Philosophy of Psychology. Lisa Bortolotti
reason as well as man’ (Hume 1739/2007, 118).
Another challenge, which is the focus of this chapter, is to deny optimism about human rationality. Non-human animals are not rational, but humans are not rational either. In the memorable beginning of his essay ‘An Outline of Intellectual Rubbish’, Russell wrote:
Man is a rational animal – so at least we have been told. Throughout a long life I have searched diligently for evidence in favour of this statement. So far, I have not had the good fortune to come across it, though I have searched in many countries spread over three continents. (Russell 1961, 45)
Russellian pessimism about human rationality is echoed in a particularly influential psychological research programme in the 20th century, the heuristics and biases research programme, led by Amos Tversky and Daniel Kahneman. The studies in this programme revealed ‘systematic and severe errors’ (Tversky & Kahneman 1974, 1124) in human reasoning, which seem to have a ‘bleak implication for human rationality’ (Nisbett & Borgida 1975, 935). These studies caused heated debates on human rationality, sometimes dubbed ‘rationality wars’ (Samuels, Stich, & Bishop 2002), both at the theoretical and conceptual level and at the empirical and experimental level. The main aim of this chapter is to examine the relevant psychological studies to see if they really do have bleak or pessimistic implications for human rationality.
We start by presenting a definition of rationality (Section 1.2) before turning to the relevant psychological studies, in particular the ones from the heuristics and biases programme, which reveal a systematic failure to reason according to the rules of logic, probability, and decision-making (Section 1.3). The results of these studies support a pessimistic view of human rationality (Section 1.4). However, objections to the pessimistic interpretation were raised by Gerd Gigerenzer and his ecological rationality research programme (Section 1.5). The objections from this programme are significant because they bring to the fore important ideas that enhance our understanding of human rationality/irrationality. However, we shall argue that these objections do not refute the pessimistic interpretation. Although Gigerenzer’s objections do not refute pessimism, they lead us to examine an important issue: whether the difference between optimism and pessimism about rationality hangs on different accounts of the aim of cognition (Section 1.6). In the end, we defend a moderate form of pessimism, according to which humans are not as rational as we might have thought, before appreciating the results of the psychological studies on reasoning.
1.2 Clarifying Rationality
Are humans rational? To answer this question, we need to investigate two sets of issues: (1) ‘What does it mean to be rational?’, ‘What is necessary for a person to be rational?’, and ‘What is sufficient for a person to be rational?’ (the philosophical questions about rationality); and (2) ‘Do humans satisfy a sufficient condition for being rational?’, ‘Do humans fail to satisfy a necessary condition for being rational?’, and ‘What do empirical studies say about these issues?’ (the psychological questions about rationality).
We discuss the philosophical questions in this section and the psychological questions in the next section.
‘Rationality’ means different things in different contexts (Bortolotti 2014). We will focus on a narrow sense of rationality here: that is, rationality in the context of reasoning. In particular, we (tentatively) accept what Edward Stein (1996) calls ‘the standard picture of rationality’, according to which rationality consists in reasoning in accordance with the rules of logic, probability, and decision-making (see Box 1A).
BOX 1A: The Standard Picture of Rationality
Definition:
‘According to this picture, to be rational is to reason in accordance with principles of reasoning that are based on rules of logic, probability theory, and so forth. If the standard picture of reasoning is right, principles of reasoning that are based on such rules are normative principles of reasoning, namely they are the principles we ought to reason in accordance with.’ (Stein 1996, 4)
Example of deductive reasoning:
Jessie believes that either Paula got an A in geometry or Vanessa did, and also that Vanessa did not get an A but got a D instead. From these premises, Jessie concludes that it is Paula who got an A. Jessie is rational: her logical reasoning is in accordance with a rule of logic that one can infer P from (P or Q and not Q) (disjunctive syllogism).
Example of probabilistic reasoning:
Felix assigns the probability 0.8 to the idea that it will rain tomorrow based on the televised weather forecast that there is an 80% chance of rain tomorrow. Felix also assigns the probability 0.9 to the idea that it will rain tomorrow and so he will go to the gym for a workout. Felix is irrational: his probabilistic judgment violates what is known as the ‘conjunction rule’; the probability of an event A occurring, P(A), cannot be less than the probability of A and another event B occurring at the same time, P(A&B).
But how often should human agents reason in accordance with these rules to be considered rational agents? If the answer is ‘Always’, then a rational person wouldn’t be allowed to make any mistakes; this requirement is far too demanding. It is obvious that human agents are not rational according to such a stringent requirement. A realistic view would take into account the fact that rational agents can make reasoning errors, such as the occasional performance error that is attributed to some interfering factor (e.g., lack of concentration) (see Box 1B), but they do not make systematic reasoning errors. Thus, rationality is compatible with occasional reasoning errors, but it is not compatible with systematic reasoning errors.
BOX 1B: Competence vs Performance
When you attempt a task and fail, one of two things may be happening: either you do not have the capacity to accurately perform the task due to some knowledge-gap, or you have the capacity to accurately perform the task but fail to perform it accurately on a particular occasion due to external factors (such as a distraction).
Example: Gina and Tommaso were asked to calculate the square of four by their teacher. Gina did not answer at all because she did not know that the square of a number is the number multiplied by itself. Gina lacked the competence to answer the question correctly. Tommaso answered ‘twelve’. He knew how to make the square of a number and he was familiar with the four times table, but he was distracted by a sudden noise outside the classroom and gave the wrong answer. Tommaso made a performance error.
When Cohen says that ‘ordinary human reasoning […] cannot be held to be faultily programmed’, he means that human reasoning competence is intact (human agents can apply the rules of good reasoning in ideal conditions), although reasoning performance may be imperfect (human agents may make errors in applying the rules of good reasoning due to external factors).
The standard picture of rationality is accepted by many, perhaps most, philosophers, and reasonably so. After all, there is a key sense of ‘rational’ that describes people who follow the rules of logic, probability, and decision-making when they reason and solve problems. There are other senses of ‘rational’, of course, such as those that describe people who are not overwhelmed by their emotions when they make decisions,