Irrationality E-Book

Lisa Bortolotti

• Herausgeber: John Wiley & Sons
• Kategorie: Geisteswissenschaft
• Serie: Key Concepts in Philosophy
• Sprache: Englisch

Seitenzahl: 311

Veröffentlichungsjahr: 2014

Beschreibung

We talk about irrationality when behaviour defies explanation or prediction, when decisions are driven by emotions or instinct rather than by reflection, when reasoning fails to conform to basic principles of logic and probability, and when beliefs lack coherence or empirical support. Depending on the context, agents exhibiting irrational behaviour may be described as foolish, ignorant, unwise or even insane.

In this clear and engaging introduction to current debates on irrationality, Lisa Bortolotti presents the many facets of the concept and offers an original account of the importance of judgements of irrationality as value judgements. The book examines the standards against which we measure human behaviour, and reviews the often serious implications of judgements of irrationality for ethics and policy. Bortolotti argues that we should adopt a more critical stance towards accepted standards of rationality in the light of the often surprising outcomes of philosophical inquiry and cognitive science research into decision making.

Irrationality is an accessible guide to the concept and will be essential reading for students and scholars interested in the limitations of human cognition and human agency.

Leseprobe

Key Concepts in Philosophy

Heather Battaly, Virtue

Joseph Keim Campbell, Free Will

Roy T. Cook, Paradoxes

Douglas Edwards, Properties

Bryan Frances, Disagreement

Douglas Kutach, Causation

Ian Evans and Nicolas D. Smith, Knowledge

Daniel Speak, The Problem of Evil

Joshua Weisberg, Consciousness

Chase Wrenn, Truth

Copyright © Lisa Bortolotti 2015

The right of Lisa Bortolotti to be identified as Author of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.

First published in 2015 by Polity Press

Polity Press

65 Bridge Street

Cambridge CB2 1UR, UK

Polity Press

350 Main Street

Malden, MA 02148, USA

All rights reserved. Except for the quotation of short passages for the purpose of criticism and review, no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.

ISBN-13: 978-0-7456-4782-1

ISBN-13: 978-0-7456-4783-8(pb)

ISBN-13: 978-0-7456-8727-8 (epub)

ISBN-13: 978-0-7456-8726-1 (mobi)

A catalogue record for this book is available from the British Library.

Library of Congress Cataloging-in-Publication Data

Bortolotti, Lisa.

    Irrationality / Lisa Bortolotti.

        pages cm

    Includes bibliographical references and index.

    ISBN 978-0-7456-4782-1 (hardback) – ISBN 978-0-7456-4783-8 (pbk.)    1.  Practical reason.    I.  Title.

    BC177.B595 2014

    128′.4–dc23

                                                                                2014013539

The publisher has used its best endeavours to ensure that the URLs for external websites referred to in this book are correct and active at the time of going to press. However, the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.

Every effort has been made to trace all copyright holders, but if any have been inadvertently overlooked the publisher will be pleased to include any necessary credits in any subsequent reprint or edition.

For further information on Polity, visit our website: politybooks.com

For Giulio Paternoster and Anna Bonaria Mameli

Acknowledgements

I am grateful for the incredible support that I received from Emma Hutchinson, Pascal Porcheron and Sarah Lambert at Polity Press. They have been very patient and very understanding.

I acknowledge the intellectual support of the Philosophy of Health and Happiness research cluster in the School of Philosophy, Theology and Religion at the University of Birmingham. The many conversations I had with students and colleagues shaped my views about rationality and human agency, and greatly inspired me. I would also like to thank Benjamin Costello for proofreading the whole manuscript and providing insightful comments.

Some of the arguments I discuss in the book have been developed in previous work, either single-authored or authored together with invaluable collaborators. I would like to thank Jillian Craigie, Matteo Mameli, Rochelle Cox, Amanda Barnier, Bert Heinrichs and Matthew Broome for working with me and for their precious insights into many of the philosophical issues surrounding irrationality.

