crucial to
thinking. I focus especially on metacognition in child
development and in the logical reasoning of adults. I then turn
in the following two sections to epistemic cognition, the subset
of metacognition most crucial to reasoning.

The Study of Metacognition

Metacognition is cognition about cognition. The study of
metacognition dates at least to Plato, but the term
“metacognition” originated in the late 1970s in developmental
and educational psychology (Brown, 1978; Flavell, 1979). The
psychological and educational literature on metacognition now
includes thousands of publications using that term and many
thousands more using related terminology. Overlapping
concepts from a variety of academic fields over the past
century include executive functioning, executive control, self-
monitoring, self-regulation, self-knowledge, introspection,
reflection, epistemic cognition, critical thinking, cognizance,
consciousness, metarepresentation, attribution, perspective
taking, and theory of mind (Campbell & Bickhard, 1986;
Demetriou, Makris, Kazi, Spanoudis, & Shayer, 2018;
Demetriou & Spanoudis, 2018; Dewey, 1910/1997; Miller,
2012; Piaget, 1977/2001; Pillow, 2012; Stanovich, 2011;
Tarricone, 2011).
Research on metacognition generally distinguishes
knowledge of cognition from regulation of cognition (Schraw &
Moshman, 1995). Knowledge of cognition refers to what
individuáis know about their own cognition, the cognition of
others, or cognition in general. Regulation of cognition involves
metacognitive activities that help control one’s inferential
processes, including planful strategy selection, monitoring of
comprehension and task performance, and evaluation of
processes and outcomes. Beyond this basic distinction,
however, the theoretical conceptualization of metacognition is a
complex matter, complicated by the diversity of the terminology
used. Researchers who use the term “metacognition,” for
example, generally focus on cognition about one’s own
cognition; researchers using the terms “attribution,”
“perspective taking,” or “theory of mind” generally focus on
cognition about other minds; and researchers referring to
“epistemic cognition” are usually most concerned with
knowledge about general issues of justification and truth. All of
this is metacognitive in the broad sense of the term, but the
diverse terminology accurately reflects the expansive and
multifaceted nature of metacognition.

Metacognition in Development
Cognition about cognition is central to the developmental theory
of intelligence formulated over the past several decades by
Andreas Demetriou and colleagues (Demetriou et al., 2018;
Demetriou & Spanoudis, 2018). In order to provide a
differentiated picture of what might broadly be called the
metacognitive realm, they highlight what they cali “cognizance”
as a crucial aspect of what they cali “reflexive consciousness.”
Consciousness is defined as “awareness of one’s own
existence and current relation with the world” as in the
statement, “I know I am here and that the object I see is blue.”
Consistent with dual process theories, it is acknowledged that
much cognitive activity is unconscious. A basic level of
consciousness includes “awareness of one’s own current object
of thought” as in “I know that I see a red apple.” Reflexive
consciousness is defined as “awareness of one’s own mental
processes (I know that I see by my eyes), knowledge (I know I
saw apples before), and abilities (I am good in remembering
what I saw before).” This includes cognizance, the portion of
reflexive consciousness “involving awareness of cognitive
processes” (Demetriou et al., 2018, all quotes from p. 1).
 Cognizance draws on self-monitoring, defined as “the mind’s
eye turning to itself,” as in “I try to balance on a seesaw” or “I
try to learn how to add up numbers.” It also draws on reflection,
defined as “mentally re-enacting past experiences, memories,
or thoughts to better understand events, actions, and thoughts
and their relations.” Finally, cognizance draws on
metarepresentation, which is defined as “encoding of the
results of self-monitoring and reflection and tagging or
redescripting them into new representations for future use.”
Research indicates that cognizance plays a role in the
emergence of executive control and in the relation of executive
control and reasoning. Cognizance, executive control, and
reasoning are seen together as the “developmental trinity of
mind” (Demetriou et al., 2018, all quotes from p. 1).
In a cross-sectional and longitudinal study motivated by this
metacognitive theory, Kazi, Kazali, Makris, Spanoudis, and
Demetriou (2019) looked specifically at how cognizance
mediates between executive control and reasoning over the
age range of 4 -1 0 years. A total of 113 children of ages 4, 6, or
8 years were assessed on a battery of cognitive and
metacognitive tasks and were assessed again two years later
(when the oldest were 10 years oíd). Cognizance was
assessed via measures of (1) awareness of the perceptual and
inferential origins of knowledge, involving the ability to
differentiate visión, hearing, and inference as distinct sources of
knowledge; (2) awareness of cognitive processes, involving the
ability to evalúate the similarity and relative difficulty of múltiple
pairs of tasks; (3) first-order theory of mind, which requires
understanding that a character would look for an object where
she believed it to be, even if that belief was known to the child
to be false; and (4) second-order theory of mind, which requires
determining what one character would believe about what a
second character believes under circumstances of incomplete
information. Executive control was assessed via measures of
(1) attention control, involving perceptual discrimination and
response inhibition; (2) cognitive flexibility, involving the flexible
pairing of labels with picture properties; and (3) working
memory, involving the backward recall of between two and six
words or digits. Reasoning was assessed via measures of (1)
inductive reasoning, involving matrices of increasing complexity
such that they required abstraction of one, two, or three
dimensions; and (2) deductive reasoning, involving the ability to
determine the implications of rules involving three simple logical
forms.
Consistent with theoretical expectations and previous
research, results indicated that even 4-year-olds have
inferential and metacognitive abilities and that between ages 4
and 10 years children show systematic age-related progress in
reasoning and in múltiple aspects of metacognition, leading to
what Kazi et al. (2019) described as “the highly organized
cognitive attainment observed at the end of childhood.” Even
second-order theory of mind was within the competence of
many (though not most) 8- and 10-year-olds. More detailed
analyses supported theoretical predictions concerning the
crucial role of cognizance in the development of executive
control and reasoning. Kazi et al. (2019) concluded that the
results “strongly suggested that there is a central core of
cognizance underlying awareness of one’s own and others’
mental states” and that this core “develops systematically
through the years, emerging from cognitive actions or
experiences and contributing to their further development, in an
ever-changing spiral of interactions with executive and
reasoning process[es].”

Metacognition in Reasoning
Research on reasoning, as we saw in Chapters 2 and 3,
provides extensive evidence for the crucial importance of
metacognition. In recent reasoning research, metacognition is
generally conceptualized within a dual process framework in
which metacognition is associated with the deliberate and
reflective processes of System 2 and/or with the resolution of
conflicts emerging from the interplay of System 2 with the
automatic inferences and intuitions of System 1 (Amsel,
Klaczynski, Johnston, Bench, Cióse, Sadler, & Walker, 2008;
Markovits, Brunet, Thompson, & Brisson, 2013; Markovits,
Thomson, & Brisson, 2015). These studies generally use
complex statistical techniques to test múltiple predictions
derived from the particular dual process model used by the
study. For present purposes, I ¡Ilústrate how such studies are
conducted without detailing the specific models, predictions,
analyses, and results.
Focusing on logical reasoning, Markovits et al. (2013)
provided what they called in their title, “direct evidence for a
dual process model of deductive inference.” Across two
studies, the authors presented college students with a series of
seemingly reasonable simple inferences that enabled
researchers to distinguish (a) a “statistical strategy,” which
intuitively accepts or rejects inferences on the basis of the
estimated likelihood of counterexamples; from (b) a
“counterexample strategy,” which recognizes that a single
counterexample undermines the validity of a deductive
argument. Though previous researchers had provided evidence
for both sorts of processes, the focus had been on showing that
reasoners found to be generally more competent were more
likely to use the counterexample strategy. The purpose of the
present research was to show that most reasoners are capable
of both strategies and to examine their interplay. It was
hypothesized that the statistical strategy was a System 1
default strategy, whereas the counterexample strategy involved
System 2 deliberation and reflection, and that reasoners
exercised some degree of agency in choosing a strategy.
Strategy use was manipulated through the imposition of strict
time constraints, which interfered with System 2 functioning.
Study 1 compared performance under time constraint with
performance when allowed as much time as needed; it also
compared participants who experienced the time constraints
before getting unlimited time to those who experienced the time
constraints after a period of unlimited time to respond. Study 2
added measures of reasoner confidence. Results were
consistent with detailed predictions made on the basis of the
dual process model.
In two additional studies, Markovits et al. (2015) presented
college students with a series of conditional inference tasks
systematically varying in contení. Some tasks used familiar
content that would easily enable students to think of
counterexamples and thereby reject invalid arguments. Some
tasks used content that, though also familiar, did not suggest
obvious counterexamples. Finally, some tasks used “abstract”
content with no meaning. Confidence in one’s judgments was
also assessed. The pattern of results was “consistent with a
dual meta-representational model, which supposes that people
have both a meta-representation of logical validity,
preferentially used to evalúate abstract content, and one based
on knowledge-based cues, which is preferentially activated with
familiar content” (Markovits et al., 2015, p. 689). In the second
study, half the students began with an experimental
manipularon intended to orient their thinking toward
counterexamples, which had previously been shown to improve
logical performance, although it does not explicitly teach about
logical principies such as necessity. Once again the pattern of
results, albeit complex, suggested that people have logical
conceptions of inferential validity, but these conceptions
operate within the larger context of a dual process model and
do not always prevail.

Metacognitive Development Beyond Childhood

Metacognitive development continúes beyond age 10 years but
rapidly becomes much less age-related and universal as
children move into adolescence. As we have seen,
metacognition is already apparent in young children, but even
in adulthood no one ever becomes, in some general way,
metacognitive. Our metacognition is always part of a larger
cognitive system and is best construed within a developmental
dual process model of cognition.
 Much of the literature on metacognitive development beyond
childhood focuses on what has come to be called epistemic
cognition (Greene et al. 2016; Moshman, 2015), which is the
aspect of metacognition of particular concern with respect to
reasoning (Moshman, 2011a). As we will see in Chapters 6 and
8, epistemic development beyond childhood is also a matter of
special concern for democracy and education. Before turning to
epistemic development beyond childhood, however, we first
consider its development across childhood.

Epistemic Development in Childhood

Epistemic cognition is the aspect or subdivisión of
metacognition concerned with knowledge about epistemology
— that is, knowledge about the justification and truth of beliefs.
Research on cognitive development in childhood has
addressed the development of epistemic cognition using a
variety of terminologies, usually without direct reference to
epistemology. I consider in this section three relatively discrete
metacognitive literatures that address important aspects of
early epistemic development (Moshman, 2015).

Development of Metalogical Understanding

As we saw in Chapter 2, logic is initially implicit in biological
self-regulation (Piaget, 1967/1971) and then in sensorimotor
action beginning in the first year of life (Langer, 1980, 1986;
Piaget, 1936/1963). The development of logical reasoning is
not about the appearance of logic but about increasing control
of our inferences and increasingly explicit knowledge about the
logic implicit in them (Moshman, 1990, 2004b). The control of
our inferences is metacognitive; the explicit knowledge about
logical necessity and proof is, specifically, epistemic.
Developmental research on knowledge about logic suggests
a sequence of four stages of metalogical understanding
(Moshman, 1990, 2015). Stage 1 (Explicit Content/lmplicit
Inference) is typical of children under the age of 6 years.
Children in this stage can tell you what they are thinking about
(for example, toys, animals, colors, sizes) but they are not
aware they are making inferences, though of course they are
making inferences all the time (as we saw in Chapter 2). As a
result, preschool children do not distinguish conclusions from
premises, do not attribute inferences to others, and do not
recognize inference as a potential source of knowledge (Miller,
2012; Pillow, 2012; Sodian & Wimmer, 1987). In Stage 2
(Explicit Inference/lmplicit Logic), associated with later
childhood, children are aware not only of what they are thinking
about but also of their inferences. They recognize conclusions
as the outcome of inferences, distinguish conclusions from
premises, recognize that others also make inferences, and
understand that, as a result of inference, people may come to
know things they have not observed or been told. The explicit
attention to inferences involves an implicit appreciation of their
logical form and necessity.
Stage 3 (Explicit Logic/lmplicit Metalogic) may be considered
the adult level of metalogical understanding, involving what
Markovits et al. (2015) cali “a metacognitive representaron of
abstract validity” (p. 681). Stage 3 competencies begin to be
seen about age 11 or 12 years and are routinely seen in
adolescents and adults, though people of all ages (as we saw
in Chapter 2) commonly fall short of logical norms. In Stage 3,
explicit awareness of logical form and necessity makes it
possible to distinguish the validity of an argument (does the
conclusión follow necessarily from the premises?) from the
truth of its individual premises and conclusión (Moshman &
Franks, 1986). This enables hypothetico-deductive reasoning,
central to Piaget’s conception of formal reasoning (Chapter 2).
The explicit attention to logic entails an implicit metalogical
understanding of logic as a domain of necessity distinct from
empirical science. Stage 4 (Explicit Metalogic) is the stage of
logicians, who reflect on the nature and justification of formal
logical systems, their interrelations with each other and with
natural human languages, the nature and scope of logic, and
the relation of logic to Science and other domains.
The description of these stages provides an important insight
into the process of development that accounts for them. In each
stage transition, what was previously implicit becomes an
explicit object of reflection. In Chapter 7, we will retum to the
development of metalogical understanding as part of a more
general account of the processes that account for progress in
reasoning and rationality.

Preschool Theory of Mind

Extensive research with young children has made it clear that
they know much about cognitive and other aspects of the mind
from very early ages. In fact, by the age of 4 years, knowledge
of mind is sufficiently organized and has sufficient explanatory
power that 4-year-olds are commonly said to have a “theory of
mind” (Alien & Bickhard, 2018; Miller, 2012; Moshman, 2015;
Pillow, 2012; Sodian & Kristen, 2016).
In contrast to research using the term “metacognition,” which
generally focuses on people’s knowledge about and regulation
of their own cognition, research using the term “theory of mind”
generally focuses on knowledge about other minds, especially
with regard to other people’s knowledge and inferential
processes. Despite the distinction in terminology and research
traditions, most research on what is called “theory of mind” falls
squarely within the broader realm of metacognition as cognition
about cognition. Correspondingly, research on what is called
“metacognition” is widely interpreted as showing that people, at
least beyond the age of 4 years, organize their metacognitive
knowledge into theories (Schraw and Moshman, 1995). As we
saw in the previous section, the metacognitive theory of
development proposed by Demetriou and colleagues
(Demetriou & Spanoudis, 2018; Demetriou et al., 2018; Kazi et
al., 2019) uses the term cognizance to include knowledge of
both one’s own and other minds.
The most popular measure of theory of mind has been the
false belief task, which assesses the understanding that other
people will act on the basis of their own beliefs, even if those
beliefs are false. Extensive research has shown that most 4-
year-olds, and virtually all 5-year-olds, understand this,
whereas 3-year-olds generally do not (Alien & Bickhard, 2018;
Miller, 2012; Moshman, 2015; O’Madagain & Tomasello, 2019;
Pillow, 2012; Sodian & Kristen, 2016; Wellman, Cross, &
Watson, 2001, Wimmer & Perner, 1983). In a typical study, a
child sees a hidden object moved from one location to another
and then must predict where another person, who was aware of
its original location but did not see it moved, will look for it.
Children of age 4 generally expect the other person to look for
the object where the person saw it hidden, even though it is no
longer there, whereas 3-year-olds generally expect the other
person to look for the object where the child knows it to be,
apparently unable to conceive that the other person could have
a different, and false, belief.
Understanding that people may have different beliefs and
that beliefs may be false raises questions of truth and
justification. Such questions initially arise with respect to
particular beliefs but ultimately lead to questions about
knowledge in general. Thus, in the 4-year-old’s theory of mind
we have what may be seen as the beginnings of epistemic
cognition (Alien & Bickhard, 2018; lordanou, 2016; Moshman,
2015; Sodian & Kristen, 2016).
Many developmental researchers have provided evidence
they interpret as showing that theory of mind, or at least
important aspects of it, can be seen in children under 4 years
(Miller, 2012; Moshman, 2015; Sodian & Kristen, 2016). Some
have argued for theory of mind in infancy (for reviews, see
Sodian, 2011; Sodian & Kristen, 2016). The term apparently
originated in the study of nonhuman species (Premack
&Woodruff, 1978), and there is ongoing disagreement about
how to interpret evidence in very young children and in other
species. But there is a strong theoretical consensus that human
children by the age of 4 years have important insight into the
minds of others, including the knowledge that people may have,
and act on, false beliefs.

Constructivist Theory of Mind

Bucking the trend to find theories of mind in younger and
younger children, some developmentalists have provided
evidence for further development of theory of mind in later
childhood. In particular, the insights of older children suggest
that they, unlike younger children, recognize the active,
interpretive nature of the mind as a constructor of knowledge.
Research on epistemic development beyond age 4 years
provides a picture of progress toward constructivist theories of
mind. The minds of very young children are already active,
interpretive, and constructive, but only as they get older do
children increasingly recognize and theorize about this, thus
constructing a constructivist theory of mind (Carpendale &
Chandler, 1996; Chandler, Hallett, & Sokol, 2002; Lalonde &
Chandler, 2002; Miller, 2012; Moshman, 2015; Pillow, 2012).
Michael Chandler and colleagues (Carpendale & Chandler,
1996; Chandler, Hallett, & Sokol, 2002; Lalonde & Chandler,
2002) have studied the development of knowledge about
subjectivity and interpretive diversity beyond the recognition of
false belief at age 4. Carpendale and Chandler, for example,
presented children ranging from 5 through 8 years with a
variety of tasks in which two puppets disagreed with each other
about matters of interpretation involving ambiguous figures
(such as the classic duck-rabbit, which can be seen as a duck
or a rabbit), lexical ambiguity (such as “wait for a ring,” which
could refer to the ring of a telephone or a ring to wear), and
ambiguous referential communication (leaving two
possibilities). The children also encountered deviant
interpretations (such as perceiving the duck-rabbit as an
elephant) and disagreements about matters of taste (such as
whether a soup tasted good or bad). Younger children showed
some ability to recognize and explain differences in taste but
were mostly unable to recognize the legitimacy of different
interpretations. The 7- and 8-year-olds showed substantially
greater understanding that two interpretations may both be
legitímate in cases of ambiguíty. They usually recognized,
however, that the deviant interpretations could not be justified.
The explanations of the older children showed an emerging
insight that legitímate interpretations, unlike matters of taste,
are grounded in evidence and reasons concerning the object to
be interpreted.
Related research has shown similar developmental trends
across childhood (Chandler et al., 2002; Miller, 2012;
Moshman, 2015; Pillow, 2012; Wainryb, Shaw, Langley,
Cottam, & Lewis, 2004). Epistemic development beyond basic
theory of mind appears to involve two steps. The first is the
recognition that differences in belief are not only a matter of
someone not knowing what someone else knows and thereby
having a false belief. Two people may have different
perspectives on the same thing and therefore perceive or
interpret it differently. Thus, children increasingly understand
the relativity of perspective. But development is not simply
progress toward epistemological relativism. Older children
come to recognize that, in some cases, despite legitímate
diversity of interpretation, there may be a rational basis for
judgment. The duck-rabbit may reasonably be seen as either a
duck or a rabbit, but there is no justification for calling it an
elephant.
Epistemic development in childhood is thus, in large part,
progress in knowledge about the constructive powers of the
mind. As we will see in Chapter 7, the constructive powers of
the mind are the basis for a constructivist account of
development in which the actions of organisms are seen as
playing a major role, along with heredity and environment, in
developmental progress. The emergence of a constructivist
theory of mind, which recognizes the constructive powers of the
mind, thus enables increasing control of one’s own
development.

Epistemic Development Beyond Childhood

As we have seen, older children know far more about
knowledge than younger children, who already know enough by
age 4 years to be said to have a theory of mind. What remains
to develop? As we will now see, there is extensive evidence
that epistemic development beyond childhood proceeds from
(1) simplistic notions of truth and falsity, to (2) increasing
recognition of the subjectivity and relativity of knowledge, to (3)
increasing appreciation of the possibility of rational judgment
despite subjectivity. But wait. Isn’t that the developmental trend
we just observed in childhood? Indeed it is, and this is no small
issue (lordanou, 2016). Before we turn to specific theories and
research concerning later epistemic development, we need to
look in a more general way at the lifespan pattern.

Epistemology Across the Lifespan

Epistemic cognition has generally been seen as an advanced
and late-developing form of cognition and has generally been
studied in adolescents and college students. Young
adolescents are assumed in this literature to be
epistemologically naíve. The focus is on describing this
epistemic starting point and successive levels of epistemic
development. But research on children makes it clear that the
epistemic cognition of young adolescents, far from being a
starting point, is the sophisticated outcome of an ongoing
developmental process.
Chandler et al. (2002) provided an elegant and convincing
resolution to the problem of coordinating these two literatures.
Epistemic development, they argued, occurs in two cycles,
each proceeding through three corresponding stages to create
a developmental spiral. The developmental trend is recursive,
with precisely two iterations, one in childhood and one beyond.
Child development is at the “retail” level, and development
beyond childhood is at the “Wholesale” level. Epistemic
development in childhood moves from a naíve objectivism to a
“retail” subjectivism that acknowledges genuine differences in
interpretation and disagreements that cannot be rationally
resolved, but only in particular cases that come to one’s
attention. The subsequent understanding that some
interpretations are better justified than others and that some
beliefs are just plain wrong is also achieved by children only at
a “retail” level, thus completing the first cycle of development in
late childhood.
Adolescents retain the hard-won epistemic insights of
childhood but recognize “Wholesale” questions of knowledge,
including abstract issues of truth, justification, objectivity, and
subjectivity. At this higher level they begin with simplistic
conceptions of truth and objectivity but often come to recognize
“the dangerous prospect that diversity of opinion is somehow
intrinsic to the knowing process” (Chandler et al., 2002, p. 162).
Adolescent subjectivism, unlike that of childhood, threatens the
very basis for knowledge. But some people, at least some of
the time, show evidence of development beyond radical
subjectivism to a rationalist theory of knowledge, the third stage
of the second cycle.

Theoríes of Epistemic Development Beyond

Childhood
There are a variety of theories and research traditions in the
study of epistemic development beyond childhood (Moshman,
2015). Although they differ substantially in methodology, they
all posit a sequence of steps, each of which represents
progress over the previous one. The number of steps varies
across theories but can readily be reduced, for each of the
major theories, to the three basic stages that Chandler et al.
(2002) proposed for both child development and (at a higher
level) later development. However, even if adolescents and
adults cycle through the same issues as children at a more
abstract level, research shows an important difference between
the two cycles. In contrast to epistemic development in
childhood, which is universal and strongly tied to age, epistemic
development beyond childhood varies greatly across
individuáis in its rate and extent.
 The study of epistemic cognition is generally traced back to
a longitudinal study by William Perry (1970/1999) of “intellectual
and ethical development” in college. Students were interviewed
annually across their four years of college about their
educational experiences. Except for this general focus, the
interviews were unstructured and open-ended. The results
were interpreted as illustrating nine “positions” that emerge in a
natural progression, though some students develop more
rapidly than others and some fail to develop beyond the second
or third position. The first three positions represent increasingly
complex forms of dualism, involving a sharp distinction between
right and wrong. Diversity of opinion and cognitive uncertainty
are increasingly seen as possible and legitímate across the
three positions but are attributed to confusion (position 2) or
temporary ignorance in cases where authorities have not yet
determined the truth (position 3). Some students move beyond
this to posit, outside the realm of authority, a distinct epistemic
realm in which diversity of opinion is expected and legitímate
(position 4), but that realm often expands to swallow the
dualistic realm of truth and falsity, leading (in position 5) to a
“radical reperception of all knowledge as contextual and
relativistic” (p. 121). Development beyond this (positions 6 -9 ) is
seen by Perry as a matter of commitment, responsibility, and
personal identity.
Patricia King and Karen Kitchener (1994) provided a
developmental theory of “reflective judgment” that drew on the
work of Perry (1970/1999) but was also rooted in philosopher
John Dewey’s (1910/1997) conception of “reflective thinking”
and motivated by educational concerns with “critical thinking.”
They too relied on interviews, but their interviews were
specifically structured to focus on matters of intellectual
diversity, justification, certainty, and truth. Thousands of
adolescents and adults in dozens of studies have participated
in such interviews, including some who were interviewed
múltiple times over periods of up to ten years. The results are
generally consistent with a sequence of seven stages, though
most people provide judgments associated with two or three
adjacent stages, and progress through the stages (as with
Perry’s positions) is highly variable in rate and extent. The first
three stages, labeled “pre-reflective,” proceed from simplistic
conceptions of absolute truth (stage 1), to increasing
recognition that knowledge may not be immediately available
(stage 2), to acknowledgment that personal opinions must
sometimes be respected until certainty can be attained (stage
3). The next two stages, designated “quasi-reflective,” involve
increasing recognition of the uncertainties and ambiguities of
knowledge (stage 4), leading to a conception of knowledge
(stage 5) as “contextual and subjective” due to being “filtered
through a person’s perceptions and criteria for judgment” (p.
14).
Thus far, King and Kitchener’s (1994) theory is generally
consistent with that of Perry (1970/1999), but it diverges
beyond stage 5. Perry took the radical subjectivism,
contextualism, and relativism of his fifth position as the
outcome of epistemic development and presented his
remaining four positions as matters of personal choice and the
development of identity. King and Kitchener, in contrast,
proposed that some individuáis achieve “reflective thinking,”
which acknowledges that knowledge is constructed but seeks
justifications and better interpretations through the coordination
of múltiple perspectives and contexts (stage 6) and, in its most
advanced form (stage 7), construes knowledge as “the
outcome of a process of reasonable inquiry,” which generates
conclusions that may be “defended as representing the most
complete, plausible, or compelling understanding of an issue on
the basis of the available evidence” (p. 14). Thus, King and
Kitchener’s seven stages, like Perry’s nine positions, can be
divided into three levels, with the difference that King and
Kitchener, unlike Perry, posited a post-subjectivist level that
recognizes the possibility of rational judgment despite
subjectivity.
Annick Mansfield and Blythe Clinchy (2002) assessed the
epistemic cognition of children at about 10 years and then
followed them longitudinally with further assessments three
years later and again three years after that, when they were
about 16. Participants were presented with vignettes in which
protagonists disagreed with each other and were then asked a
series of questions about the disagreement, their own view,
whether certainty could be achieved, and whether both
protagonists could be right. Despite substantial differences
across individuáis and vignettes, they found a general age
trend in “epistemological positions” from an “objective” concern
with facts, to a “subjective” respect for opinions, to an
“integrated” appreciation of both. Although Mansfield and
Clinchy’s criteria for attributing advanced epistemologies were
less stringent than those of King and Kitchener (1994), leading
to what appears to be somewhat earlier development, their
three positions correspond well with the three general levels
into which King and Kitchener divide their seven stages,
indicating theoretical agreement about the basic course of
epistemic development.
Michael Chandler, in addition to his work reported earlier on
the emergence of a constructivist theory of mind in middle
childhood, also studied epistemic development in adolescence
(Chandler, 1975, 1987; Boyes & Chandler, 1992; Chandler,
Boyes, & Ball, 1990), including its role in identity formation.
Children beyond age 4 years often doubt the truth of particular
claims, but such doubts are “retail” because they are “case-
specific.” Adolescents, in contrast, may face “that Wholesale
brand of unassuageable doubt that ultimately poisons the
whole well of certainty, by calling into radical question the very
prospects for any kind of trustworthy knowledge whatsoever”
(Chandler et al., 1990, p. 378). This may lead to “skepticism,”
an “epistemic posture” similar to the subjectivist intermedíate
level proposed by previous theorists. Some proceed beyond
this to “postskeptical rationalism,” which rejects “unbridled
relativism” in favor of acknowledging “rationally based truth
claims” and “arguably good reasons for choosing one course of
action over another” (Chandler et al., 1990, p. 380). Chandler’s
account of epistemic development beyond childhood is thus
broadly consistent with those of King and Kitchener (1994) and
Mansfield and Clinchy (2002). Research independently
assessing epistemic cognition and personal identity suggested
that identity formation is a distinct process partly accounted for
by progress in epistemic cognition (Boyes & Chandler, 1992;
Chandler et al., 1990), rather than a substitute for a post-
subjectivist epistemology (as in Perry, 1970/1999).
Deanna Kuhn, known for her research on both scientific
reasoning (as discussed in Chapter 3) and argumentation (to
be discussed in Chapters 5 and 8), has long seen
“epistemological understanding” as crucial to both (Kuhn, 2005;
Kuhn, Cheney, & Weinstock, 2000; Kuhn & Weinstock, 2002).
In contrast to the extensive interviews and complex coding
schemes of other theorists, Kuhn et al. (2000) simply presented
scenarios in which two characters disagreed and asked
participants whether only one was right or whether both “could
have some rightness,” and, if the latter, whether one view could
be better or more right than the other. Responses that only one
character could be right were coded as “absolutist.” Responses
that neither view could be better or more right than the other
were coded as “multiplist.” Finally, responses that one view
could be better even if neither is absolutely right or wrong were
coded as “evaluativist.” Although responses differed
systematically across topics, Kuhn (2005; Kuhn et al., 2000;
Kuhn & Weinstock, 2002) proposed a post-childhood
developmental sequence that proceeds in different domains at
different rates and to different extents but always from
absolutism to multiplism to evaluativism.
Absolutists, in Kuhn’s (2005) developmental analysis, have
already moved beyond the naíve “realism” of early childhood in
which beliefs are assumed to be “faithful copies of reality . . .
received directly from the external world” (p. 30).
Acknowledging the possibility of false belief, absolutists
nevertheless assume that such beliefs can only result from
“inadequate or incorrect information” and “are easily correctable
by reference to an external reality” (Kuhn, 2005, p. 31). In a
case of disagreement, someone must be wrong. But some
adolescents come to question this, at least in some domains,
and move to a multiplist epistemology in which knowledge is a
subjective function of the knower rather than a causal outcome
of the known. Mind enters the picture. But now, everyone is
right. Some individuáis, however, at least in some domains,
show further progress in the “extended task of coordinating the
subjective and the objective elements of knowing” (Kuhn, 2005,
p. 32), thus achieving a post-relativist epistemology of
evaluativism, which recognizes that “some opinions are in fact
more right than others, to the extent they are better supported
by argument and evidence” (Kuhn, 2005, p. 32). Development
thus proceeds, as in other theories of epistemic progress
beyond childhood, from an objectivist conception of knowledge
as facts, to a subjectivist conception of knowledge as opinions,
to a rationalist conception of knowledge as “judgments, which
require support in a framework of alternatives, evidence, and
argument” (Kuhn, 2005, p. 32).
The major theories of epistemic development beyond
childhood, as we have just seen, postúlate a sequence that can
readily be seen as a transition from objectivism to subjectivism
to rationalism. Objectivism is an epistemology (theory of
knowledge) that holds there to be ultímate truths that are
directly observable, provable, or known to infallible authorities.
Subjectivism is an epistemology in which knowledge is seen as
constructed from, and thus determined by, one’s point of view.
Rationalism is an epistemology in which reasons short of proof
provide bases for belief and action. A closer look at relevant
theories and evidence, however, complicates the picture of
epistemic cognition in múltiple ways.

Complexities of Epistemic Cognition

Epistemic cognition cannot be understood simply as a three-
stage developmental sequence proceeding from objectivism to
subjectivism to rationalism. In this section I note several ways
in which this picture is too simple (Greene et al., 2016;
Moshman, 2015). I conclude, however, that the three-stage
model remains fundamental not only to understanding
epistemic cognition but more generally to understanding the
nature of rationality and development.

Complications and More Complications

One major complication of the three-stage model of epistemic
development is the fact that the three-stage sequence can be
seen both in childhood and in development beyond childhood.
As discussed, Chandler et al. (2002) proposed that epistemic
development is best understood as a developmental spiral in
which the transition from objectivism to subjectivism to
rationalism occurs twice, first at a “retail” level of particular
cases and then at a “Wholesale” level of knowledge in general.
But the complexities of epistemic cognition and development go
beyond this.
A second complication is that there are substantial individual
differences in the rate and extent of development, especially
with respect to the second cycle. Even in childhood,
development proceeds more rapidly in some children and less
rapidly in others. Nevertheless, virtually all children achieve a
basic theory of mind by their fifth birthday and then proceed
beyond that toward an increasingly constructivist theory of
mind, with age-related milestones along the way in their
understanding of viewpoint diversity, relativity of perspective,
and the possibility of rational judgment. Development beyond
childhood, in contrast, is much less age-related and universal.
Whether and at what rate it proceeds varies greatly across
individuáis, depending in large part on social contexts that may
promote or hinder epistemic development (a matter to which we
retum in Chapters 7 and 8).
To further complícate matters, general stages of
development cannot account for the evidence that epistemic
cognition is to a large extent domain-specific. As discussed
earlier in this chapter, epistemic development in the domain of
logic is a matter of increasing comprehension of the logical
necessity of deductive inferences. Although the three-stage
model accounts for the development of causal, principled, and
precedent-based reasoning, which are central to the domains
of science, morality, and social convention, respectively (see
Chapter 3), it does not seem applicable to the development of
metalogical understanding (Moshman, 2015). Even where the
three-stage model applies, moreover, development may
proceed at different rates in different domains and for different
topics within domains, and may cease at different stages. The
transition from objectivism to subjectivism, for example, may
occur earlier with respect to matters of sociology, morality, or
history than with respect to matters of physics or chemistry
because the role of subjectivity is more obvious in the former
cases. But subjectivism may persist longer in the former cases
than the latter and may never be overcome, whereas
subjectivism in the physical sciences, if achieved at all, may be
more quickly replaced by a post-subjectivist rationalism. Such
differences, and their educational implications, have been
explored in many studies (Greene et al., 2016).
Finally, as dual process theories remind us, individual
judgments and behavior in any particular context reflect a
variety of automatic processes and intuitions, with results that
may or may not reflect the individual’s most sophisticated
epistemic insights. Optimal reasoning based on one’s best
epistemic understanding is possible, but reasoning is in large
part a function of heuristics and biases automatically activated
by subtle features of context and task demands. Epistemic
development enhances the quality of our best reasoning, but
regardless of epistemic level, we do not always function at our
epistemic best.

Recognizing Subjectivity and Transcending

Subjectivism
Despite all the complexities of epistemic cognition, the three-
stage sequence from objectivism to subjectivism to rationalism
shows up in study after study, plays a major role in reasoning,
and seems crucial to understanding human rationality and
development. Following objectivism, the sequence consists of
two transitions. The first transition involves the recognition of
inherent subjectivity, with the subjectivist implication that what
appeared to be the truth is simply our point of view, which is
one of many points of view. The second transition recognizes
the possibility of rational judgment despite subjectivity, thus
transcending subjectivism to restore a role for objectivity.
With respect to reasoning, the recognition of subjectivity has
the potential to undermine our motivation (Kuhn, 2005). Why
think hard when all ideas are just subjective opinions, no better
or worse than any other opinions? In transcending subjectivism,
we see that there is reason after all to pursue the truth, even if
it is not so readily and permanently achieved as we once
thought. Advanced epistemic cognition is important not only in
directing our reasoning but also in motivating it.
The three-stage sequence is consistent with a general
conception of rationality as metasubjective objectivity
(Moshman, 2011a). Objectivity is achieved not by overcoming
all subjectivity but rather by reflecting on and compensating for
particular subjectivities. New subjectivities, and associated
limits on our objectivity, are always waiting to be discovered,
and then to be overcome, leading to greater objectivity. But
even when we transcend a particular subjective perspective,
we do so from some meta-perspective, which is itself
subjective. Transcending subjectivity is an ongoing
developmental process, not a final achievement. We will
address this in more detail in Chapter 7.

