A Theory of Causation in the Social and Biological Sciences

A. Reutlinger

• Publisher: Palgrave Macmillan
• Year: 2013
• Print ISBN: 9781137281036
• eBook ISBN: 9781137281043
• Topic: Philosophy
• Subtopic: Social Philosophy

This first full length treatment of interventionist theories of causation in the social sciences, the biological sciences and other higher-level sciences the presents original counter arguments to recent trends in the debate and serves as useful introduction to the subject.

Part I

Causation in the Special Sciences and the Interventionist Theory of Causation

1

Causation in the Special Sciences

The aim of this book is to provide a theory of causation in the special sciences (that is, a theory of causation in the social sciences, the biological sciences and other higher-level sciences). I attempt to reach this goal by pursuing a negative target and a positive target: the main negative target is to argue against a currently influential theory of causation, the interventionist theory of causation. The book will focus mainly on the interventionist theory developed in James Woodward’s (2003) book Making Things Happen. Counter-arguments against the interventionist theory will attempt to show that the central concept of the interventionist theory of causation – that is, the concept of a possible intervention – is immensely problematic. For this reason, I will argue that the interventionist theory of causation is not tenable. The main positive target of the book consists in replacing the interventionist approach with my own explication of causation in the special sciences, the comparative variability theory of causation.

In understanding the project, it is crucial to clarify what precisely is meant by ‘providing a theory of causation’. In current philosophy of science, there are several different philosophical projects – and corresponding questions and tasks – called ‘theory of causation’:

• first, the semantic project focuses on the truth conditions of causal statements and the analysis of causal concepts (e.g. Lewis 1973b; Woodward 2003);
• second, the metaphysical project is concerned with locating the truth-makers of causal statements in the (physical) world (e.g. Dowe 2000; the contributions in Price and Corry 2007);
• third, the methodological project consists in a rational reconstruction of those methods for scientific testing of causal statements that are used in the special sciences (for influential works on methodology, cf. Cartwright 1989, 1999, 2007; Reiss 2008; Russo 2009; for causal modelling approaches to causal methodology, cf. Spirtes et al. 2000; Pearl 2000; Williamson 2005).

It is common to distinguish these three projects and it has been emphasized by influential voices in the debate on causation how important it is to be aware of the ambiguity between talking of a ‘theory of causation’ and asking the corresponding question ‘What is causation?’ In his classic book The Cement of the Universe, Mackie famously draws the distinction between these three philosophical investigations of causation as follows:

My treatment is based on distinctions between three kinds of analysis, factual, conceptual, and epistemic. It is one thing to ask what causation is ‘in the objects’, as a feature of the world that is wholly objective and independent of our thoughts, another to ask what concept (or concepts) of causation we have, and yet another to ask what causation is in the objects so far as we know it and how we know what we know about it. (1980: viii–ix, emphasis added, similarly pp. 1–2)

More recently, Phil Dowe distinguishes a semantic and a metaphysical project:

As it is the case with many philosophical questions, our question ‘What is causation?’ is ambiguous, and consequently the philosophy of causation legitimately involves at least two distinct tasks. […] We begin by considering these two approaches to the task of philosophy. The first is conceptual – to illucidate our normal concept of causation. The second is empirical – to discover what causation is in the objective world. (2000: 1)

I agree with Mackie and Dowe that the question of what causation is needs to be disambiguated by clarifying whether one pursues the conceptual, the metaphysical, or the methodological project with respect to causation. I assume that these projects neither exclude each other nor are they entirely independent; rather, they should be regarded as being complementary modes of philosophical research. However, in this book attention will be restricted to the semantic project, with the exception, in Chapter 6, where a metaphysical question about causation in the special sciences is addressed. The reason for addressing a metaphysical question is that interventionists claim that their proposed truth conditions for causal statements, and especially their key concept of an intervention, does substantial work for describing the nature of causation.

The method of conceptual analysis has often been associated with clarifying the meaning of everyday discourse and testing the result of these analyses by confronting them primarily (or even exclusively) with our common-sense intuitions. Yet, the conceptual, or semantic, project with respect to causation is not restricted to everyday concepts: especially in philosophy of science, one wants to know how causal notions are used in scientific contexts by scientists. In philosophy of science, analysing the concept of causation and providing truth conditions for causal statements is best reconstructed as an explication in the sense introduced by Rudolf Carnap (Carnap 1950; also Quine 1960: s. 53). In his Logical Foundations of Probability, Carnap characterizes an explication as follows:

The task of an explication consists in transforming a given more or less inexact concept into an exact one or, rather, in replacing the first for the second. We shall call the given concept (or the term used for it) the explicandum, and the exact concept proposed to take the place of the first (or the term proposed for it) the explicatum (1950: 3, original emphasis).

According to Carnap (1950: 7), an explication of, for instance, causation is adequate if and only if (iff) it conforms to the following criteria to a satisfying degree:

1. The explicatum has to apply to paradigm cases of the explicandum;
2. The explicatum has to be stated in an ‘exact form’; that is, in a ‘well-connected system of scientific concepts’ (which include concepts such as laws of nature);
3. The explicatum has to be (potentially) fruitful for empirical scientific research;
4. The explicatum should be as simple as possible; that is, it should be as simple as the more important requirements 1, 2, and 3 permit.