Finally, I acknowledge the support of an Arts and Humanities Research Council Fellowship (‘The Epistemic Innocence of Imperfect Cognitions’, grant number AH/KOO3615/1) in the preparation of chapters 2 and 4 of the book.

Introduction

Judgements of irrationality are a constant feature of our social practices and, on many occasions, they also play an important role in our mutual interactions. This book is an attempt to understand what we take irrationality to be; what grounds our judgements of irrationality; and to what extent irrationality compromises human agency.

The book has three main aims. The first aim is descriptive. I examine some of the dominant conceptions of irrationality, and highlight elements of continuity and discontinuity among them. In order to achieve this aim, I review some recent debates in philosophy and cognitive science that centre on the conformity of human behaviour to different norms of rationality.

The second aim is to challenge the assumption that human agency is rational agency. Irrational behaviour should not be seen as the exception, as an unexpected deviation from the norm that needs to be explained away, but as a core feature of the behaviour of human agents. Acknowledging that irrationality is pervasive allows human agents to develop strategies to reduce the potentially negative effects of irrationality on the acquisition of knowledge, on the pursuit of their goals and on flourishing more generally.

The third aim is to reassess the relationship between the cognitive and the affective, and between intuitive or associative processes and deliberative ones. The recent empirical literature invites us to re-evaluate the role of emotions in attitude formation, problem-solving, and decision-making. The discussion will provide some reasons to be sceptical of the claim that rationality is about disregarding emotional reactions. Rational human agents form true beliefs, have consistent preferences and make good choices. This means trusting emotions and intuitions in some contexts, and adopting a more reflective and deliberative stance in other contexts.

By the end of the book, I will have sketched a psychologically realistic, but not entirely pessimistic, account of the relationship between human agency and rationality.

The Project

What is irrationality, and why is it a key concept? The main motivation for writing a book on irrationality is to clarify what we mean by the concept when we use it in an academic context and in everyday language. There is no definition of irrationality that is universally accepted, and one may feel that irrationality is a moving target. It is a concept that escapes clarification because it is used to express disapproval towards behaviour that is sanctioned, where the type of behaviour to be sanctioned and the reasons why it is sanctioned change according to the aspects of human agency that are found to be problematic in each context.

What is human agency? Agents are those who act. When we have human agents in mind, we assume that they have the capacity to act by deliberating on the best course of action for them. This means that they have interests dictated by their beliefs and desires, and on the basis of those interests they select goals for themselves. They form the intention to pursue those goals, and search for the best means of achieving their ends. Human agency is often defined by rationality. It is widely accepted that human agents distinguish themselves from non-human agents (such as non-human animals) or from human non-agents (such as people in a persistent vegetative state) by a set of capacities, among which are: their capacity to accurately represent the world via a coherent set of largely true beliefs; the capacity to set themselves appropriate goals and make good choices by exercising reason; the capacity to make plans by choosing the best means to achieve their goals; the capacity to act in accordance with their plans; the list could go on.

The rationality of human agents is characterized by norms: their beliefs must be true and form a coherent set; their intentions to act must be governed by the correct identification of those means that are likely to further their ends; their actions must be consistent with their intentions. When philosophers describe human agency, they recognize some of its limitations but emphasize that human agents share a background of rationality, are not at the mercy of emotional reactions and gut feelings, and use reflection to pursue goals that advance their interests.

Judgements of irrationality typically express disapproval towards some aspects of the behaviour of human agents. When do we deem an agent's behaviour to be irrational? The following are common examples:

When beliefs are badly supported by evidence and conflict with the science of the day.

When reasoning fails to conform to reasoning standards shaped by basic principles of logic and probability.

When decisions are driven by emotions or instinct rather than by deliberation.

When intentions do not reflect beliefs and desires.

When actions badly match intentions.

When the appropriate means to achieving a goal are not selected.

When goals are not pursued consistently.

Although these are just a few cases in which judgements of irrationality are made, each case involves the infringement of a different set of norms. Depending on the context and on the extent to which their behaviour deviates from the norm, agents exhibiting irrational behaviour may be regarded as foolish, ignorant, unwise or even mad. When their departure from rational norms is significant, their very agency is called into question.