Conclusión
In this chapter we have supplemented the earlier consideration
of various forms of reasoning (Chapters 2 and 3) with a direct
look at metacognition, the basis for thinking, and especially the
subset of metacognition known as epistemic cognition, the
basis for reasoning. We have seen that metacognition,
including epistemic cognition, develops rapidly over the first 11
or 12 years of life, with the potential for further development,
especially in epistemic cognition, long beyond that. This is fully
consistent with the pattern seen in the development of various
forms of reasoning. As we will discuss further in Chapter 7,
cognitive development can be seen as progress toward
rationality, thus distinguishing it from other cognitive changes. A
distinction can be made between basic development, referring
to the universal progress in rationality across the first 12 years
of life, and advanced development, referring to further progress
that is highly variable across individuáis.
But there remains something missing in this picture. As is
typical in the literatures of cognitive and developmental
psychology, we have been looking almost exclusively at the
reasoning and epistemic cognition of individuáis. We now tum
to argumentation, which I construe as collaborative reasoning.
5
ARGUMENTATION AS
COLLABORATIVE REASONING

We have seen in Chapters 2 through 4 that, over the course of

childhood, we all construct múltiple forms of reasoning—
including logical, scientific, principled, and precedent-based
reasoning— and the metacognitive abilities and epistemic
insights that support such reasoning. By the age of 12 or 13
years, all normal humans have rational abilities and
understandings at a level beyond that of any child of 8 or 9
years, though such children are themselves functioning at a
level beyond any 4-year-old, who in turn is the product of four
years of ongoing development.
It is equally clear, however, that teenagers and adults,
regardless of age, often fall far short of optimal reasoning.
There are substantial individual differences in the levels of
rational competence achieved, but even those most capable of
advanced reasoning and epistemic understanding often reach
conclusions guided largely by automatic inferences and biases.
Dual process theory has been presented as a theoretical
framework that accounts for both human rational competence
and the routine and systematic failures in actual human
reasoning.
In recent years, many psychologists and philosophers have
argued for reconceptualizing reasoning as a fundamentally
social process of argumentation (Laden, 2012; Mercier &
Sperber, 2017). Hugo Mercier and Dan Sperber (2017), in
particular, have proposed that reasoning evolved to serve
social purposes, thus explaining the well-documented failures
of individual reasoning. Mercier and Sperber go too far, in my
view, in downgrading individual rationality and the role of logic
(Moshman, 2018; see Chapter 7). But they are right that
argumentation is crucial to reasoning in ways that the study of
reasoning has largely failed to address. In considering
argumentation in this chapter, I am not just moving to a new
topic but expanding the discussion to a whole new dimensión of
reasoning.
Argumentation is a rational social process of generating and
responding to reasons. That is, it is a process of reciprocal
justification. Argumentation often takes the form of debate, in
which each side attempts to prevail by showing the other to be
wrong. Even then, argumentation is a rational process in that it
relies on reasons (Aikin & Talisse, 2019).
But argumentation is not only, or even primarily, a matter of
debate. My concern in this chapter is with what I deem the ideal
form of argumentation, a process of collaborative reasoning
aimed at achieving the truth, or at least the most justified
conclusión, which may differ from any of the initial positions.
I begin with a case of collaborative logical reasoning, in
which groups tum out to be more logical than individuáis. I then
review research showing that groups commonly show better
reasoning than individuáis. But group reasoning, obviously, can
also go wrong. The final portion of the chapter considers the
circumstances under which collaborative reasoning is most
likely to be productive and the ways we can enhance its
rationality.

Collaborative Reasoning in a Logic Task

In this section I report a study of logical reasoning in which
groups performed more rationally than individuáis. The study
involved the selection task (presented briefly in Chapter 2 and
again in more detail below), which was introduced in the 1960s
and quickly carne to be seen as a major challenge to usual
conceptions of human rationality (Wason & Johnson-Laird,
1972). The task requires participarte to select the card or cards
(out of a set of four) that must be turned to test a simple
hypothesis. The solution is to turn the two cards that could
falsify the hypothesis (one of which is a less obvious choice
than the other) and not turn either of the other two cards
(though one of those is tempting). Múltiple studies showed that
only about 10% of college students selected the correct two
cards (Wason & Johnson-Laird, 1972). Performance was better
on some variations of the task, leading to theoretical
controversy about how to account for the complex pattern of
results that accumulated over the 1970s and 1980s (Moshman,
2011a).

The Selection Task and Me

My personal connection to the selection task goes back to the
1970s when I was a gradúate student in developmental
psychology doing research on the development of logical and
scientific reasoning in adolescence. Research on the selection
task seemed to suggest that most adults did not understand the
basic logic of hypothesis testing that was central to Piaget’s
conception of formal operations, the stage of cognitive
development attained in early adolescence. My dissertation
research (Moshman, 1979) used a variation of the selection
task to show that there was developmental progress in
hypothesis testing over the course of adolescence but that
failures of logic remained common even among college
students.
I also worked the selection task into my teaching and for
decades routinely introduced it by having students come up
with individual answers and then challenging them to discuss
the task with each other to reach a consensus solution (with the
threat that no one could leave until they did). My only
intervention was to record all initially selected card
combinations on the blackboard, tally the number of students
selecting each, and occasionally survey the class to document
shifts of opinion on the blackboard. Although the vast majority
of students initially showed the same incorrect selection
patterns routinely seen in research, most classes converged on
the correct selection pattern well before the end of the class
period. Peer pressure was obviously not prevailing given that
initially popular response patterns were ultimately rejected.
What I observed instead was a rational social process in which
individuáis presented differing views about which cards should
or should not be turned and a variety of arguments for or
against these views, with the strongest arguments, in most
cases, ultimately prevailing.
In the early 1990s I sought to document the remarkable
success of group reasoning on the selection task in
collaboration with Molly Geil, a student who conducted the
research described below and wrote up the quantitative part of
the results for her master’s thesis. We presented the study at
the 1994 meeting of the Jean Piaget Society in Chicago but
had difficulty getting it published. It was rejected by an
education journal and then by a journal focusing specifically on
Science education, in large part because it did not provide
evidence of student learning.
By the time of the second rejection, however, I had become
aware of a new journal entitled Thinking and Reasoning that
seemed an ideal outlet for the study, which, we had tried to be
clear, was about group reasoning, not individual learning. The
two initial reviewers identified themselves as Henry Markovits
and David O’Brien, both major contributors to the study of
logical reasoning. Both appreciated the quantitative
documentation of rational group performance on the notoriously
difficult selection task but wanted to see more qualitative
analysis of the actual processes of argumentation that led to
group success. A revised versión of the manuscript was finally
accepted for publication in 1997 and published the following
year.
The increasing interest in group reasoning in recent years is
reflected in the citation rate of Moshman and Geil (1998), as
shown by Google Scholar (as of February 2020). In each of the
10 years following its publication (1998-2007), the article was
cited between one and seven times, with an average rate under
four citations per year and no reason to expect that figure to
increase. In each of the next 7 years (2008-2014), however,
the article was cited between 5 and 17 times, averaging about
13 citations per year. Over the subsequent 5 years (2015-
2019), citation rates ranged from 18 to 30, with an average over
23 citations per year, reaching 30 in 2019.

A Study of Group Reasoning

The Moshman and Geil (1998) study engaged 143 college
students, mostly education majors, in reasoning about the
original versión of the selection task. Each student was
presented with a sheet of paper picturing four cards that
showed, from left to right, E, K, 4, and 7. Each card, they were
informed, had a letter on one side and a number on the other.
Below the cards was this hypothesis: “If a card has a vowel on
one side then it has an even number on the other side.”
Students were asked to select the card or cards they would
need to turn to determine conclusively whether the hypothesis
was true or false for the set of four cards, and to justify their
response in writing.
The correct response is to select E (a vowel) and 7 (an odd
number). The vowel must be turned because an odd number
on the other side would falsify the hypothesis; the odd number
must be turned because a vowel on the other side would falsify
the hypothesis. Neither of the other cards needs to be turned
because neither could falsify the hypothesis. If the hypothesis is
not falsified by turning the E and 7 cards, then it is true
regardless of further evidence.
The challenge of this task is that it requires insight and effort
to recognize the crucial importance of falsification and consider
each card with that in mind. In the absence of sufficient
metalogical understanding and self-regulation, we succumb to
a natural verification bias, seeking to support the hypothesis (if
vowel, then even) by turning the E (a vowel) to look for an even
number, and perhaps also the 4 (an even number) to look for a
vowel. Thus, the most common responses to the selection task
are to select only the E or only E and 4. The challenge is to
recognize that the 7 must also be turned because it could falsify
the hypothesis, and the 4 need not be turned because it could
not falsify the hypothesis.
Of the 143 participants, 32 were asked to solve the task
individually. The remaining 111 were assigned to 20 groups of
5 or 6 students, with each group asked to reach a consensus.
In half of those groups, each student was asked to solve the
task individually prior to group discussion. All sessions were
videotaped.

Evidence for Coilective Rationality

Results for the 32 students who only solved the task
individually were consistent with the literature. Although they
were told to take as much time as necessary and understood
that they needed to remain for the allotted 50 minutes to earn
class credit and would be expected to justify their response in
writing, only three selected the correct response of E and 7.
The most common responses were E and 4 (selected by 14)
and E only (selected by 5), with no other response pattern
selected by more than two individuáis.
For the 20 groups, on the other hand, the results were
dramatically different. All 20 reached a consensus within 45
minutes. The correct response of E and 7 was selected by 15
groups, with 3 selecting E only; 1 selecting E, 4, and 7; and 1
selecting all four cards. Whether students provided individual
responses prior to group discussion turned out to make no
difference; correct responses were provided by eight of the ten
groups in which individuáis first responded individually, and
seven of the ten in which they did not.
Further analysis focused on the ten groups for which
individual responses were provided prior to discussion. In no
case did all students in a group make the same initial
selections. In seven groups there was at least one person who
initially selected the correct combination of E and 7. This was
never the most common initial response, but it nevertheless
prevailed in five of these seven cases. Even more interesting,
the correct response also prevailed in all three of the remaining
groups despite the fact that not a single individual in those
groups initially selected that combination of cards.
Examination of the videotapes showed that students
routinely challenged each other to justify their selections and
encouraged each other to consider consequences and
alternatives. Final selections reflected voluntary agreement
based on emerging logical insights rather than passive
conformity to views favored by the most peers or by someone
deemed more expert than others. To ¡Ilústrate the rational and
constructive quality of discussion, we summarized and quoted
at length from the discussions in two of the three groups (out of
the ten that made initial individual selections) that reached the
correct conclusión despite no individual initially doing so.

Collaborative Reasoning: Group 1

Group 1 consisted of Alice, Ben, Carol, Dan, and Earl (these
are not their real ñames, but Alice and Carol were female and
the others male). Alice initially selected the E, 4, and 7 cards to
be turned; the others all selected just E and 4. As discussion
began, everyone agreed that E should be turned to see if there
was an even number on the other side, 4 should be turned to
see if there was a vowel, and the other two cards need not be
turned. Alice laughed off her original selection of the 7 as a
mistake and joined Ben, Carol, and Dan in reinforcing each
other’s statements as to the relevance of E and 4 to the
hypothesis and the irrelevance of K and 7. But then Alice
reconsidered:

Alice: Maybe 7 has a vowel on the other side.

Ben: It could, but as far as this hypothesis here, it just doesn’t
matter.
Alice: But if it has . . .
 Dan: It just says if it has a vowel on one side.
Atice: Yeah, but it says if it has a vowel on one side, then it
has an even number on the other side.
Dan: So maybe we’re wrong.
Carol: [surprised and excited] Oh, that’s truel

Everyone agreed that this complicated the issue, leading

Ben to propose that turning the E and 4 would “test” the
hypothesis, and finding an even number and a vowel,
respectively, on the other sides would “support it as opposed to
proving it,” because “we couldn’t prove it unless we turned over
all of them.” But after some discussion of the mechanics of
turning and keeping track of four cards, Dan suddenly
questioned the need to turn K, and Alice agreed it was
unnecessary because the hypothesis was only “concerned with
vowels.” Then:

Carol: But what if it has an even number on the other side?

Dan: It doesn’t say anything about. . .
Alice: It doesn’t say that if it’s a consonant. . .
Dan: It just says if it has a vowel. It doesn’t say if it has a
consonant it can’t have any of them.
Carol: That’s true
Dan: I don’t think we have to turn over K.

Apparently convinced, Carol now wondered whether 4 is

also unnecessary, noting, “it doesn’t say that if it has an even
number on one side it has a vowel on the other,” and then
concluding, “really we don’t need to turn over 4.” Ben joined her
in explaining that the conditional hypothesis is not reversible:

Carol: You don’t have to turn over 4 because it [the

hypothesis] says if it has a vowel on one side it has an
even number on the other side.
Ben: It doesn’t say if it has an even number on one side it
has a vowel on the other.

Alice and Dan continued to insist that the 4 must be turned,

however, and when Carol tried to explain that the 4 “could have
a consonant on the other side and it still wouldn’t . . Alice
interjected, “Yeah, but we need to check it because it is an
even number, so we have to find out if it has a vowel on the
other side.” Carol responded, “Yeah, I guess,” and Alice,
defending the choice of E, 4, and 7, went on to explain that the
E must be turned because “we have to find out if it’s an even
number,” and the 7 must be turned because “we need to find
out if it’s a consonant or a vowel because if it’s a vowel then it’s
false.”
In the literature of the selection task, the response of E, 4,
and 7 has long been seen as showing a partial insight into the
crucial importance of falsification (Wason & Johnson-Laird,
1972). Alice’s explanation illustrates this nicely. On one hand,
she still explains the selection of E and 4 on the basis of a
search for cards combining a vowel with an even number
(which seemingly verify the hypothesis If vowel, then even): we
turn the vowel to look for an even number and we turn the even
number to look for a vowel. But she has recognized, and
convinced the others, that the 7 must also be turned, even
though it could not turn out to be a vowel with an even number,
because it could turn out to be a vowel with an odd number,
which would falsify the hypothesis. The partial insight is
recognizing that any card that could falsify the hypothesis must
be turned. What remains is the full insight of recognizing that
only cards that could falsify the hypothesis need to be turned.
But Ben and Carol now had at least an inkling of this,
enough so they continued to question the need to turn the 4.
When Ben wondered about the implications of it having a
consonant on the reverse side, Carol suggested that turning the
4 would “prove” the hypothesis if there were a vowel on the
reverse side and “wouldn’t do anything” if there were a
consonant. “It would either prove it or it wouldn’t do anything,”
she concluded. “The 7 and the E are the only ones that can
disprove it.” There was a very long pause at this point, during
which Alice looked back at the instructions, murmuring “to
determine conclusively.” Then:

Dan: Okay, we have to turn over E for sure, right?

Carol: Yeah.
Dan: Because it has a vowel on one side and we need to find
out if it has an even number on the other. K we don’t have
to worry about, because it doesn’t say anything about. . .
Ben: . . . consonants . . .
Dan: . . . having a consonant.
Atice: It doesn’t say if it has a consonant it’s odd, or
whatever.
Dan: And 4 . . .
Ben: I think we need to turn 4.
Dan: I think we have to turn over 4 because . . .
Earl: It’s the same as E, really.
Alice: It’s the same as E, yeah, we know it’s an even number
so we have to find out if it has . . .
Dan: Well, maybe we don’t. [He pauses, then proceeds
slowly, with Carol nodding and murmuring assent.] If it
has a consonant on one side it doesn’t matter if it has an
odd or an even number. So it really doesn’t matter if we
look at 4. Does it? Do you see that?
Alice and Ben [simultaneously]: I see what you’re saying.
Dan: It can tell us where that’s right, but it can’t tell us it’s
wrong.
Carol: Yeah.
Dan: And 7 I think we have to turn over. 'Cause we need to
find out if that has a vowel.
Carol: Because it can prove right or wrong.
Dan: Because it can prove it wrong.

The group then confirmed that agreement had been reached

and reflected on the process that had generated this
consensus:

Ben: So, are we narrowing it down to E and 7 this time?

Dan: I think so.
Carol: I think it should be E and 7 now.
 Dan. I do too.
Ben: That’s pretty interesting to watch us all concur.
Alice: I wouldn’t have come up with this if we hadn’t, you
know, talked about it.
Carol: I know, I was totally set on E and 4.
Ben: We all were.

Recall that Alice originally selected 7 in addition to E and 4.

With respect to 7, one person eventually succeeded in
convincing the group that it must be turned. But everyone fully
agreed at the start that of course the 4 must be turned, just as
surely as the E must be turned and for the same reason: to
verify the hypothesis. With respect to the 4, then, the group
managed to talk itself out of a unanimous misconception. Each
of the five students independently wrote on the final task sheet
that only E and 7 should be turned and provided a written
explanation consistent with the group’s final arguments.
Who was responsible for the final selection of E and 7? Four
of the five students (everyone but Earl) fully shared
responsibility for the group response. Alice was the only
student who initially proposed turning 7 and was the first to
point out that it could have a vowel on the reverse side. Ben
was the first to distinguish “supporting” a hypothesis from
“proving” it, though he mistakenly argued that proof required
turning all four cards. Dan then took the lead in showing that K
need not be turned. Carol expanded on Dan’s argument to
make the case that 4 also need not be turned, initially
convincing Ben but not Alice or Dan. Alice then noted that 7
could falsify the hypothesis. Carol noted later that 4 could not
disprove the hypothesis whereas E and 7 could. Finally, Dan
established that the group had agreed to turn 7 (in addition to
E) and provided what turned out to be the decisive argument
for not turning 4. Through collaborative reasoning, the four
students generated a higher level of understanding than any of
them alone achieved.
Collaborative Reasoning: Group 2
Group 2 consisted of Frank, Gwen, Hal, Ike, and Jay (again the
ñames are fictitious, but Gwen was female and the others
male). Hal initially proposed to turn only the E; Frank proposed
E, 4, and 7; and the other three proposed E and 4. Jay began
the discussion by proposing to turn just E and 4; Gwen agreed.
Frank agreed on turning E and 4 but noted his selection of 7 as
well. He argued that E must be turned because it is a vowel,
and K need not be turned “because we’re not concerned with a
consonant having an odd or even; we’re talking about just the
vowel card.” With respect to the other side of the 4, he started
to explain, “if that’s a vowel, then . . but Hal interrupted, “see,
that doesn’t have to be a vowel, though.” “Yeah, it does,”
responded Gwen, but Hal continued to explain his view, noting
the irreversibility of the hypothesis: “All that that’s saying is that
all the vowels have to have even numbers on the other side.
But it’s not saying that, like, the number 4 has to have a vowel
on the other side.” Frank and Gwen now agreed, and Hal
continued:

The only thing that you have to turn over is E, but by

turning over E you’re just supporting the hypothesis,
you’re not proving. The only way you can prove that thing
right is if you have all the vowels facing up and then you
turn all of them over. You know what l’m saying? That’s
the only way you can prove that hypothesis correct.

Frank now asserted, “you’d have to turn over 7.” Hal

responded, “you’d have to turn over all of the vowels and they’d
have to be facing up too. You couldn’t have one facing the
other way because you wouldn’t know.” Gwen and Jay agreed,
and the discussion moved on to the requirements of the task
and then returned to the need to turn E and the irrelevance of K
and 4. Frank was not done, however, and finally got his point
across:

Frank: But we would have to turn over the 7, then. Because a

 vowel, like an I or an O, would throw us off.
Gwen: You’re right, we wouldn’t be sure of the 7.
Frank: So, the E and the 7?
Gwen: Yeah.
Hal: Yeah, that’s a good point.
Gwen: It is a good point. [Pause.] So we’ve decided E and 7
because we want to know if E has an even number under
it and if 7 has a vowel under it or a consonant.
Frank: That would disprove it.
Gwen: Right. That’s interesting.

Frank and Gwen concluded the session by clarifying that if

the 7 has a vowel under it, then the hypothesis is false, which is
why the 7 must be turned. Each of the five students then
indicated the choice of E and 7 on the final task sheet and
provided an explanation consistent with the group’s final
argument. Three of the five played major roles in the
discussion. Frank was a persistent advócate for turning the 7,
finally convincing the others. Hal made the case that the 4 need
not be turned, though initially as part of an argument for just
turning E. Gwen insisted at first on turning 4 but was convinced
by Hal’s argument against this. Later, she was the first to
appreciate the significance of Frank’s argument for turning 7
and helped him clarify it and convince the others (Moshman &
Geil, 1998).

Additional Evidence for Collective Rationality

Extensive research in cognitive, developmental, social, and
educational psychology demonstrates the rational advantages
of collaborative reasoning— that is, argumentation in
collaborative contexts. I begin this section with a brief overview
of theory and research concerning the valué of collaborative
cognition and then turn specifically to studies that, like
Moshman and Geil (1998), show the rationality of group
reasoning in comparison to individual reasoning, including
some that look specifically at the conditions favoring collective
rationality.

Rational Advantages of Collaborative Cognition

Research on collaborative learning, memory, judgment,
problem solving, decision-making, and reasoning provides
extensive evidence for the cognitive valué of collaborative
cognition, including peer argumentation (Baillon, Bleichrodt,
Liu, & Wakker, 2016; Bearison & Dorval, 2002; Cur§eu, Mesiec,
Pluut, & Lucas, 2015; Doise & Mugny, 1984; Laughlin, 2011;
Mercier, Boudry, Paglieri, & Trouche, 2017; Mercier & Sperber,
2017; Moshman & Geil, 1998; Piaget, 1932/1965b; Tomasello,
2014; Zillmer & Kuhn, 2018). With respect to the Wason
selection task, for example, research with individuáis has
generally shown that cues about the importance of falsification
are less helpful than one might expect, but it appears that
groups profit more from such cues than do individuáis
(Augustinova, 2008). Although the selection task is likely
beyond young children (more below), there is no doubt that
children engage in meaningful argumentation beginning in the
preschool years and show rapid progress in collaborative
reasoning (Domberg, Kóymen, & Tomasello, 2018).
Much of the research on peer argumentation has focused
especially on showing that collaborative reasoning promotes
the learning and development of those who engage in it (Baillon
et al., 2016; Doise & Mugny, 1984; Laughlin, 2011, Chapter 7;
Piaget, 1932/1965b). Part of what develops, moreover, is the
ability to engage more productively in collaborative cognition of
various sorts. This has important implications for both
development and education, to be addressed in Chapters 7 and
8 respectively.

Group Rationality
I turn now to the more specific question of the rationality of
groups engaged in collaborative reasoning as compared to the
rationality of individuáis reasoning on their own. It is well
established that under the right circumstances group
performance is more competent than that of the average
individual. Patrick Laughlin (2011) reviewed and organized the
literature on group problem solving, broadly defined to include
decision-making, reasoning, and related cognitive processes,
including decades of his own research. He concluded (Laughlin
2011, pp. 141-142) that group tasks can be ordered on a
continuum from “intellective” (involving matters of logic and/or
clear facts) to “judgmental” (where there is no “generally
accepted demonstrably correct answer”). In judgmental tasks a
consensus may suffice; in intellective tasks the objective is
truth. The intellective/judgmental continuum reflects an
underlying continuum of “demonstrability.” High demonstrability
requires not only a shared logic and sufficient information, but
also the ability and opportunity of correct group members to
convince others and the ability of the others to recognize
correct solutions. The greater the demonstrability of a particular
solution, the smaller the proportion of group members needed
to convince the group to adopt that solution. Demonstrability
makes it possible for a minority to convince a majority, and high
demonstrability makes it possible for a single correct individual
to convince everyone else. Thus for highly intellective tasks,
groups perform as rationally as their most rational member.
In the most intellective tasks, Laughlin (2011) noted, groups
may even outperform the best of an equivalent number of
individuáis, using more complex and insightful strategies that
enable them to succeed on fewer triáis. This is consistent with
Moshman and Geil’s (1998) finding that groups may identify the
correct solution to the selection task even if none of their
members were able to do so individually. Here we see the
strongest evidence for the constructive power of peer
argumentation.
How could a group construct a better answer than any
individual within it? The final resolution may be a combination
of the best ideas produced by individuáis, even if only one
person originally comes up with some of those ideas. In
Moshman and Geil (1998), for example, Group 2 ultimately
accepted Frank’s argument for turning the 7 and Hal’s
argument against turning the 4, though each was initially alone
in his view. Similarly, Group 1 concluded that the 7 must be
turned, although only Alice initially made this selection.
It is noteworthy, however, that Group 1 also reached a
consensus against turning 4, despite the fact that its five
members unanimously agreed at the start that the 4 must be
selected. No single individual, moreover, was responsible for
this new understanding. Carol was the first to propose that the
4 should not be turned, but her arguments drew on Dan’s case
against turning K, Ben’s earlier distinction between “supporting”
and “proving” the hypothesis, and Alice’s insight with respect to
7 about the crucial importance of falsification. It was Dan,
moreover, who ultimately convinced the others, with Carol’s
assistance, that 4 need not be turned. The group decisión not
to turn 4 was not a choice among conflicting individual ideas. It
was a novel insight co-constructed by four members of the
group. At its most Creative and powerful, the rationality of
groups is not just a matter of selecting and combining the best
ideas. It also involves the social construction of new ideas.

Conditions of Collective Rationality

Research looking specifically at the conditions of collective
rationality has shown that the valué of collaborative reasoning
is due not just to the pooling of ideas but to the process of
argumentation (Baillon et al., 2016). Cur§eu, Jansen, and
Chappin (2013), for example, divided 617 college students into
176 small groups— 91 of them “collaborative” and 85
“consultative.” Each group was presented with the same set of
ten decision-making tasks adapted from tasks commonly used
in the literature on human heuristics and biases (introduced in
Chapter 1). In the present versions, the logically or statistically
normative solution in each case was presented as one of
several options. Collaborative groups were instructed to reach
a consensus choice for each task. In the consultative groups, a
group leader was randomly selected to make the group choice
on each task after getting input from each of the other group
members, who were not permitted to discuss the task with each
other.
Analysis focused on the extent to which groups achieved
“synergy,” defined as “an objective gain in group performance
as compared to summed individual performances” (Cur§eu et
al., 2013, p. 2). A group process can be said to show weak
synergy if the group performs better than its average member,
and strong synergy if it performs better than its best member.
Cur§eu et al. (2013) predicted that the collaborative groups
would show greater synergy than consultative groups because
the opportunity to engage in discussion would enhance rational
decision-making. This prediction was confirmed. The
collaborative groups generally achieved at least weak synergy
and were significantly more likely than the consultative groups
to do so. Moreover, the collaborative groups, on average, carne
significantly closer than the consultative groups to strong
synergy, though most did not achieve it.
This research provides specific evidence for the role of
collaborative reasoning in the rationality of groups. With a slight
shift in terminology, we can say that, at least under some
circumstances, microdemocracies outreason
microdictatorships, raising the question of whether and how we
might extend the present analysis to the societal level. We take
up that question in Chapter 6.
Researchers have also addressed the question of access to
relevant information by designing studies that systematically
manipulated such access. Laughlin (2011) reviewed research
in which information favoring a suboptimal decisión was
distributed to all group members, whereas information favoring
an optimal decisión was provided only to individuáis. He found
a clear pattern: Such groups generally made suboptimal
decisions because they focused on the shared information and
failed to integrate the additional information known only to
particular individuáis. But the research showed two ways of
enhancing rationality. First, groups were more likely to make
the optimal decisión if the task was presented as a problem to
be solved rather than a judgment to be made. Second, groups
also did better if individuáis were appointed as “experts” on the
unique information each possessed, thus “providing social
validation for information known only to one person” (Laughlin,
2011, p. 142). This research also raises questions for Chapter
6. How can we enhance the ability of democratic institutions to
function in a problem-solving mode and take relevant expertise
into account?

The Rational Social Ideal

Despite the evidence for group rationality, groups can also be
irrational, sometimes less rational than their average member,
sometimes even doing terrible things that most or all of their
members wish they hadn’t (Moshman, 2007b, 2011b). This
raises a series of questions, some of them addressed in part by
the research just reviewed: What aspects of group functioning
promote rationality? How can we arrange and support rational
groups? How can we extend this to rational social systems?
In this section I consider five features of the two groups from
Moshman and Geil (1998), and other such groups, that seem
important for the rational outcomes observed in that study.
These features are mostly obvious things that would usually go
without saying, but this is the place to make them explicit. The
discussion in each case notes extensions to democratic social
systems, to be considered further in the next chapter and
beyond. The focus for now, however, is the nature of these
groups as rational social ideáis.

Small Group
For a start, the two groups were small enough that everyone
could get repeated opportunities to speak to the group,
everyone could hear from everyone else, and individuáis with
differing views could engage in sustained argument that others
were free to join. No one was required to speak, but everyone
was free to do so múltiple times, and most actively participated.
Obviously, the opportunity for this sort of free discussion
declines as the size of the group increases. Group discussion
procedures, such as Robert’s Rules of Order, enable groups of
several dozen to approximate the ideal of free discussion. For a
group of several hundred, however, it is inevitable that
opportunities to speak will be strictly limited, even if everyone
ultimately gets a turn; moreover, sustained argumentation
among a small number of individuáis will be impossible due to
the number of people waiting their turn to speak.
For a larger social system, face-to-face discussion among all
persons is obviously impossible. This raises serious questions
about the possibility of scaling up the benefits of group
rationality to social systems. Can the members of a society be
said to be engaged in collaborative reasoning? How can a
social system operate rationally?

Individual Competence
A second characteristic of the two groups is that all participants
were college students. How would younger individuáis have
done? Based on research on the development of logical
reasoning in individuáis, it seems likely that rational group
performance on the selection task could be found even if the
members of the group were young teens, but not if they were
children under 11 or 12 years. The problem is not just that none
of the children would initially have the correct solution. The
problem is that the children would lack the hypothetico-
deductive reasoning necessary to grasp the logic of the task
even if (as with the adult groups in Augustinova, 2008) they
were provided with helpful insights.
In addition to developing competence in logical and other
forms of reasoning, children also show substantial and
predictable development in argumentation ability over the first
11 or 12 years of life. Argumentation competence, it should be
clear, is social as well as cognitive. To argüe well, one must (1)
recognize the need to provide reasons for one’s ¡deas, which
includes understanding what counts as a reason and what
others will find convincing; (2) recognize the need to refute
alternative ideas, which requires attending to others and
understanding their ideas; and (3) recognize the possibility of
higher-order argumentation, which includes responding to
refutations of one’s own ideas, anticipating such refutations,
acknowledging strengths and weaknesses in a variety of ideas,
including one’s own, and working with others to construct better
ideas. Children make substantial progress in the first set of
competencies beginning in the preschool years, with ongoing
development across childhood in the first two sets, and show
some capacity for higher-order argumentation by the age of 11
or 12 years. People of all ages fall far short of ideal
argumentation, however, with substantial differences across
individuáis in the rationality of their argumentation, due in large
part to differences in epistemic cognition. Beyond the age of
about 12 years, individual differences in rational argumentation
are more related to education and experience than to age
(Kuhn, 1991,2005).
Group rationality, as we have seen, is not just the sum of the
rationality of the individuáis composing the group (Baillon et al.,
2016; Cur§eu et al., 2013; Cur§eu et al., 2015; Laughlin, 2011).
Nevertheless, even in collaborative reasoning, individual
competence matters. For social systems concerned with
argumentation, education aimed at the promotion of individual
rationality is important, though part of a larger picture.

Respect for Reasons

A third characteristic of both groups is respect for reasons,
which was seen throughout and created an ideal process of
argumentation. Participants provided reasons for their
selections, attended to the reasons provided by others,
responded to those reasons, and considered objections to their
own reasons. They attempted to convince others to change
their selections and were open to being convinced to change
their own, but no one was pressured or forced to conform to a
majority view just to create a unanimous group conclusión.
Aiming for a correct solution, the consensus they sought to
achieve was a justified consensus, not just an end to the
argument. Outside the realm of logic, it may be more difficult for
groups to approximate ideal rational functioning, but small
groups may nevertheless encourage greater respect for
reasons than would be seen in individuáis on their own.
At the level of social systems, respect for reasons is
complicated not only by group size but by considerations of
ideology, identity, and expertise. When argument turns to
complex social issues, ideologies and identities may commit us
to favor or oppose particular proposals without adequate
consideration of relevant reasons. There are also complex
issues concerning the role of experts and the democratic use of
expertise. We retum to all of these matters in the next chapter.

Respect for Persons

Reasons are, of course, provided by persons. Respect for
reasons requires respect for persons, which is also manifested
in the present two groups. Argumentation in each was
conducted in such a way that everyone had ampie opportunity
to speak, though no one was required to do so; everyone’s
ideas were taken seriously, which is not to say they were
supported; and participants disagreed with each other’s ideas
without attacking each other personally. As we have seen,
there is evidence that groups operating on a basis of equal
respect for all participants may operate more rationally, on
average, than those set up with a single authority receiving
inputfrom others (Cur§eu et al., 2013).
Liberal democracies are founded on a moral basis of equal
respect for all persons. The case for equal respect, however, is
not just moral or governmental. There is an epistemic case for
equal respect for persons based on its contribution to rational
argumentation and the progress of knowledge. In the next
chapter we consider the extensión of this insight to social
systems.

Intellectual Freedom
Finally, the two groups that have been the focus of this chapter
were characterized by intellectual freedom, not as an ideology
but as a natural condition of rational argumentation. Nobody
needed to condemn censorship because it was taken for
granted that all persons were free to express all views about
the cards to be selected.
But the implicit commitment to intellectual freedom was more
subtle than just a commitment to free speech. Participants did
not feel free to talk about anything they pleased. Without
anyone setting any limit, the conversation naturally remained
on topic, focused on the logic of the task at hand. The
participants shared a common goal of determining the pattern
of cards to be selected and understood that to maximize their
chances of identifying the logically correct solution they should
limit their speech to the topic at hand but set no limits on
viewpoints regarding the logic of the task or the cards to be
turned. Incorrect views of which cards to turn disappeared not
because those who held them were silenced or intimidated but
because those who held them were convinced to change their
views.
There remains the question of intellectual freedom in social
systems, where, in múltiple forums and institutions, we face the
interplay of ideologies and identities, often with respect to
complex issues requiring special expertise. Over the next three
chapters, we will return to intellectual freedom repeatedly as we
examine its role at a societal level (Chapter 6) and consider its
special importance for development (Chapter 7) and education
(Chapter 8). As we will see, outside the realm of an open forum
for free expression, there are many social contexts where
speech can legitimately be restricted on the basis of topic but
not viewpoint. We will also see that intellectual freedom,
beyond freedom of expression, also includes access to
information and ideas.
Conclusión
Despite the focus of reasoning research on individuáis,
theorists such as Habermas (1990) and Piaget (1995,
1932/1965b) have long argued for the importance of
collaborative reasoning. Even at his most biological, Piaget
(1967/1971) was always clear that logical operations, far from
being just the property of individual organisms, are no less a
property of social interaction:

In the realm of knowledge, itseems obvious that

individual operations of the intelligence and operations
making for exchanges in cognitive cooperation are one
and the same thing, the “general coordination of actions”
to which we have continually referred being an
interindividual as well as an intraindividual coordination
because such “actions” can be collective as well as
executed by individuáis.
(p. 360)

Research has strongly supported these theoretical positions.

But argumentation among individuáis becomes increasingly
difficult as the number of individuáis increases. In small groups,
under the right circumstances, people naturally work together in
what may be called microdemocracies, collaborating to
determine the truth and make the most justifiable choices. But
democratic social systems on a larger scale are more complex
because individuáis cannot all speak to each other directly.
Even where argumentation is possible, it may be undermined
by considerations of ideology and identity and by lack of access
to relevant expertise. We take up these matters in the next
chapter, as we extend the discussion of social rationality to the
societal level.
6
DEMOCRACY AS COLLABORATIVE
RATIONALITY

In this chapter I consider democracy as a social system in

which individuáis collaborate in rational social action. Such a
system requires deliberation, and many theorists of democracy
have argued that deliberative democracy is the ideal form of
democracy. But democracy, especially in its deliberative
versions, requires rational agents. If people are less rational
than is commonly assumed, democracy may not be possible.
And there is plenty of reason for doubt about human rationality.
I begin this chapter with a definition of deliberative
democracy as a form of democracy. Next, drawing on research
and theory reviewed in Chapters 2-5, I argüe that people are
indeed sufficiently rational for democracy to work, and that
psychological research on group rationality (Chapter 5)
specifically supports deliberative democracy as the ideal form
of governance. I then look at issues of ideology, identity,
expertise, and intellectual freedom. I conclude that deliberative
democracy is an ongoing challenge, but not an impossibility,
and there is no better alternative.
For psychologists concerned with human reasoning and
rationality, this chapter extends the discussion beyond the
study of individual cognition (Chapters 2 -4 ) and reasoning in
small groups (Chapter 5) to the rationality of social systems. As
we will see in Chapters 7 and 8, rational social systems not
only reflect the rationality of the individuáis within them but play
important roles in the development and promotion of individual
rationality.
 For theorists of democracy, and anyone else concerned with
democracy, this chapter presents a psychologically-based
visión of deliberative democracy. This picture is generally
consistent with the literature of deliberative democracy that has
flourished in philosophy, law, political Science, and related
disciplines since the 1980s, rooted in neo-Kantian political
moralities, such as those of Habermas (1996) and Rawls
(1971, 2001). The present approach, however, is rooted more
directly in the psychological realities of cognition, development,
and peer interaction.

