2 The ‘What is…?’ question

2.1 Philosophy and science

We will consider some different attempts to answer the question ‘What is an emotion?’. Because we shall often need to refer to this question in what follows, I shall call it the ‘What is…?’ question. Before we investigate some of the ways in which philosophers have attempted to answer it, we should consider what an answer might look like.

What might a scientific answer to the ‘What is…?’ question tell us about emotion, for example, those offered by neurophysiologists and psychologists? A neurophysiologist might answer the ‘What is…?’ question by explaining how emotional occurrences are realised in patterns of activity in the brain and nervous system. A cognitive psychologist might investigate how emotions influence the way in which we think and act. These scientific accounts can be tested by experiment; in addition, they can be assessed in the light of established theories about how the brain works, and about how people think and behave. How does a philosophical answer to the ‘What is…?’ question relate to these scientific accounts of the nature of emotion?

A standard way to answer this question is to say that, while science is concerned to describe and explain the way in which the world actually works, the role of philosophy is to take a step back – to analyse and to criticise the assumptions that underlie our everyday beliefs and our scientific theories. Sometimes it is suggested that the primary concern of philosophy is to understand the concepts that we use in describing and explaining the world: to give an account of our concept emotion is to explain what counts as a case of emotion, for example, by identifying some feature that all cases of emotion have in common. But we might also expect a philosophical account of emotion to tell us something about the broader body of beliefs and assumptions that underlies our everyday talk about emotions – as I attempted to do in section 1. Not all these beliefs and assumptions are concerned with what an emotion is or bear on our concept of emotion. Nevertheless, they help to determine how we think and talk about emotions. Some philosophers refer to this broader set of beliefs and assumptions as our ‘conception’ of emotion (Johnston 1987, p. 60–1). In what follows, I shall use the term ‘conceptual question’ rather loosely to refer both to questions about concepts and to questions about conceptions.

What kinds of question might a philosopher ask about our conception of emotion? First, they might set out to provide a descriptive account; that is, they might set out to describe what we ordinarily conceive an emotion to be. An account of this kind should enable us to recognise the kinds of assumption that we are making when we describe someone as experiencing an emotion, and to understand how these assumptions fit together with other very fundamental beliefs that we tend to share – for example, beliefs about the nature of rationality or freedom.

Once these assumptions have been made explicit, it is possible to criticise them. We might find that our conception of emotion is rather untidy – smudging distinctions that it would be helpful to draw. More radically, we might find that it is nonsensical, or that it fails to cohere with other fundamental assumptions that we make about ourselves or the world. In this situation, a philosopher might take a step further, and offer a revisionary account – suggesting ways in which we should modify our conception of emotion, in order to highlight certain distinctions or to eliminate any inconsistencies.

This is not the only basis on which a philosopher might develop a revisionary account. A philosophical account of the emotions might set out the conception of emotion that we ought to adopt if we are to conceive of emotions in the way empirical investigation shows them to be. An account of this kind will answer not only to conceptual considerations but also to the evidence of everyday experience and current scientific theory. It may lead us to the conclusion that we need to revise some fundamental assumptions about the nature of the emotions. Indeed, in extreme cases, such an account may imply that we ought to jettison our concept of emotion altogether.

What this shows is that the answer to a conceptual question can sometimes depend on the answer to an empirical question. But the relationship between empirical and conceptual questions can also run in the opposite direction. I can best explain this with an example. Suppose that a documentary programme is advertised as follows. ‘In the middle ages, many people suffered from demonic possession. Now Professor X explains what demonic possession is!’ Intrigued, you make time for the documentary. On the programme, Professor X explains that people who were thought in the middle ages to be possessed by demons were in fact suffering from a form of schizophrenia. Perhaps you are satisfied by the explanation. Even so, you may feel that the advertisement for the programme rather overstated the case. Professor X did not explain what demonic possession is. Rather, Professor X has explained something else, that is, why certain people were once thought to be possessed by demons. Indeed, rather than saying what demonic possession is, the professor's explanation implies that there is no such thing as demonic possession.

But why should we resist the claim that Professor X has shown that demonic possession just is a form of schizophrenia? The problem is that our concept of demonic possession rests on certain assumptions – for example, the assumption that there are malevolent beings that are capable of controlling our actions. This is not to say that our concept could not be modified a little; but there are limits to the extent to which a concept can be modified before it becomes unrecognisable. At the very least, it might be thought that the concept of demonic possession implies control by an external agent. If so, schizophrenia is just not the right sort of thing to be demonic possession.

In the same way, if someone offers an answer to the ‘what is…’ question, we can ask whether they have offered an account of what emotion is, or whether they have given us an account of something else – perhaps even an account that implies that there is no such thing as emotion. Once again, this need not imply that our concept of emotion cannot be revised a little. But to count as an answer to the ‘what is…’ question, an account must characterise emotion in a way that is at least recognisable, given our everyday concept.

If this is right, conceptual questions and empirical questions are not always independent of each other. Nevertheless, they are different kinds of question. In what follows, I shall assume that, broadly speaking, philosophical inquiry aims to answer conceptual questions, relying primarily on reflection and argument; while scientific inquiry aims to answer empirical questions, and does so by appealing to experimental evidence.

However, this distinction between philosophy and science needs to be softened in two ways. First, it should not be taken to imply that philosophy and science are two quite independent enterprises, which can proceed in isolation from each other. As we have seen, scientific theory provides one way to test the adequacy of our unreflective conception of emotion. Conversely, philosophical reflection on our concepts is not always an end in itself, but can be used as a tool, helping to provide a framework in which scientific investigation can take place.

Secondly, in actual practice, these boundaries are not treated as sacrosanct. Sometimes, the earliest attempts to propose an empirical account of a phenomenon are very general in nature, and try to make progress by piecing together clues from everyday experience, as well as what little experimental work has been done. In this situation, it is not uncommon for a philosopher to suggest a speculative empirical account. Conversely, scientists often raise and discuss conceptual questions as they develop their theories.

As we shall see, not all the theorists that we will encounter characterise their projects in quite the same way. We will need to bear this in mind while investigating their accounts, and especially when we assess the objections that they make to each other.