It is an especially challenging task to identify norms of rationality because different notions of rationality are deployed in the literature. Many distinguish between norms of instrumental and substantive rationality. The former are concerned with processes, for instance, the processes by which an attitude is formed or a decision is reached. The latter are concerned with ends, for instance, the ends that should be pursued in forming attitudes or making choices. Another common distinction is about whether the success criterion for rational agency is epistemic or pragmatic. The rational agent is either the one who gets things right (e.g., the one with true beliefs), or the one for whom things go well (e.g., the one with beliefs that are conducive to happiness or flourishing). There is considerable overlap between these two criteria because, in many contexts, epistemic rationality is a source of pragmatic benefits, and well-being fosters attitudes that contribute positively to the acquisition of true beliefs. However, epistemic and pragmatic considerations can conflict, and the discussions in this book will reflect on irrational beliefs that are pragmatically beneficial, and on rational beliefs that lead to failure.

The Methodology

Throughout the book, the philosophical debates surrounding irrationality are approached in a way that is informed by current psychological studies of different aspects of human agency. One may wonder why the philosopher should pay attention to the ever-growing body of empirical data about human cognitive capacities, and whether she can contribute to the interpretation of empirical data in any substantial way.

There are different ways of understanding the relationship between philosophy and the sciences, and some encourage a very narrow view of the remit of philosophy. Either the philosopher is called to shed light on those issues that non-mature sciences struggle to deal with, or she is merely asked to disentangle conceptual confusions. On the former view, when a scientific discipline becomes mature enough to deal with a particular area of investigation, the role of the philosopher is exhausted. On the latter view, philosophy has an important but merely critical role. Fortunately, there is a third way (Stone and Davies 1993). Science and philosophy, with their often distinct focus and methodology, can genuinely engage in a two-way interaction and advance understanding of a number of thorny problems.

According to this latter option, philosophy and the cognitive sciences are on equal footing: they inform and constrain each other. And this is the methodological framework that shapes the contents of this book. Being well informed about the latest results of empirical studies on human irrationality allows us to rule out some views about the rationality of human agency. Conversely, philosophical reflections on the relationship between evidence and beliefs, beliefs and desires, intuition and choice, and so on, prevent us from hypothesizing empirical connections that lie on conceptual confusions and suggest new avenues for empirical investigation.

Some Themes

There are several themes that run throughout the book. One well-accepted view I shall challenge is that one needs to be rational in order to be an agent. In chapter 1, we have some clear examples of this idea. If a young woman has very long showers but declares that saving water is very important to her, we may be puzzled as to what she genuinely believes. Inconsistent behaviour is not just a mistake to be rectified. It threatens one's status as an agent, and that is why we feel entitled to doubt that the woman's sincere assertion is a guide to what she believes. In chapter 2, the view that there is a tight relationship between agency and rationality is shown to have tangible consequences for those who are diagnosed with a psychiatric disorder. In our society people with some psychiatric diagnoses who report bizarre beliefs or act in ways that seem inexplicable or self-defeating are often thought to have lost the capacity for autonomous choice. As a result, they are no longer held fully responsible for their actions.

A second, related view I shall challenge is that one needs to be rational in order to be a successful agent. The analysis of the relationship between rationality (intended as the agent's satisfaction of the relevant epistemic standards) and success (intended as the fulfilment of goals that increase the agent's well-being) underlies all the debates presented in the book. Does irrationality necessarily lead to failure? I hope that the discussion will shed some light on the complex relationship between rationality and success, and highlight in which contexts they go hand in hand and in which contexts they conflict.

A third view to be resisted is that one needs to engage in reflective deliberation to be a rational and successful agent. One common way to distinguish human agency from non-human agency is to point out that human agents can control the effects of emotional and motivational forces on their behaviour, and the philosophical orthodoxy tells us that failure to let reason prevail over emotion is the main obstacle to living a good life and making wise choices. Versions of this idea are discussed in chapters 3 and 4. Even in one of the most distinctively human capacities, the making of moral judgements, it has been shown that the role of raw affective responses is significant. Agents may choose a course of action following their unconscious reactions (e.g., disgust or arousal), and then make up reasons to vindicate their choices. Reason is at the service of emotion, not the other way round. Moreover, strong motivational factors interfere with attitude formation; that is, people believe what they desire to be true, not what they have evidence to believe. For instance, many believe that they are better than average at most things, and have inflated conceptions of their skills and talents.