Deliberative Democracy
Deliberative democracy is a form of liberal democracy that
takes as its ideal the achievement of consensus through
rational processes of argumentation. Deliberative democracy,
as we will now see, is founded on respect for persons and
respect for reasons. It is generally seen as an ideal but, at least
in small peer groups, it is also a social fact.

Democracy: Respect for Persons

Democracy is commonly seen as a system in which individuáis
vote and the majority rules. But majorities may vote to ban
minority religious and political views or to act against minority
ethnic, racial, sexual, or other groups. Rule of the majority is a
threat to minorities. If rule of the majority is all there is to
democracy, then democracy is a danger to minorities of all
sorts, and ultimately to everyone.
Liberal democracy may be defined as a social system based
on equal respect for all persons. Equal respect for all persons
entails a right of each to particípate in democratic decision-
making. This includes a right to have your vote count equally to
others, which is the basis for the democratic principie of
majority rule. But equal respect for all persons also entails
respect for the fundamental rights of each individual, which
constrains what democratic majorities may legitimately do
(Rawls, 1971, 2001). These fundamental rights are typically
specified in democratic constitutions protecting freedoms of
belief and expression, legal equality, and rights to privacy and
due process (Dworkin, 1985, 1996, 2011; Fallón, 2018; Niño,
1996; West Virginia v. Barnette, 1943).
At a societal level, respect for persons requires respect in
the abstract for persons one has never met and will never
meet, with the expectation that one’s respect will be
reciprocated. Thus, liberal democracy assumes equal and
mutual respect as abstract principies (Gutmann & Thompson,
2004). Principies of equal and mutual respect do not require
that people be identical or experience identical circumstances,
but many have argued that democracy is undermined by radical
economic inequality (Taylor, 2019).

Deliberative Democracy: Respect for Reasons

Many theorists of democracy have argued for some versión of
what has been called deliberative democracy (Aikin & Talisse,
2019; Báchtiger et al., 2018b; Elstub & McLaverty, 2014;
Gutmann & Thompson, 2004, 2018; Niño, 1996). A deliberative
democracy is a form of liberal democracy that takes as its ideal
the achievement of consensus through rational processes of
argumentation. As a form of liberal democracy, deliberative
democracy is rooted in respect for persons. What distinguishes
deliberative democracy is that it is equally rooted in respect for
reasons. Respect for reasons, and associated processes of
argumentation, serve to generate better understanding and
better decisions. Such respect is thus in the best interest of the
social system (Estlund & Landemore, 2018; Landemore, 2013).
By recognizing persons as rational agents, moreover, respect
for reasons furthers respect for persons. To respect persons is
not just to allow them to vote. Full respect for persons requires
serious consideration of their beliefs and reasons.
Ideally, deliberation leads to consensus in cases where
consensus can be reached. When it fails to do so, majority rule
remains the basis for making decisions that must be made.
Even when a decisión must be made by majority vote,
however, it should be clear to all that the majority could be
wrong and that dissent may remain justified.

Deliberative Democracy as an Ideal

Deliberative democracy is generally seen as an ideal to which
societies should aspire. No actual society fully conforms to the
ideal of deliberative democracy, so a description of any society
as a deliberative democracy is, at best, an idealization of its
actual functioning. Some societies come closer to the ideal than
others but, even for those that fall far short, the ideal of
deliberative democracy provides useful guidance in identifying
shortcomings and enhancing deliberation and democracy.
Deliberative democracy rests on a conception of persons as
rational agents. This conception is also an ideal. Individuáis,
like societies, routinely fall far short of rational norms. To refer
to people as rational agents is an idealization that highlights an
important feature of human psychology but ignores others.
Even if the idealization misses part of reality, however, it can
serve as an ideal in identifying irrationality and promoting
human development.
Idealizations and ideáis serve important roles (Appiah,
2017). In chemistry, the ideal gas law has long played a central
role in understanding and predicting the motion of gases,
despite the fact that no real gas precisely conforms to its
fundamental assumptions. Similarly, at higher levels of
complexity, idealizations of persons as rational agents and of
social systems as deliberative democracies are important in
understanding human behavior and social functioning. In
addition, at the psychological and social levels, ideáis serve a
normative function. Ideáis of rational agency and deliberative
democracy guide us in recognizing where we fall short in
matters of both truth and morality and in promoting
developmental progress.
Deliberative Democracy as a Social Fact
Deliberative democracy is not just an ideal, however. It is also a
social fact. As we saw in Chapter 5, research on group
functioning shows that, under the right circumstances, small
groups often function as microdemocracies in which individuáis
collaborate through processes of argumentation that fully
respect both persons and reasons. Democratic deliberation at
this level is not just a normative ideal but a very cióse
approximation to actual human behavior. The problem of
deliberative democracy, then, is not whether it is possible at all
but whether and how we can scale up from the
microdemocracies observed in cognitive, developmental, and
social psychology to the political democracies postulated as
theoretical or moral ideáis for large social systems (Báchtiger &
Wegmann, 2014; Parkinson, 2018).
Historically, there are many intermedíate cases of
deliberative democracy in small societies. The most famous
example may be ancient Athens (Saxonhouse, 2006). Athens
was not a liberal democracy, in that women and slaves were
excluded from the deliberative Assembly, but for those
Athenians recognized as full citizens, the Assembly was a
democratic site for free and equal deliberation. Many small
European societies for centuries after that relied on deliberative
democratic gatherings of free men. Similarly, deliberative
democracy was historically central to governance within many
indigenous groups of the Americas (Taylor, 2019). To this day
in the United States, many New England towns rely on town
meetings for local governance.
But what about groups too large for face-to-face meetings?
Deliberation is increasingly compromised as group size
increases from dozens to hundreds to thousands to tens of
thousands (as in ancient Athens). For a nation with a
population in the millions, there is substantial basis for
questioning even the possibility of achieving anything that
approximates a deliberative democracy. But even at very large
scales, some social systems do better than others. As we will
see, psychological research provides good reason for retaining
deliberative democracy as a normative ideal and useful
guidance in our efforts to approximate it better.

Is Democracy Psychologically Possible?

There is plenty of reason for pessimism about the future of
deliberative democracy. Many have questioned whether human
beings are capable of deliberative democracy, or any form of
democracy at all. The editors of the Oxford Handbook of
Deliberative Democracy began their introductory chapter
(Báchtiger et al., 2018a), as might be expected, by noting that
deliberative democracy has become “a flourishing field”
characterized by theoretical advances, innovative practice,
empirical study, and “robust and productive” criticism. But they
acknowledged immediately that “all is not rosy”:

As we bring this Handbook to fruition, the world at large

appears to be moving in some disconcerting anti-
deliberative and anti-democratic directions. Post-truth
politics is the antithesis of deliberative democracy.
Resurgent authoritarian and populist leaders in many
countries have little interest in deliberation— except to
suppress it. Even where deliberation is not repressed, we
too often see levels of political polarization that signal
inabilities to listen to the other side and reflect upon what
they may have to say.
(P- 1)

The rise of deliberative democracy as a theoretical ideal has

not been accompanied by a rise in deliberative democracy.
Some have worried, looking at current trends, that democracy,
deliberative or otherwise, is simply on the way out (Taylor,
2019). Deliberative democracy may be the ideal social system
for ideal rational agents, but perhaps humans fall so far short of
rational ideáis that they cannot sustain deliberative democratic
social systems. Rosenberg (2014), reviewing relevant
psychological research in social, cognitive, and developmental
psychology, concluded:

[P]eople tend to think in simple terms. They prefer

homogeneous categories and look for single concrete
causes and effects. Moreover, people construct these
categorical and causal linkages in a typically flawed way.
They tend: (a) to focus on what is apparent, concrete or
more cognitively available; (b) to process this information
either by utilising cognitive short cuts or by uncritically
applying prior personal experience or culturally available
explanations; and (c) unconsciously to skew final
judgements so as to maintain their self-esteem or prior
evaluations of persons or groups. In sum, individuáis’
thinking appears to lack the objectivity, complication,
integration and abstraction to enable them to appreciate
the complexity of social problems and to imagine
effective, novel ways of addressing them.
(p. 112)

Similarly, Karpowitz and Mendelberg (2018), reviewing the

literature of political psychology, concluded that people express
“high levels of confidence” in their own false beliefs and are
highly resistant to correction. Failing to judge new information
objectively, “they are strongly influenced by biases, such as the
characteristics of the messenger” and subject to “motivated
reasoning— the biased processing of information based on pre-
existing emotions and attitudinal predispositions.” Specifically,
people “tend to be forgiving and uncritical of information or
arguments that fit their predispositions and unforgiving and
hypercritical of information that contradicts their beliefs.”
Exposure to “a balanced set of arguments” may lead those with
“strong pre-existing attitudes” to “double down on their prior
opinions.” In general, “political judgments are made quickly,
emotionally, and outside of conscious awareness, then
rationalized with reasons afterwards.” That is, “judgment
typically comes before reasons— the exact reverse of the
decision-making processes anticipated by deliberative theory”
(p. 539, emphasis in original).
Neither Rosenberg (2014) ñor Karpowitz and Mendelberg
(2018), however, advocated giving up on deliberative
democracy, ñor do I (see also De Brasi, 2018). I argüe in this
section, based on evidence reviewed in Chapters 2 through 5,
that people are sufficiently rational for deliberative democracy
to be the ideal form of collective governance. Without denying
the evidence for systematic and pervasive human irrationality, I
suggest three reasons for cautious optimism. First, as seen in
Chapters 2-4, people are not as utterly irrational as some have
claimed. On the contrary, adults at their best are rational in
ways beyond the competence of children, and even children
show increasing rationality over the course of childhood.
Second, as seen in Chapter 5, argumentation under the right
circumstances enhances reasoning. Deliberative democracies
draw in múltiple ways on the rational powers of argumentation
to operate more rationally than the individuáis who compose
them. Finally, as discussed in Chapter 8, there are many ways
democratic societies can promote rationality, and thus
democracy, through education. Human rationality, then, is a
psychological fact, a social fact, and a reality that can be
substantially enhanced through education.

Human Rationality as a Psychological Fact

Human irrationality is an obvious and well-documented fact of
life, but human rationality is no less a fact, as shown by
extensive research in cognitive and developmental psychology.
Rationality can be seen in the logic intrinsic to biological self-
regulation, in the sensorimotor functioning of infants, and in the
justifiable logical and causal inferences of preschool children.
Progress in rational agency over the first 12 years of life shows
a universal and predictable trajectory. By the age of about 4
years, children’s knowledge of mind is sufficiently organized
and useful that theorists credit them with a theory of mind,
including an understanding that beliefs can be false and that
people act on the basis of their own beliefs even when those
beliefs are false. Young children recognize direct observation
and testimony from others as sources of knowledge. Beginning
about age 6 years, they also recognize that they and others
make inferences, which can also be a source of knowledge. By
the age of 8 or 9 years, they show rigorous conceptual
knowledge about the logic of classes, relations, and numbers
and a constructivist theory of mind that recognizes and reflects
on interpretive diversity. And beginning about age 11 or 12
years, they bring their deductive abilities to the realm of
hypothetical possibilities, enabling advanced forms of logical,
scientific, moral, and social conventional reasoning, and
serious reflection on the justification of beliefs and the nature of
knowledge and truth.
Nothing in the developmental evidence, however, negates
the extensive evidence for adult irrationality. The most
systematic theoretical effort to coordínate the evidence for both
rationality and irrationality has been the development since the
1970s of dual process theories. In a basic versión of dual
processing, System 1 is automatic and intuitive whereas
System 2 is deliberate and reflective. The heuristics and biases
of System 1 are often adaptive, but they may lead us astray.
System 2 deliberation and reflection may overcome our biases
and unjustified assumptions but may also introduce alternative
biases and assumptions that serve to maintain our theoretical
beliefs and personal identities. Thus, rationality is not simply a
reflective override of our automatic processes and intuitions.
Dual process theories help us see how rationality is possible,
how it develops, and why it remains an ongoing challenge for
everyone.
The complexities of dual processing, however, should not
obscure the underlying developmental reality: All normal human
beings beyond the age of about 12 years are capable of
functioning at a level of rational agency beyond what would be
possible for virtually any child under about 10 years, for any
member of any other known species, or for any artifact known
to humans or likely to be created in the near future (but for the
far future, see the discussion of Star Trek’s Mr. Data in Chapter
9). Human reasoning falls far short of rational perfection in
múltiple ways, but the rational agency of human minds, as we
saw in Chapters 2-4, is a well-documented psychological fact.

Human Rationality as a Social Fact

Karpowitz and Mendelberg (2018) wrote: “If nothing else,
political psychologists are likely to agree on one conclusión:
deliberation is an unnatural state of affairs for most citizens” (p.
538). But what they mean by deliberation is political
argumentation about abstract social issues. Deliberation in the
broader sense of serious argumentation within a group about
matters at hand is a natural and common state of affairs for all
normal human beings beyond early childhood, and often leads
to more justifiable beliefs and actions. Research on group
cognition has shown that groups often perform better than
individuáis, and that rational processes of argumentation play a
central role in group rationality. Under ideal circumstances,
groups may reason better than even the most rational individual
within them.
Of course, not all groups function well even under favorable
circumstances, and actual circumstances are often far from
ideal. Nevertheless, social interaction under the right
circumstances can greatly promote good reasoning, often
better than what most individuáis could muster on their own.
Those circumstances include respect for reasons, respect for
persons, and intellectual freedom. The potential for rationality,
which is real but limited in individuáis, can be greatly enhanced
through argumentation (as discussed in detail in Chapter 5).
This provides a second reason for cautious optimism about the
possibility of democracy, and a psychological basis for
specifically advocating deliberative democracy.

Education for Rationality

Dual process theories of reasoning and research on
argumentation as collaborative reasoning provide bases for
some degree of optimism about human rationality. There is no
guarantee, however, that human beings naturally tend to be
sufficiently rational, even in groups, for deliberative democracy.
On the contrary, there is plenty of reason to worry that human
rationality is not up to the task of democratic self-governance.
Fortunately, rationality is not simply a fixed ability beyond our
control that may turn out to be insufficient for our needs. If we
need greater rationality for democracy to work, as it appears we
do, there is much we can do to promote rationality. The single
most important thing we can do, I suggest, is to make rationality
the central goal of education.
The promotion of rationality, however, is not just the teaching
and learning of some assortment of discrete facts or skills. To
be rational is to be a rational agent, and rational agency is the
outcome of developmental processes. Education for rationality,
then, requires promoting the development of rational agency.
The development of rational agency has already been
addressed in Chapters 2 -3 with respect to múltiple forms of
reasoning and in Chapter 4 with respect to metacognition and
epistemic cognition, but the developmental portions of those
chapters focused primarily on describing the pattern of
developmental progress. In Chapter 7, I provide a more
systematic lifespan account of the development of rational
agency that cuts across múltiple epistemic domains and forms
of reasoning and focuses on the process of developmental
change. Chapter 8 then presents a conception of education for
rationality that recognizes the constructive and social aspects
of the developmental processes responsible for progress in
rational agency.
Education for rationality must also promote the rationality of
group functioning both within small groups and within
democratic society. This includes a variety of social and civic
competencies and valúes that are also the outcome of
developmental processes (more in Chapter 7) and can be
promoted through education aimed at fostering development
(Chapter 8).
Actual educational systems fall far short of what they could
and should accomplish. Indeed, much of what passes for
education in most societies, especially at elementary and
secondary levels, is a process of indoctrination that undermines
rationality and civic engagement. But as we will see in Chapter
8, there is good reason to believe that education, properly
conceived and conducted, can enhance both the rational
agency of individuáis and the rationality of argumentation within
democratic and deliberative systems.

Rationality and Identity

To say deliberative democracy is humanly possible is not to say
it is easily achieved or readily sustained. On the contrary,
deliberative democracy faces many formidable challenges,
perhaps the most daunting and crucial of which is the urgent
need “to tame the wild beast of identity” (Maalouf, 2001, p.
157).

Identity as a Theory of Oneself

Kwame Anthony Appiah, a philosopher who specializes in
ethics and identity, tells the story of a Jewish man who is
marooned on a desert island (2018, p. 41). When rescuers
arrive after many years, they observe that he has put up three
structures. “What are these buildings?” they ask. “This one is
my house,” he responds, “and this one is the synagogue I go
to.” “And that one,” he adds, “is the synagogue I don’t go to.”
We see here the social nature of identity, its psychological
centrality, and its potential to divide (Killen & Rutland, 2011;
Moshman, 2011a). Identity brings us together in ways crucial to
how we define ourselves individually, but often at the cost of
setting us in opposition to others, sometimes so strongly that
our opposition becomes increasingly central to our identities.
In the humanities and social sciences, as well as in ordinary
discourse, the term “identity” is typically used to refer to one’s
affiliation with some social group, including national, ethnic,
racial, sexual, political, and religious groups (Appiah, 2005,
2018; Moshman, 2007a). Within developmental psychology,
however, the term typically refers to a relatively mature form of
self-conception and personal commitment first seen in
adolescence (Erikson, 1968; Moshman, 2011a; for related
philosophical conceptions, see Korsgaard, 2009; Parfit, 1984).
The two uses of the term are sometimes distinguished by
referring respectively to “social identity” and “personal identity.”
Further analysis, however, undermines any sharp distinction.
Our social identities are in large part a matter of identification—
a psychological process involving personal interpretations,
choices, and commitments with regard to our múltiple
overlapping group affiliations; correspondingly, our personal
identities are constructed in social contexts and in large part
oriented toward prioritizing and coordinating our múltiple group
connections (Moshman, 2007a, 2011a).
With this in mind, we may define identity as an explicit theory
of oneself as a person (Moshman, 2011a). Let me elabórate on
that definition to show how it coordinates the personal and
social aspects of identity. This will help clarify both the
psychological importance of identity and its danger to
deliberative democracy.
To begin, an identity is a self-conception— that is, a
conception of oneself. But people have many self-conceptions.
The challenge of identity is to unify those self-conceptions, or at
least identify and coordínate the most important, in a way that
enables you to make sense of yourself. Thus, an identity is not
just any conception of yourself, ñor is it simply a collection of
self-conceptions. Rather, your identity is a theory of yourself.
What does it mean to construe identity as a theory? A theory
is a structure of knowledge that organizes and explains some
domain or set of facts. A theory of yourself is a relatively
coherent self-conception that explains your behavior and
experiences. But even young children have structures of
knowledge sufficiently organized and explanatory to qualify as
theories, including theories of themselves. For example,
children’s understanding of false belief is widely taken to show
a theory of mind as early as age 4 years (as discussed in
Chapter 4). But this is not an identity. To have an identity in the
developmental sense associated with Erik Erikson (1968; see
Moshman, 2011a), it is not sufficient to have a theory of
yourself posited by psychologists as implicit in your behavior.
An identity is an explicit theory of oneself.
But not just any explicit theory of oneself is an identity. A
theory of one’s anatomy and physiology may be biologically
insightful and medically useful, but it is not an identity. A theory
of the external forces that cause you to behave in particular
ways may be psychologically insightful and personally useful,
but it is not an identity. An identity is, specifically, an explicit
theory whereby you see yourself as a person. To see yourself
as a person is, at mínimum, to see yourself as a rational agent
— that is, as having reasons for your actions. Even if your
reasons are not good reasons you are still a rational agent
(qualitatively), though less rational (quantitatively) than if you
had good reasons. Some identities are more rational than
others, but identity always involves an effort to see yourself as
a rational agent.
To see yourself as a rational agent, moreover, is largely a
matter of seeing yourself as a rational agent among other
rational agents, which includes seeing yourself as a moral
agent among other moral agents. It is largely in social
interactions within communities of rational and moral agents
that we provide reasons for our beliefs and actions in efforts to
explain and justify them. Our theories of self are constructed
and become increasingly explicit in social contexts. They
enable us to explain and justify ourselves to ourselves and
others.
But in large societies there are far too many rational agents
for us to have direct relations with each as an individual. We
address the complexity of the social world by categorizing
people along a variety of dimensions for a variety of purposes
and theorizing about the resulting social categories. We see
ourselves and others as affiliated in a variety of ways with a
variety of groups, and we see some of these affiliations as
defining who we are in relation to others, who they are in
relation to us, and what moral responsibilities we have to
whom.

Dangers ofldentity
Strong identities may enhance our rational agency by providing
reasons for action based on our reflective theories of ourselves
as persons. Identity is also, however, a threat to rationality.
Evidence threatening your identity threatens your fundamental
explanation and justification of who you are as a person. We
naturally tend to ignore, deny, critique, or reinterpret evidence
contrary to our theories, preferring to seek and focus on
evidence that shows we are right, which we accept with much
less scrutiny. This verification bias is routine even for theories
that don’t matter much to us (as in the Wason selection task
discussed in Chapters 2 and 5). It is greatly exacerbated when
the theories in question are the ideologies that define our
identities— our deepest theories of who we are as persons and
as groups.
And that’s just the start of the problem. To the extent that
identity is social, not just individual, societies are subject to a
process of dichotomization in which, at the extreme, a single
identity dimensión, such as race, religión, or political ideology,
comes to be seen by all as the primary basis for identity (Klein,
2020). As a result, people see themselves and others as fitting
into a small number of primary categories within that single
dimensión— in the extreme just two, so that everyone is either
one of “us” (whoever “we” may be) or one of “them.” This can
lead to processes of dehumanization in which “they” come to
be seen as less than fully human, sometimes leading to
violence, ethnic cleansing, and genocide in cases where those
seen as subhuman are also seen as posing a threat to us
(Maynard, forthcoming; Moshman, 2004a, 2007b, 2011b).
Even in cases that do not generate violence, identity-related
processes of dichotomization and dehumanization seriously
undermine deliberative democracy. Dichotomization reduces
argumentation from an effort of múltiple individuáis to
determine the truth and agree on a course of action to what is
perceived instead as a contest between two ideologies and/or
identities. This undermines the effort to address pluralism in a
rational manner, a central concern of deliberative democracy
(Cinalli & O’FIynn, 2014). Dehumanization of the other further
undermines deliberation. Although we are usually well aware
that our own groups are internally diverse, we see other groups
as relatively homogeneous, thus underestimating the rational
agency of individuáis within them and downgrading their
personhood. If “they” are mostly automatons driven by hatred,
propaganda, and peer pressure, then there is no basis for them
to particípate as individuáis in democratic deliberation. That
leaves argumentation to take place mostly within ideological
groups that don’t talk to each other, likely leading to greater
polarization of ideological belief.
Dehumanization, which denies the personhood of the other,
is part of a more general phenomenon of denial (Bardon,
2020), often driven by our need to maintain our conceptions of
ourselves and our groups as moral agents regardless of how
we actually treat others. The verification bias in hypothesis
testing may be an automatic response, but when denial serves
to maintain our moral identities it is often active and strategic,
albeit less than fully conscious. In dual process terms, we
cannot rely on System 2 to make us rational by overruling
System 1 because System 2 is itself engaged in processes of
denial necessary to preserve our moral identities— our
conceptions of ourselves as moral agents. The result is that we
often have, to varying extents, false moral identities— that is,
conceptions of ourselves and our groups as much more moral
than we actually are (Moshman, 2004a, 2011a).
Rationalist Identity
There will always be identities, and with them processes of
dichotomization, dehumanization, and denial that undermine
our efforts to create deliberative democracies. One part of our
response, to be explored in the next section, must be to create
and maintain rational social institutions that engage us in
collaborative reasoning to overcome our individual limitations.
But we should also consider the rationality of identity and the
possibility that some identities are more rational and/or more
supportive of societal rationality. The challenge is to identify
and encourage better identities without undermining the very
concept of identity by specifying what identities everyone must
have.
One characteristic of a good identity is that it is relatively
rational. Drawing on the usual criteria for evaluating a theory, a
theory of self might be considered more rational than some
alternative theory of self if it is more coherent or more
consistent with evidence. But these are difficult things to judge.
All human identities likely fall far short of theoretical ideáis, and
none of us is in a good position to rate the relative rationality of
our diverse identities.
Rather than focus on whether or to what extent an identity is
rational, I suggest we focus instead on whether it is rationalist
— that is, whether commitment to rationality is among the core
commitments of the identity in question. A rationalist identity is
an explicit theory of oneself that highlights and valúes one’s
rationality. To the extent that you are committed to rationality,
you may be more willing to consider disconfirming evidence
and critical arguments because, whatever other commitments
you may have, determining the truth and doing the right thing
are no less important to how you conceive who you are.
Changing your mind on the basis of evidence or logic confirms
your theory of yourself as the rational agent you want to be. We
will address rationalist identity further in Chapter 7 as a
potential outcome of advanced development and will consider
the promotion of rationalist identity in Chapter 8 as an important
part of educational efforts to promote rational agency without
dictating or manipulating identity commitments.
We can’t simply elimínate the wild beast of identity because
our identities, however savage, are our best visions of who we
are. The challenge is to regúlate ourselves on the basis of the
epistemic aim to seek and acknowledge the truth, however
much we wish to evade or deny it, including truths about
ourselves and about our moral obligations to others.

Rational Institutions
Even with progress in rationality and greater commitment to it,
individual people will always fall woefully short of rational
ideáis. Deliberative democracy requires rational institutions. At
the very least, such institutions must allow for democratic
deliberation in order to achieve the epistemic benefits of
argumentation. In a large society, however, deliberative
democracy cannot consist of a single open forum in which
every individual participates fully and equally in every decisión.
Thus, we must recognize deliberative democracies as
deliberative systems coordinating the ongoing operations of
rational institutions within them (Parkinson, 2018).
One issue of particular importance and concern is that
rational deliberation on various topics requires various sorts of
expertise (Post, 2012). But to gain the benefits of expertise we
must rely on experts, which seems to require vesting authority
in an elite group, contrary to basic norms of democracy (Brown,
2014; Rosenfeld, 2019; Taylor, 2019). If it could be shown that
a system of rule by experts would be more rational than a
deliberative democracy, we would indeed be faced with a
difficult choice between expertise and democracy. But experts
are only experts on particular topics, and expertise itself is the
outcome of democratic deliberation among experts. As we will
see, the challenge is to incorpórate expertise into democratic
functioning. What we need are (a) rational institutions
committed to democratic deliberation and (b) a deliberative
system that respects and maintains such institutions. In this
section, I consider some rational institutions crucial to any
large-scale deliberative democracy.

Legislatures
It is often said that we need more politicians willing and able to
“reach across the aislé.” In the United States Senate, there is
literally an aislé separating the seats of Democrats and
Republicans, the two major political parties. In many other
legislative bodies around the world, the aislé is metaphorical
but the ideological divide between the major political parties is
no less real. To reach across the aislé is to seek some common
ground or basis for collaboration, or at least meaningful
discussion.
In legislatures dominated by two political parties composed
of party loyalists who vote the party line, reaching across the
aislé may be the best anyone can do. But an occasional breach
of the aislé is not good enough for deliberative democracy. In
the deliberative ideal, every member of the society speaks to
and hears everyone else. In societies too large for this,
representatives are chosen to meet as members of legislative
institutions small enough for all members to meet together for
deliberation. Organizing a legislature to operate as an ongoing
battle between two political parties or ideological factions
undermines democratic deliberation among the legislators and
thus undermines deliberative democracy. We need to elimínate
both the literal aislé and the ideological dichotomization it
represents.
But is that possible? In the United States, the House of
Representatives has no literal aislé but is divided no less
sharply than the Senate between Democrats and Republicans.
In both the House and the Senate (which together make up
Congress, the legislative branch of the federal government), the
two parties meet separately in party caucuses to decide on
plans, select party leaders, and engage party solidarity. Turning
to the states, we find the same pattern of a bicameral
legislatura that operates largely as an ongoing battle between
the same two major parties.
But there is one notable exception. The Nebraska state
legislatura is the only unicameral state legislatura in the U.S. It
was specifically designed in the 1930s with múltiple features to
foster deliberative democracy by enhancing transparency and
downplaying party politics. Nebraska senators operate
somewhat more rationally than other state legislators not
because of their individual qualities but because they operate
within a system set up to encourage genuine deliberation,
which is good for everyone.
The unicameral legislatura (informally known as the Unicam)
was conceived and promoted by George Norris, who
represented Nebraska in the U.S. Congress from 1903 to 1943,
first in the House of Representatives and later in the Senate.
Nebraskans voted in 1934 to replace their traditional bicameral
legislatura with a unicameral legislatura, which began operation
in 1937. The Unicam has a single chamber consisting of 49
senators representing 49 legislative districts. Every bilí is
assigned to a committee of senators, who hold a public hearing
at which anyone may speak. Bills approved by committee may
be scheduled to be heard on the floor, where all 49 senators
may particípate at each stage of deliberation. In contrast to
bicameral legislaturas, there are no secret processes or
reconciliation committees to resolve differences in bilis passed
by two chambers. To prevent quick passage of laws without
adequate discussion and debate, legislation must be approved
three times over an extended period, usually weeks or months,
with deliberation prior to each vote. In addition, although bilis
can be amended by a majority of those present, approval at
each of the three stages of debate requires 25 votes (or more
in the case of a filibuster), even if some of the 49 senators are
absent or abstain.
State senators are elected on a nonpartisan basis. All
candidates run in a single primary; the top two go forward to the
general election. These are often a Republican and a
Democrat, but they are often two Republicans and can be (at
least in Lincoln or Omaha) two Democrats. The operation of the
Unicam is also, ideally, nonpartisan. There are no party
caucuses. It is common for senators to work closely with each
other on potential bilis without regard to party affiliation and to
find Creative ways to get 25 or more votes from the 49
senators.
How does this work in practice? Of course, there is party
politics. But sometimes the system enables senators to go
beyond that. I present here one notable recent example.
In March 2017, as president-elect of the Academic Freedom
Coalition of Nebraska (AFCON), I provided the Unicam’s
Education Committee with written testimony concerning
AFCON’s opposition to a bilí designed to actívate the
Americanísm committee of every school district across
Nebraska. These committees were established by a 1949
Nebraska law driven by the rising anti-communist hysteria that
a few years later carne to be called McCarthyism. I testified that
AFCON saw the Americanism law as an infringement on the
academic freedom of teachers and students and on the
academic integrity of Nebraska schools. Specifically, I argued
that for a state legislature to determine curriculum and set
methods of instruction, as this law did, is micromanagement,
which in the present academic context is indoctrination.
Compulsory Americanism is, paradoxically, just plain un-
American; genuine patriotism, I insisted, cannot be compelled
or coerced and does not require Americanism committees.
Quoting from a historie decisión of the U.S. Supreme Court
striking down mandatory flag ceremonies (West Virginia v.
Barnette, 1943), I argued that we should not underestimate “the
appeal of our institutions to free minds” (p. 641).
The bilí had strong support from the most conservative
Republicans who, together with relatively moderate
Republicans, had a substantial majority of the legislative seats.
Nevertheless, like similar bilis in previous years, it failed to pass
due to liberal opposition and the absence of party discipline in
the nonpartisan Unicam.
When I heard in January 2019 that a Republican State
senator was once again dusting off Nebraska’s Americanism
law, I expected the worst, but was pleasantly surprised to find
that the new bilí had some noteworthy positive features. The bilí
was subsequently improved, moreover, by several
amendments responding to liberal objections. In fact, in its final
versión, the bilí essentially repealed the Americanism law and
replaced it with a civics education law. The new law would
elimínate the Americanism committees and require instead that
the school board of each school district appoint a “committee
on American civics.” The difference was not just linguistic. The
new committees were responsible for assuring that civics
education is taken seriously, but their mandate showed
considerable respect for the autonomy of students, faculty, and
school systems.
The final versión of the bilí was supported by senators
spanning the political spectrum. It passed 44-2, with one
senator abstaining and two absent, and was signed by the
governor. By the standards of academic freedom to be
discussed in Chapter 8, the new law still micromanages
curriculum, instruction, and assessment. But compared to the
law it replaced, it is far more flexible in its requirements, far less
indoctrinative in its language, and overall much more respectful
of intellectual freedom. In an age of intense dispute over
matters of history, government, and social responsibility, the
nonpartisan structure and deliberative requirements of the
Nebraska Unicam enabled it to reach near-consensus on a
conception of civics education far more academically and
morally justifiable than the compulsory Americanism it replaced.

Courts of Law
Legislatures pass laws, but the judicial system is generally
responsible for determining whether someone has broken the
law. In legal systems following English common law (including
the United States and Cañada), a criminal trial generally
involves both a judge and a jury of peers. The judge is
responsible for conducting a fair trial ¡n accord with applicable
legal procedures and ¡nstructing the jurors about applicable
laws and legal standards. The exercise of such responsibility
requires legal expertise.
The rationale for the jury system is that democratic
standards of liberty and equality will be compromised if people
can be convicted of crimes and sentenced in secret processes
controlled by a legal elite. Instead, triáis are generally public
and the determination of guilt, in particular, is made by a group
of jurors (generally 12) who can be considered peers of the
accused in that they are selected from the public at large. The
jurors generally do not have, and do not need, legal expertise,
because the judge instructs them regarding legal
considerations. The role of the jury is to attend to trial
testimony, determine the facts of the matter, and reach a
conclusión as to whether those facts show a violation of the
law, as interpreted and explained by the judge.
Although jurors do not need legal expertise, the legal system
relies on their rationality. One might reasonably wonder, then,
given that people are not equally rational, why we need a jury
of 12 people. Why not select the most rational and fair person
from the pool of potential jurors and let that person serve alone
as the jury? The jury selection process already begins with a
pool of potential jurors and eliminates those who appear unable
to understand the relevant facts or law or likely to have a
conflict of interest. Even if we can’t always tell who will be the
very best juror, surely we could extend the jury selection
process to settle on someone better than average.
The problem is that we know from extensive research that
even the most rational people routinely fall far short of rational
standards (Chapters 2 -4 ) and that groups often show higher
levels of rational performance than individuáis (Chapter 5). The
system of múltiple jurors dates back centuries before this
psychological research but seems entirely consistent with its
basic findings.
 Even groups do not always function rationally, of course. If a
few members of the jury confidently assert some ¡nterpretation
and conclusión, others may go along because it sounds
plausible, then others because of peer pressure or a desire to
reach closure, leading to a perfunctory hearing of alternative
and dissenting views followed by a vote for the majority
conclusión. But English common law generally requires a
unanimous decisión for the jury to reach a verdict (otherwise
there is a “hung jury,” which may lead to a new trial). The
requirement of unanimity means dissenters cannot simply be
outvoted; they must be convinced. Thus, deliberation is
enhanced, which may lead to a fuller recall and consideration of
the facts than any juror alone could muster and more
comprehensive and justified interpretations than those initially
reached by most or all of the individual jurors. Research on
juries indicates that, as expected, “group discussion helps
jurors individually and collectively to clarify their positions and
conclusions, and increases their certainty that they are
reaching the right verdict” (Hans, 2007, p. 586). Juries required
to reach a unanimous verdict have been found more likely to
begin with a “general discussion of the evidence” and to “pay
more attention to those who hold minority views” (Hans, 2007,
p. 587). Dissenters have been found to “particípate more in the
discussion and have more influence” when consensus must be
achieved (Hans, 2007, p. 587).
Thus, the jury system (1) relies on peer decisions for moral
reasons related to the nature of democracy; (2) relies on
argumentation among múltiple peers for epistemic reasons
related to group rationality; and (3) generally requires
consensus in order to maximize the epistemic advantages of
collaborative reasoning. The jury is a small deliberative
democracy operating within a court system that provides the
legal expertise to connect the work of the jury to the larger
social system of deliberative democracy of which the judicial
system is a part. The court is not itself a deliberative
democracy. Those present do not all have an equal right to
speak or make decisions. But the operation of the court system
provides the legal expertise that enables deliberative
democracy to work across múltiple levels within a deliberative
and democratic system.
But why only one judge? In most cases it is presumed that
the requirements of law are clear to anyone sufficiently expert
in the law. But sometimes the state of the law is itself uncertain.
Even legal experts may disagree about the most justifiable
interpretation of a vague law, the applicability of a particular law
to the case at hand, or the best way to resolve apparent
inconsistencies between applicable laws. Experts may also
disagree about whether particular laws or government actions
are consistent with constitutional requirements, which may
themselves be vague enough to justify múltiple interpretations.
In such cases, our best hope for rational legal judgment is
deliberation among experts, which is generally achieved
through a process of appeal to higher courts, up to a supreme
court with múltiple members.
Thus, the state of the law, in cases of controversy, is itself
determined through a process of deliberation, but in this case
the deliberation takes place among experts. Cases decided by
a high court may be controversial, and some decisions may be
denounced as political, rather than legal, and thus ¡Ilegitímate
(Fallón, 2018). Nevertheless, there are rational grounds for
favoring some decisions over others (Amar, 2012; Dworkin,
1985, 1996, 2011; Fallón, 1987, 2018). As a result, judges from
diverse backgrounds with diverse political views often reach
consensus on legal matters (Fallón, 2018). Even when they
don’t, the expectation of written justification and the effort to
find a rationale convincing to a majority encourage rational
judgment.
In a legal system that maintains itself over many
generations, a key consideration in settling any case is
consistency with past cases. Even the highest court takes its
own previous decisions as an important constraint on present
decisions. Precedents may be overruled if they are deemed
clearly wrong, but the default expectation is to rule in accord
with them or provide specific reasons for new exceptions or
new limits on their applicability. Thus, although precedents are
not the only basis for legal reasoning, precedent-based
reasoning (as discussed in Chapter 3) plays a key role in law.
In relying on precedent, law establishes itself as a rational
multigenerational social system where previous decisions
become social conventions that play a role in present
deliberations (Amar, 2012; Moshman, 2015).