The goal in this book is to provide an explication of causation that is adequate for the special sciences. In particular, I want to test and challenge the adequacy of the currently very influential and widely received interventionist theories of causation (Pearl 2000; Hitchcock 2001; Woodward 2003) as an explication of causation in the special sciences. Focus will be mainly on James Woodward’s interventionist theory of causation, because it is the most influential philosophical account of causation in an interventionist vein. To present only a few paradigmatic and well-known examples of the success and fruitfulness of interventionist theories of causation in other areas of philosophy, the interventionist theory of causation has been applied:

• in the debate on mechanistic explanation: Glennan (2002), Craver (2007), Weber (2008);
• in philosophy of biology: Waters (2007);
• in the debate on mental causation (and higher-level causation, in general): Campbell (2007), Shapiro and Sober (2007), and Shapiro (forthcoming);
• in the debate on metaphysics of causation: Eagle (2007), Menzies (2007); and
• in the debate on laws and explanation in the special sciences: Woodward and Hitchcock (2003), and Leuridan (2010).

Moreover, as will be shown in detail, interventionist ideas are popular among social scientists and among philosophers of social science. It is presumably no exaggeration to say that interventionist theories of causation are widely used in philosophy of science. One might even be tempted to call interventionism the new orthodoxy in this area.

So, what do interventionists claim with respect to the truth conditions of causal statements? Interventionists hold, roughly, that X is a cause of Y iff there is a possible intervention on X that changes Y. An intervention is a manipulation of the cause – and only of the cause – which is supposed, in principle, to be possible; that is, an intervention is not restricted to the capacities of human agents to intervene. It should be emphasized once again that the goal of this book is, primarily, to discuss interventionist theories and, ultimately, to replace them with a theory (the comparative variability theory) that, it will be argued, is a better candidate for the job. For these purposes, the examples of causal statements are restricted to paradigmatic cases in the social sciences. This restriction is motivated by two reasons:

• interventionists often use examples from the social sciences, especially from economics, to show the adequacy of their theories; and
• social scientists themselves tend to adopt interventionist theories of causation.

In other words, if one takes interventionism to be an explication of causal concepts that are used in paradigmatic causal statements in the social sciences, one makes a maximally strong and fair case for interventionist theories of causation. So, let us begin with a list of paradigmatic causal statements in the social sciences:

Statement 1:	Caldwell’s model of child survival in developing countries formulates the causal claim that a mother’s education C1 (measured in number of schooling years) and a father’s socio-economic status C2 (measured by income of the fathers) are causes of child mortality E (cf. Russo 2009: 26–30);
Statement 2:	Low GNP per capita causes a high rate of infant mortality (Little 1998, 2000, cited in Bartelborth 2007: 136);
Statement 3:	An increase in supply of a commodity – while the demand for it stays the same – causes a decrease in the price (Cartwright 1989: 149f; Kincaid 2004: 177);
Statement 4:	‘In 1973 OPEC drastically reduced the amount of oil that it supplied to the world, and in short order there were long lines at the gas pumps in the USA and a much higher price for gasoline. When the gasoline price rose, sales of large gas-guzzling automobiles declined. At the same time, exploration of new oil sources increased and eventually the known world oil reserves remained steady, despite earlier widespread predictions that oil reserves would soon be depleted’ (Kincaid 2004: 168);
Statement 5:	The growth of money in an economy has the capacity to raise the general level of prices (Reiss 2008: 266f.);
Statement 6:	‘Changes in real income cause changes in the money stock’ (Hoover 2001: 46);
Statement 7:	The Iraqi invasion of Kuwait caused the 1990/91 recession in the United States (cf. Hoover 2001: 1);
Statement 8:	Low social status causes poor health conditions such that the higher one’s social status, the better one’s health (cf. Cartwright 2009: 411);
Statement 9:	An increase in the inflation rate causes a decrease in the unemployment rate (cf. Reiss 2008: 169, 192);
Statement 10:	Commercial hacienda systems tend to agrarian revolt, and plantation systems tend to lead to labour reforms (cf. Kincaid 1996).

Before discussing which criteria of adequacy these paradigmatic causal claims suggest, three disclaimers should be added with respect to these examples of causal statements.

• The book will not be concerned with questions about ontological and explanatory reductionism: in the social sciences and their philosophy, issues of reduction are often discussed under the label ‘methodological individualism’ (cf. Heath 2005, Kincaid 1996, Hedström and Ylikoski 2010). Neither the related micro–macro distinction will be discussed, nor the exact nature of social macro-entities investigated. Similarly, the questions as to whether these macro-entities can be understood to be causally related at all and whether there is ‘downward causat...

Table of contents

• Cover
• Title
• Copyright
• Contents
• List of Figures
• Acknowledgements
• Part I Causation in The Special Sciences and The Interventionist Theory of Causation
• Part II What is Wrong with Interventionist Theories
• Part III An Alternative Theory of Causation in Special Sciences
• Notes
• Bibliography
• Index