Synopsis

The book has four chapters with a similar structure. In each chapter, I consider a conception of irrationality that applies to one aspect of human agency. I ask which normative standards an agent's behaviour is taken to violate when it is judged as irrational in that sense. Next, I review some of the classic and contemporary debates centred on the relevant conception of irrationality. Finally, I discuss the relationship of the relevant conception of irrationality with what we value about human agency.

Given the vastness, diversity and multi-disciplinarity of the literature on irrationality, I can only scratch the surface here. The reader will have an opportunity to continue exploring each conception of irrationality and the debates surrounding it by following the suggestions for further reading at the end of each chapter.

Chapter 1 – Irrationality and Interpretation

Unlike clouds, rocks and plants, human agents have mental states with content (intentional states), and their intentional states drive their actions. When we want to understand what an agent is doing, or predict what she will do next, we rely on the intentional states we can ascribe to that agent based on her verbal and non-verbal behaviour. Lena is reaching for the fridge because she wants a beer. Lena will go to the supermarket to buy a beer if there is no beer left in the fridge. The practice of explanation and prediction of behaviour in intentional terms is often called ‘interpretation’. An influential view in the philosophy of mind tells us that, in order for ascriptions of beliefs and desires (e.g., ‘There is beer in the fridge’, ‘I want a beer’, ‘My local supermarket sells beer’) to support interpretation, agents need to have a coherent system of largely true beliefs that fit nicely with their other intentional states and are reflected in their behaviour. Such a notion of epistemic rationality is thought to be a necessary condition for ‘interpretability’. The idea is that there is a ‘rationality constraint’ on interpretation (Bortolotti 2005a, 2005b). For interpretation to work, beliefs need to be largely true and form a coherent system. If beliefs are false or there are inconsistencies in an agent's beliefs, interpretation fails. The behaviour of those who violate the epistemic norms of truth and consistency becomes inexplicable and unpredictable, and because of that their agency is called into question. If one cannot have largely true and coherent belief sets, then one cannot have beliefs at all.

The conception of irrationality I discuss in chapter 1 concerns the violation of those assumptions and expectations that allegedly ground interpretation. In that chapter, the rationality constraint on interpretation will be reviewed and found wanting, and alternative conceptions of what grounds our attempts at understanding and predicting each other's behaviour will be explored.

Chapter 2 – Irrationality and Mental Health

Human agents can succeed in achieving their goals and flourish, or struggle and find themselves in situations that are deeply distressing. Some of the agents who struggle and experience distress attract a psychiatric diagnosis. Tom is very preoccupied that people at work talk about him when he is not around, and believes that they want to get him fired. This thought is so distressing that he cannot concentrate on anything else during the day, and he cannot sleep properly at night. If he comes to the attention of a psychiatrist, he may be diagnosed with persecutory delusions. On what basis is mental illness diagnosed? Diagnostic manuals available to clinicians characterize disorders on the basis of their symptoms, and the symptoms are often defined as failures of epistemic rationality. For instance, psychotic symptoms such as delusions are defined as ‘fixed beliefs’, that is, beliefs that are not responsive to evidence.

It is often argued that irrationality and insanity are inextricably linked, and mental illness just is a breakdown of rationality and self-knowledge, ultimately undermining one of the most cherished prerogatives of human agency, autonomous decision-making. The conception of irrationality I discuss in chapter 2 concerns the violation of the epistemic norms that allegedly underlie normal functioning. In that chapter, I challenge the simplistic assimilation of insanity with irrationality, and resist the idea that mental illness necessarily compromises autonomous agency and responsibility for action.