Epistemic Institutions and Professions

Epistemic institutions and professions are those centrally
concerned with truth and justification. These include Science,
education, and journalism. Epistemic institutions may be part of
the government, supported by government, or independent of
it; regardless, they are crucial to democracy (Rosenfeld, 2019).
It is important that scientists, educators, journalists, and
others in knowledge-oriented professions understand their role
in the larger deliberative and democratic system. It is equally
important that everyone understand the reliance of democracy
on truth and the epistemic nature of institutions claiming to
provide such truth. I address these matters in Chapter 8, in
which I consider the many ways in which academic institutions
serve democracy and the means by which they do so.

Intellectual Freedom in Society

Free speech was widely deemed central to democracy in
ancient Athens (Saxonhouse, 2006) because it served equal
participation in governance among those permitted to
particípate in discussion at the Assembly (women and slaves
were excluded). But what about states too large for direct
democracy? In such states, governing— a process engaged in
by citizens— becomes government— an abstract entity that may
be a threat to individuáis. In large-scale indirect democracies,
then, guarantees of free speech serve as a protection of
individual rights against government power. Nevertheless, free
speech remains a right to particípate, as emphasized in
deliberative democracy. Beyond democratic participaron and
individual rights, free speech is also critical to the progress of
knowledge, as discussed in Chapter 5 with respect to group
discussion.
Thus, free speech is important for reasons of democracy,
individual rights, and the pursuit of truth (Ash, 2016; Moshman,
2017, 2020). But free speech is not sufficient. Deliberative
democracy requires intellectual freedom, which includes
freedoms of belief, expression, discussion, inquiry, and access
to information and ideas. I will retum to intellectual freedom in
Chapter 7 with respect to its role in the development of
rationality and will address it most systematically in Chapter 8
with regard to the centrality of intellectual freedom for
education. As we will see, the regulation of speech by
educational, governmental, and other authorities is often
necessary and justified, but authorities of all sorts routinely
infringe without justification on the intellectual freedom
necessary for education, democracy, and intellectual
development. A central principie for addressing the legitimacy
of restrictions on free expression is that restrictions on speech
should generally not be based on the content of what is said
and especially not on objections to the speaker’s point of view
(Strossen, 2018).

Conclusión
Deliberative democracy extends the benefits of argumentation
to entire societies, at least in principie. Real societies, however,
fall far short of deliberative and democratic ideáis. That may
always be the case, but we could do much better if education
were aimed more directly and effectively at the promotion of
rational agency and civic responsibility. In Chapter 8, I consider
how education at all levels could be more effective in this
regard.
 Central to my conception of education for rationality and
democracy is that such education must focus on enabling and
encouraging developmental progress. Before we turn to
education, then, we look first, in Chapter 7, at processes and
patterns of development.
7
THE RATIONAL CONSTRUCTION OF
RATIONAL AGENCY

In this chapter we consider the development of rationality. We

have already addressed this in earlier chapters, which
considered the development of logical reasoning (Chapter 2); of
causal/scientific, principled/moral, and precedent-based
reasoning (Chapter 3); and of metacognition, including
epistemic cognition concerning the nature of justification and
truth (Chapter 4). The present chapter goes beyond the earlier
treatment in three major ways.
First, the aim here is to provide an account of development
that goes beyond description to explanation. Chapters 2 -4
focused mostly on describing developmental sequences. The
present focus (especially in the second section) is on the
rational and constructive process of change that explains
developmental progress. This focus on process is critical not
only for explaining development but also for educational efforts
to promote the development of rationality (the topic of Chapter
8).
Second, the earlier accounts of stages or levels of
development were each specific to particular forms or aspects
of reasoning and rationality (such as logical reasoning, causal
reasoning, and epistemic cognition). The aim here (especially
in the third section) is to look more broadly at general patterns
in the emergence and progress of rational agency. This
suggests a distinction between basic development in childhood,
leading to a universal level of maturity achieved about age 12
years, and advanced development in adolescence and beyond,
which is much more individualized.
Finally, Chapters 2 -4 focused on individual development.
Following up on Chapters 5 and 6, we now consider the role of
argumentation and democratic institutions in development (with
more on education to follow in Chapter 8). In addition, I propose
in the final section that groups and societies may themselves
show developmental progress.
I begin with an evolutionary theory of human reasoning. The
critique of this theory leads directly to the more developmental
and constructivist account that undergirds the rest of the
chapter.

“The Enigma of Reason”

The most important and influential book on reasoning since
Daniel Kahneman’s (2011) Thinking: Fast and Slow has
probably been Hugo Mercier and Dan Sperber’s (2017) The
Enigma of Reason. In reviewing it for the journal Human
Development, I praised it for directing attention to the social
aspects of reasoning typically ignored in the cognitive and
developmental literatures but argued that the proposed theory
ignored extensive developmental research and, as a result,
misconstrued the nature and role of both logic and reflection
(Moshman, 2018; see also, Noveck, 2019; Mercier & Sperber,
2019). After providing a summary and critique of Mercier and
Sperber’s theory in this section, I then provide in the next
section a constructivist developmental account of human
reasoning and rationality that is more consistent with dual
process theories and, I argüe, with available evidence from
cognitive and developmental psychology.

Mercier and Sperber’s Theory

Mercier and Sperber (2017) provided an evolutionary account
of human reasoning, crediting Charles Darwin with “the
realization that whatever traits humans share as a species are
not gifts of the gods but outcomes of biological evolution.
Reason, being such a trait, must have evolved” (p. 1).
Specifically, they argued, reasoning evolved to serve the social
purpose of argumentation.
In elaborating their evolutionary account, Mercier and
Sperber (2017) questioned the distinction made by dual
process theories “between intuition and reasoning as if these
were two quite different forms of inference,” proposing instead
that “reasoning is itself a kind of intuitive inference” (p. 7). They
maintained that logicians, philosophers, and psychologists have
greatly overemphasized the role of reflection in human
inference because they themselves “commonly resort to higher-
order arguments as part of their trade,” which is “hardly typical
of the population at large” (Mercier & Sperber, 2017, p. 152).
Reflection is dismissed as a “professional twist of mind” rather
than “a basic human trait” (Mercier & Sperber, 2017, p. 152).
Reasoning, Mercier and Sperber (2017) argued, is a matter
of argumentation, rather than individual thinking, which leads
them to discount the role of logic:

Whereas reason is commonly viewed as a superior

means to think better on one’s own, we argüe that it is
mainly used in our interactions with others. We produce
reasons in order to justify our thoughts and actions to
others and to produce arguments to convince others to
think and act as we suggest. We also use reason to
evalúate not so much our own thought as the reasons
others produce to justify themselves or to convince us.
Whereas reasoning is commonly viewed as the use of
logic, or at least some system of rules to expand and
improve our knowledge and our decisions, we argüe that
reason is much more opportunistic and eclectic and is not
bound to formal norms.
(P- 7)

The role of logic in reasoning continúes to be disparaged

throughout the book. “Logic,” the authors maintain, “tells us
neither how we reason ñor how we should reason” (Mercier &
Sperber, 2017, p. 174). Logic is deemed pragmatic and
contextual, sometimes helpful to reasoning, but playing no
special role. As one subtitle insists, “In Reasoning, Logic Is a
Heuristic Tool” (Mercier & Sperber, 2017, p. 158).

Critique
No serious developmentalist doubts the valué of an
evolutionary perspective, but nearly all reject simplistic
evolutionary models that explain reasoning and other
psychological phenomena on the basis of a direct causal link
from evolution to genes to behavior. What is missing is
development (Lerner & Overton, 2017; Lickliter & Witherington,
2017).

The human species is indeed a product of evolution and

can only be understood in light of its evolution. But it is no
less true that a person is a product of development and
can only be understood in light of that development. This
makes little difference if development is driven by
evolutionary instructions encoded in our genes. But
research over the past half-century has made it
increasingly clear that development can only be
understood as a constructive process that coordinates
emerging structural relations across genetic,
physiological, behavioral, environmental, social, and
cultural levels. Development is not determined by, and
cannot be reduced to, genes or evolution.
(Moshman, 2018, p. 61)

As discussed in Chapters 2 and 4, research on the

development of logical reasoning and metalogical
understanding shows that logic is implicit in norms of biological
self-regulation and can be seen, long before the emergence of
language, in the increasingly coordinated sensorimotor
cognition of infants. We also saw that preschool children
routinely make logical inferences and that, beginning about age
6 years, children recognize inference as a potential sources of
knowledge and make systematic progress as they go on to
construct increasingly reflective understandings of the
difference between logical and empirical knowledge. As we will
see in the next section, the development of metalogical
understanding appears to be an ongoing constructive process
in which knowledge implicit in inference at any level becomes
an object of reflection, which generates more explicit
knowledge about logic, which guides higher levels of reasoning.
Reflection is not just a “professional twist of mind.” On the
contrary it is a universal feature of human cognition and, in its
strongest form, a fundamental process of developmental
change, especially with respect to logic.
The problem goes even deeper. The dismissal of logic is
rooted in a general skepticism about the epistemic basis of
reasoning:

Mercier and Sperber (2017) see reasoning as

metacognitive in some ways, but they do not see it as
epistemic. In their view, reasoning evolved as a pragmatic
instrument to serve particular social purposes, not to
determine the truth, so any truth that results is mostly
incidental. Justification is crucial to argumentation, but
justification, in the proposed theory, is neither based on
logic ñor aimed at truth.
This view is in sharp contrast to that of most
philosophers, and especially epistemologists, who see
truth as a necessary condition of knowledge. But Mercier
and Sperber understand that they are challenging
standard philosophical views. What they don’t seem to
realize is that they are also challenging the epistemic
cognition of ordinary people (Moshman, 2015). By the
age of 4 years, children recognize that beliefs can be
false; for many years after that they construct increasingly
sophisticated conceptions of truth, justification, and
knowledge. This epistemic development is reflected in
increasingly sophisticated argumentation. It is not clear
 how the proposed theory can account for this. Why would
evolution generate a universal course of individual
development toward false and useless ideas about logic
and truth?
(Moshman, 2018, p. 63)

I concluded that the proposed theory offers much of valué but is

best viewed as complementary to standard views, rather than
as an alternative theory:

With regard to the emphasis on argumentation, reasoning

is both social and individual, with reciprocal influences
across múltiple levels. Perhaps we have underestimated
the social aspect of reasoning as a collaborative activity,
a claim worth considering, but there is no need to
disparage the epistemic agency and development of
individuáis and much evidence to remind us not to do so.
With regard to development, evolutionary accounts
must acknowledge and explain developmental processes,
sequences, and outcomes. People are not collections of
evolved traits. The conceptualization of reason as a trait
draws our attention away from its development and thus
undermines our understanding of its nature, including its
relation to logic.
Finally, reasoning is indeed much more than logic, and
the present work helps us see better the many ways in
which this is so. But reasoning sometimes properly relies
on strict deduction and its associated possibilities,
impossibilities, and necessities. Logic is crucial to much of
our reasoning whether we know it or not, and we’re better
off if we know it.
(Moshman, 2018, p. 64)

Constructivist Accounts of Development

A comprehensive account of rational agency must explain its
development. Nativist accounts highlight the genes, which are
seen as having a direct causal role. In the extreme (as in
Mercier & Sperber, 2017), genes are so strongly predictive of
mature outcomes that development is essentially ignored.
Behavior is assumed to be shaped directly by the evolutionary
history of the species and explained on that basis.
The standard alternative to nativism is empiricism, which
highlights learning from the environment. In the extreme,
evolution is essentially ignored on the assumption that it
provides nothing more than a blank slate for learning.
Development, in turn, is reduced to nothing more than a
process of learning whatever the physical or social environment
presents or teaches.
Going beyond both nativism and empiricism, and even the
sum of both, is interactionism, which recognizes that
development usually or always involves an interaction of
nativist and empiricist factors. At the very least, the interaction
is statistical. The effects of heredity depend on one’s
environment, and the effects of any given environment depend
on the genome of the species and/or individual differences in
genes. Beyond such statistical interactions, moreover,
interactionists often note that development involves an ongoing
interaction of organism with environment, and thus (one might
assume) of genes with environment.
But here we run into a whole new complication. Organisms
are not simply collections of genes. The organism that interacts
with its environment is itself the outcome of a developmental
process involving, at the least, complex interactions of
hereditary and environmental factors. A mature organism, and
even a maturing one, cannot be reduced to its genes any more
than to its environmental history. It is an emergent entity. And
as the emerging organism achieves the ability to act, its
behavior increasingly plays a role in development that cannot
be reduced to genes, environments, or even in interaction of
both.
Psychologists who emphasize the constructive role of the
organism in its own development are constructivists. Nearly all
developmentalists acknowledge the role of constructivist factors
¡n development, along with nativist and empiricist factors, and
¡n that sense all developmentalists are constructivists as well as
nativists and empiricists. But many see the role of the organism
as central and are thus constructivist in a stronger sense.
Constructivism is a theoretical perspective emphasizing the
active role of the organism in the construction of knowledge.
Constructivist theories take a variety of forms (Moshman,
1982). Rational constructivism (Moshman, 2011a) construes
the constructive process of development as a rational process
directed by an increasingly rational agent. Rational
constructivism is rooted in the developmental epistemology of
Jean Piaget (1975/1985, 1977/2001) and has been highly
influential in subsequent research and theory on cognitive and
moral development (Demetriou & Spanoudis, 2018; Gibbs,
2019; Kazi et al., 2019; Kohlberg, 1981, 1984; Kuhn, 2005;
Kuhn et al., 1988; Moshman, 2004b, 2015; Nucci, 2001, 2014;
O’Madagain & Tomasello, 2019; Tomasello, 2014; Turiel, 1983,
2008). In rational constructivism, the developing organism (at
least in some species) increasingly approximates a rational
agent, acting on the basis of reasons of its own, not simply
doing what it is caused to do by forces beyond its control. The
rational agent that emerges from development is, moreover,
increasingly responsible for its own further development.
Rational constructivism is an idealized view of development
corresponding to rational agency as an idealized view of
personhood, rational agents as idealized persons, and
deliberative democracy as an idealized social system. We are
all subject to forces beyond our control; organisms and social
systems change in a variety of ways due to a variety of
nonrational and irrational forces. As we saw in Chapters 2-4,
however, there is clear evidence for developmental progress in
rationality. The most convincing explanation for this is that
development, at its normative core, is a rational process
directed by rational agents.
Reflection and Coordination as Developmental
Processes
Piaget (1967/1971, 1977/2001) provided an account of
cognitive development as an ongoing process of coordination
and reflection by an increasingly rational agent.

[L]ogico-mathemat¡cal structures owe their origin neither

to physical experiment ñor to instinctive or hereditary
transmission but are derived by reflective abstractions
from the general coordinations of action and, farther back
still, from nervous coordinations, and so on back to the
most widely generalized of the organizing functions in life.
(Piaget, 1967/1971, p. 342)

Many theorists have highlighted and examined the role of

reflection in development (Alien & Bickhard, 2018; Campbell &
Bickhard, 1986; Karmiloff-Smith, 1992; Piaget, 1977/2001). A
good example is the role of reflection in the development of
metalogical understanding. I present here once again the four
stages of metalogical understanding originally presented in
Chapter 4, focusing now on the role of reflection in explaining
the transition from each stage to the next.
As we saw in Chapter 2, logic is initially implicit in biological
self-regulation (Piaget, 1967/1971) and then in infants’
sensorimotor actions (Langer, 1980, 1986; Piaget, 1936/1963).
The development of logical reasoning does not culminate in
logic. On the contrary, logic can be found as far back as we
look. What develops is our metacognitive control of our
inferences and our explicit knowledge about the logic implicit in
them (Moshman, 1990, 2004b; 2015). Stage 1 of metalogical
development is the stage of explicit content and implicit
inference (typical in children under 6 years). Stage 1 thinkers
can tell you what they are thinking about but they are not aware
of the inferences intrinsic to their cognitive processes and use
of language. They do not distinguish conclusions from
premises, do not attribute inferences to others, and do not
recognize inference as a potential source of knowledge (Miller,
2012; Pillow, 2012; Sodian & Wimmer, 1987).
The transition to Stage 2 (beginning about age 6 years)
involves a process of reflection on inference, rendering explicit
what was earlier implicit in action. But we can only reflect on
inferences from the standpoint of an implicit logic that
addresses their normative properties. To attain Stage 2
metalogical understanding we must construct an implicit logic
that enables us to assess our increasingly explicit inferences.
Stage 2 thinkers recognize conclusions as the outcome of
inferences, distinguish conclusions from premises, recognize
that others also make inferences, and understand that, as a
result of inference, people may come to know things they have
not observed or been told. The explicit attention to inferences
involves an implicit appreciation of their logical form and
necessity.
The transition to Stage 3 begins about age 11 or 12 years
and remains a work in progress for most people. It involves a
process of reflection on logic itself. Explicit awareness of logical
form makes it possible to distinguish (a) the abstract issue of
whether a conclusión follows necessarily from some set of
premises from (b) the more concrete challenge of evaluating
the truth of individual premises and conclusions. This enables
hypothetico-deductive reasoning, which greatly extends the
scope of our rationality. The explicit attention to logic entails an
implicit metalogical understanding of logic as a domain of
necessity distinct from empirical science.
Given the challenges of Stage 3, most people probably leave
their most advanced metalogical understandings implicit. The
transition to Stage 4, which may be limited largely to logicians,
involves reflections on (a) the nature and justification of formal
logical systems, (b) their interrelations with each other and with
natural human languages, (c) the nature and scope of logic,
and (d) the relation of logic to science and other domains.
Logicians are experts on logic, but this does not mean they
invented it. What they are expert at is something we all use
and, to varying extents, understand. Logicians have simply
achieved a higher level of reflection on logic, which enhances
understanding but raises whole new sets of questions to be
addressed through further reflection.
The development of metalogical understanding is thus
explained as a process of reflection through múltiple levels
(Campbell & Bickhard, 1986; Moshman, 1990). In each stage
transition, what was previously implicit becomes an explicit
object of reflection.
Consider another example, which illustrates the interrelated
roles of reflection and coordination as developmental
processes. The development of epistemic cognition in domains
such as science and morality, as we saw in Chapter 4, shows a
recurring pattern of three basic stages. The first, objectivist,
stage involves explicit knowledge about the world. Implicit in
any act of knowing is the subjective perspective of the knower,
but objectivist knowers construe the known as a given without
recognizing that it is a function of their subjective perspective.
The transition to the subjectivist stage involves a process of
reflection on subjectivity. Explicit awareness of one’s
subjectivity undermines one’s earlier conception of the
objectivity of knowledge, which represents an advance in
understanding. Taken to a subjectivist extreme, however, the
commitment to subjectivity undermines the possibility of
objectivity and thus of knowledge.
The transition to the rationalist stage can be seen as a
coordination of objectivity and subjectivity at a higher level of
reflection that acknowledges both sufficiently to consider their
relations. Regardless of the details, it seems clear that
developmental progress in epistemic cognition is a constructive
process. It may be influenced by genes and environments but
is not driven by either, or even by an interaction of both.
Developmental progress through reflection and coordination
is not a gradual accumulation of knowledge but the generation
of a series of discrete and increasingly equilibrated stages
(Piaget, 1975/1985). The philosopher Thomas Nagel (1986) put
¡t this way:

We can add to our knowledge of the world by

accumulating information at a given level— by extensive
observation from one standpoint. But we can raise our
understanding to a new level only if we examine that
relation between the world and ourselves which is
responsible for our prior understanding, and form a new
conception that includes a more detached understanding
of ourselves, of the world, and of the interaction between
them.
(p.5)

Peer Interaction as a Developmental Context

Children obviously learn from their peers, and much of what
they learn is useful and important. But children also learn from
adults, including their parents and teachers, who may often be
a source of better information and guidance because they are
more mature and experienced. What is special about peer
interaction as a context for developmental progress is not that
children can learn from their peers but that they can develop by
interacting with peers (Piaget, 1932/1965b, 1995). Of course,
children also develop in the course of interacting with adults
(O’Madagain & Tomasello, 2019; Tomasello, 2014), but there is
reason to believe that interaction with peers plays a unique and
crucial developmental role.
The role of peer interaction in development was central to
Piaget’s (1932/1965b) classic early work on moral
development. Children routinely learn rules of behavior from
parents and other authorities with the ability to enforce such
rules. But genuine morality requires more than following rules
enforced by those in power. Genuine morality requires respect
for persons, including the ability to engage in reciprocal
relations among rational agents who recognize each other as
social equals. The development of what Piaget called an
“autonomous” morality is a constructive process of coordination
with others and reflection on one’s relations with them, which
can only take place in the context of peer interaction.
Piaget and many theorists since have extended this
conception of the crucial importance of peer interaction for
moral development into a more general conception of the
special role of peer interaction, including argumentation, in the
development of reasoning and rationality (Kuhn, 2005; Kuhn et
al., 2016; Moshman, 2011a; Piaget, 1995). Respect for others
as persons and peers requires respect for reasons. Respect for
reasons and persons includes both paying attention to the
reasons others have for their beliefs and actions and providing
others with reasons for one’s own beliefs and actions.
The development of epistemic cognition, for example, is
surely enhanced by, and arguably requires, peer interaction. It
is natural to construe what we perceive as the objective truth.
We recognize our perceptions as subjective when we engage
seriously with peers who have different perceptions. Peer
interaction also enables us to coordínate our subjectivities to
obtain a greater objectivity, and to reflect on this process in
ways that promote our epistemic cognition. Peer interaction is
also crucial preparation for participation in deliberative
democracy in part because it promotes development of the
social skills and understandings necessary for deliberative
democracy, including those associated with argumentation.
Peer interaction is not just interaction with others one’s own
age. Strictly speaking, peer interaction is the reciprocal social
interaction of rational agents who are, and perceive themselves
to be, equal in authority, status, and power. Such interaction
ideally takes place under conditions in which all are, and
perceive themselves to be, equally free to express their views,
hear from others, and fully engage in discussion. We have
already seen that intellectual freedom is crucial for good
reasoning (Chapter 5) and deliberative democracy (Chapter 6).
In creating genuine peer interactions, intellectual freedom also
enables and promotes developmental progress. There is much
more to be said about this, but I will defer further discussion of
¡ntellectual freedom to Chapter 8, in which I address its role in
education, including the development of rationality.

Development Across and Beyond Childhood

In Chapters 2 -4 I discussed, in much more detail than some
readers may have deemed necessary, developmental changes
in múltiple domains of reasoning and rationality, including
logical reasoning, scientific reasoning, principled reasoning,
precedent-based reasoning, and epistemic cognition. In this
section I aim instead to propose a general pattern of
development that we see in all domains (Moshman, 2009b,
2011a). There appears to be a period of basic development
over the first 12 years in which natural human competencies
are constructed in predictable, age-related sequences by all
normal persons, followed by a period of advanced development
in which the direction and extent of development are much
more individualized, not inevitable, and no longer tied to age.

Basic Development
Cognitive development has long been seen as an age-related
sequence of cognitive competencies. Piaget provided the best-
known theoretical account of that sequence but there were
important theories of cognitive development before him and
there have been many since (Demetriou & Spanoudis, 2018;
Kazi et al., 2019). Based on extensive empirical evidence,
virtually all theories posit qualitative transformations in a
consistent age-related sequence over the first 12 years of life.
Most theorists of cognitive development since Piaget share his
rational constructivist conception of the developmental process
but differ in the precise processes and developmental contexts
they emphasize and in their conceptualization of cognitive
competence. For present purposes, we need not dwell on these
theoretical differences. Our focus here is on the basic empirical
picture of emerging competencies, about which there is greater
consensus. In what follows I summarize evidence reviewed in
Chapters 2-4.
A major trend in cognitive development since the 1960s has
been the rapidly expanding literature on infant cognition, which
can readily be interpreted as showing unexpected logical,
causal, and moral competencies implicit in the behavior of pre-
linguistic children. As language develops, children’s logical,
causal, and moral inferences become increasingly observable.
Their behavior also begins to show evidence that they have
metacognitive knowledge about minds and cognitive
processes.
Beginning about age 4 years, children’s knowledge of mind
is sufficiently organized and explanatory that theorists routinely
credit them with a “theory of mind” (Miller, 2012; Moshman,
2015; Pillow, 2012; Sodian & Kristen, 2016). This includes an
explicit understanding that beliefs may be false and that people
act on the basis of their own beliefs even when they are false
(O’Madagain & Tomasello, 2019). Of course there are
individual differences in the rate of development. The basic
theory of mind typical of 4-year-olds does not appear suddenly
on each child’s fourth birthday. It is sometimes seen in 3-year-
olds, especially those within a few months of turning 4, and
sometimes not yet seen in 4-year-olds, especially those who
have turned 4 recently. But basic theory of mind is virtually
never seen in 2-year-olds and is virtually universal in 5-year-
olds. Despite some developmental variation, virtually all 5-year-
olds have attained a level of rationality virtually never seen
before the age of 3 years.
Beginning about age 6 years, children show awareness that
they are making inferences, attribute inferences to others, and
recognize inference as a source of knowledge for themselves
and others. Their explicit attention to inferences enables them
to distinguish conclusions, the outcomes of inference, from
premises. Implicit in their reasoning is a new appreciation of
logical form and necessity (Moshman, 1990, 2015). Here again,
development does not happen in an instant or always at the
exact same age, but biologically normal children in normal
human social environments construct the same basic rational
competencies in the same sequence at about the same age.
By the age of 8 or 9 years, children have constructed a
constructivist theory of mind, which recognizes the active and
interpretative nature of the mind as a constructor of knowledge.
This includes understandings of subjectivity that would have
been literally unthinkable just a few years earlier (Miller, 2012;
Moshman, 2015; Pillow, 2012).
Finally, around the age of 11 or 12 years children begin to
demónstrate forms and levels of reasoning and rationality
virtually never seen before the age of 10 years. These include
hypothetico-deductive reasoning, explicit conceptions of
inferential validity, reflective coordination of theories and
evidence, third-party perspective taking, principled forms of
moral reasoning, and reflective epistemologies (Moshman,
2011a, 2013). Most young adolescents become somewhat
more consistent over the next few years in applying their most
advanced reasoning and insights, but no one ever achieves
anything cióse to consistently rational cognition. As we have
seen in coordinating developmental research with dual process
theories, advanced competencies often supplement earlier
ones without replacing them. Young teens at their best far
surpass the rationality of 9-year-old children, but people of all
ages often show behavior that falls far short of this standard.

Advanced Development
Development generally continúes beyond the age of 12 years
but further development is much more specific to particular
individuáis and circumstances and much less related to age.
Some adults sometimes show forms of logical, scientific, or
moral reasoning beyond what would ever be seen in a 13-year-
old. Some adults understand the relation of social systems to
interpersonal morality in ways that would rarely be seen in a
young teen. Some achieve advanced levels of metacognitive
self-regulation or make progress in epistemic cognition toward
reflective understandings of the subjectivity of knowledge, and
some go beyond that in coordinating subjectivity and objectivity
in ways that would mystify almost any young teen (and most
older individuáis as well). Some people construct rationalist
identities that highlight and valué rationality, justification, and
truth.
In the extensive literature on advanced forms of human
cognition and rational agency, however, I have never seen any
evidence for any form or level of reasoning or rationality that is
achieved by all normal adults but rarely seen in young teens.
This is in sharp contrast to the pattern of basic development in
childhood just reviewed, where it is easy to show múltiple
examples of rational competencies routine at some given age
that are rarely or never seen just a few years earlier. Thus, in
understanding the development of rational agency, it seems
important to make a distinction between the basic development
of the first 12 or so years of life, which is universal and age-
related, and advanced development in adolescence and
beyond, which is much less predictable.
Adults generally share stereotypes of adolescents as subject
to egocentrism, peer pressure, and irrational risks; failing to
give appropriate weight to future consequences; and commonly
operating on the basis of impulse and intuition rather than
reflective reasoning. With respect to all of these cognitive
crimes, adolescents are guilty as charged. What is routinely
overlooked, however, is that adults are guilty of all the same
crimes. Some people are more rational than others, but beyond
age 12 we cannot predict rationality from age.
In the absence of psychological evidence, it is sometimes
assumed on the basis of brain research that, because human
brains arguably continué to develop at least until the age of 25
years, 25-year-olds have attained levels of rational agency not
seen in older teens, who in turn have attained levels not seen in
young teens. But our brains change throughout our lives, in
large part as a result of our individual actions and
environments. There is no evidence to support the assumption
of general progress in rational agency during adolescence and
beyond due to brain maturation or the corresponding
assumption that there is some age at which adults have
generally attained a level of rationality beyond the capacity of
adolescent brains (Moshman, 2011a, 2013).
Nevertheless, as already noted, there is clear evidence for
development in adolescence and beyond. Some people
achieve advanced levels of epistemic cognition, for example,
that are rarely or never seen in young teens. Some adults
coordínate moral and social perspectives, principies, and
precedents in reflective ways unseen in young teens. Adult
groups sometimes coordínate their arguments and perspectives
in reflective forms of argumentation not seen in a middle school
class. Development beyond the age of 12 or 13 years is best
seen as advanced development, differing from the basic
development of childhood in that it is no longer universal and
closely tied to age. Basic development can be expected for any
normal individual in any normal human environment. The
nature and extent of advanced development is much more
dependent on specific individual and social circumstances,
such as formal education.
All of this requires some rethinking of what we mean by
psychological maturity and of the natural tendency of adults to
perceive adolescents as immature.

Immaturity is a meaningful and important biological

concept, . . . not just a label for those not yet like us. It is
not sufficient to assume we know it when we see it. To
say an organism is immature implies there is something
still to develop. But there is no evidence for a universally
achieved state of maturity yet to be reached beyond
adolescence. What we perceive as immaturity is the
possibility of further development. Development often
continúes long beyond childhood and arguably remains
possible throughout adulthood, but development beyond
childhood, unlike child development, is not the attainment
of a natural and universal state of maturity.
(Moshman, 2013, p. 169)
 Understanding the distinction between basic and advanced
development leads to a somewhat counterintuitive conclusión
about education, which will play an important role in the next
chapter. Basic development will take place in any normal
human environment, so to ensure basic development it may
suffice to see that children develop in supportive environments
in which they receive adequate nourishment, attention,
cognitive challenges, social opportunities, etc. But advanced
development requires something more; it can only be expected,
or at least is far more likely, in environments that specifically
promote it. Education for rationality should of course begin
early and be a serious consideration in elementary education,
but it is arguably most important in middle school and beyond,
where progress in rationality can no longer be taken for
granted.

Development of Rational Self-Governance

To be a rational agent is to believe and act on the basis of
reasons, rather than being driven by external causes. To be a
rational agent, then, is to be self-governing. Development can
thus be construed as progress in rational self-governance. In
individuáis, the development of rational self-governance may
be seen as the development of personhood. In interpersonal
groups and social systems, the development of rational self-
governance is the development of deliberative democracy.

Personhood
Who counts as a person? Psychologically, a person is at
mínimum a rational agent, so one criterion for personhood is
rational agency. Even a grasshopper is a biological agent. But
although the grasshopper acts, it does not have reasons for
what it does. We may ask why the grasshopper jumps, but the
expected response is a causal theory of elementary behavior,
not an account of the grasshopper’s beliefs and valúes. The
grasshopper is an agent, but presumably not a rational agent.
 What about a lizard? A parrot? A dog? A chimpanzee? A 9-
month-old human ¡nfant? A 2-year-old child? Whether some or
all of these are rational agents is not clear, depending on how
we define rational agency and on empirical facts about various
species and developmental milestones. What does seem clear,
especially from research on young children’s metacognition and
theories of mind (Chapter 4), is that human beings attain
rational agency by the age of 4 years. More plausibly, I think,
children are rational agents in some minimal sense long before
that, and attain a higher level of rational agency at about age 4
years.
But democracy requires equal respect for all persons, which
entails legal equality in fundamental rights and liberties,
including the right to vote. If the criterion for personhood is
nothing more than rational agency, it seems to follow that we
must accord full rights to all human beings who have reached
the age of 4 years and arguably also to even younger children
and some animals. The more defensible alternative is to
distinguish levels of rationality, and thus levels of personhood,
and set a threshold of full personhood. We must avoid setting
the threshold too high, however. Based on the developmental
research reviewed in this book, it seems reasonable to set the
criterion of full personhood at the basic level of rational agency
attained by all normal human beings in all human cultural
contexts about the age of 12 or 13 years. We must of course
promote development beyond that, but a more stringent
criterion of full personhood would exelude many people of all
ages and thus undermine democracy.
We must of course respect the personhood of all rational
agents. For human children, such respect enables further
development of rational agency to the point of full personhood.
Even where it doesn’t, however, there is a strong case to be
made that at least some mammals, and probably some birds,
are entitled to some degree of respect for their own choices.
We must also recognize our moral obligations to all who can
suffer, even if they are not rational agents. At the very least, we
have an obligation not to cause suffering needlessly, and
arguably we have a moral obligation to relieve suffering in
some cases. Rational agents have a moral responsibility to
treat each other as rational agents, but there are moral
obligations that extend beyond that.
Nevertheless, rational agency is central to personhood.
Democracy requires a criterion of full personhood carrying full
entitlement to fundamental rights and liberties. Developmental
research suggests a criterion that distinguishes persons
beyond the age of 12 or 13 years from children. Many have
proposed that the voting age be lowered to 16 years (Hart &
Youniss, 2018). Psychological research, I suggest, justifies the
more radical step of lowering not just the voting age but the age
of majority, and lowering it not just to 16 years but rather to 14
years. This is not to say 14-year-olds should be expected to
function on their own without social support. We all need social
support. But even young teens are much more like adults than
children, and many young teens are more rational, by any
measure, than many adults. What modern societies take to be
the problems of adolescence, it appears, are largely due not to
biologically immature brains but to the paradoxical social status
of being psychologically an adult in a culture that treats you as
a child (Moshman, 2011a, 2013).

