Irrationality

Lisa Bortolotti

• Publisher: Polity
• Year: 2014
• Print ISBN: 9780745647838, 9780745647821
• eBook ISBN: 9780745687278
• Edition: 1
• Topic: Psychology
• Subtopic: Cognitive Psychology & Cognition
• Index: Psychology

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.

1
Irrationality 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 because it involves at least three entities that are potentially connected to one another in a causal way: a speaker, an interpreter and a relevant feature of their shared environment. By engaging in this form of triangulation, that is, responding to the speaker and reacting to the environment, the interpreter gradually starts to understand the speaker's language, but attributions of meaning to utterances and attributions of content to mental states cannot begin unless some prior assumptions are in place. Ruling out the possibility that speakers have obviously false beliefs in this sort of context is a reasonable assumption for the radical interpreter. (And that is why, when a rabbit appears, ‘Gavagai!’ is taken to mean ‘Look, there's a rabbit!’ as opposed to ‘Look, there's an elephant!’)

As interpretation is a challenging task when very little background information is available, the interpreter's working assumptions play an important role. According to Davidson, in order to make hypotheses about the meaning of unfamiliar utterances, the interpreter needs to assume that the speaker will have true beliefs (Davidson 1980: 238) and that the speaker's beliefs will be connected to one another according to principles of good reasoning that are informed by basic principles of logic (Davidson 1985: 346). The success of radical interpretation is hostage to the truth condition and the logic condition. When combined, t...

Table of contents

• Cover
• Key Concepts in Philosophy
• Title page
• Copyright page
• Dedication
• Acknowledgements
• Introduction
• 1: Irrationality and Interpretation
• 2: Irrationality and Mental Health
• 3: Irrationality and Choice
• 4: Irrationality and the World
• Conclusion: Irrational Human Agents
• Bibliography
• Index