Chapter 3 – Irrationality and Choice

In their typical portrayal, developmentally mature human agents make choices by deliberating. They weigh up reasons and evidence for and against a certain course of action. Differently from human infants, ants and clams, the behaviour of mature human agents is not necessarily driven by instinct, but shaped by reason. Carolyn needs to buy a house. She has to choose whether to buy a house in the city centre or a house in the suburbs. The house in the centre is smaller and more expensive, but closer to where Carolyn works. The house in the suburbs is bigger and cheaper, but farther away from work and the city attractions. Carolyn values her time above all things and she does not want to spend one hour commuting every day. So she chooses the house in the city centre. Human agents tell similar stories about many of their important choices in life and justify what they end up doing by reference to reasons and values. But it is not always clear how they actually made their choice. What if Carolyn chose the smaller house because she loved how it looked, and then rationalized her choice by appealing to the benefits of the shorter commute?

The conception of irrationality I discuss in chapter 3 concerns the relationship between reflective deliberation and emotions or affect in attitudes formation and decision-making. The wisdom of a human agent's attitude and choices can be characterized epistemically or pragmatically, as the agent tracking the truth and matching expert judgement, or as the agent endorsing attitudes and making choices that contribute to her long-term well-being. The question is whether reliance on unconscious urges and raw feelings causes the agent to make the wrong choice. Is a course of action selected on the basis of careful deliberation always better than a course of action selected on the basis of gut feelings? In the chapter, I shall show that the main challenge for human agents is not to control the emotions, but to know when to be guided by their emotions, and when to stop and think instead.

Chapter 4 – Irrationality and the World

The rational agent forms beliefs, reconstructs memories and develops narratives that are constrained by evidence and free from motivational influences. This reliance on evidence underlies two enterprises that have been regarded as distinctive of human agency as the capacity to control emotions: the acquisition of scientific knowledge leading to the successful manipulation of nature, and the development of self-narratives to interpret the past, guide the future and ground autonomous decision-making. Both enterprises are aimed at tracking the truth and are supposed to be responsive to the feedback of the physical environment and social context surrounding the agent.

The conception of irrationality I discuss in chapter 4 involves the alleged clash between epistemic and pragmatic ends. The epistemically rational human agent is like a good scientist: she forms beliefs that are well supported by evidence, and revises or abandons them when counter-evidence becomes available. She takes into account all the relevant evidence, avoiding biases and acknowledging her limitations. Her beliefs may not be always true, but they are supported by the past and present evidence available to her. The human agent who is epistemically irrational does not meet these standards and may let her desires influence her beliefs, reconstructing her memories to maintain a better concept of herself, and building self-narratives that are unreliable as a result.

Although it is commonly believed that such failures of rationality compromise both knowledge and happiness, this is not always the case, and cognitions that violate key norms of epistemic rationality may have not only pragmatic but also epistemic benefits in the context of the limitations of human cognition. This leads to a consideration that seems supported by the literature examined in the whole book: epistemic rationality is an aspiration for human agents, but when its often demanding standards are not met, then it is still possible for humans to exercise agency, and to exercise it with some degree of success. One measure of the achievement of human agency is the extent to which agents can navigate the world around them notwithstanding their considerable limitations.

1Irrationality and Interpretation

In his last novel, entitled One, No One and One Hundred Thousand, Italian playwright Luigi Pirandello (1867–1936) tells the story of Vitangelo Moscarda. One day, in front of the mirror, Moscarda notices that his nose is not entirely straight. This simple observation kick-starts a series of revelations that deeply unsettle him. For instance, Moscarda realizes that his wife sees him as someone silly and superficial, very differently from how he sees himself, and that his colleagues and friends have also constructed images of him which do not correspond to what he takes to be his ‘true’ self. Moscarda decides to destroy all his ‘false’ selves one by one by acting in a way that is inconsistent with, and systematically defies, people's expectations. For instance, everybody assumes that he cares for the business his father bequeathed to him, but instead he gets rid of it. In the end, Moscarda's seemingly erratic behaviour is considered as a sign of folly and he is confined to a mental institution. People in Moscarda's life do not realize that Moscarda's actions are part of a plan. They are exposed to behaviour that does not make sense given what they have always known about Moscarda. Their interactions with him are made difficult by the unpredictability of his behaviour. They cannot explain Moscarda's behaviour on the basis of the information they already have, and they come to judge him as irrational.