Interpersonal and Societal Development

What about groups? Like individuáis, groups often grow and
change. But do groups develop? Neither growth ñor change is
necessarily a good thing, but if groups potentially develop,
there may be reason to promote the development of groups,
not just of the individuáis who compose them.
Development may be defined as a type or pattern of change
that is extended over time, self-regulated, qualitative, and
Progressive (Moshman, 2011a). In order to think more
rigorously about developmental change we must consider each
of these four criteria.
First, and perhaps most obvious, developmental change
takes place over an extended period of time. In the case of
individual development, that means developmental change
takes place over some portion of the lifespan. In the case of
very young infants, that may be just a matter of weeks or even
days, considering the period of change relative to their current
age. For older individuáis, development generally refers to
change over a period of months or years, typically years for
adolescents and adults. Small groups may also develop over a
period of months or years; for large social systems
developmental change is generally observed over years or
decades.
Second, developmental change is directed or regulated by
internal organismic or systemic processes rather than being
imposed by others or strongly shaped by environmental forces.
Thus, development is not simply a matter of learning many
things over a long period of time. Nativists see development as
a process directed by the genes. Rational constructivists see
development as the outcome of the rational and constructive
processes of a self-regulating organism. The rational
constructivist view also enables us to see developmental
change in enduring groups and social systems.
Third, developmental change involves a succession of
qualitatively distinct structures or modes of functioning.
Growing larger is not, in itself, a developmental change. Among
individuáis, for example, development can be seen in the
successive stages of metalogical understanding discussed
earlier in the chapter or in the transition from objectivist to
subjectivist to rationalist epistemologies noted above and
discussed in Chapter 4. Among groups, similarly, development
involves a transition to something new and different, not just
more of the same.
Finally, to deem a change developmental we must show that
it involves progress toward something better. A key purpose of
the concept of development is to distinguish Progressive
changes from changes that are arbitrary, neutral, regressive,
pathological, or specific to particular environments or cultures.
In its original biological meaning, development is progress
toward maturity, a conception that also works well in children’s
psychological development, where we can specify sequences
of rational competencies all normal children construct over the
course of their first 12 or so years. As we have repeatedly
seen, however, there is no state of maturity in the future of all
adolescents. Nevertheless, we can speak meaningfully of the
development of rational agency beyond childhood in at least
some individuáis provided we can specify what they are
constructing and how it represents an advance over what carne
before. Similarly, in the case of groups or societies there is no
universal state of maturity toward which they naturally tend.
Nevertheless, it seems reasonable to suggest that transitions to
higher levels of deliberation and democracy can be seen as
progress in rational self-governance and moral respect for
persons. To the extent that Progressive transitions are also
extended, self-regulated, and qualitative, they qualify as
development.
Do we really need to restrict the use of the term
“development” in these ways? I suggest it is useful to do so.
Individuáis, groups, and social systems routinely change in a
variety of ways. If we referred to all such changes as
development, the term would serve no purpose. By limiting
development to changes that are extended, self-regulated,
qualitative, and Progressive, we identify an important
subcategory of change that transforms the individual, group, or
social system, and that we may, in some cases, choose to
promote.
It is common, for example, to speak of “economic
development” when measures of economic activity show gains.
Development in the strict sense, however, means something
more than growth. To speak meaningfully of economic
development, there must be a self-regulated and qualitative
transition in economic structures or modes of functioning that
can reasonably be seen as progress.
In Development as Freedom, the Nobel prize-winning
economist Amartya Sen (1999) argued that development is
furthered by conditions of liberty and respect for human
agency. Freedoms of speech, press, and assembly, along with
other civil liberties and opportunities, promote the economic
development of nations. Sen rejected the common assumption
that freedom must be sacrificed to make development possible
and is thus a luxury to be enjoyed only in economically
advanced societies. He showed instead that freedom is crucial
to economic progress. Freedom, however, is not just the means
of development but equally its end. Development is not just a
matter of increasing wealth. Wealth, argued Sen, is valuable for
the freedoms it permits. We can choose whether or not to have
or do whatever we can afford. Genuine development, then,
enhances our capabilities and options. Development, in Sen’s
analysis, is progress toward freedom.
Although Sen (1999) was concerned with the economic
development of nations, his analysis applies to individual
development as well. Both individuáis and nations are active
agents in their own development, not resources to be
developed by others. People cannot develop other people, ñor
can nations develop other nations. We can promote the
development of others, but only by enhancing their agency, not
by restricting it. Restrictions on agency are in fact
counterproductive; they generate resistance because
consciousness of agency and commitment to freedom are
universal. Thus, Sen concluded, we must respect freedom in
order to promote development, which, it turns out, is the
promotion of freedom. For both individuáis and nations,
freedom is both the means and the end of development.
One reason to promote the development of individual
rational agency is to enhance the ability of individuáis to
particípate in deliberative and democratic groups and social
systems. It is likely, however, that such participaron is crucial to
individual development. Participaron in childhood groups both
requires and supports moral reasoning, for example.
Participaron in social systems both requires and supports
precedent-based reasoning. This may seem paradoxical but it
is simply developmental. We develop within groups and
societies, with complex reciprocal influences across individual,
group, and societal levels, and with meaningful potential for
developmental change at each of these levels.

Conclusión
The emergence of rationality is neither a maturational process
directed by genes ñor a process of learning from one’s
environment, though genes and environments are obviously
important. Moreover, it cannot be explained as an interaction of
these two factors, though hereditary and environmental factors
are deeply interconnected. Rather, rationality is constructed
over time by increasingly rational agents through rational and
social processes of reflection and coordination. Subjectivity is
inevitable, but reflection on subjectivity enables progress to
metasubjective forms of objectivity and increasingly rational
self-governance at individual, social, and societal levels. The
rational agency of an individual or group, I conclude, is the
outcome of a rational process of construction whereby rational
agents, individually and collectively, enhance their rationality.
This picture of rational agents becoming increasingly rational
sounds pretty good, but of course it’s an idealization. In the real
world, rational agency and rational developmental processes
are much more elusive than we would like them to be, which
means there is plenty of room for improvement. We now
consider how we can promote rationality by promoting its
development.
8
EDUCATION FOR RATIONALITY

Education serves many purposes but is widely seen as

concerned in large part, if not most centrally, with the promotion
of rationality (Moshman, 2009b). If rationality were simply a
collection of learnable skills, the challenge of education for
rationality would be to identify those skills and determine how
best each of them can be taught. Extensive research on
development and education has made it clear, however, that
rationality cannot be reduced to a set of discrete skills and
cannot be learned by a direct process of social transmission.
On the contrary, rationality develops through constructive
processes and is best fostered by education aimed at
promoting development.
As we saw in Chapter 7, peer interaction encourages
processes of reflection and coordination central to development
and thus plays a central role in the development of rationality.
In this chapter we extend this insight to consider the role of
peer argumentation in education for rationality. The discussion
then turns to the prevalence of indoctrination in educational
curricula in fields including Science and history and to
indoctrinative modes of instruction that circumvent or
undermine rational agency rather than promoting its
development. As a general response, I suggest a
reconceptualization of academic freedom as intellectual
freedom in academic contexts, including freedoms of teaching,
learning, and inquiry. Academic institutions that respect the
academic freedom of all engaged in academic work are most
likely to succeed in promoting the development of rational
agency. The chapter concludes with an overview of the skills,
knowledge, and valúes of an ideal rational agent.

Critical Thinking
Critical thinking has long been seen by philosophers,
educators, and others as a crucial part of education and
arguably its most central goal. Although critical thinking has
many definitions, it is widely seen as involving a concern for
reasons and thus as closely connected to rationality. To think
critically is to seek, respect, provide, and evalúate reasons.
Thus critical thinking is, essentially, reasoning.
Harvey Siegel (1988, 1997, 2018) specifically advocated
what he called “the reasons conception” of critical thinking,
“which holds that the critical thinker is one who is appropriately
moved by reasons” (Siegel, 1988, p. 2, emphasis in original).
He suggested that critical thinking is the “educational cognate”
of rationality. That is, “critical thinking involves bringing to bear
all matters relevant to the rationality of belief and action.” Thus,
“education aimed at the promulgation of critical thinking is
nothing less than education aimed at the fostering of rationality
and the development of rational persons” (Siegel, 1988, p. 32).
Critical thinkers are logical when they ought to be, but it
should be clear that critical thinking is much more than making
logical inferences. It should also be clear (especially from
Chapter 2) that there is no need to teach logic, at least not at
the level of basic inferences. The basic logical inferences
required in the notorious selection task (see Chapters 2 and 5),
for example, are all routine in early childhood. What develops
later, and what we ought to promote, is metalogical
understanding and control of inferences.
Richard Paul (1990) distinguished “weak” from “strong”
conceptions of critical thinking and advocated the latter.
Students who learn to think critically, he warned, generally
apply their critical abilities mostly to ideas they disagree with,
thus enabling them to maintain their own ideas. Such thinking is
critical only in a weak sense of the word. Critical thinking in the
strong sense includes recognition of one’s own ideological
assumptions and identity commitments and active efforts to
subject one’s own ideas and those of one’s groups to rigorous
critique. The concept of strong critical thinking thus recognizes
that people naturally tend to protect their ideologies and
identities and that education for critical thinking must enable
and encourage students to understand and overcome those
tendencies.
Siegel (1988) similarly questioned whether someone who is
capable of critical thinking but rarely thinks critically, or does so
only in self-serving ways, is really a critical thinker. Critical
thinking, he argued, is not just the ability to assess reasons.
Critical thinking is the actual assessment of reasons, which
cannot be explained by ability alone. The ideal critical thinker
has “certain attitudes, dispositions, habits of mind, and
character traits” that together comprise “the critical spirit”
(Siegel, 1988, p. 39). Critical thinkers are not just able to
assess reasons properly; they are disposed to do so. To have a
critical spirit is to have a certain kind of character, one that (a)
“is inclined to seek, and to base judgment and action upon,
reasons;” (b) “rejects partiality and arbitrariness;” (c) “is
committed to the objective evaluation of relevant evidence;”
and (d) “valúes such aspects of critical thinking as intellectual
honesty,. . . objectivity, and impartiality” (Siegel, 1988, p. 39). It
is not enough that one is able to seek reasons or judge
impartially; the critical thinker is committed to doing so,
regardless of self-interest. To have a critical spirit is to be
“inclined to seek reasons and evidence; to demand justification;
to query and investígate unsubstantiated claims” (Siegel, 1988,
p. 39). Critical thinkers ideally engage their reason assessment
skills in all appropriate contexts, even when their own actions
and most deeply held convictions are challenged. The critical
spirit is most fundamentally “a deep commitment to and respect
for reasons” (Siegel, 1988, p. 39). To have a critical spirit is “to
valué good reasoning, and to be disposed to believe and act on
its basis” (Siegel, 1988, p. 39, emphasis in original).
Consideration of the critical spirit suggests the importance of
teaching in what Siegel (1988) called “the critical manner.”

The critical manner is that manner of teaching that

models and reinforces the critical spirit. A teacher who
utilizes the critical manner seeks to encourage in his or
her students the skills, habits and dispositions necessary
for the development of the critical spirit. This means, first,
that the teacher always recognizes the right of the student
to question and demand reasons; and consequently
recognizes an obligation to provide reasons wherever
demanded. The critical manner thus demands of a
teacher a willingness to subject all beliefs and practices to
scrutiny, and so to allow students the genuine opportunity
to understand the role reasons play in the justification of
thought and action. . . . [T]he teacher must submit her
reasons to the independent evaluation of the student.
(Siegel, 1988, p. 45)

Deanna Kuhn (2019) conceptualized critical thinking as,

most fundamentally, a matter of argumentation. Critical thinking
is not so much an individual property one has to some degree
as something one does with others: “Rather than a fixed
attribute of an individual, critical thinking is a dynamic activity
that may show itself to a greater or lesser extent and in varying
forms depending on the affordances of the context” (p. 161).
That is, critical thinking is collaborative reasoning. In the course
of thinking critically with others, Kuhn has found, individuáis
may develop in ways that enhance their ability to think critically
on their own. This insight is central to her argumentation
curriculum, discussed in the next section.
Consistent with all of these conceptions, critical thinking can
be seen as central to rational agency. To take critical thinking
as the primary goal of education is to conceive education as the
promotion of rational agency. As we saw in the previous
chapter, rational agency is the outcome of a developmental
process. To promote rationality, then, we must promote the
processes central to its development.

Education as the Promotion of Development

As we saw in Chapter 7, the development of rationality is a
constructive process in which rational agents further their own
development through reflection and coordination. To promote
the development of rationality we must engage students in
processes of reflection and coordination. This is not just
important in elementary education, where development in
rationality is relatively rapid and universal. It is no less
important in secondary and higher education, where the aim is
to have students go beyond the “mature” level of rational
agency typically achieved around the age of 12 years. There is
no universal state of maturity beyond that, but there is plenty of
room for further development.
As we also saw in Chapter 7, peer interaction is a crucial
developmental context in which reflection and coordination
promote advances in rational agency. Such interaction is
routine across childhood but opportunities for more advanced
argumentation and development are less universal. Education
for rationality must provide such opportunities.
Kuhn (2005) advocated “education for thinking,” and
proposed that inquiry and argument are central to such
education. Inquiry refers especially to scientific inquiry, which is
both empirical and theoretical (see Chapter 3). It is not enough
that students leam established scientific facts. They should
learn how scientific knowledge is generated, including
processes of seeking and interpreting evidence, proposing and
evaluating theories, and participating in a community of inquiry.
The point is not that they will learn to rely only, or even mostly,
on the results of their own original research. We all rely on
scientists and other experts. But to understand Science and
expertise, and thereby make good decisions about what to
believe, we must understand the nature of Science and the
basis for expert conclusions. Education for thinking promotes
knowledge of inquiry by engaging students in inquiry.
Education for thinking also engages students in
argumentation, which has been central to Kuhn’s subsequent
work (Kuhn, Feliciano, & Kostikina, 2019; Zillmer & Kuhn, 2018;
see also Dacey, 2020; Mercier, Boudry, Paglieri, & Trouche,
2017). In Argüe with Me: Argument as a Path to Developing
Students’ Thinking and Writing, Kuhn, Hemberger, and Khait
(2016) wrote directly for teachers. In a further effort to make her
work even more accessible, Kuhn (2018a) has also published a
student manual written directly for middle school students. In
Buiiding Our Best Future: Thinking Critically About Ourseives
and Our World, she provided a detailed set of steps for
undertaking productive argumentation, and furthering
associated reasoning and writing skills, in middle school and
beyond.
Buiiding Our Best Future presents 44 topics in four
categories. The categories are “A Personal Future” (including
topics such as course selection, nutrition, alcohol, and money
management), “A Community Future” (e.g., school curriculum,
age of majority, taxes and services, eider care), “A National
Future” (e.g., voting rights, health insurance, abortion, capital
punishment), and “A World Future” (e.g., immigration, foreign
aid, the United Nations, weapons of mass destruction).
Students may also choose topics of their own.
For each agreed topic, two positions are identified and each
student chooses a position, leading to a división into two teams.
The teams ultimately compete in a “showdown” of
argumentation including a series of face-to-face individual
encounters. In preparation for the much-anticipated showdown,
students engage in a sequence of activities in both small and
large groups. Guided by the manual, they generate reasons,
evalúate them, identify questions, seek relevant evidence,
consider opposition arguments, formúlate counterarguments,
and prepare “comebacks” (rebuttals) to opposition
counterarguments. For topics in the manual, lists of questions
are provided, with online access to answers that students can
draw on to construct and enhance their arguments.
After the showdown, in a phase called “sharing your
thinking,” students write individual short essays defending their
current position. Regardless of whether their position has
changed, their depth of understanding facilitates their writing
and is reflected in the quality of the final product. For a greater
challenge, students with differing views may collaborate on an
essay explaining the issue and presenting the arguments and
evidence for both sides.
There is also a Teachers Edition of Building Our Best Future
(Kuhn, 2018b), which includes most of what appeared in the
student edition plus explanatory notes and instructional
suggestions. A concluding chapter, unique to the Teachers
Edition, briefly reviews the substantial published evidence for
the effectiveness of the curriculum and provides guidance on
assessing progress in the reasoning and writing of individual
students.
This curriculum promotes good reasoning and writing by
engaging students deeply in the practice of argumentation.
Argumentation promotes reasoning by engaging students in
reflection on and coordination of diverse claims and reasons.
The rich structure of arguments in which they particípate also
serves to make their subsequent writing increasingly rational
and convincing.
The argumentation curriculum also has the potential to
revitalize democratic deliberation. It highlights a variety of social
and political issues that students may recognize as sources of
controversy and engages them in rational discourse about such
matters. Thus even as students develop individually, the
potential for democracy also develops.
For the curriculum to reach its full potential, however, it must
go beyond debate. The prospect of a “showdown” between
opposing teams is highly motivating but insufficient for
deliberative democracy, which requires an orientation toward
achieving a rational consensus to the extent that this is
possible. Thus, the collaborative essay written by students with
differing views seems to me of particular importance. In some
cases, collaboration may result in agreement on at least some
key points. Even where it doesn’t, there is valué in the process
of reaching agreement on an essay that both writers see as
reasonable in its framing of the issue, its presentation of
differing viewpoints, and its coverage of relevant arguments
and evidence. When an essay like this is successful, the writers
don’t just agree to disagree but construct a more advanced
understanding of the nature of their disagreement.

Indoctrination in Curriculum and Instruction

For education to promote the development of rationality, it is
not enough to add a bit of argumentation or other opportunities
for development. Real schools everywhere, especially at the
elementary and secondary levels, are to a large extent centers
of indoctrination, which is arguably not education at all, and
certainly not the kind that promotes rationality or development
(Moshman, 2009a; Ross, 2015).
Two forms of indoctrination can be distinguished, one related
to curriculum and the other to instruction. A curriculum may be
regarded as indoctrinative if it has been devised at least in part
on nonacademic grounds to serve nonacademic purposes,
such as getting students to be more patriotic or to support
governmental or other policies. Regardless of the curriculum,
moreover, instruction may be indoctrinative if it is designed to
coerce belief rather than appealing to students’ rational agency.
In circumventing or undermining the rational agency of
students, indoctrination fails to respect both reasons and
persons, and is thus objectionable on both educational and
moral grounds (Callan & Arena, 2009; Siegel, 1988). Education
for rationality is the educational ideal in that it both respects and
promotes students’ rational agency.
I provide in this section examples of indoctrination in both
curriculum and instruction. I begin with science education and
then turn to aspects of education that raise issues of identity.

Science Education
Science education may fail to promote rationality to the extent
that the curriculum has been shaped by political and religious
pressures. I provide here, as an example, a brief history of
Science education in U.S. elementary and secondary schools
(Laats & Siegel, 2016; Moshman, 2009a). As I will note,
however, even when the curriculum is scientifically justified,
Science education fails to promote rational agency if it is
presented in a manner that indoctrinates students.
In the early 20th century, although there was a consensus
among scientists that species evolve over long periods of time,
the teaching of evolution was seen by most Americans as a
threat to the Christian conception of divine creation of a finite
number of fixed “kinds” of organism; many also perceived a
threat to the Biblical chronology that set the age of the Earth at
6,000 to 10,000 years. The highly publicized 1925 trial of John
Scopes for teaching evolution brought the matter to national
consciousness. Scopes was convicted, and it remained ¡Ilegal
to teach evolution in many states for the next several decades.
Even in states that did not ban the teaching of evolution, the
topic was generally deemed too controversial for the
curriculum.
In 1957, the Soviet Union launched Sputnik, the Earth’s first
artificial satellite, raising concerns in the United States about
trailing in the race for space and, more generally, falling behind
in Science. A resurgence of interest in science education led to
a series of new curricula. As a result, beginning in the early
1960s, biology education increasingly recognized the central
role of evolution in explaining life. Creationists objected, but the
U.S. Supreme Court ruled in 1968 that laws banning the
teaching of evolution were an unconstitutional establishment of
religión.
Opponents of evolution education responded by developing
a new versión of creationism that omitted all mention of God
and supernatural causation. New laws mandated that if public
schools included evolution in the Science curriculum they must
provide “balanced treatment” by devoting equal attention to the
new “scientific creationism.” Those laws were struck down by
federal courts, however, and ultimately in 1987 by the U.S.
Supreme Court, which concluded that “scientific creationism”
had no scientific basis and that balanced treatment laws thus
served no purpose other than to promote Christianity.
Foes and skeptics of evolution now developed a new theory
known as “intelligent design,” which postulated that some
biological systems are so “irreducibly complex” that they could
not have evolved from simpler systems by a process of natural
selection. Thus, they could only be the product of an intelligent
designen Although the intelligent designer was not identified as
God, a federal court ruled in 2005 that intelligent design was no
more scientific than earlier versions of creationism, and it
became clear that efforts to mandate its inclusión in the
curriculum were constitutionally doomed.
But efforts to undermine evolution education continué and,
since the turn of the century, have been expanded to include
attacks on teaching about climate change. New “academic
freedom” laws recognized the right of individual teachers to
discuss the “strengths and weaknesses” of evolution and
alternative views, and extended this to múltiple aspects of
Science deemed problematic, including climate Science. The
problem here is not that such laws protect the academic
freedom of teachers but that they single out Science and limit
academic freedom to particular topics and theories. Teachers
should be free to present the strengths and weaknesses of all
ideas in all areas of study, and students should be encouraged
to think critically in all their classes. Academic freedom does
not permit a teacher to ignore an approved curriculum or to
indoctrinate a captive audience of students in his or her
religious or political views. It does, however, leave teachers
free to present relevant arguments and alternatives and to
promote critical thinking. As I will discuss in detail later in the
chapter, academic freedom is for everyone engaged in
teaching and learning. That certainly includes teaching and
learning about the origin, history, current state, and potential
futures of the Earth and its inhabitants. But it must include the
rest of Science as well, and the rest of education too.
It might be argued that science classes should simply
require students to believe in evolution because this is simply a
matter of believing the truth. Evolution is a well-established
scientific theory with strong empirical support; the same cannot
be said for any creationist alternative. Depending on exactly
how we define science and what versión of creationism we
consider, creationist theories are either bad science (because
they have been disconfirmed) or not science at all (because
they make no testable predictions).
But science education is not a matter of instilling true or
justified beliefs. Evolution education ought to be aimed at
understanding evolution, including the evidence that supports it.
This may be expected in most cases to lead to belief in
evolution, because the evidence is so strong, but it is ultimately
for students to determine what they believe. Although respect
for reasons requires teachers to present a scientifically
justifiable curriculum, respect for persons equally requires
teachers to recognize the right of students to determine their
own beliefs. Biology should be taught in a manner that respects
all students, and their parents, regardless of their beliefs.
Students should learn what scientists believe, and why, but the
school cannot require them to change their own beliefs. Helping
students understand evolutionary explanations, and the
associated evidence, is fully consistent with respect for their
ultímate right to believe as they will.
Education for rationality cannot require belief in evolution,
climate change, or anything else. Coercing a student to believe
something is indoctrination, even if the belief is true. We can
require students to demónstrate understanding and, in some
cases, can try to convince them to hold particular beliefs, but in
the end it is for them to determine what to believe. Respect for
reasons and persons requires an academically defensible
curriculum but forbids us to undermine rational agency by
coercing belief.

Education and Identity

These considerations of curriculum and instruction are relevant
far beyond science education. In many domains, issues of
indoctrination are deeply intertwined with issues of identity.
The teaching of history provides many such examples. In
schools around the world, history is taught in large part with the
purpose of creating patriotic citizens, which often means
manipulating and framing what is taught so as to enhance
nationalist identities and support for the current government
and the ideologies of those in power (Moshman, 2009a, 2020).
Most nations, for example, were founded in large part through
processes of ethnic cleansing, including the elimination of
indigenous groups with national identities of their own, but
history education generally downplays, omits, or denies such
uncomfortable aspects of nationhood (Moshman, 2007b,
2011b).
Many other areas of the curriculum are similarly
compromised by ideological pressures and efforts to mold
student identities. To the extent that the curriculum is not the
outcome of academic decisions by teachers and other experts
qualified and motivated to make such decisions, teaching and
learning are distorted in ways that compromise student
understanding and thus their capacity for rational decision-
making. Even in the case of a problematic curriculum, there is
some potential for rational agency and development if teachers
and students are free to challenge the curriculum. All too often,
however, both teachers and students understand the dangers
of challenging curricular expectations on matters of ideology
and identity.
Education may also be hindered by students harassing each
other, often on the basis of ethnic, racial, religious, or political
differences central to their identities. Such harassment
undermines the mutual respect required for democratic
deliberation, including argumentation among classmates. But
school efforts to crack down on whatever they deem to be hate
speech raise problems of their own, undermining the rights of
belief and expression that are critical to argumentation
(Moshman, 2009a, 2020).
Issues of intellectual freedom in education are not limited to
elementary and secondary schools. Governmental, political,
and religious pressures have compromised higher education for
centuries, and continué to do so. Many of the current
controversies relate to matters of social identity. Racial and
ethnic minorities, political conservatives, supporters of
Palestinian rights, and many others have claimed that their
freedom of expression is restricted by a variety of institutional
regulations and social pressures. Such concerns are often well
justified, both within the academic context and on the campus
at large (Lukianoff, 2014; Whittington, 2018).

Transforming Schools
Greater emphasis on peer argumentation, as discussed earlier
in the chapter, has great potential for making instruction less
indoctrinative by enhancing the role of students in their own
learning. But student argumentation takes place within a larger
context, which may greatly compromise it in two ways. First, if
argumentation takes place within an indoctrinative curriculum it
may fail to enhance rational agency. Second, control of speech
by school officials, including teachers, may also undermine
argumentation, even when it is intended to promote
argumentation by enforcing civility.
Education for rationality requires a context of intellectual
freedom. But the classroom is obviously not a forum for free
speech by anyone on any topic. We need to look more
seriously and rigorously at the nature and role of intellectual
freedom in education, which requires a reconceptualization of
academic freedom.
Intellectual Freedom in Education
As we have seen, what passes for education is often in large
part a matter of indoctrination, especially in elementary and
secondary education. This is inconsistent with the ideal of
education for rationality, which requires an academic context of
intellectual freedom in which rational agents can learn and
develop.
Extending previous discussions about the role of intellectual
freedom in collaborative reasoning (Chapter 5), deliberative
democracy (Chapter 6), and the development of rational
agency (Chapter 7), I now consider the central role of
intellectual freedom in any educational institution or program
aimed at the promotion of rationality. What we need to counter
indoctrination in curriculum and instruction, I suggest, is a
coherent and well-justified conception of academic freedom as
the intellectual freedom to do academic work, including
freedoms of teaching, learning, and inquiry (Moshman, 2009a,
2017). To the extent that an educational institution or program
embraces academic freedom in this sense, it engages the
rational agency of those involved and has the potential to
promote better understanding and the further development of
rational agency.

Academic Freedom as the Freedom to Do Academic

Work
Academic freedom, I suggest, is best defined as the freedom to
do academic work (Moshman, 2017). Academic work includes
teaching, learning, and inquiry, all of which require intellectual
freedom. Thus, academic freedom may be defined as
intellectual freedom in academic contexts, including freedoms
of teaching, learning, and inquiry (Moshman, 2009a).
Academic freedom, thus defined, is central to at least six
distinct but overlapping traditions and literatures (Moshman,
2017). First, and most obvious, is the centuries-long tradition of
academic freedom for college professors, which originated in
European higher education and is represented in the United
States in the policies of the American Association of University
Professors (AAUP). In addition to freedoms of teaching and
research, the literature in this tradition also addresses
academic governance as the collective exercise of academic
freedom (Finken & Post, 2009).
A second tradition addresses the intellectual freedom of
college students, often as part of a larger concern with their
individual rights and civil liberties. This literature usually
encompasses students’ intellectual freedom both within
academic contexts (where I would cali it academic freedom)
and in their personal lives and political activities on campus and
beyond (Lukianoff, 2014; Whittington, 2018).
A third tradition addresses intellectual freedom in elementary
and/or secondary education. Restrictions on intellectual
freedom at these levels are usually the result of community
pressure, administrative action, and/or mandates from
governing boards or legislatures, all of which typically infringe
on the academic freedom of both teachers and students (Lent
& Pipkin, 2013).
Fourth, legal and constitutional provisions may protect some
aspects of academic freedom. In the United States, for
example, First Amendment case law since the 1920s has
provided some protection for intellectual freedom in public
education, reaching a high point in the 1960s and 1970s and
then declining as a result of decisions since the 1980s.
Students and teachers have First Amendment rights as prívate
individuáis in their personal and political lives but generally
cannot count on the First Amendment to protect their academic
freedom (Driver, 2018; Moshman, 2009a; Ross, 2015).
Fifth, a great deal of academic work takes place in libraries
or through access to their collections. This includes, but is not
limited to, academic libraries. Librarians have been strong
defenders of free access to information and ideas and have
developed detailed policies to protect the intellectual freedom of
their patrons and the integrity of their collections and operations
(ALA Office for Intellectual Freedom, 2015).
Finally, many professional and disciplinary organizations
have policies that explain and defend what they commonly cali
“intellectual freedom” in a variety of academic contexts. Topics
addressed in these policies and related literatures include
evolution, sexuality, morality, religión, and national histories
(Moshman, 2009a).
The six traditions are largely isolated from each other, in part
because they differ in focus and in part because they use
different terminology. But they overlap in sharing a principled
concern with the intellectual freedom of those who engage in
academic work. Calling this academic freedom highlights that
overlap and helps clarify its nature. Academic freedom is the
intellectual freedom to teach, learn, or inquire in institutions
devised for these purposes.

Academic Roles and Contexts

Education for rationality requires respect for the rational
agency, and thus the academic freedom, of students. Such
education is inconsistent with indoctrination, which aims to
coerce rather than convince. Accordingly, teachers are
responsible for identifying or preparing an academically
justifiable curriculum and teaching it in academically justifiable
ways that respect students’ rational agency, and thus their
academic freedom. The academic freedom of teachers with
respect to curriculum and instruction is both justified and limited
by their academic responsibility to respect and promote the
rationality of their students, and thus their academic freedom.
Academic freedom, then, protects the academic relation of
teacher and student that distinguishes education from
indoctrination. In addition to protecting the rights of individual
teachers and students, academic freedom protects the process
of education by orienting it toward rationality.
Academic work also includes inquiry, broadly defined to
encompass everything from formal research intended for
publication to informal processes of seeking, interpreting, and
discussing information and ¡deas in order to reach justifiable
conclusions. Freedom of inquiry is crucial not only to the pursuit
and publication of formal research but also, as we have seen,
to forms of student learning that engage students’ rational
agency and promote its further development.
It should be clear from this discussion that academic
freedom is not just free speech for all. Neither classrooms ñor
scholarly journals are simply forums for freedom of expression.
This is not to say that intellectual freedom is less important in
academic contexts than in contexts of democratic deliberation.
On the contrary, intellectual freedom is at least as crucial in
academic contexts as in any other. But academic work requires
a subtler conception of intellectual freedom that recognizes the
aims of teaching, learning, and inquiry, and the crucial role of
expertise (Post, 2012).
A classroom differs from a forum for free expression in at
least three crucial ways. First, there is the identification of a
teacher, who is deemed more expert than the students and has
the responsibility and authority to organize the use of class time
to achieve the academic goals of the course. Even when
teachers allot time for student discussion, they frame and direct
that discussion. Second, class sessions are generally devoted
to specific topics within the broader topic of the course. Neither
teachers ñor students are free to talk about whatever they
want. Finally, because academic work is concerned with
knowledge, truth, and the promotion of rationality, it demands a
focus on understanding and justification. Both students and
teachers are expected to explain and justify their ideas and
both may be evaluated on the quality of their explanations and
justifications.
It is clear, then, that teachers may restrict student speech on
the basis of its content in order to keep the class focused on
the topic at hand. Moreover, they not only may but must
discrimínate on the basis of content in evaluating the quality of
student work, assigning higher grades to students who show
better understanding of important concepts and provide
stronger justification of their ideas. Concern for intellectual
freedom does not prevent teachers from giving students failing
grades on the basis of what they say or write. In this respect,
the classroom is neither a forum for free expression ñor a
deliberative democracy.
There is no academic reason, however, for discrimination on
the basis of viewpoint. As discussed earlier in the chapter with
respect to evolution education, students should be free to
believe what they choose but may be evaluated on the basis of
their understanding and their ability to justify their views.
Intellectual freedom in academic contexts, then, is fully
consistent with discrimination on the basis of the content of
expression but not with discrimination on the basis of viewpoint.
Viewpoint neutrality is crucial to intellectual freedom generally
(Moshman, 2020) and to associated legal standards such as
the First Amendment (Strossen, 2018). A standard of viewpoint
neutrality shows respect for persons and reasons and furthers
the promotion of rationality agency.
The standard of viewpoint neutrality applies equally with
respect to research and publication. Respect for intellectual
freedom is crucial to the academic integrity of a scholarly
journal and of the field it represents, but no academic
publication is a forum for free speech. Journals are limited to
particular areas of study and only publish work that meets
academic standards as determined by expert peer reviewers.
Decisions about what to publish are not content neutral; they
are based on the content of the submitted manuscript.
Nevertheless, the process may be deemed a system of
selection, rather than a regime of censorship, provided the
selection process is viewpoint neutral. Ideally, manuscripts are
rejected if they fail to justify their conclusions, regardless of
viewpoint, and accepted for publication if they are found to
meet academic standards, regardless of viewpoint. Authors
may be required to justify conclusions reached in the
manuscript and remove conclusions they cannot justify, but
they should not be expected to remove viewpoints simply
because reviewers deem those viewpoints false or offensive.
Content discrimination is inherent in the process, but viewpoint
discrimination would undermine the academic integrity of the
journal.

Academic Freedom as individual, Collective, and

Institutional
Academic freedom is the freedom to do academic work, which
is done by individuáis, groups, and institutions. Thus, academic
freedom is exercised individually, collectively, and
institutionally. Education for rationality requires that students
have the intellectual freedom to engage actively in learning
through inquiry and argumentation. This requires that teachers
have the academic freedom to teach an academically
defensible curriculum in a non-indoctrinative manner. The
academic freedom of teachers includes both the authority of
individual teachers with respect to their own classes and the
authority of the collective faculty to assure a coherent and
defensible curriculum across a department or larger academic
unit. The academic work of teachers and students, moreover,
takes place within institutions which themselves require the
academic freedom to conduct themselves in accord with
academic norms of intellectual freedom.
When a college or school system resists political or religious
pressures on the curriculum, for example, it is defending its
academic freedom as an academic institution. But this doesn’t
mean administrators or governing boards control the
curriculum. The academic institution must itself respect the
academic freedom of its faculty and students, the collective
faculty must respect the academic freedom of individual
teachers and students, and individual teachers must respect
the academic freedom of their students. Academic freedom is
not a special power or privilege of some elite individual, group,
or institution. It is a system of intellectual freedom that respects
the rational agency of all involved in the academic enterprise
and the goals of promoting the progress of knowledge and the
rationality of students.
Academic freedom is thus central to the academic integrity
of an academic institution. Intellectual freedom is what
distinguishes education and research from indoctrination and
propaganda. Without academic freedom, an academic
institution is not really academic after all, and thus not what it
claims to be.

Argumentation and Intellectual Freedom

In presenting her argumentation curriculum to teachers, Kuhn
(2018b) noted that adults may be concerned about having
young adolescents address controversial social and political
issues in school. But the students, Kuhn observed, “are entirely
comfortable and eager to talk about issues that they believe
really matter” (p. 207). All that is needed, she has found, is an
occasional reminder to criticize ideas, not people.
Higher education is sometimes singled out as special with
respect to the need for intellectual freedom. There is no good
reason, however, to distinguish secondary from higher
education in this regard, and intellectual freedom is crucial to
the construction of rational agency even in elementary schools
(Chapter 7).
Even if students are comfortable and capable, however, their
parents may fear the process of argumentation will somehow
undermine their children’s identities. A clear account of
academic freedom that includes the right of students not to be
indoctrinated may help convince them that they need not be
threatened by a curriculum that gets their children thinking
about matters they deem not debatable for reasons of ideology
and identity.

The Rational Ideal

If the aim of education is to promote rationality, what outcomes
would indícate we are succeeding? What would graduates
ídeally be like? I suggest here some aspects or components of
what might be called the rational ideal. The rational ideal is not
a state of maturity that everyone, or indeed anyone, fully
achieves. It is an ideal intended to help orient our educational
efforts. Its components include skills, knowledge, and valúes—
some individual, some interpersonal, and some civic.
As discussed throughout this chapter, we cannot require
students to adopt particular beliefs and valúes. Academically,
this would stifle, rather than promote, rational agency. Morally,
it would fail to respect students as persons (Siegel, 1988; Laats
& Siegel, 2016). Requiring students to believe what we deem
true and to valué what we deem good is inconsistent with
education for rationality, which requires respect for the rational
agency of students.
Nevertheless, it is proper and important to have educational
and developmental goals and to find productive and moral
ways of maximizing progress toward them. With that in mind,
drawing on discussion throughout the book, I sketch here some
components of a rational ideal.