1.1.    Predicting and Explaining Behaviour

In the novel, Pirandello wanted to say something about the relativity and fragility of personal identity, but, on a simpler level, the fictional story of Moscarda can help flesh out an influential view in the philosophy of mind. The idea is that only when behaviour conforms to basic norms of rationality can it be explained and predicted by reference to beliefs, desires and other mental states with content (hereafter, jointly referred to as intentional states). The view I am going to explore in this chapter is that interpretation fails when behaviour is irrational. Intentional agency is the capacity to represent the world and be guided by those representations in action. Human agents are intentional agents, and intentional agents are necessarily rational agents.

In a coffee shop, you see a little girl staring at a piece of chocolate cake and you think to yourself, ‘She wants to eat that piece of cake.’ You interpret the behaviour of the girl, that is, you explain her behaviour by ascribing an intentional state to her (in this case, a desire for the cake). When agents attempt to make sense of what other agents are doing or predict what they will do next, they see them as intentional agents, and they see their movements and utterances as motivated by their intentional states. The practice of interpretation includes the prediction and explanation of behaviour on the basis of those states. Consider the following examples: the waiter is friendly because he hopes to get a good tip; the boy is jumping with excitement because his grandmother will take him to the toy shop; the coach will replace a midfielder with a striker because the match is almost over and her team needs to score a goal.

The form of irrationality I shall explore in this chapter is attached to behaviour that seems to defy interpretation, and make explanations of past behaviour and inferences from observed behaviour to future behaviour either difficult or impossible. In their distinct but partially overlapping contributions, philosophers Donald Davidson and Daniel Dennett attempt to capture the key aspects of the process by which agents interpret behaviour. They are committed to the view that part of what makes it true that an individual has a certain belief or desire is that the individual would be interpreted as having that belief or desire by a suitably located interpreter. Davidson and Dennett are also committed to the view that, in order to explain and predict behaviour in intentional terms, the interpreter relies on certain assumptions, and the central one is that the behaviour measures up to standards of rationality. These commitments are the tenets of interpretationism, the view that the practice of interpretation sheds light on the nature of intentional agency.

1.1.1.    The Intentional System Theory

What notion of rationality are interpretationists working with? In Dennett's essay ‘True Believers’ (1981), we find the following account of rationality: an agent is rational if she has the beliefs and desires she ought to have and if she does what she ought to do in order to achieve her goals, given those beliefs and desires. What beliefs and desires an agent ought to have is a controversial issue in itself. Dennett suggests that rational agents believe all the available truths that are relevant to their interests, and that they desire what they believe would be good for them. In Dennett's intentional system theory, the assumption of rationality is necessary to develop a predictive strategy for human agents.

According to Dennett, there are three strategies human agents use in predicting behaviour. The physical stance gives some insight into the behaviour of the system to be predicted on the basis of the physical constitution of that system, the physical nature of what might affect the system and the knowledge of the laws of physics. This is the strategy observers adopt to predict at what temperature the water will boil and whether it will rain on Sunday. If they adopt the design stance instead, the behaviour of the system is predicted on the basis of how the system has been designed to behave. Observers use the design stance to predict when the alarm clock will ring or what will happen if they press the ESC key on the keyboard of a personal computer. They could predict the same events by adopting the physical stance, and thus by appealing to the physical constitution of the alarm clock and the computer, but the complexity of such systems makes the design stance more convenient and economical. If a malfunction occurs, then they need to revert to the physical stance, because the system stops behaving as it was designed to behave. The intentional stance requires that the system whose behaviour is to be predicted is regarded as a rational agent with intentional states. Interpreters have to think about which, say, beliefs and desires the system ought to have, given its goals. Then they can predict that the system will act to achieve its goals according to the beliefs and desires ascribed to the system. In this context, assuming rationality is assuming that the system will have the beliefs and desires it ought to have, and that it will do what it ought to do in order to achieve its goals given those beliefs and desires. The intentional stance is the default strategy for the interpretation of the behaviour of human agents.