Skills
For a start, the ideal rational agent is good at reasoning,
including logical, scientific, principled, and precedent-based
reasoning. But we cannot promote rational agency by teaching
children to make correct inferences. Young children already
make correct inferences about a variety of logical, causal,
moral, and social matters (Chapters 2-3). Progress toward the
rational ideal requires the construction of self-regulatory
abilities and epistemic cognition (Chapter 4).
Rational agents ideally particípate rationally in groups
(Chapter 5). But teachers cannot símply tell students what to do
in groups. We enhance our abílity to contribute to group
rationality by engaging in actual argumentation with real peers
about topics that matter to us in a context of intellectual
freedom. Such experiences facilítate construction of the
knowledge and valúes that support rational group functioning.
The rational ideal also includes rational agents reasoning
collaboratively in contexts of deliberative democracy (Chapter
6). This includes practical skills, such as knowing where, when,
and how to vote (Hart & Youniss, 2018). But the ideal member
of an ideal society is a rational voter who also contributes to
society in ways that go beyond voting. This requires
developmental progress in civic knowledge and valúes.

Knowledge
The ideal rational agent understands both the inherent
subjectivity of knowledge and the possibility of rational
judgment despite this (Chapter 4). But we cannot promote
rational agency by telling people what to believe about
knowledge. Knowledge about knowledge is constructed in the
course of inquiry and argumentation through processes of
reflection and coordination (Chapter 7).
It would be impossible and counterproductive for anyone to
rely entirely on the deliberate and reflective processes
associated in dual process theories with System 2 (Chapter 1).
Psychological functioning in daily life requires efficient
heuristics and adaptive biases. Nevertheless, the ideal rational
agent understands enough about dual processing to identify
and address implicit biases, including social biases such as
racism or sexism. In particular, the ideal rational agent
recognizes and works to overcome the universal bias toward
verification of one’s beliefs and hypotheses (Chapters 5 and 6).
Rational agency ideally includes knowledge about the nature
of identity and the role of identity commitments. We should not
expect others to abandon their deeply-held views upon
encountering ours, even if we have good reasons for our views.
Ideal rational agents, however, may convince each other to
modify their divergent views in ways that come closer to a
justified consensus on at least some issues or considerations
(Chapter 5). Regardless of their own ideologies and identities,
ideal rational agents understand the dangers of
dichotomization, dehumanization, and denial (Chapter 6).
Rational agents ideally show advanced understanding
concerning the epistemologies of logic, science, morality, and
social systems (Chapter 4). They coordínate these rigorously
and effectively in reasoning about complex matters. They
understand the nature and importance of expertise and its
relation to democratic deliberation (Chapter 6).
Rational agents ideally understand the valué of
argumentation in seeking the truth. Ideal rational agents seek
opportunities for collaborative reasoning (Chapter 5).
Finally, the ideal rational agent understands that the
complexity of social problems makes strong disagreement
virtually inevitable and resists the natural assumption that
anyone with sharply differing views must be either ignorant or
evil. Such understanding is critical to deliberative democracy
(Chapter 6).

Valúes
The rational ideal is to valué reasons and persons. We valué
reasons because respect for reasons is what makes us rational
agents, which is how we choose to see ourselves. We valué
persons, at least in part, because it is persons that have
reasons. In valuing reasons we valué inquiry, argument, and
expertise. In valuing persons, we valué liberty, justice, and
mutual respect. In valuing reasons and persons together, we
highlight our rational agency.
The rational ideal is also to valué truth (Moshman, 2015).
Rationality can be merely instrumental, aimed at achieving
whatever goals one might have. The rational ideal, however, is
epistemic, valuing the truth for its own sake, expecting reasons
for belief, and willing to question identity-defining beliefs for the
sake of getting to the truth. As we saw earlier, theorists of
critical thinking see active truth-seeking as fundamental to
critical thinking in its “strong sense” (Paul, 1990) and to what
Siegel (1988) calis the “critical spirit.”
To be rational is also, ideally, to valué rational agency. This
includes valuing one’s own rational agency. The rational ideal is
a rationalist identity— a theory of oneself that highlights and
takes pride in one’s rational agency, construing this as central
to one’s personhood. The rational ideal also includes valuing
the rational agency of others— seeing them as persons with
whom one can engage in argumentation, sometimes as part of
a larger deliberative democracy.
No one comes cióse to rational perfection, but we are all
rational agents capable of making progress in rational agency,
individually and collectively. We should valué that too.

Conclusión
Education should aim to promote rational agency, keeping in
mind an ideal rational agent not as a goal we expect to attain
but as an ideal for which we strive. In the final chapter I aim to
examine this ideal further, beginning by looking directly at its
antithesis.
9
REASONS AND PERSONS

We live in a world of persons— of agents who have reasons for

what they believe and do. Of course we also live in a physical
and causal universe of objects, forces, and realities beyond our
control. But we routinely see ourselves and each other as
rational agents acting on the basis of reasons of our own, not
simply caused to behave by physical, biological, psychological,
or social forces. This world of reasons and persons would not
exist without us, but it is nonetheless real to the extent that we
make it real, and presumably consistent with the complex
causal universe from which it emerges (Witherington, 2011).
The world of reasons and persons is the world of rationality,
autonomy, personal identity, and moral responsibility
(Korsgaard, 2009; Parfit, 1984). This is the world in which we
have the opportunity to construct and promote our individual
and collective rationality, rather than just believing whatever we
happen to believe and doing whatever comes naturally (see
Chapters 7 and 8). And this is the world in which we have the
opportunity to create and maintain deliberation and democracy
(see Chapters 5 and 6), rather than just be buffeted by social
forces within our various groups and social systems.
The opposite of a world of reasons and persons is the world
of the Borg. The Borg is a collective entity in the Star Trek
universe that suppresses the individual personhood of those
who compose it and those it assimilates. The prospect of the
Borg is frightful, and not just in science fiction. We may
ourselves be, or become, the Borg. In the first two sections of
this chapter we confront the Borg and consider whether, as
they never stop reminding us, resistance is futile. The
remaining two sections of the chapter then consider, through a
combination of history and science fiction, fundamental
questions of who counts as a person and what counts as a
reason.

Becoming the Borg

The Borg is a collective entity composed of biomechanical life
forms with limited individual agency who arrive in huge cubic
starships from a remóte quadrant of the galaxy. The product of
thousands of centuries of development, their powers are
formidable. First seen in a 1989 episode of the televisión series
Star Trek: The Next Generation, which takes place in 2365, the
Borg played a major role in later episodes of that show and also
in the subsequent series Star Trek: Voyager, the movie Star
Trek: First Contact, and the 2020 series Star Trek: Picard.

We Borg
The Borg is a fearsome threat because it assimilates all
societies and civilizations it encounters, incorporating diverse
humanoid species and their knowledge and technologies in its
ongoing program of self-improvement. “We are the Borg,” they
inform those they encounter. “You will be assimilated.
Resistance is futile.” This seems intended less as an effort to
threaten or demoralize than as a simple statement of fact,
which makes it all the more threatening and demoralizing.
Individual Borg drones are humanoid in appearance but with
a variety of mechanical implants and attachments, typically
including a mechanical eye and a mechanical arm. The
transformation of individual humans and members of other
sentient species into drones involves the injection of
nanoprobes, microscopic robotic devices that take control at a
cellular level and subsequently play a role in biological and
mechanical maintenance and repair.
The standard Borg ship is a giant cube, highly decentralized,
with no command center and extreme redundancy of functions.
Like the Borg collective, it can remain functional even if much of
it is destroyed. There are estimated to be millions of such ships,
mostly in the distant Delta Quadrant, where the Borg originated
and remain based as of the late 24th century. Borg ships can
reach warp speeds exceeding those of Galaxy-class Federation
starships, such the Enterprise, and have the power to destroy
them. The Borg also maintain a network of transwarp corridors
and hubs through which Borg ships achieve transwarp
velocities that enable them to reach other regions of the galaxy,
including Federation space (which includes Earth). Wherever
they go, they assimilate entire species and civilizations.
Individual drones have no ñames. They are typically
organized in small groups serving particular functions, with
group members going by designations such as Third of Five or
Seven of Nine. There is no group leader, ñor does the group
report to anyone in particular. Beyond the group level, the Borg
are linked in a hive mind such that all individuáis are in constant
contact with all others, never alone with their own thoughts.
They speak and act only on behalf of the collective. They have
no beliefs or reasons of their own. When they announce that
we will be assimilated, it’s nothing personal. The progress of
the Borg collective is all that matters for them and, they
assume, for us.

Resistance is Futile
There is another sense to “We Borg.” We ourselves may be the
Borg. In the real world, I suggest, the threat is not that the Borg
will arrive from a distant región of space and assimilate us, but
that we ourselves are already the Borg, to a considerable
extent, and naturally tend to become more so.
Have we already been assimilated? Is resistance indeed
futile? As we will now see, a major theorist of dual processing
has proposed that creating and maintaining a world of persons
with reasons of their own requires active reflection on
rationality.
“The Robot’s Rebellion”
In The Robot’s Rebellion, Keith Stanovich (2004) applied dual
process theory to the question of respecting and maintaining
our individuality in the face of something much like the Borg
and its nanoprobes. Sticking to Science, rather than Science
fiction, he addressed the fearsome threat of what he called
“replicators,” of which there are two types.

Genes and Memes

The first set of replicators are our genes. Their goal is to
replícate themselves. They valué us only to the extent that we
serve that goal. They don’t care about us as persons.
The second set of replicators are memes, units of cultural
information. They too have no goal other than to replícate
themselves. Such replícation may have little relation to
epistemic valúes of truth or justifiability. They just want to go
viral. They don’t care about reasons.
In considering the goals and valúes of our genes and
memes, I hasten to add, I am speaking loosely and
metaphorically, as if they had some degree of rational agency.
Of course they don’t. When I say that our genes and memes
don’t care about persons or reasons, I don’t mean to suggest
they care more about something else. What Stanovich (2004)
wants us to keep in mind is that they don’t care about anything.
The robot’s rebellion begins with understanding that our genes
and memes are not rational agents and do not care about us as
rational agents. To the extent that they direct our behavior, we
are robots.
Stanovich’s (2004) analysis seems to me somewhat
reductionist (see Chapter 7) in its conception of the direct
causal role of discrete genes and memes in determining
behavior. His conception of our predicament, however, seems
to me generally sound. We should not be robots in Service to
mindless genes and memes. We should not be part of the
Borg. But we are, unless we rebel.
Metarationality
Is resistance futile? Stanovich (2004) does not think so. The
possibility of rebellion lies in dual process theory. We may be
robots, by which he means “vehicles designed for replicator
propagation.” But “we are the only robots who have discovered
that we have interests that are separate from the interests of
the replicators” (Stanovich, 2004, p. xii). The rebellion has two
steps, corresponding to the two replicators.
The first step begins with the recognition that our interests as
persons are not necessarily those of our genes. Our genes
direct automatic processes that serve to replícate themselves.
These processes compose System 1 of a generic dual process
model, which can be seen as the genes keeping us on a short
leash through automatic processes and intuitions. But System 2
allows us to conceptualize our own interests and to override our
automatic impulses by thinking about how best to serve those
interests.
The second step in the rebellion is to recognize that our
thinking is infested with memes, the cultural baggage acquired
from our societies precisely because of its ability to replícate
itself, a process that does not depend on truth or justifiability. It
is not enough to think instrumentally about how best to achieve
what we take to be our goals given our current beliefs and
valúes. We must engage in reasoning, taking truth as an ideal,
requiring reasons from ourselves and others. That is, we must
reflect critically on our beliefs and valúes, coordinating them in
such a way as to achieve a broader and more stable
equilibrium, a deeper level of justification. Humans, Stanovich
(2004) insists, can and must “gain control of their lives in a way
unique among lifeforms on Earth— by rational self-
determination” (p. xiv).
This requires “meta-rationality” (Stanovich, 2004, p. 265).
Rationality must critique and control itself. We must go beyond
the instrumental rationality of serving what we take narrowly
and without question to be our valúes, desires, and purposes.
We need critical thinking in its strongest forms to achieve more
advanced levels of equilibrium. In the words of economist and
philosopher Amartya Sen (1999),

It is the power of reason that allows us to consider our

obligations and ideáis as well as our interests and
advantages. To deny this freedom of thought would
amount to a severe constraint on the reach of our
rationality.
(P• 272)

We can’t transcend our minds, one might think. But

development, as we saw in Chapter 7, is an ongoing process of
transcending our current mind through processes of reflection
and coordination. Development beyond childhood is not
automatic or inevitable, but progress in rational agency remains
possible and, as argued in Chapter 8, should be the primary
goal of education. The rebellion, it turns out, begins in early
childhood and makes remarkable progress by the age of 12 or
13 years. The challenge is to keep it going.

Borg”
In “I, Borg,” a 1992 episode of Star Trek: The Next Generation,
the Federation starship Enterprise, under the command of
Captain Jean Luc Picard, responds to what seems to be a
distress cali from a nearby moon as they chart an unexplored
star system. Commander William Riker, the first officer, beams
down with security chief Worf and chief medical officer Beverly
Crusher. They find that a small Borg starship has crashed,
leaving four drones dead and the fifth unconscious and
seriously injured.
Dr. Crusher insists on treating him, as she would anyone
needing her medical assistance. Captain Picard is highly
concerned because the Borg are known to retrieve lost drones,
living or dead, but finally agrees to have the Borg drone
beamed up to the Enterprise and isolated in a detention cell
surrounded by a subspace damping field to prevent any
communication between the drone and the collective. Dr.
Crusher tends to him there, with help from chief engineer
Geordi La Forge, who sets up the damping field and replaces
the drone’s damaged mechanical implants. Meanwhile, Picard,
La Forge, and other crew members devise a plan to destroy the
Borg collective from within by introducing into the drone an
invasive program involving a topological anomaly that would
spread through the collective after the drone is retrieved.
Crusher is appalled by what she deems a plan for genocide,
but the other sénior staff believe the extraordinary threat posed
by the Borg leaves no alternative.
The Borg drone finally regains consciousness and explores
its cell, looking for an access terminal so it can interface with
the collective. The search is unsuccessful, but Geordi La Forge
later cautiously enters the cell to install a power conduit that will
provide the energy the Borg uses for nourishment.
“We are Borg,” the drone announces. “You will be
assimilated. Resistance is futile.”
When Geordi asks who “we” refers to, the drone simply
repeats, “We are Borg.” Geordi points out that the drone is
alone and singular and asks if it has a ñame or means of
identification. The drone responds, “Third of Five.” Geordi
remarks that the ñame does suit him.
In a later meeting with the Borg drone, Geordi introduces Dr.
Crusher. When the drone asks what a doctor is, she replies that
a doctor heals the sick and repairs the injured. The drone
responds that the sick and injured are reabsorbed and others
take their place. Dr. Crusher explains that she saved his life
and, when he asks why, describes it as her duty.
The Borg drone asks Geordi his designation, leading Dr.
Crusher to explain that they have ñames, not designations. The
drone wonders about having a ñame and Geordi asks if he
wants one. They settle on “Hugh” and make introductions: ‘Tm
Beverly.” Tm Geordi.” “We are Hugh.”
Shortly after, as Geordi examines Hugh’s eyepiece, Hugh
notes matter-of-factly that when they are assimilated they will
have eyepieces too. Dr. Crusher responds that they don’t want
to be assimilated. Hugh remarks on how quiet it is, noting
specifically the absence of other voices. He adds that on a Borg
ship there are always thousands of voices because they all
have each other’s thoughts in their minds. Dr. Crusher
suggests that Hugh is lonely.
Having second thoughts about the plan to use Hugh to
destroy the Borg, Geordi goes to Ten Forward, the Enterprise
bar, where he talks with the bartender Guinan, a recurring
character played by Whoopi Goldberg. Guinan appears human
but is actually an El-Aurian at least five centuries oíd. She is
known for listening and is generally mild and nonjudgmental,
but she has a history with the Borg, which virtually destroyed
her people, leaving her among a remnant of survivors scattered
around the galaxy. She had already cautioned Picard strongly
that the Borg would come to find their lost drone and would
assimilate or destroy the Enterprise. Now, hearing Geordi refer
to the drone as Hugh, she is incredulous: “You named the
Borg?”
Geordi insists she go talk to Hugh. When she responds that
she has nothing to say, he suggests she listen. She resists, but
later goes. Standing outside the detention cell and observing
the drone through the forcé field, she remarks that he doesn’t
look so tough. “We are Borg,” he says, adding, after some
prodding from her about assimilation and resistance, that
resistance is futile.
“Resistance is not futile,” she replies, noting that her people
resisted the Borg and some survived, albeit scattered and
isolated. To her amazement, Hugh reflects on the possibility
that resistance is not futile and then comments on her being
lonely, adding, “We are also lonely.”
Later, as the crew proceed with their plan, Geordi tells Hugh
that they are studying him because the Federation is interested
in learning about other species. Hugh responds, “We assimilate
species. Then we know everything about them.” Geordi
explains that he and others do not wish to be assimilated
because they valué their individuality and sense of self. In
response to Hugh’s questions he acknowledges that
individuation can be lonely and explains friendship. Hugh
concludes that what he is describing is “like Geordi and Hugh.”
In a subsequent discussion of the plan to destroy the Borg,
Geordi tells Captain Picard of his reservations. Picard likens the
drone to a laboratory animal and strongly suggests that Geordi
“unattach.” But later, Guinan shows up to urge Picard to meet
with Hugh, arguing that he should at least look him in the eye
before using “this person” to destroy his race. Picard interrupts:
“It’s not a person, dammit, it’s a Borg.”
Eventually, however, he does meet with Hugh. Drawing on
his personal experience with the Borg, he successfully poses
as a fellow member of the collective and states that the
Enterprise and its culture will be assimilated. To Hugh’s
objection that they do not wish it, he responds that their wishes
are irrelevant. To Hugh’s observation that they will resist,
Picard responds that resistance is futile. When Hugh replies
that resistance is not futile and that some have escaped, Picard
insists they will be found and assimilated, and when Hugh asks
about Geordi, describing him as his friend, Picard responds that
Geordi too will be assimilated and that Hugh must cooperate in
the assimilation. “I will not,” he says, and when Picard reminds
him that he is Borg he replies, “No, I am Hugh.”
Picard is taken aback by Hugh’s rejection of Borg aims and
astonished by his use of the first person. Having heard what he
didn’t want to hear, he concludes that the Enterprise cannot
proceed with the plan to use Hugh as an instrument to destroy
the Borg.
The sénior staff meet to consider alternative options, which
include erasing Hugh’s memory of everything that happened on
the Enterprise and then beaming him down to the moon to be
found by the Borg. They decide instead to beam him down with
his memory intact in the hope that his reconnection to the Borg
will result in his new knowledge of self and sense of
individuality spreading rapidly through the collective. But Dr.
Crusher points out that he may not want to go. They agree that
the choice is his.
Picard and Geordi explain to Hugh that it is up to him to
choose whether to beam down and be retrieved by a Borg ship,
which is approaching, or seek asylum and stay on the
Enterprise. “Choose what I want,” says Hugh in wonderment,
trying to grasp the concept. He wants to stay with Geordi,
whom he regards as his friend. But he knows the Borg will
come for him, endangering Geordi and others. In what
Stanovich (2004) would cali meta-rationality, he recognizes his
good reasons for wanting to stay but overrides them on the
basis of further reflection, choosing to beam down despite what
he wants. On the transporter pad about to beam down, he tells
Captain Picard, “I do not want to forget that I am Hugh.” At a
loss for any response, Picard simply says “energize,”
instructing the transporter chief to beam Hugh down.
Subsequent episodes of Next Generation showed that
Hugh’s link to other Borg in his ship resulted in a spread of
individuality that, at least initially, radically disrupted their
collective functioning, with dire consequences for former drones
unable to function as persons, many of whom fell under the
spell of an unscrupulous leader. The long-term consequences
of Borg individuality played a central role in the 2020 series
Picard, which takes place decades later.
Is there hope for the Borg, or for us? To avoid becoming the
Borg, or at least to be less like them, we must respect reasons
and persons. But to do that we must address difficult issues of
who counts as a person and what counts as a reason.

Who Counts as a Person?

Ancient Athens is properly lauded for its direct democracy and
equal respect for all who participated, including equal
opportunity to speak (Saxonhouse, 2006). It was simply a
matter of common sense, however, that slaves and women did
not count as persons entitled to equal respect or freedom of
speech. Looking back now, almost everyone reading this book
will probably consider it obvious that no one should be a slave
and that women should count fully as persons. But the question
of who counts as a person and in what sense remains
fundamental to any social system. It is inconsistent with
democracy for some persons to count more or less than others.
All those who count as persons, at least in some full sense of
personhood, must count equally. But who counts as a person?

“Standing Bear Becomes a Person”

I first learned of Standing Bear around 1980 when I read the
chapter “Standing Bear Becomes a Person” in Dee Brown’s
(1970) classic Bury My Heart at Wounded Knee: An Indian
History of the American West (see Starita, 2008, for a more
comprehensive account). Wounded Knee was the South
Dakota site of an 1890 massacre in which U.S. troops shot
helpless members of a Minneconjou Lakota community of
several hundred, killing most of them. It was far from the only
such massacre. But in the expansión and consolidation of the
United States, mass killing was not the only, or even the
primary, means of genocide:

From the late nineteenth century to the mid-twentieth

century, the governments of the United States and
Cañada ran boarding schools for indigenous children from
dozens of tribes (Adams, 1995; Churchill, 2004). The
intent was to elimínate indigenous cultures by severing
the link between generations and assimilating the children
to American society. No one would be killed, in theory, but
the indigenous cultures would disappear with the passing
of the present generation.
The students, accordingly, were isolated from their
families and communities. They received “white man’s”
ñames to replace those by which they had been known.
They were issued clothing appropriate for their gender in
“civilized” society. Boys, often to their great dismay, had
 their long hair cut. They were required to leam English
and forbidden to speak their native languages. Christian
beliefs and middle-class valúes dominated the curriculum.
History was taught as the progress of civilization in the
Americas since 1492. Girls were trained in domestic skills
and boys in agricultural and industrial skills so they could
function in society.
The motto was “Kill the Indian, save the man.” Indian
schools proudly illustrated their success with before-and-
after photographs that showed young savages
transformed into civilized Americans. The children were
“saved.” What was “killed” was the social identity that
would otherwise have linked them to their childhoods and
ancestral cultures.
(Moshman, 2011b, p. 925)

In the era of genocidal massacres such as Wounded Knee,

the boarding schools were supported by liberáis as the more
humane approach to what was seen as an inexorable historical
process. The “vanishing” of the “Indian” was deemed inherent
to the progress of civilization, a matter of “manifest destiny.”
The natives and their cultures would be assimilated. Resistance
was futile.
But resistance was not always futile. The Ponca are a small
tribe with a long history on land by the N¡obrara River in what is
now northeast Nebraska, near the South Dakota border. Their
number dropped as low as 200 after a smallpox epidemic
wiped most of them out in the winter of 1800-1801. Over the
following decades they suffered many hardships, including
repeated attacks by other tribes, until eventually they were
informed that their land had been given away by the U.S.
government in a treaty with a more troublesome tribe and were
forced in 1877 to move to “Indian Country,” in what is now
Oklahoma, despite the strong objections of the Ponca chiefs,
including Standing Bear. Many died along the way, especially
children, as they walked 600 miles south over the course of two
months, often in harsh weather, on what became known as the
Ponca trail of tears. (The term “trail of tears” originated in
connection with earlier remováis of the Cherokee and other
nations from the Southeastern U.S. to lands beyond the
Mississippi.)
Within 18 months, a quarter of the tribe had perished, mostly
of disease in what the Ponca called “the Warm Country.”
Acceding to his teenage son’s dying wish, Standing Bear
promised to bury his bones back by the Niobrara. The other
chiefs thought the trip back beyond the capability of the
surviving Ponca, but Standing Bear set off with 30 men,
women, and children and returned to Nebraska, in another
harrowing trip. There, he and his people were apprehended by
the U.S. military under the command of General George Crook,
who had orders to return them to Indian Country.
General Crook, based at Fort Omaha, was responsible for
the Department of the Platte, a región stretching from Nebraska
west to Montana, north to Cañada, and south to Texas. He was
the veteran of more than two decades of military campaigns
against western tribes, and had long seen the natives and their
ways as a primitive roadblock to the progress of civilization.

But in recent years, he had grown weary of the broken

treaties, the unprovoked massacres of women and
children, the corruption in Washington, the moral
bankruptcy of the reservation agents. And after more and
more contact . . ., [h]e began to see the Indian as a
person whose beliefs and culture had sustained him for
centuries, someone who could not be forced to adopt
radically new valúes and traditions in a few short years.
(Starita, 2008, p. 118)

Crook would follow his orders if he had to, but he did not
wish to. He contacted Thomas Henry Tibbles, assistant editor
of the Omaha Daily Herald, who interviewed Standing Bear and
wrote extensively about the plight of the Ponca, leading to an
upsurge in popular sympathy and national publicity. Tibbles
also recognized the possibility of a lawsuit based on the 14th
Amendment and convinced two of Omaha’s top lawyers to take
the case pro bono (without pay). One, John Lee Webster, was
an expert on constitutional matters. The other, Andrew Jackson
Poppleton, was general counsel of the Union Pacific Railroad
and later the first president of the Nebraska Bar Association.
Meanwhile, General Crook, still sitting on his orders to get the
Ponca to Indian Country, could only delay so long. Apparently
at his suggestion, Standing Bear’s attorneys sought a writ of
habeas corpus from Judge Elmer Dundy requiring General
Crook to bring Standing Bear to court and respond to Standing
Bear’s charge that he was unlawfully detained.
Dundy issued the writ and the trial took place in May 1879.
After the lawyers for Standing Bear and for the government had
completed their arguments, Judge Dundy announced that
Standing Bear had requested permission to make a closing
statement and, though this was unprecedented, he had granted
the request. Standing Bear faced the audience, holding out his
arm, and then turned to the judge (assisted by a translator):

That hand is not the color of yours, but if I pierce it, I shall
feel pain. If you pierce your hand, you also feel pain. The
blood that will flow from mine will be of the same color as
yours. I am a man. The same God made us both.
(Starita, 2008, p. 151)

Turning back to the audience, Standing Bear went on to tell

a story of standing with his wife and little girl on the bank of a
wide and impassable river, its waters rising rapidly, with cliffs all
around, and finally, scanning the cliffs in desperation, spotting a
steep, rocky path to safety. But as they climb the path, waters
rushing in behind them,

a man bars the passage.. . . If he says that I cannot pass,

I cannot. The long struggle will have been in vain. My wife
and child and I must return and sink beneath the flood.
We are weak and faint and sick. I cannot fight.
Turning again to face the judge, he added, “You are that man”
(Starita, 2008, p. 151).
There was silence in the court. The judge struggled to
maintain his composure. Some women in the back could be
heard weeping. General Crook leaned forward on the table, his
head in his hands. Then he stood up, walked across the
courtroom, and shook Standing Bear’s hand.
Ten days later, Judge Dundy released his detailed ruling. He
noted that if sympathy were decisive he would have ruled for
Standing Bear immediately after the closing arguments. But the
case must be decided on the basis of law, and given the state
of the law, the ruling did not challenge governmental authority
over tribes. It implicitly accepted without question that, for
Indians, the only alternatives to mass killing were (1) ethnic
cleansing by the removal of tribes to reservations or Indian
Country or (2) the assimilation of individuáis into American life.
Drawing on arguments from Standing Bear’s attorneys, he
carefully responded point by point to the government’s legal
arguments. There were two key aspects to the ruling.
First, Dundy rejected the government’s argument that the
writ of habeas corpus was ¡Ilegitímate because Standing Bear
was not a Citizen. Careful examination of the law made it clear
that any person, regardless of citizenship, could file for a writ of
habeas corpus. Was Standing Bear a person? Dundy quoted
from Webster’s dictionary a definition of a person as, “a living
soul; a self conscious being; a moral agent; especially a living
human being; a man, woman, or child; an individual of the
human race” (Starita, 2008, p. 155). He found this definition
satisfactory for legal purposes and concluded that an Indian is
a person and that the writ was thus proper.
Second, without questioning governmental authority over
native tribes, Dundy argued that the law supported Standing
Bear’s case that natives can expatríate from their tribes and the
facts of the case showed that he had done so and had a history
of farming, educating his children, and living a praiseworthy
American life. Thus, he was protected by the Fourteenth
Amendment to the U.S. Constitution, adopted in 1868, three
years after the Civil War, the first section of which reads:

All persons bom or naturalized in the United States, and

subject to the jurisdiction thereof, are citizens of the
United States and of the state wherein they reside. No
state shall make or enforce any law which shall abridge
the privileges or immunities of citizens of the United
States; ñor shall any state deprive any person of life,
liberty, or property, without due process of law; ñor deny
to any person within its jurisdiction the equal protection of
the laws.

The interpretation of the 14th Amendment was and remains

a matter of considerable uncertainty and dispute, but Judge
Dunby concluded that there was no valid legal basis for
Standing Bear’s imprisonment and ordered his release. He and
other Ponca eventually returned to their homelands, though not
without further difficulties. After much deliberation, the
government decided not to pursue an appeal, worried that the
U.S. Supreme Court might uphold the decisión and set a
national precedent that would undermine its Indian policy.
Psychologically, Standing Bear had already been a rational
agent, and thus a person. Now he was a person in the eyes of
the law, with a right to choose for himself.

“The Measure of a Man”

Nearly five centuries later, in 2365, the starship Enterprise
arrives at Starbase 173, a new outpost of the United Federation
of Planets (in “The Measure of a Man,” a 1989 episode of Star
Trek: The Next Generation). Commander Bruce Maddox, an
expert on robotics, comes aboard and takes immediate notice
of Lt. Commander Data, an android who serves as a sénior
officer, who likewise takes notice of Maddox. Maddox explains
to Captain Picard that he evaluated Data’s application to
Starfleet Academy, and Data adds that Maddox cast the solé
vote against his admission on grounds that he was not a
“sentient being.”
Maddox explains that he has made it his life’s work to
understand Data, who was created by an eccentric genius in
work that has never been replicated, and is cióse to a
breakthrough that will enable the production of múltiple
androids like Data. The next step in his research requires him
to “disassemble” Data, which will include downloading the
memories from his positronic brain to the Starbase 173
Computer. Mr. Data, who has devoted substantial effort to
studying himself, is initially intrigued. After questioning Maddox
about technical specifics, however, he concludes that Maddox
lacks the knowledge to carry out the procedures safely and
successfully. Picard tells Maddox he cannot permit the
research.
Maddox responds with a step that he says he had hoped
would not be necessary. He produces Starfleet orders
transferring Data from the Enterprise to Starbase 173, where
he would be under Maddox’s command. Data later informs
Picard that he will not submit to the procedure. Looking for legal
options, Picard consults Captain Philippa Louvois, who has
been appointed to set up and head the new Judge Advócate
General’s office for the sector. She says Mr. Data can refuse to
undergo the procedure, but the transfer cannot be stopped.
Picard worries that once Maddox is in charge of Data he will do
whatever he wants. Is there no other option? There is always
an option, responds Louvois. Data can resign.
Data promptly resigns from Starfleet. He describes himself to
Maddox as “the culmination of one man’s dream,” an addition
to “the substance of the universe.” “If by your experiments I am
destroyed,” he explains, “something unique, something
wonderful will be lost. I cannot permit that.”
At a subsequent meeting with Louvois and Picard, Maddox
dismisses Picard’s concern about Mr. Data’s rights as a
sentimental reaction to Data’s human appearance. “If it were a
box on wheels,” he insists, he would not be facing this
opposition, asking them to imagine a future in which every ship
in Starfleet has a Data on board. He argües that Data cannot
resign because he is not a person. He is the property of
Starfleet. He cannot refuse the procedure any more than the
ship’s Computer can refuse a refit.
After looking into the matter, Captain Louvois rules that the
law supports Maddox’s position. Pressed by Picard, she tells
him that if he challenges the decisión she must arrange a
hearing but reminds him that she does not yet have sufficient
legal staff to conduct the hearing. When Picard insists there
must be regulations to cover the situation, she replies that there
are. Mr. Data may appeal, she will serve as judge, Picard will
represent him, and Commander Riker, the next-ranking officer,
will make the case against him. Riker protests that he cannot
do so because Data is his comrade and he regards him as a
friend, but the judge threatens that she will dismiss the appeal if
he declines the role or does not take it seriously.
Riker prepares for the hearing by studying Data’s technical
schematics. As the hearing begins, he calis Data to the stand
and demonstrates that Data is a machine, questioning him
about his storage and processing capacity, illustrating his great
strength. He removes Data’s hand and shows it to the judge.
Finally, based on his technical research, he does something
few knew was even possible. Reaching behind Mr. Data, he
turns him off.
The demonstration is stunning, and Picard requests a
recess. He goes to Ten Forward, where Guinan tends bar, to
express his dismay about the case. Guinan notes that if
Maddox somehow succeeds in creating a multitude of Datas
they would all be very valuable, adding, “and now he’s been
ruled the property of Starfleet. That should increase his valué.”
How so, asks Picard. Guinan responds that there have been
disposable people in the history of many worlds who do the
dirty work that no one else wants to do because it is too difficult
or too hazardous. No one need think about the welfare of such
people or worry about how they feel. The replication of Data,
now that he is Starfleet property, would create whole
generations of disposable people. “You’re talking about
slavery,” says Picard, to which Guinan responds in her usual
mild manner that this characterization seems a bit harsh.
But Captain Picard does not think this is too harsh, and
when the trial resumes he acknowledges immediately that Data
is a machine, as are we all, though of different types, and calis
him to the witness stand. He asks Data about several items he
packed after resigning. Why did he pack a display of his
medals? What purpose do they serve? Data responds that he
just wanted them, wondering, “Is that vanity?” Asked about a
book Picard gave him, he says he valúes it as a reminder of
friendship and Service. Finally, there is a holocube of Tasha
Yar, an Enterprise security chief who died dramatically in a
previous episode. At first Data declines to discuss it, explaining
only “I gave my word.” Picard convinces him Tasha would have
wanted him to speak under these circumstances. Data finally
says she was “special” to him, adding, “we were intímate.”
Picard now calis Commander Maddox to the stand and
questions him about the criteria that must be met to qualify as a
“sentient being” and thus a person with rights. Maddox refers to
intelligence, self-awareness, and consciousness. Picard
challenges Maddox to prove to the court that he, Picard, is
sentient. Maddox scoffs. Picard asks whether Data is
intelligent. Maddox acknowledges that Data is intelligent in that
“it” is capable of learning, understanding, and coping with new
situations. And self-awareness?
Picard asks Data what he is doing. “I am taking part in a
legal hearing to determine my rights and my status,” replies
Data. “Am I a person or property?” And what is at stake? “My
right to choose, perhaps my very life.” Data is clearly intelligent
and self-aware, concludes Picard, meeting two of Maddox’s
three criteria. And consciousness? How does anyone know
Data lacks whatever that is?
But Maddox insists his research on Data is crucial so that he
can “leam from it and construct more.” How many more? “As
many as are needed,” responds Maddox, suggesting the
number could be in the hundreds, even in the thousands. A
single Data is a curiosity, says Picard, maybe even a wonder,
“but thousands of Datas, isn’t that becoming a race? And won’t
we be judged by how we treat that race?” Reminding the court
that Starfleet was founded to seek out new life, he concludes:
“Well there it sits, waiting.”
“It sits there looking at me,” says the judge after a moment of
uncertainty, “and I don’t know what it is.” She complains that
she is not competent to address matters of metaphysics or
questions best left to saints and philosophers, but accepts that
she must make a ruling. Data is a machine, she grants, but he
is not the property of Starfleet. The basic issue, which she
suggests everyone has been dancing around, is whether Data
has a soul, adding that she doesn’t know whether he does or
whether she herself has one. “But I have got to give him the
freedom to explore that question himself. It is the ruling of this
court that Lieutenant Commander Data has the freedom to
choose.”
Mr. Data promptly chooses not to undergo Maddox’s
procedure but urges him to continué his research, which he still
finds intriguing. “He is remarkable,” says Maddox after Data
leaves. “You didn’t cali him ‘it,’” observes the judge.
Riker stays away from the later celebration as Mr. Data
resumes his position as an officer on the Enterprise. Data
seeks him out and Riker explains he wouldn’t feel welcome
because he argued so strongly against Data. But Data already
knows Riker did that so the appeal would not be summarily
dismissed. “That action injured you and saved me,” he says. “I
will not forget it.” Commander Riker calis him a wise man. “Not
yet, sir” replies Mr. Data, “but with your help, I am learning.”