Even though, in principle, it is possible to predict the behaviour of human agents on the basis of the first two predictive strategies, the physical and the design stance, the intentional stance is not superfluous. Dennett thinks that a hypothetical external observer of human behaviour, say, a curious Martian, would miss something important if he did not adopt the intentional stance. The Martian would not be able to recognize those patterns of behaviour that support generalizations and predictions within the intentional stance. By adopting the physical or the design stance, and thinking exclusively in terms of physical constitution or biological function, the Martian would fail to provide the simplest explanation of the behaviour of human agents, and to see the patterns that enable human agents to understand each other and coordinate with one another (e.g., people are sad when they receive bad news; if you invite someone for dinner at your place, they will likely bring a bottle of wine or some chocolate; people who stand next to the bus stop believe that the bus will come soon and they will catch it when it comes).

Interpreters could use the intentional strategy to predict the behaviour of any individual, and they do sometimes talk about the behaviour of their pets or of plants, clocks or clouds in intentional terms (e.g., ‘The tomato plant wants sunlight’, ‘The dog believes that the squirrel is up in the oak tree’). But the success of the intentional strategy in the prediction of the behaviour of human agents is greater than its success in the prediction of the behaviour of domestic animals, plants, clocks or clouds. Human agents use the intentional strategy all the time with each other, and typically are very good at explaining and predicting the behaviour of fellow humans on the basis of the assumptions they make about intentional systems. For Dennett, this is not surprising: through evolution, humans are designed to be rational.

What are the consequences of irrationality for interpretation? For Dennett, prediction on the basis of the intentional stance is always compromised in cases of irrationality. That said, different forms of irrationality might invite different responses. Consider obviously false perceptual beliefs, such as those caused by hallucinations. For instance, Delia who is intoxicated experiences hallucinations which give rise to false perceptual beliefs about her surrounding environment, such as: ‘There is a flashing light in front of me.’ Delia's behaviour can be explained in terms of the effects of alcohol or drugs on visual perception. It is successfully explained on the basis of faulty mechanisms underlying perception and predicted not via the intentional, but the physical stance.

In cases of seemingly inconsistent or self-deceiving behaviour, Dennett suggests that interpreters cannot make clear, determinate attributions of beliefs and desires, and their predictions are going to be unstable too. Suppose Mary is self-deceived. She keeps telling herself that her son Jimmy did not rob the bank even though she has plenty of evidence that he did. She cannot admit to herself that Jimmy is guilty of a serious crime. Which belief should an interpreter ascribe to Mary? Is it possible to reliably predict what she will do next, based on the belief that Jimmy is innocent? Indeterminate ascriptions of beliefs and desires and tentative predictions can be made when deviations from rationality are temporary and occasional. They constitute the exception to the rule. If behaviour is pervasively and systematically irrational, then no attribution and no prediction can be made via the intentional stance.

1.1.2.    Radical Interpretation

In Inquiries into Truth and Interpretation (1984), Davidson also argues for a very influential version of the rationality constraint on interpretation. He invites us to consider a special case of belief ascription, the case of radical interpretation. Interpretation is radical when someone attempts to understand the utterances and the behaviour of an individual who speaks a totally unknown language and is potentially embedded in a very different cultural background from that of the interpreter. Imagine that an anthropologist finds a previously unknown tribe on a remote island, and wants to understand their language in order to study their way of life. Radical interpretation starts with a set of causal interactions among speaker, interpreter and their shared environment. The classic example in the literature is that of a speaker seeing a rabbit and uttering ‘Gavagai!’ (Quine 1960). Even if the interpreter has never heard this utterance before, she can reasonably guess that it means something like ‘Ho, a rabbit!’ or ‘Look, there's a rabbit!’ because the shared environment provides a telling clue.

The set of causal interactions is called triangulation