What Counts as a Reason?

Respect for reasons is crucial to argumentation and lies at the
heart of rational belief and action, and thus of personhood. But
does respect for persons and reasons mean we must accept all
beliefs and actions as equally justified and all reasons as
equally strong? Is knowledge simply a matter of perspective? Is
truth ultimately relative to our various cultural, political, and
religious worldviews? Are there many truths, or none at all? Are
we now, as many have suggested, in a “post-truth” universe of
“alternative facts”? Such questions were central to the
discussion of epistemic cognition and development in Chapter
4, in which I argued on the basis of extensive evidence that
radical subjectivism is not the last word, developmentally or
epistemologically (see also Moshman, 2015). For another
approach to these questions, I go back to a 1968 episode of the
original Star Trek.
In “For the World is Hollow and I Have Touched the Sky,” the
starship Enterprise encounters an isolated asteroid 200 miles in
diameter. Analysis shows that it is moving on an independent
course, at sub-light speed, correcting for gravitational forces. It
is composed, moreover, of a hollow outer shell and breathable
air within. Further analysis of the asteroid’s trajectory shows
more: It is on course to collide in 396 days with a planet
inhabited by over 3 billion people.
Captain Kirk, Mr. Spock, and Dr. McCoy beam inside to
investígate. The asteroid turns out to be a multigenerational
ship that has been en route for at least 10,000 years. Its current
inhabitants, however, many generations removed from those
who initially populated the huge ship, believe themselves to be
on a planet called Yonada. The three officers meet “the High
Priestess of the People.” This being classic Star Trek, she is
young, beautiful, scantily ciad, and, it soon turns out, ready for
love. She introduces them to “the Oracle of the People,” the
source of all knowledge about the will of “the creators.”
Later, Spock reminds Kirk of the prime directive, which
forbids interference with other cultures. Kirk concludes,
however, that the prime directive is overridden in this case by
the clear and present danger of planetary collision.
Suddenly, an oíd man, who has apparently sought them out,
observes that they are not of Yonada. “No, we’re from outside
your world,” replies Kirk. The oíd man asks where that is. “Up
there,” replies Kirk, “outside, up there, everywhere.” The man
nods, knowingly. Many years ago, he tells them, he climbed the
mountains, which is strictly forbidden. “Things are not as they
teach us,” he laments, “for the world is hollow and I have
touched the sky.” Then he drops dead before them. The Oracle
has formidable powers.
Meanwhile, the High Priestess has been falling in love with
McCoy, much to his (and perhaps Kirk’s) surprise. But she
remains serious about her responsibilities to her people, which
include enforcing the truth of the creators as revealed by the
Oracle. Sacrilege is punishable by death. “I hope you men of
space, of other worlds, hold truth as dear as we do,” she says
to McCoy. “We do,” he replies.
While McCoy dallies with the High Priestess, Kirk and Spock
try to determine how to adjust the course of the asteroid ship
but are repulsed by the Oracle and barely escape with their
lives. In the end, armed with nothing but the truth, Kirk
confronts the High Priestess. “Now listen to me,” he demands
repeatedly, insisting she hear what she vehemently rejects as
“your truth of your world.” What she and her people have
believed, he maintains, is false. Yonada is a giant spaceship
that has gone perilously off course. The High Priestess listens
but seemingly still rejects it all. “Why should the truth be kept
from us? Why should the creators keep us in darkness?”
The episode reaches its epistemic climax as she abruptly
runs off to consult the Oracle, which immediately accuses her
of having listened to the words of the nonbelievers. “They said
they spoke the truth,” says the High Priestess, but the Oracle
dismisses whatever they said as “their truth.” In response to her
suggestion that truth is “truth for all,” the Oracle maintains there
can be no other truth for her than “the truth of Yonada,” which is
her truth. But this no longer satisfies her. “I must know the truth
of the world,” she insists.
In the end, Spock and Kirk correct the flaw in the ship’s
control system and get it back on course for the planet it was
meant to reach. McCoy asks the High Priestess to return to the
Enterprise with him. But despite her love for him and her new
knowledge of the world, she remains with her people,
explaining that she understands the great purpose of the
creators and will honor it.
This episode strongly rejects epistemological relativism. We
all have our own ideas, but we don’t have our own truths. The
prime directive requires respect for other cultures but it does
not render all their beliefs correct. Group beliefs can be wrong,
and ignorance can be deadly. We should all seek “the truth of
the world.”
I’m pretty sure the eight billion people of Earth are not living
inside an asteroid, but we may well be on course toward
disaster. We can’t count on Captain Kirk and Mr. Spock to
come back in time and beam down to set us right. Our only
hope is to enhance our own rational functioning, individually
and collectively.

Conclusión
We live in a world of reasons and persons, but only if we
choose to do so. We are also the Borg, except to the extent
that we choose to be something more. We are vehicles for
genes and memes, except to the extent that we choose to
valué ourselves as individual persons. We become and remain
rational agents to the extent that we exercise and reflect on our
rational agency, individually and collectively.
We will not achieve a utopia of justification, truth, justice, and
democracy, but we have the psychological competence to
make real progress toward such ideáis and to maintain them at
a high level. But this requires that we work together, which
requires mutual respect. Without respect for persons, respect
for reasons cannot be fully realized.
But we cannot just get together in groups and work things
out. We must devise and respect institutions that make
collaborative reasoning possible and that direct our
argumentation toward truth and rational decisión making. The
ideal social institution, the sort of group or society within which
rational and epistemic ideáis are best realized, is deliberative
democracy. The corresponding educational ideal is the
promotion of rational agency, which requires a focus on the
developmental construction of knowledge and reasoning under
conditions of intellectual freedom.
The Borg have already arrived, but resistance is not futile.
GLOSSARY

Academic The intellectual freedom to do academic work, including

freedom teaching, learning, and inquiry.
Agent An actor or subject; one who acts or knows.
Agent, rational One who believes or acts on the basis of reasons, preferably
good reasons.
Argument A justification, especially a complex one.
Argumentation A social process of rational debate or collaborative
reasoning.
Belief Roughly, acceptance of a proposition as true.
Cognition Knowledge and associated inferential processes, including
perception, representation, conceptualization, interpretation,
memory, thinking, and reasoning.
Constructivism A theoretical perspective emphasizing the active role of the
organism in the construction of knowledge.
Coordination A process of developmental change involving structural
integration and reorganization.
Deductive Logical reasoning in the strict sense of reaching only
reasoning logically necessary conclusions.
Dehumanization A process in which those categorized as other come to be
seen as less than persons.
Democracy, A form of liberal democracy that takes as its ideal the
deliberative achievement of consensus through rational processes of
argumentation.
Democracy, A social system based on equal respect for all persons.
liberal
Denial Rejection of the truth, especially to maintain one’s moral
identity.
Development A type or pattern of change that is extended, self-regulated,
qualitative, and Progressive.

Dichotomization A process by which the multidimensional complexity of social

identity is reduced to two categories.
Domain, A broad field or topic of knowledge that may or may not
cognitive correspond to an academic discipline.
Domain, A cognitive domain with a distinct epistemology, including a
epistemic distinct conception of justification and/or truth and a
correspondingly distinct form of prototypical reasoning.
Dual process A theory of thinking or reasoning that distinguishes (a)
theory automatic and/or intuitive processes from (b) deliberate
and/or reflective processes.
Empiricism A conception of development that highlights learning from
the environment.
Epistemic Knowledge about matters of epistemology—that is,
cognition knowledge about the justification and truth of beliefs.
Epistemic The development of epistemic cognition.
development
Epistemic A cognitive domain with a distinct epistemology, including a
domain distinct conception of justification and/or truth and a
correspondingly distinct form of prototypical reasoning.
Epistemology The study of knowledge, especially with regard to normative
matters of justification and truth.
Hypothetico- Deductive reasoning from premises deemed to be
deductive hypothetical or false.
reasoning
Identity An explicit theory of oneself as a person.
Identity, moral An explicit theory of oneself as a moral person.
Identity, rational An explicit and justifiable theory of oneself as a person.
Identity, An explicit theory of oneself that highlights and valúes one’s
rationalist rationality.
Inference Going beyond the data.
Intellectual Freedoms of belief, expression, discussion, inquiry, and
freedom access to information and ¡deas.
Justification A reason for belief or action, especially as offered in a
context of argumentation.
Knowledge Roughly, justified true belief.
Logic Strict norms of good reasoning, especially rules of deductive
necessity.
Logical Truth based on logical proof rather than empirical evidence.
necessity
Logical Reasoning in accord with norms of logic.
reasoning
Metacognition Cognition about cognition, including (a) conceptual
knowledge about knowledge and inference and (b)
procedural knowledge implicit in cognitive self-regulation.
Metasubjective Explicit or theoretical knowledge about logic.
objectivity
Metalogical Progress in objectivity through (subjective) reflection on and
understanding reconstruction of one’s subjectivity.
Metasubjectivity Knowledge about or reflection on subjectivity.
Moral A theory concerning the justification and truth of moral
epistemology beliefs and the corresponding rationality of moral action.
Moral identity An explicit theory of oneself as a moral person.
Morality Respect or concern for the rights or welfare of others.
Nativism A conception of development that highlights the direct causal
role of genes.
Norm A rule, principie, or ideal against which action, inference,
belief, or reality can be evaluated.
Normative Referring to norms, and thus to what ought to be.
Objectivism An epistemology that holds there to be ultímate truths that
are directly observable, provable, or known to infallible
authorities.
Objectivity Truth in relation to reality regardless of subjective
perspective.
Peer interaction Reciprocal social interaction of rational agents who are, and
perceive themselves to be, equal in authority, status, and
power.
Person At mínimum, a rational agent.
Perspective A (subjective) orientation or point of view.

Perspective Seeing from perspectives other than one’s own.

taking
Precedent A judgment that is (to some degree) binding on future
judgments in analogous cases.
Precedent- Reasoning on the basis of precedents.
based
reasoning
Principie An abstract norm whose application to cases requires
rational judgment rather than strict imposition of a concrete
rule.
Principled Reasoning on the basis of explicit principies, especially
reasoning moral principies.
Progressive Change that can be normatively judged to make progress
change toward maturity, rationality, or some other justifiable end.
Qualitative Change in kind, not just amount.
change
Rational Minimally, acting on the basis of reasons. More stringently,
having good reasons for one’s beliefs and actions.
Rational agency The competence to believe or act on the basis of reasons.
Rational agent One who believes or acts on the basis of reasons, preferably
good reasons.
Rational A form of constructivism in which organisms are seen as
constructivism rational agents making developmental progress.
Rational identity An explicit and justifiable theory of oneself as a person.
Rationalism A postsubjectivist epistemology in which reasons short of
proof provide bases for beliefs and actions.
Rationalist An explicit theory of oneself that highlights and valúes one’s
identity rationality.
Rationality The quality of having reasons, especially good reasons.
Reason A rational basis, and potential justification, for belief or
action.
Reasoning Epistemologically self-regulated thinking—that is, forms of
 thinking, such as argumentation, that are aimed at reaching
justifiable and/or true conclusions.

Reasoning, Logical reasoning in the strict sense of reaching only

deductive logically necessary conclusions.
Reasoning, Deductive reasoning from premises deemed to be
hypothetico- hypothetical or false.
deductive
Reasoning, Reasoning in accord with norms of logic.
logical
Reasoning, Reasoning on the basis of precedents.
precedent-
based
Reasoning, Reasoning on the basis of explicit principies, especially
principled moral principies.
Reasoning, Reasoning aimed at empirical justification and truth.
scientific
Reflection A metacognitive process of consciousness and
developmental change.
Science Empirical study and theoretical explanation of the world.
Scientific Reasoning aimed at empirical justification and truth.
reasoning
Self-regulated Change that is directed or regulated from within.
change
Social A norm of behavior binding only within a particular social
convention system.
Subjectivism An epistemology in which knowledge is seen as constructed
from, and thus determined by, one’s point of view.
Subjectivity Perspective or point of view, involving the contribution of
mind to knowledge.
Theory A structure of knowledge that organizes and explains some
domain or set of facts.
Thinking The deliberate application and coordination of one’s
inferences to serve one’s purposes, including problem
solving, decisión making, judgment, planning, and various
 forms of reasoning.
Third-party Seeing one’s reciprocal relation to another person from the
perspective (meta)perspective of a real or hypothetical third party.
taking

Truth Objectivity, as a matter of concrete fact or epistemic ideal.

Understanding Theoretical knowledge.
Verification bias A natural tendency to seek and attend to evidence
supporting one’s beliefs and hypotheses.
REFERENCES

Adams, D. W. (1995). Education for extinction: American Indians and

the boarding school experience, 1875-1928. Lawrence: University
of Kansas Press.
Aikin, S. F., & Talisse, R. B. (2019). Why we argüe (and how we
should): A guide to political disagreement in an age of unreason
(2nd ed.). New York: Routledge.
ALA Office for Intellectual Freedom. (2015). Intellectual freedom
manual (9th ed.). Chicago: American Library Association.
Alien, J. W. P., & Bickhard, M. H. (2018). Stage fright: Internal
reflection as a domain general enabling constraint on the
emergence of explicit thought. Cognitive Development, 45, 77-91.
Amar, A. R. (2012). America’s unwritten constitution: The precedents
and principies we live by. New York: Basic Books.
Amsel, E., Klaczynski, P. A., Johnston, A., Bench, S., Cióse, J.,
Sadler, E., & Walker, R. (2008). A dual-process account of the
development of scientific reasoning: The nature and development
of metacognitive intercession skills. Cognitive Development, 23,
452—471.
Appiah, K. A. (2005). The ethics of identity. Princeton, NJ: Princeton
University Press.
Appiah, K. A. (2017). As if: Idealization and ideáis. Cambridge, MA:
Harvard University Press.
Appiah, K. A. (2018). The lies thatbind: Rethinking identity. New York:
Norton.
Ash, T. G. (2016). Free speech: Ten principies for a connected world.
New Haven, CT: Yale University Press.
Augustinova, M. (2008). Falsificaron cueing in collective reasoning:
Example of the Wason selection task. European Journal of Social
Psychology, 38, 770-785.
Báchtiger, A., Dryzek, J. S., Mansbridge, J., & Warren, M. E. (2018a).
Deliberative democracy: An introduction. In A. Báchtiger, J. S.
Dryzek, J. Mansbridge, & M. E. Warren (Eds.). The Oxford
handbook of deliberative democracy (pp. 1-31). New York: Oxford
 University Press.
Báchtiger, A., Dryzek, J. S., Mansbridge, J., & Warren, M. E. (Eds.).
(2018b). The Oxford handbook of deliberative democracy. New
York: Oxford University Press.
Báchtiger, A., & Wegmann, A. (2014). “Scaling up” deliberation. In S.
Elstub & P. McLaverty (Eds.), Deliberative democracy: Issues and
cases (pp. 118-135). Edinburgh: Edinburgh University Press.
Baillon, A., Bleichrodt, H., Liu, N., & Wakker, P. P. (2016). Group
decisión rules and group rationality under risk. Journal of Risk and
Uncertainty, 52, 99-116.
Baldwin, J. M. (1895). Mental development in the child and the race.
New York: Macmillan.
Bardon, A. (2020). The truth about denial: Bias and self-deception in
science, politics, and religión. New York: Oxford University Press.
Barrouillet, P., & Gaillard, V. (Eds.). (2011). Cognitive development
and working memory: A dialogue between neo-Piagetian theories
and cognitive approaches. New York: Psychology Press.
Barrouillet, P., & Gauffroy, C. (2013). Dual processes and mental
models in the development of conditional reasoning. In P.
Barrouillet & C. Gauffroy (Eds.), The development of thinking and
reasoning (pp. 95-121). New York: Psychology Press.
Bearison, D. J., & Dorval, B. (2002). Collaborative cognition: Children
negotiating ways ofknowing. Westport, CT: Ablex.
Bloom, P. (2013). Just Babies: The origins ofgood and evil. New York:
Crown.
Bloom, P. (2016). Against empathy: The case for rational compassion.
New York: HarperCollins.
Boyes, M. C., & Chandler, M. (1992). Cognitive development,
epistemic doubt, and identity formation in adolescence. Journal of
Youth and Adolescence, 21, 737-763.
Braine, M. D. S., & O’Brien, D. P. (Eds.). (1998). Mental logic.
Mahwah, NJ: Erlbaum.
Brown, A. L. (1978). Knowing when, where, and how to remember: A
problem in metacognition. In R. Glaser (Ed.), Advances in
instructional psychology. New York: Halsted Press.
Brown, D. (1970). Bury my heart at Wounded Knee: An Indian history
ofthe American West. New York: Bantam.
Brown, M. B. (2014). Expertise and deliberative democracy. In S.
Elstub & P. McLaverty (Eds.), Deliberative democracy: Issues and
cases (pp. 50-68). Edinburgh: Edinburgh University Press.
Bryant, P. E., & Trabasso, T. R. (1971). Transitive inferences and
 memory in young children. Nature, 232, 456-458.
Callan, E., & Arena, D. (2009). Indoctrination. In H. Siegel (Ed.),
Oxford handbook of philosophy of education (pp. 104-121).
Oxford, UK: Oxford University Press.
Campbell, R. L., & Bickhard, M. H. (1986). Knowing levels and
developmental stages. Basel, Switzerland: Karger.
Carey, S., & Smith, C. (1993). On understanding the nature of
scientific knowledge. Educational Psychologist, 28, 235-251.
Carpendale, J. I. M., & Chandler, M. J. (1996). On the distinction
between false belief understanding and subscribing to an
interpretive theory of mind. Child Development, 67, 1686-1706.
Cauley, K. M. (1992). Children’s construction of the logic of arithmetic.
Journal of Applied Developmental Psychology, 13, 65-74.
Cesana-Arlotti, N., Martín, A., Téglás, E., Vorobyova, L., Cetnarskí, R.,
& Bonatti, L. L. (2018). Precursors of logical reasoning in preverbal
human infants. Science, 359, 1263-1266.
Chandler, M. J. (1975). Relativism and the problem of epístemologícal
lonelíness. Human Development, 1 8 ,171-180.
Chandler, M. J. (1987). The Othello effect: Essay on the emergence
and eclípse of skeptícal doubt. Human Development, 30, 137-159.
Chandler, M. J., Boyes, M., & Ball, L. (1990). Relativism and statíons
of epistemic doubt. Journal of Experimental Child Psychology, 50,
370-395.
Chandler, M. J., Hallett, D., & Sokol, B. W. (2002). Competing claims
about competing knowledge claims. In B. K. Hofer & P. R. Pintrich
(Eds.), Personal epistemology: The psychology of beliefs about
knowledge and knowing (pp. 145-168). Mahwah, NJ: Erlbaum.
Chapman, L. J., & Chapman, J. P. (1959). Atmosphere effect re-
examíned. Journal of Experimental Psychology, 58, 220-226.
Churchill, W. (2004). Kill the Indian, save the man: The genocidal
impact of American Indian residential schools. San Francisco: City
Lights Books.
Cinalli, M., & O’FIynn, I. (2014). Pluralísm and deliberative democracy.
In S. Elstub & P. McLaverty (Eds.), Deliberative democracy: Issues
and cases (pp. 82-97). Edínburgh: Edínburgh University Press.
Cur§eu, P. L., Jansen, R. J. G., & Chappín, M. M. H. (2013). Decisión
rules and group rationality: Cognitive gain or standstíll? PLOS
One, 8(2), e56454. doí: 10.1371/journal.pone.0056454
Cur§eu, P. L., Mesíec, N., Pluut, H., & Lucas, G. J. M. (2015).
Cognitive synergy in groups and group-to-individual transfer of
decision-making competencíes. Frontiers in Psychology, 6, 1375.
 doi: 10.3389/fpsyg.2015.01375
Dacey, A. (2020). Come now, let us reason together: Cognitive bias,
individualism, and interactionism in critical thinking education.
Informal Logic, 40, 47-76.
De Brasi, L. (2018). Citizenry incompetence and the epistemic
structure of society. Unisinos Journal of Philosophy, 19, 201-212.
De Neys, W. (2012). Bias and conflict: A case for logical intuitions.
Perspectives on Psychological Science, 7, 28-38.
De Neys, W. (2018). Bias, conflict, and fast logic: Towards a hybrid
dual process future? In W. De Neys (Ed.), Dual process theory 2.0
(pp. 47-65). New York: Routledge.
Demetriou, A., Makris, N., Kazi, S., Spanoudis, G., & Shayer, M.
(2018). The developmental trinity of mind: Cognizance, executive
control, and reasoning. WIREs Cognitive Science, doi:
10.1002/wcs.1461
Demetriou, A., & Spanoudis, G. (2018). Growing minds: A
developmental theory of intelligence, brain, and education. New
York: Routledge.
Dewey, J. (1997). How we think. Mineóla, NY: Dover. (Original work
published 1910)
Doise, W., & Mugny, G. (1984). The social development of the
intellect. Oxford: Pergamon.
Domberg, A., Kóymen, B., & Tomasello, M. (2018). Children’s
reasoning with peers in cooperative and competitive contexts.
British Journal of Developmental Psychology, 36, 64-77.
Driver, J. (2018). The schoolhouse gate: Public education, the
Supreme Court, and the battle for the American mind. New York:
Pantheon.
Dworkin, R. (1985). A matter of principie. Cambridge, MA: Harvard
University Press.
Dworkin, R. (1996). Freedom’s law: The moral reading ofthe American
Constitution. Cambridge, MA: Harvard University Press.
Dworkin, R. (2011). Justice for hedgehogs. Cambridge, MA: Harvard
University Press.
Elstub, S., & McLaverty, P. (Eds.). (2014). Deliberative democracy:
Issues and cases. Edinburgh: Edinburgh University Press.
Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton.
Estlund, D., & Landemore, H. (2018). The epistemic valué of
democratic deliberation. In A. Báchtiger, J. S. Dryzek, J.
Mansbridge, & M. E. Warren (Eds.), The Oxford handbook of
deliberative democracy (pp. 113-131). New York: Oxford
 University Press.
Evans, J. St. B. T. (2007). Hypothetical thinking: Dual processes in
reasoning and judgement. New York: Psychology Press.
Evans, J. St. B. T. (2017). Thinking and reasoning: A very short
introduction. Oxford: Oxford University Press.
Evans, J. St. B. T., & Stanovich, K. E. (2013). Dual-process theories of
higher cognition: Advancing the debate. Perspectives on
Psychological Science, 8, 223-241.
Fallón, R. H., Jr. (1987). A constructivist coherence theory of
constitutional interpretation. Harvard Law Review, 100, 1189—
1286.
Fallón, R. H., Jr. (2018). Law and legitimacy in the Supreme Court.
Cambridge, MA: Harvard University Press.
Feiman, R., Hartshorne, J. K., & Barner, D. (2019). Contrast and
entailment: Abstract logical relations constrain how 2- and 3-year-
old children interpret unknown numbers. Cognition, 183, 192-207.
Finken, M. W., & Post, R. C. (2009). For the common good: Principies
of American academic freedom. New Haven, CT: Yale University
Press.
Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new
area of cognitive-developmental inquiry. American Psychologist,
34, 906-911.
Gelman, R., & Gallistel, C. R. (1978). The child’s understanding of
number. Cambridge, MA: Harvard University Press.
Gibbs, J. C. (2019). Moral development and reality: Beyond the
theories of Kohlberg, Hoffman, & Haidt (4th ed.). Oxford: Oxford
University Press.
Gopnik, A. (2009). The philosophical baby: What children’s minds tell
us about truth, love, and the meaning of Ufe. New York: Farrar,
Straus, and Giroux.
Gopnik, A. (2012). Scientific thinking in young children: Theoretical
advances, empirical research, and policy implications. Science,
337 (6102), 1623-1627.
Goswami, U. (1992). Analógica! reasoning in children. New York:
Psychology Press.
Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing
inferences during narrative text comprehension. Psychological
Review, 101, 371-395.
Greene, J. A., Sandoval, W. A., & Bráten, I. (Eds.) (2016). Handbook
of epistemic cognition. New York: Routledge.
Gutmann, A., & Thompson, D. (2004). Why deliberative democracy?
 Princeton, NJ: Princeton University Press.
Gutmann, A., & Thompson, D. (2018). Reflections on deliberative
democracy: When theory meets practice. In A. Báchtiger, J. S.
Dryzek, J. Mansbridge, & M. E. Warren (Eds.), The Oxford
handbook of deliberative democracy (pp. 900-912). New York:
Oxford University Press.
Habermas, J. (1990). Moral consciousness and communicative action.
Cambridge, MA: MIT Press.
Habermas, J. (1996). Between facts and norms: Contributions to a
discourse theory of law and democracy. Cambridge, MA: MIT
Press.
Haidt, J. (2001). The emotional dog and its rational tail: A social
intuitionist approach to moral judgment. Psychological Review,
108, 814-834.
Hans, V. P. (2007). Deliberation and dissent: 12 Angry Men versus the
empirical reality of juries. Chicago-Kent Law Review, 82, 579-589.
Hart, D., & Youniss, J. (2018). Renewing democracy in young
America. Oxford: Oxford University Press.
Helwig, C. C. (1995). Adolescents’ and young adults’ conceptions of
civil liberties: Freedom of speech and religión. Child Development,
66, 152-166.
Helwig, C. C., Ruck, M. D., & Peterson-Badali, M. (2014). Rights, civil
liberties, and democracy. In M. Killen & J. G. Smetana (Eds.),
Handbook of moral development (2nd ed.) (pp. 46-69). New York:
Psychology Press.
Henle, M. (1962). On the relation between logic and thinking.
Psychological Review, 69, 366-378.
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from
childhood to adolescence. New York: Basic Books. (Original work
published 1955)
Inhelder, B., & Piaget, J. (1964). The early growth of logic in the child:
Classification and seriation. London: Routledge. (Original work
published 1959)
lordanou, K. (2016). From theory of mind to epistemic cognition. A
lifespan perspective. Frontline Learning Research, 4 , 106-119.
Johnson-Laird, P. N. (1983). Mental models. Cambridge, MA: Harvard
University Press.
Johnson-Laird, P. N. (2006). How we reason. Oxford: Oxford
University Press.
Johnson-Laird, P. N., & Byrne, R. M. J. (1991). Deduction. Hillsdale,
NJ: Erlbaum.
Kahneman, D. (2011). Thinking, fast and slow. New York: Farrar,
Straus, & Giroux.
Karmiloff-Smith, A. (1992). Beyond modularity: A developmental
perspective on cognitive science. Cambridge, MA: MIT Press.
Karpowitz, C. E., & Mendelberg, T. (2018). The political psychology of
deliberation. In A. Báchtiger, J. S. Dryzek, J. Mansbridge, & M. E.
Warren, The Oxford handbook of deliberative democracy (pp.
535-555). New York: Oxford University Press.
Kazi, S., Kazali, E., Makris, N., Spanoudis, G., & Demetriou, A. (2019).
Cognizance in cognitive development: A longitudinal study.
Cognitive Development, 52.
Killen, M., & Rutland, A. (2011). Children and social exclusión:
Morality, prejudice, and group identity. Malden, MA: Wiley-
Blackwell.
King, P. M., & Kitchener, K. S. (1994). Developing reflective judgment:
Understanding and promoting intellectual growth and critical
thinking in adolescents and adults. San Francisco: Jossey-Bass.
Klaczynski, P. A. (2000). Motivated scientific reasoning biases,
epistemological beliefs, and theory polarization: A two-process
approach to adolescent cognition. Child Development, 71, 1347-
1366.
Klaczynski, P. A. (2011). Age differences in understanding precedent-
setting decisions and authorities’ responses to violations of deontic
rules. Journal of Experimental Child Psychology, 1 0 9 ,1-24.
Klaczynski, P. A., & Felmban, W. S. (2019). When (and when not) to
make exceptions: Links among age, precedent setting decisions,
and argument evaluation. Journal of Genetic Psychology, 180,
170-184.
Klein, E. (2020). Why we’re polarized. New York: Avid Reader Press.
Kohlberg, L. (1981). The philosophy of moral development. San
Francisco: Harper & Row.
Kohlberg, L. (1984). The psychology of moral development. San
Francisco: Harper & Row.
Korsgaard, C. M. (2009). Self-constitution: Agency, identity, and
integrity. Oxford: Oxford University Press.
Koslowski, B. (1996). Theory and evidence: The development of
scientific reasoning. Cambridge, MA: MIT Press.
Koslowski, B. (2013). Scientific reasoning: Explanation, confirmation
bias, and scientific practice. In G. J. Feist & M. E. Gorman (Eds.),
Handbook ofthe psychology of science (pp. 151-192). New York:
Springer.
Koslowski, B., Marasia, J., Chelenza, M., & Dublin, R. (2008).
Information becomes evidence when an explanation can
incorpórate it into a causal framework. Cognitive Development, 23,
472—487.
Kuhn, D. (1989). Children and adults as intuitive scientists.
Psychological Review, 96, 674-689.
Kuhn, D. (1991). The skills of argument. Cambridge, UK: Cambridge
University Press.
Kuhn, D. (2005). Education for thinking. Cambridge, MA: Harvard
University Press.
Kuhn, D. (2018a). Building our best future: Thinking critically about
ourselves and our world (student ed.). New York: Wessex.
Kuhn, D. (2018b). Building our best future: Thinking critically about
ourselves and our world (teachers ed.). New York: Wessex.
Kuhn, D. (2019). Critical thinking as discourse. Human Development,
62, 146-164.
Kuhn, D., Amsel, E., & O’Loughlin, M. (1988). The development of
scientific thinking skills. San Diego, CA: Academic Press.
Kuhn, D., Cheney, R., & Weinstock, M. (2000). The development of
epistemological understanding. Cognitive Development, 15, 30 9 -
328.
Kuhn, D., Feliciano, N., & Kostikina, D. (2019). Engaging
contemporary issues as practice for citizenship. The Social
Studies, 110, 207-219.
Kuhn, D., Hemberger, L., & Khait, V. (2016). Argüe with me: Argument
as a path to developing students’ thinking and writing (2nd ed.).
New York: Routledge.
Kuhn, D. & Weinstock, M. (2002). What is epistemological thinking and
why does it matter? In B. K. Hofer & P. R. Pintrich (Eds.), Personal
epistemology: The psychology of beliefs about knowledge and
knowing (pp. 121-144). Mahwah, NJ: Erlbaum.
Laats, A., & Siegel, H. (2016). Teaching evolution in a creation nation.
Chicago: University of Chicago Press.
Laden, A. S. (2012). Reasoning: A social picture. Oxford: Oxford
University Press.
Lalonde, C. E., & Chandler, M. J. (2002). Children’s understanding of
interpretation. New Ideas in Psychology, 20, 163-198.
Landemore, H. (2013). Democratic reason: Politics, collective
intelligence, and the rule of the many. Princeton, NJ: Princeton
University Press.
Landy, J. F., & Royzman, E. B. (2018). The moral myopia model: Why
 and how reasoning matters in moral judgment. In G. Pennycook
(Ed.), The new reflectionism in cognitive psychology: Why reason
matters (pp. 70-92). New York: Routledge.
Langer, J. (1980). The origins of logic: From six to twelve months. San
Francisco: Academic Press.
Langer, J. (1986). The origins of logic: One to two years. Orlando, FL:
Academic Press.
Laughlin, P. R. (2011). Group problem solving. Princeton, NJ:
Princeton University Press.
Lent, R. C., & Pipkin, G. (2013). Keep them reading: An anti-
censorship handbook for educators. New York: Teachers College
Press.
Lerner, R. M., & Overton, W. F. (2017). Reduction to absurdity: Why
epigenetics invalidates all models involving genetic reduction.
Human Development, 6 0 ,107-123.
Lickliter, R., & Witherington, D. C. (2017). Towards a truly
developmental epigenesis. Human Development, 60, 124-138.
Lukianoff, G. (2014). Unlearning liberty: Campus censorship and the
end of American debate. New York: Encounter.
Maalouf, A. (2001). In the ñame of identity: Violence and the need to
belong. New York: Arcade.
Mansfield, A. F., & Clinchy, B. M. (2002). Toward the integration of
objectivity and subjectivity: Epistemological development from 10
to 16. New Ideas in Psychology, 20, 225-262.
Markovits, H. (2013). The development of abstract conditional
reasoning. In P. Barrouillet & C. Gauffroy (Eds.), The development
of thinking and reasoning (pp. 71-91). New York: Psychology
Press.
Markovits, H. (Ed.). (2014). The developmental psychology of thinking
and decision-making. New York: Psychology Press.
Markovits, H. (2020). The development of logical reasoning. In G. W.
Noblit (Ed.), Oxford research encyclopedia of education. Oxford:
Oxford University Press. doi:
10.1093/acrefore/9780190264093.013.873
Markovits, H., Brunet, M.-L., Thompson, V., & Brisson, J. (2013).
Direct evidence for a dual process model of deductive inference.
Journal of Experimental Psychology: Learning, Memory and
Cognition, 39, 1213-1222.
Markovits, H., Thompson, V., & Brisson, J. (2015). Metacognition and
abstract reasoning. Memory & Cognition, 43, 681-693.
Maynard, J. L. (forthcoming). Ideologies, identities and speech in
 atrocity crimes. In B. Hola, H. N. Brehm, & M. Weerdesteijn (Eds.),
Oxford handbook of atrocity crime. Oxford: Oxford University
Press.
McCulloch, W. S. (1961). What is a number, that a man may know it,
and a man, that he may know a number? General Semantics
Bulletin, 26-27, 7-18. [Reprinted in W. S. McCulloch (Ed.). (1965).
Embodiments of mind (pp. 1-18). Cambridge, MA: MIT Press.]
McCulloch, W. S., & Pitts, W. H. (1943). A logical calculus of the ideas
immanent in nervous activity. Bulletin of Mathematical Biophysics,
5, 115-133. [Reprinted in W. S. McCulloch (Ed.). (1965).
Embodiments of mind (pp. 19-39). Cambridge, MA: MIT Press.]
Mercier, H., Boudry, M. Paglieri, F., &Trouche, E. (2017). Natural-born
arguers: Teaching how to make the best of our reasoning abilities.
Educational Psychologist, 5 2 ,1-16.
Mercier, H., & Sperber, D. (2017). The enigma of reason. Cambridge,
MA: Harvard University Press.
Mercier, H., & Sperber, D. (2019). Replies to critics. Teorema, 38,
139-156.
Miller, S. A. (2012). Theory of mind: Beyond the preschool years. New
York: Psychology Press.
Moshman, D. (1979). Development of formal hypothesis-testing ability.
Developmental Psychology, 15, 104-112.
Moshman, D. (1982). Exogenous, endogenous, and dialectical
constructivism. Developmental Review, 2, 371-384.
Moshman, D. (1990). The development of metalogical understanding.
In W. F. Overton (Ed.), Reasoning, necessity, and logic:
Developmental perspectives (pp. 205-225). Hillsdale, NJ:
Erlbaum.
Moshman, D. (1995). The construction of moral rationality. Human
Development, 38, 265-281.
Moshman, D. (2004a). False moral identity: Self-serving denial in the
maintenance of moral self-conceptions. In D. K. Lapsley & D.
Narvaez (Eds.), Moral development, self, and identity (pp. 8 3 -
109). Mahwah, NJ: Erlbaum.
Moshman, D. (2004b). From inference to reasoning: The construction
of rationality. Thinking & Reasoning, 10, 221-239.
Moshman, D. (2007a). Social identity and its discontents. Journal of
Applied Developmental Psychology, 28, 184-187.
Moshman, D. (2007b). Us and them: Identity and genocide. Identity, 7,
115-135.
Moshman, D. (2009a). Liberty and learning: Academic freedom for
 teachers and students. Portsmouth, NH: Heinemann.
Moshman, D. (2009b). The development of rationality. In H. Siegel
(Ed.), Oxford handbook of philosophy of education (pp. 145-161).
Oxford, UK: Oxford University Press.
Moshman, D. (2011a). Adolescent rationality and development:
Cognition, morality, and identity (3rd ed.). New York: Psychology
Press.
Moshman, D. (2011b). Identity, genocide, and group violence. In S. J.
Schwartz, K. Luyckx, & V. L. Vignoles (Eds.), Handbook of identity
theory and research (pp. 917-932). New York: Springer.
Moshman, D. (2013). Adolescent rationality. Advances in Child
Development and Behavior, 45, 155-183.
Moshman, D. (2015). Epistemic cognition and development: The
psychology of justification and truth. New York: Psychology Press.
Moshman, D. (2017). Academic freedom as the freedom to do
academic work. AAUP Journal of Academic Freedom, 8.
www.aaup.org/JAF8/academic-freedom-freedom-do-academic-
work#.Wcqm7l6QzuR
Moshman, D. (2018). Reasoning, logic, and development: Essay
review of The enigma of reason by Hugo Mercier and Dan
Sperber. Human Development, 61, 60-64.
Moshman, D. (2020). “Hate speech,” free speech, and group violence.
In R. J. Sternberg (Ed.), Perspectives on hate: How it originates,
develops, manifests, and spreads (pp. 203-224). Washington, DC:
APA Books.
Moshman, D., & Franks, B. A. (1986). Development of the concept of
inferential validity. Child Development, 57, 153-165.
Moshman, D., & Geil, M. (1998). Collaborative reasoning: Evidence for
collective rationality. Thinking & Reasoning, 4, 231-248.
Nagel, T. (1986). The view from nowhere. New York: Oxford University
Press.
Neimark, E. D. (1970). Model for a thinking machine: An information-
processing framework for the study of cognitive development.
Merrill-Palmer Quarterly of Behavior and Development, 16, 3 45 -
368.
Neimark, E. D. (1975). Intellectual development during adolescence. In
F. D. Horowitz (Ed.), Review of child development research (Vol.
4, pp. 541-594). Chicago: University of Chicago Press.
Neimark, E. D. (1987). Adventures in thinking. San Diego: Harcourt
Brace Jovanovich.
Neimark, E. D., & Santa, J. L. (1975). Thinking and concept
 attainment. Annual Review of Psychology, 2 6 ,173-205.
Neimark, E. D., & Slotnick, N. S. (1970). Development of the
understanding of logical connectives. Journal of Educational
Psychology, 61, 451-460.
Niño, C. S., (1996). The constitution of deliberative democracy. New
Haven, CT: Yale University Press.
Noveck, I. (2019). From Queensbury rules to argumentative theory: A
review of Mercier & Sperber’s The enigma of reason. Teorema,
38, 121-138.
Nucci, L. P. (2001). Education in the moral domain. Cambridge, UK:
Cambridge University Press.
Nucci, L. P. (2014). The personal and the moral. In M. Killen & J. G.
Smetana (Eds.), Handbook of moral development (2nd ed.) (pp.
538-558). New York: Psychology Press.
O’Madagain, C., & Tomasello, M. (2019). Joint attention to mental
content and the social origin of reasoning. Synthese. doi:
10.1007/sl 1229-019-02327-1
Overton, W. F. (Ed.). (1990). Reasoning, necessity, and logic:
Developmentalperspectives. Hillsdale, NJ: Erlbaum.
Parfit, D. (1984). Reasons and persons. Oxford: Oxford University
Press.
Parkinson, J. (2018). Deliberative systems. In A. Báchtiger, J. S.
Dryzek, J. Mansbridge, & M. E. Warren (Eds.). The Oxford
handbook of deliberative democracy (pp. 432-446). New York:
Oxford University Press.
Paul, R. W. (1990). Critical thinking: What every person needs to
survive in a rapidly changing world. Rohnert Park: CA: Center for
Critical Thinking and Moral Critique, Sonoma State University.
Pennycook, G. (2018). A perspective on the theoretical foundation of
dual process models. In W. De Neys (Ed.), Dual process theory
2.0 (pp. 5-27). New York: Routledge.
Perry, W. G. (1999). Forms of intellectual and ethical development in
the college years: A scheme. San Francisco: Jossey-Bass.
(Original work published 1970)
Piaget, J. (1954). The construction of reality in the child. New York:
Basic Books. (Original work published 1937)
Piaget, J. (1963). The origins of intelligence in children. New York:
Norton. (Original work published 1936)
Piaget, J. (1965a). The child’s conception of number. New York:
Norton. (Original work published 1941)
Piaget, J. (1965b). The moral judgment of the child. New York: Free
 Press. (Original work published 1932)
Piaget, J. (1971). Biology and knowledge: An essay on the relations
between organic regulations and cognitive processes. Chicago:
University of Chicago Press. (Original work published 1967)
Piaget, J. (1972). Judgment and reasoning in the child. Totowa, NJ:
Littlefield, Adams. (Original work published 1928)
Piaget, J. (1974). Understanding causality. New York: Norton. (Original
work published 1971)
Piaget, J. (1985). The equilibraron of cognitive structures: The central
problem of intellectual development. Chicago: University of
Chicago Press. (Original work published 1975)
Piaget, J. (1995). Sociological studies. New York: Routledge. (Original
works published 1928-1977)
Piaget, J. (2001). Studies in reflecting abstraction. Hove, UK:
Psychology Press. (Original work published 1977)
Piaget, J., & Inhelder, B. (1975). The origin of the idea of chance in
children. New York: Norton. (Original work published 1951)
Pillow, B. H. (2012). Children’s discovery of the active mind:
Phenomenological awareness, social experience, and knowledge
about cognition. New York: Springer.
Post, R. C. (2012). Democracy, expertise, and academic freedom: A
First Amendment jurisprudence for the modern state. New Haven,
CT: Yale University Press.
Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a
theory of mind? Behavioral and Brain Sciences, 1, 515-526.
Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard
University Press.
Rawls, J. (2001). Justice as fairness: A restatement. Cambridge, MA:
Harvard University Press.
Ricco, R. B. (2015). Development of reasoning. In R. M. Lerner (Ed.),
Handbook of child psychology and developmental Science (7th
ed., Vol. 2) (pp. 519-570). Hoboken, NJ: Wiley.
Ricco, R. B., & Overton, W. F. (2011). Dual systems competence <— >
procedural processing: A relational developmental systems
approach to reasoning. Developmental Review, 3 1 ,119-150.
Richard, M. (2019). Is reasoning a form of agency? In M. Balcerak
Jackson & B. Balcerak Jackson (Eds.), Reasoning: New essays on
theoretical and practical thinking (pp. 91-100). New York: Oxford
University Press.
Rips, L. J. (1994). The psychology of proof: Deductive reasoning in
human thinking. Cambridge, MA: MIT Press.
Rosenberg, A., & Mclntyre, L. (2020). Philosophy of science: A
contemporary introduction (4th e<±). New York: Routledge.
Rosenberg, S. W. (2014). Citizen competence and the psychology of
deliberation. In S. Elstub & P. McLaverty (Eds.), Deliberative
democracy: Issues and cases (pp. 98-117). Edinburgh: Edinburgh
University Press.
Rosenfeld, S. (2019). Democracy and truth: A short history.
Philadelphia: University of Pennsylvania Press.
Ross, C. J. (2015). Lessons in censorship: How schools and courts
subvert students’ First Amendment rights. Cambridge, MA:
Harvard University Press.
Saxonhouse, A. W. (2006). Free speech and democracy in ancient
Athens. New York: Cambridge University Press.
Scholnick, E. K., & Wing, C. S. (1991). Speaking deductively:
Preschoolers’ use of If in conversation and in conditional
inference. Developmental Psychology, 27, 249-258.
Scholnick, E. K., & Wing, C. S. (1992). Speaking deductively: Using
conversation to trace the origins of conditional thought in children.
Merrill-Palmer Quarterly, 3 8 ,1-20.
Scholnick, E. K., & Wing, C. S. (1995). Logic in conversation:
Comparative studies of deduction in children and adults. Cognitive
Development, 10, 319-345.
Schraw, G., & Moshman, D. (1995). Metacognitive theories.
Educational Psychology Review, 7, 351-371.
Sells, S. B. (1936). The atmosphere effect: An experimental study of
reasoning. Archives of Psychology (Columbia University), 29, 3 -
72.
Selman, R. L. (1980). The growth of interpersonal understanding:
Developmental and clinical analyses. New York: Academic Press.
Sen, A. (1999). Development as freedom. New York: Knopf.
Siegel, H. (1988). Educating reason: Rationality, critical thinking, and
education. New York: Routledge.
Siegel, H. (1997). Rationality redeemed? Further dialogues on an
educational ideal. New York: Routledge.
Siegel, H. (2018). Education’s epistemology: Rationality, diversity, and
critical thinking. Oxford: Oxford University Press.
Siegel, S. (2019). Inference without reckoning. In M. Balcerak Jackson
& B. Balcerak Jackson (Eds.), Reasoning: New essays on
theoretical and practica! thinking (pp. 15-31). New York: Oxford
University Press.
Smetana, J. G. (2011). Adolescents, families, and social development:
 How teens construct their worlds. Malden, MA: Wiley-Blackwell.
Smetana, J. G., Jambón, M., & Ball, C. (2014). The social domain
approach to children’s moral and social judgments. In M. Killen &
J. G. Smetana (Eds.), Handbook of moral development (2nd ed.,
Vol. 2) (pp. 23-45). New York: Psychology Press.
Smith, L. (1993). Necessary knowledge: Piagetian perspectives on
constructivism. Hillsdale, NJ: Erlbaum.
Smith, L. (2002). Reasoning by mathematical induction in children’s
arithmetic. Oxford: Pergamon.
Sobel, D. M., & Kirkham, N. Z. (2007). Bayes nets and babies: Infants’
developing statistical reasoning abilities and their representation of
causal knowledge. Developmental Science, 10, 298-306.
Sodian, B. (2011). Theory of mind in infancy. Child Development
Perspectives, 5, 39-43.
Sodian, B. (2018). The development of scientific thinking in preschool
and elementary school age: A conceptual model. In F. Fischer, C.
A. Chinn, K. Engelmann, & J. Osborne (Eds.), Scientific reasoning
and argumentation: The roles of domain-specific and domain-
general knowledge. New York: Routledge.
Sodian, B., & Kristen, S. (2016). Theory of mind. In J. A. Greene, W.
A. Sandoval, & I. Bráten (Eds.), Handbook of epistemic cognition
(pp. 68-85). New York: Routledge.
Sodian, B., & Wimmer, H. (1987). Children’s understanding of
inference as a source of knowledge. Child Development, 58, 4 2 4 -
433.
Staffel, J. (2019). Attitudes in active reasoning. In M. Balcerak Jackson
& B. Balcerak Jackson (Eds.), Reasoning: New essays on
theoretical and practica! thinking (pp. 44-67). New York: Oxford
University Press.
Stanovich, K. E. (2004). The robot’s rebellion: Finding meaning in the
age ofDarwin. Chicago: University of Chicago Press.
Stanovich, K. E. (2011). Rationality and the reflective mind. Oxford:
Oxford University Press.
Starita, J. (2008). “I am a man”: Chief Standing Bear’s journey for
justice. New York: St. Martin’s Griffin.
Strossen, N. (2018). Hate: Why we should resist it with free speech,
not censorship. Oxford: Oxford University Press.
Tarricone, P. (2011). The taxonomy of metacognition. New York:
Psychology Press.
Taylor, A. (2019). Democracy may not exist, but we’ll miss it when it’s
gone. New York: Metropolitan Books.
Thompson, V. A., & Newman, I. R. (2018). Logical intuitions and other
conundra for dual process theories. In W. De Neys (Ed.), Dual
process theory 2.0 (pp. 121-136). New York: Routledge.
Tibbetts, E. A., Agudelo, J., Pandit, S., & Riojas, J. (2019). Transitive
inference in Polistes paper wasps. Biology Letters, 15(5),
20190015. doi: 10.1098/rsbl.2019.0015
Tomasello, M. (2014). A natural history of human thinking. Cambridge,
MA: Harvard University Press.
Trippas, D., & Handley, S. J. (2018). The parallel processing model of
belief bias: Review and extensions. In W. De Neys (Ed.), Dual
process theory 2.0 (pp. 28-46). New York: Routledge.
Turiel, E. (1983). The development of social knowledge: Morality and
convention. Cambridge, UK: Cambridge University Press.
Turiel, E. (2008). The development of children’s orientations toward
moral, social, and personal orders: More than a sequence in
development. Human Development, 51, 21-39.
Wainryb, C., Shaw, L. A., Langley, M., Cottam, K., & Lewis, R. (2004).
Children’s thinking about diversity of belief in the early school
years: Judgments of relativism, tolerance, and disagreeing
persons. Child Development, 75, 687-703.
Wason, P. C., & Johnson-Laird, P. N. (1972). Psychology of
reasoning: Structure and content. Cambridge, MA: Harvard
University Press.
Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of
theory-of-mind development: The truth about false belief. Child
Development, 72, 655-684.
West Virginia State Board of Education v. Barnette, 319 U.S. 624
(1943).
Whittington, K. E. (2018). Speak freely: Why universities must defend
free speech. Princeton, NJ: Princeton University Press.
Wilkins, M. C. (1928). The effect of changed material on ability to do
formal syllogistic reasoning. Archives of Psychology, 16,1-83.
Wimmer, H., & Perner, J. (1983). Beliefs about beliefs: Representation
and constraining function of wrong beliefs in young children’s
understanding of deception. Cognition, 1 3 ,103-128.
Witherington, D. C. (2011). Taking emergence seriously: The centrality
of circular causality for dynamic systems approaches to
development. Human Development, 54, 66-92.
Woodworth, R. S., & Sells, S. B (1935). An atmosphere effect in formal
syllogistic reasoning. Journal of Experimental Psychology, 18,
451—460.
Zentall, T. R., Peng, D., & Miles, L. (2019). Transitive inference in
pigeons may result from differential tendencies to reject the test
stimuli acquired during training. Animal Cognition, 22, 619-624.
Zillmer, N., & Kuhn, D. (2018). Do similar-ability peers regúlate one
another in a collaborative discourse activity? Cognitive
Development, 45, 68-76.
AUTHOR INDEX

Adams, D. W. 142
Agudelo, J. 22
Aikin, S. F. 68, 84
ALA Office for Intellectual Freedom 129
Alien, J. W. P. 57, 58, 107
Amar, A. R. 99
Amsel, E. 37, 54
Appiah, K. A. 85, 90, 91
Arena, D. 124
Ash, T. G. 100
Augustinova, M. 76, 79

Báchtiger, A. 15, 84, 86

Baillon, A. 76, 77, 80
Baldwin, J. M. 24
Ball, C. 43, 47
Ball, L. 62
Bardon, A. 93
Barner, D. 33
Barrouillet, P. 10, 11
Bearison, D. J. 76
Bench, S. 54
Bickhard, M. H. 52, 57, 58, 107, 108
Bleichrodt, H. 76
Bloom, P. 42, 43, 45
Boudry, M. 76, 122
Boyes, M. C. 62, 63
Braine, M. D. S. 18, 22, 23, 29
Bráten, I. 12
Brisson, J. 11, 54
Brown, A. L. 52
Brown, D. 142
Brown, M. B. 94
Brunet, M.-L. 11, 54
Bryant, P. E. 22, 24
Byrne, R. M. J. 18, 29

Callan, E. 124
Campbell, R. L. 52, 107, 108
Carey, S. 37
Carpendale, J. I. M. 58
Cauley, K. M. 33
Cesana-Arlotti, N. 18
Chandler, M. J. 13, 58-60, 62-64
Chapman, J. P. 27
Chapman, L. J. 27
Chappin, M. M. H. 77
Chelenza, M. 37
Cheney, R. 63
Churchill, W. 142
Cinalli, M. 93
Clinchy, B. M. 62
Cióse, J. 54
Cottam, K. 59
Cross, D. 57
Cur§eu, P. L. 76-78, 80, 81

Dacey, A. 122
De Brasi, L. 88
De Neys, W. 30
Demetriou, A. 12, 52, 53, 57, 107, 111
Dewey, J. 52, 61
Doise, W. 76
Domberg, A. 76
Dorval, B. 76
Driver, J. 129
Dryzek, J. S. 15
Dublin, R. 37
Dworkin, R. 42, 84, 99

Elstub, S. 84
Erikson, E. H. 91
Estlund, D. 85
Evans, J. St. B. T. 11, 18, 25, 30

Fallón, R. H., Jr. 84, 99

Feiman, R. 33
Feliciano, N. 122
Felmban, W. S. 47
Finken, M. W. 128
Flavell, J. H. 52
Franks, B. A. 26, 56

Gaillard, V. 10
Gallistel, C. R. 32
Gauffroy, C. 11
Geil, M. 69, 75-77, 79
Gelman, R. 32
Gibbs, J. C. 42-45, 47, 107
Gopnik, A. 21, 40
Goswami, U. 48
Graesser, A. C. 6
Greene, J. A. 12, 13, 55, 64, 65
Gutmann, A. 84

Habermas, J. 82, 83
Haidt, J. 45
Hallett, D. 13, 58
Handley, S. J. 30
Hans, V. P. 98
Hart, D. 115, 133
Hartshorne, J. K. 33
Helwig, C. C. 44
Hemberger, L. 122
Henle, M. 18, 28, 29

Inhelder, B. 18, 23-26, 33, 36, 40

lordanou, K. 58, 59

Jambón, M. 43, 47
Jansen, R. J. G. 77
Johnson-Laird, P. N. 10, 18, 28, 29, 36, 40, 68, 72
Johnston, A. 54

Kahneman, D. 7, 10, 11, 29, 33, 103

Karmiloff-Smith, A. 107
Karpowitz, C. E. 87-89
Kazali, E. 53
Kazi, S. 52-54, 57, 107, 111
Khait, V. 122
Killen, M. 91
King, P. M. 13, 6 1 ,6 2
Kirkham, N. Z. 33
Kitchener, K. S. 13, 61, 62
Klaczynski, P. A. 11, 30, 47, 54
Klein, E. 92
Kohlberg, L. 44, 45, 47, 107
Korsgaard, C. M. 91, 135
Koslowski, B. 37
Kostikina, D. 122
Kóymen, B. 76
Kristen, S. 57, 58, 111
Kuhn, D. 13, 37, 40, 63, 65, 76, 80, 107, 110, 121-123, 131,
132

Laats, A. 124, 132

Laden, A. S. 67
Lalonde, C. E. 58
Landemore, H. 85
Landy, J. F. 45
Langer, J. 18, 21, 56, 108
Langley, M. 59
Laughlin, P. R. 76-78, 80
Lent, R. C. 128
Lerner, R. M. 104
Lewis, R. 59
Lickliter, R. 104
Liu, N. 76
Lucas, G. J. M. 76
Lukianoff, G. 127, 128

Maalouf, A. 90
Makris, N. 52, 53
Mansbridge, J. 15
Mansfield, A. F. 62
Marasia, J. 37
Markovits, H. 11, 18, 25, 26, 29, 30, 54-56, 69
Maynard, J. L. 93
McCulloch, W. S. 19, 20, 31
Mclntyre, L. 37
McLaverty, P. 84
Mendelberg, T. 87-89
Mercier, H. 18, 31, 67, 76, 103-106, 122
Mesiec, N. 76
Miles, L. 22
Miller, S. A. 23, 52, 56-59, 108, 111, 112
Moshman, D. ix, 6, 7, 12, 13, 18, 23, 25, 26, 30, 31, 38, 40, 42,
43, 47, 52, 55-60, 64, 65, 67-69, 75-79, 91, 93, 99, 100,
103-108, 110-116, 119, 124, 126-130, 134, 142, 148
Mugny, G. 76

Nagel, T. 109
Neimark, E. D. xi
Newman, I. R. 30
Niño, C. S. 15, 84
Noveck, I. 18, 31, 103
Nucci, L. P. 43, 107
O’Brien, D. P. 18, 22, 23, 29, 69
O’FIynn, I. 93
O’Loughlin, M. 37
O’Madagain, C. 57, 107, 109, 111
Overton, W. F. 11, 12, 18, 104

Paglieri, F. 76, 122

Pandit, S. 22
Parfit, D. 91, 135
Parkinson, J. 86, 94
Paul, R. W. 120, 134
Peng, D. 22
Pennycook, G. 30
Perner, J. 57
Perry, W. G. 61-63
Peterson-Badali, M. 44
Piaget, J. 2, 3, 13, 18-21, 23-26, 29, 31-33, 36, 39, 40, 42-45,
52, 56, 68, 69, 76, 82, 107-111
Pillow, B. H. 23, 52, 56-59, 108, 111, 112
Pipkin, G. 128
Pitts, W. H. 19, 20
Pluut, H. 76
Post, R. C. 94, 128, 130
Premack, D. 58

Rawls, J. 3, 15, 84
Ricco, R. B. 11, 12, 18
Richard, M. 11
Riojas, J. 22
Rips, L. J. 18, 22, 29
Rosenberg, A. 37
Rosenberg, S. W. 87, 88
Rosenfeld, S. 94, 99
Ross, C. J. 124, 129
Royzman, E. B. 45
Ruck, M. D. 44
Rutland, A. 91
Sadler, E. 54
Sandoval, W. A. 12
Santa, J. L. xi
Saxonhouse, A. W. 86, 100, 141
Scholnick, E. K. 2 1 ,2 2
Schraw, G. 12, 52, 57
Sells, S. B. 27, 28
Selman, R. L. 43-45
Sen, A. 117, 138
Shaw, L. A. 59
Shayer, M. 52
Siegel, H. 120, 121, 124, 132, 134
Siegel, S. 11
Singer, M. 6
Slotnick, N. S. xi
Smetana, J. G. 43, 47
Smith, C. 37
Smith, L. 18, 33, 38
Sobel, D. M. 33
Sodian, B. 23, 37, 38, 40, 56-58, 108, 111
Sokol, B. W. 13, 58
Spanoudis, G. 12, 52, 53, 57, 107, 111
Sperber, D. 18, 31, 67, 76, 103-106
Staffel, J. 11
Stanovich, K. E. 11, 12, 52, 137, 138, 141
Starita, J. 142-145
Strossen, N. 100, 130

Talisse, R. B. 68, 84
Tarricone, P. 52
Taylor, A. 84, 86, 87, 94
Thompson, D. 84
Thompson, V. A. 11, 30, 54
Tibbetts, E. A. 22
Tomasello, M. 57, 76, 107, 110, 111
Trabasso, T. R. 6, 22, 24
Trippas, D. 30
Trouche, E. 76, 122
Turiel, E. 43, 47, 107

Wainryb, C. 59
Wakker, P. P. 76
Walker, R. 54
Warren, M. E. 15
Wason, P. C. 18, 28, 36, 40, 68, 72, 76, 92
Watson, J. 57
Wegmann, A. 86
Weinstock, M. 63
Wellman, H. M. 57
Whittington, K. E. 127, 128
Wilkins, M. C. 27
Wimmer, H. 23, 56, 57, 108
Wing, C. S. 21, 22
Witherington, D. C. 38, 104, 135
Woodruff, G. 58
Woodworth, R. S. 27, 28

Youniss, J. 115, 133

Zentall, T. R. 22
Zillmer, N. 76, 122
SUBJECT INDEX

abduction 35
academic freedom 96, 97, 119, 125, 127-132
Academic Freedom Coalition of Nebraska 96
academic integrity 96, 129,131
adolescents: development of 38, 55, 56, 60-64, 68; distinct
from children 112; education of 122-124, 131, 132; and
identity 91; in lifespan perspective 102, 112-115; rationality
of 112-115; reasoning of 25, 26, 44, 47, 48
agency see rational agency
analogical inference 48
android, person hood of 145-148
argumentation 3; as basis for democracy 84-86, 89, 94; as
basis for development 110; as collaborative reasoning 14,
15, 67-82; constraints on 93, 127; in education 121-124,
127, 131-134; in juries 97-99; in legislatures 95-97; as
purpose of reasoning 103-106; rationality of 67, 68
Athens 86, 100, 141, 142
autonomy see rational agency

biases see heuristics and biases

biology: as basis for development 19 -21 ,10 3-10 7; as basis for
logic 19-21; education in 124-126
boarding schools 142-143
Borg 5, 135-141, 150
bounded rationality 9-10

causal reasoning 2, 36-38, 87

causality: conceptions of 21, 39, 40; relation to reasons 9-10,
135; in science, 36-40
Cherokee 143
child development see development
civil liberties 44,117, 128; see also intellectual freedom
cognitive development 106-114; see also development
cognitive psychology 9-13, 27-30
collaborative reasoning 14,15, 67-82; see also argumentation
college student development 61
concrete operations 24-26
Constitution, U.S. 145
constructivism 106, 107
constructivist theory of mind 58, 59, 112
convention see social convention
coordination: as developmental process 20, 21, 107, 109, 110;
of inferences 6, 49; of perspectives 42, 62; of subjectivity and
objectivity 109; of theories and data 37, 41
courts 97-99
creation/evolution controversy 124-126
critical thinking 11,61, 120, 121, 125, 134
curriculum 96, 97, 124-129, 131

decision-making 77, 78, 84, 85, 150

deduction 18, 20, 24, 25, 35, 106
dehumanization 93, 133
deliberative democracy 83-100, 123
democracy: in argumentation 78, 82, 86; deliberative 83-100;
and education 89-90, 123; and expertise 94, 95, 97-100;
and identity 90-94; institutions of 94-100; and intellectual
freedom 100; liberal 15, 81, 84-86; psychological basis of
86-90; rationality of 83-100
denial 93, 133
development: beyond childhood 112-114; child 111,112;
economic, 117; epistemic 55-66; of groups, 116,117; of
identity 91, 92; lifespan, 60, 110-114; mathematical 31-33;
of metacognition 52-55; of metalogical understanding 56, 57,
107-109; moral 43-45; process of 14, 106-110; as progress
13-14, 116, 117; promotion of 119, 122-124; of reasoning
18-26, 30, 31, 39-41, 43-45, 47, 48
developmental epistemology 107
developmental psychology 12-14, 52, 75, 91
dichotomization 92, 93, 95, 133
domains, epistemic 49, 63-65
dual process theories 10-12; and rationality, 88, 89, 137, 138;
of reasoning 29-31, 41, 54, 55

education 16, 89, 90, 119-134

educational psychology 38, 52, 75
emotion 11, 42, 44, 45, 87
empathy 42, 44
empirical knowledge 32, 37, 38
empiricism 106
epistemic cognition 3, 12. 13, 51, 55-66
epistemic development 13, 51, 55-66, 109, 110
epistemic domains 49, 63-65
epistemic doubt 62
epistemology 6; developmental 107; of logic 38; of morality 42,
43; objectivist 49, 63, 109; rationalist 13, 49, 63, 64, 109; of
science 37-39; of social convention 46, 47; in Star Trek
148-150; subjectivist 49, 63-65, 109, 148-150
ethnic cleansing 93,126, 145
evolution: as basis for cognition 103-106; in science education
124-126
executive control 12, 52-54
expertise: in argumentation 78; in democracy 94, 95, 97-100;
in education 130

false belief task 53, 57, 58

First Amendment 128-130
formal logic 27, 57, 104, 108
formal operations 25-27, 40, 68
formal reasoning 24-26, 56, 104
freedom of religión 44, 125
freedom of speech see intellectual freedom

genes 104, 106, 116, 137, 138

genocide 93, 139, 142

habeas corpus 144, 145

heuristics and biases 10.11, 27, 29, 45; see a/so dual process
theories
history 126, 142
human rights 15, 115, 146
hypothesis testing: constraints on 93; logic of, 25, 36, 37, 68;
scientific 25, 26, 36, 37, 40
hypothetical reasoning 24-26, 56, 88, 108

ideáis 150; argumentive 68, 78-82, 89, 95; democratic 15,

83-88, 95; developmental 107; educational 124; rational 9,
31, 94, 120, 121, 132-134
identity: dangers of 90-94, 120, 133; and democracy 90-94;
development of 62, 63; and education 126, 127, 132; moral
93; personal 43, 61, 62, 91, 92; rationalist 93, 94, 134; social
91-93, 126, 127, 142; as theory of self 90-92
“Indians” 142-145
indoctrination 90, 96, 124-129, 131
induction 35, 53
infant competence, 18, 26, 33, 48, 58, 111
infant development, 20, 21, 39, 40
inference 6: analogical 48; automatic 6, 10-12, 28-30, 49;
awareness of 6 ,1 2, 23, 56; to the best explanation 35;
causal 40; control of 6, 12, 56; logical 18-24, 28-30, 56, 108,
120; moral 111; relation to thinking 6; reflection on 108; as
source of knowledge 53, 88; statistical 33; transitive 22, 24
inferential validity, understanding of 27, 54-56, 112
intellectual freedom 16; in argumentation 81; as basis of
academic freedom 128; conceptions of 44; in democracy
100; in development 110; in education 127-133, 150; and
viewpoint neutrality 100, 130, 131
intelligent design 125

journalism 99, 100

juries 97-99
justification, knowledge about see epistemic cognition
justification of beliefs see epistemology

Kant 3, 43, 83
knowledge about knowledge see metacognition, epistemic
cognition

learning: distinctfrom development 31, 90, 106, 116; freedom

of 125-132
legal reasoning 48, 99
legislatures 95-97
liberal democracy 15, 81, 84-86
liberáis, 97, 143
logic: abductive 35; as basis for reasoning 18, 28, 35, 103-106;
biological basis of 19-21; deductive 18, 20, 24, 25, 35, 106;
development of 18-26, 56, 57, 107-109; epistemology of 38;
formal 27, 57, 104; inductive 35, 53; informal 35; knowledge
about 56, 57, 107-109; necessity of 19, 35, 38, 56, 108;
relation to language 21, 22, 33, 34; relation to mathematics
31-33; relation to morality 42, 43; relation to rationality 35
logical reasoning: in adults 27-30; development of 18-26, 56,
57, 107-109
logico-mathematical knowledge 19, 21, 32, 33

mathematical reasoning 31-33

maturity 13, 113-117, 122
memes 137, 138, 150
memory, working 10, 11, 30, 53
metacognition 12, 51-55; as basis for thinking 12, 40, 49;
development of 52-54; as knowledge about knowledge 12; in
reasoning 54, 55; relation to epistemic cognition 12, 51, 55;
and self-regulation 52
metalogical understanding 23; construction of 107-109;
development of 56, 57, 107-109
metarationality 138
metasubjective objectivity 4, 65, 118
moral identity 93
moral reasoning: development of 43-45; as principled
reasoning 41, 42; see also morality
morality 41-45; development of 43-45, 110; epistemology of
42, 43; principled 41, 42; relation to social convention 45-47;
see also moral reasoning
motivation 11, 65
mutual respect 42, 84, 127, 134, 150

Native Americans (“Indians”) 142-145

nativism 106, 116
Nebraska 95-97, 143-145
necessity, logical, 19, 35, 38, 56, 108
norms: academic 131; as basis for reasoning 8 ,1 0, 26, 35;
democratic 85, 94; as ideáis 85; logical 35, 42, 56, 104;
moral 41, 42, 45-47; social conventional 41, 45 -47

objectivist epistemology 49, 63, 109

objectivity, metasubjective 4, 65, 118

peer interaction: in development 109, 110; in education

122-124; in jury system 97, 98; in reasoning 67-82
personal identity 43, 61, 62, 91, 92
persons 114, 115, 141-148; as democratic citizens 15, 84, 85,
142, 145; as rational agents 8, 9, 114, 115, 135; respect for
15, 42, 81,84, 85, 110
perspective taking 42-45, 112
philosophers 8, 24, 35, 105, 120
philosophy of Science 37, 38
Piaget: on biology 19-21, 31; on developmental processes
107-110; on logic 19-21, 23-26, 32; on moral development
42-44; on peer interaction 82, 109, 110; on reasoning 23-26,
32, 33, 39, 40
Ponca 143-145
precedent-based reasoning 45-48, 99, 118
principled reasoning 41-46, 84
probabilistic reasoning 33
progress 13, 14, 116, 117; see also development
qualitative change 13, 111, 116, 117

rational agency: as basis for development 107-110, 118, 122;

as basis for personhood 9, 85, 92,114, 115; as basis for
rationality 8; development of 106-114,118; role in education
121, 122, 124, 128, 129
rational constructivism 107
rational institutions 99, 100
rationalist epistemology 13, 49, 63, 64, 109
rationalist identity 93, 94, 134
rationality: of adolescents, 112-115; as aim of education
119-134; bounded 9-10; as characteristic of personhood 8;
conceptions of 8; development of 106-118; of developmental
process 14, 107-110; as good reasoning 8; of groups 67-82;
as having reasons 8; and irrationality 15, 29, 31, 88, 89; as
metasubjective objectivity 4, 65, 118; moral 43; as rational
agency 8; relation to identity 90-94, 120, 133, 134; relation to
logic 35; of social systems 83-101; in young children 111
reasoning: abductive 35; of adolescents 25, 26, 44, 47, 48;
analogical 48; causal 2, 36-38, 87; collaborative 14, 15,
67-82; conditional 21, 22, 28, 36, 55, 72; deductive 24-26,
35, 54-56, 106; development of 18-26, 30, 31, 39-41,
43-45, 47, 48; epistemic basis for 6, 7, 48, 49; formal 24-26,
56, 104; about hypotheses 25; hypothetical 24-26, 56, 88,
108; inductive 35, 53; informal 35; legal 48, 99; logical 18-30,
56, 57, 107-109; mathematical 31-33; moral 41-45;
precedent-based 45-48, 99, 118; principled 41-46, 84;
probabilistic 33; scientific 36-41; social conventional 45-48;
as subset of thinking 6, 7, 12, 48, 49; see also
argumentation; hypothesis testing
reasons 135,148-150; as basis for belief and action 92, 114;
relation to critical thinking 120, 121; relation to personhood 8,
9, 135; relation to rationality 8, 134; relation to reasoning 6,
49; respect for 80, 81, 84, 85, 121, 126
reflection: as basis for reasoning 103,105; as developmental
process 57, 107-109; in dual process theories 11,12
reflective judgment 61, 62
relativism 59, 61, 62, 150
religión 44, 92, 124-126, 129
respect: mutual 42, 84, 127, 134, 150; for persons 15, 42, 81,
84, 85, 110; for reasons 80, 81, 84, 85, 121, 126
rights: adolescent 115; democratic 15, 84, 100, 115; in
education 127-130; moral 41-44; of rational agents 115,
146-148
“Robot’s Rebellion” 137-139

science: causal explanation in 37, 38; epistemology of 37-39;

as rational institution 37, 99, 100; understanding of 122
science education 122, 124-126
scientific reasoning: as causal reasoning 36, 37; development
of 39-41; epistemology of 37-39; as hypothesis testing 36,
37; role of theory in 37
Scopes trial 124
selection task 28, 36, 68-79, 120
self-conceptions 91-93
self-regulation: as aspect of metacognition 12, 52; of biological
systems 19; in thinking and reasoning 51
self-serving biases 11, 92, 120
sentience 146, 147
skills 119, 132, 133, 142
slavery 86, 100, 142, 147
social convention: as distinctfrom morality 45-47;
epistemology of 46, 47; in law 99
social conventional reasoning: development of 47, 48; as
precedent-based reasoning 45, 46
social domain theory 47
social identity 91-93, 126, 127, 142
social interaction see peer interaction
social psychology 11, 75
social systems: as basis for social conventions 46, 47;
development of 116, 117; rationality of 79-87, 94-100
Standing Bear 142-145
StarTrek 5, 135-137, 139-141, 145-150
student rights 44, 127-130
subjectivist epistemology 49, 63-65, 109, 148-150
Supreme Court, U.S. 96, 125, 145

theory: of mind 52-54, 57-59, 64, 88, 111, 112; in scientific

reasoning 37; of self 90-92, 94, 134
thinking: as coordination of inferences 6, 49; critical 11,61,
120, 121, 125, 134; metacognitive basis of 12, 40, 49;
relation to reasoning 6, 7, 12, 48, 49
Thinking & Reasoning (journal) 30, 69
trail of tears 143
truth see epistemology
truth, knowledge about see epistemic cognition
“truth of the world” 149, 150

United States 86, 95, 97, 125, 128, 142, 145

valúes 114, 121, 132-134, 137, 138

verification bias 70, 92, 93, 133
viewpoint neutrality 130, 131
violence 93

West Virginia v. Barnette 84, 96

working memory 10, 11, 30, 53
Wounded Knee 142, 143