Counterfactual reasoning as cognitive fictionalizing

As Nicholas Rescher has pointed out, the use of hypothetical and counterfactual reasoning is a venerable practice,²⁷ and scholars have been actively' interested in explaining the pervasiveness of such reasoning, as well as its structure and operation, since at least the early' twentieth century'. The most ambitious such account probably remains Hans Vaihinger’s The Philosophy of “As If” (1911).²⁸ Over the past several decades, academic interest in this phenomenon has spread outside philosophy to influence and even transform some areas of inquiry', including some closely related to law.²⁹ Full exploration of these scholarly developments is impossible in the space available here. This chapter will just outline a few major themes of this work and explore some of the senses in which its subject is a kind of “fictionalizing.”

Hypothetical or counterfactual propositions are propositions that we know are possibly' (in the case of hypotheticals) or certainly (in the case of counter-factuals) not consistent with states of affairs in the actual world. They' are thus not factual propositions. However, they' can help us understand states of affairs in the actual world; they help us understand facts. Such propositions are necessary components of the scientific method.³⁰ The generation of scientific hypotheses for testing involves the imagination and, usually, the communication of non-actual states of affairs before they can have been factually tested. This practice, too, thus involves fictionalizing in a sense, yet most would agree that it aims at truth determination, not away from it. As shown above, criticism of a course of action also depends on at least the implication of such possibilities, if not their explicit statement. In addition, the positing of counterfactual sequences is now widely considered to be a necessary' part of causal explanation.³¹

In the mid-twentieth century', philosophers seeking an understanding of the general structure of such reasoning began describing it as involving the postulation of alternative “possible worlds.”³² The American philosophers Robert Stalnaker and David Lewis started to explore this way of analyzing counterfactual statements in the late 1960s and early 1970s, focusing on the inferences implicit in different kinds of “if-then” statements, such as “If you drop the glass, it will break” or “If I had been born in France, I would speak French better than I actually do.” Statements taking these forms had been puzzling to philosophers because of their peculiar truth status. We intuitively evaluate some statements of this kind as true. But in the case of counterfactuals like “If I had been born in France ...” we know that the antecedent, or premise, is false. (I was born in the United States.) According to traditional logic, since the premise is false, it would seem that the conclusion inferred from it, and thus the entire proposition communicated by' the sentence, must also be false. As Stalnaker put it, “A conditional sentence expresses a proposition which is a function of two other propositions, yet not one which is a truth-function of those propositions.”³³ To explain our contrary’ intuition, Stalnaker and Lewis followed earlier philosophers in suggesting that we think of conditional statements as referring not to facts about the actual world, but to possible facts, or facts about a possible world. This suggestion does not completely' solve the puzzle, however, since it is not clear how we could check whether references to “possible” things are referring properly and therefore whether the statements containing them are communicating true propositions.

Counterfactual statements share many' of their peculiarities with statements referring to nonexistent or fictional entities, like unicorns or Sherlock Holmes. Most would agree that Sherlock Holmes does not exist in the actual world, but also that some statements about him seem intuitively true, such as the statement that he lived in London. In the context of already written-about fictional characters, in contrast to counterfactual conditionals, there is an easy way to check our intuitions about truth; we can check the truth of at least some statements about Holmes by consulting the fictional narratives describing him.³⁴ Such narratives also provide a model for thinking about the possible worlds that operate in the counterfactual-conditional context. They' help justify' the use of the label “fiction” to refer to these other, counterfactual and hypothetical, possible worlds; we engage with and reason from such unwritten-about possibilities in ways resembling our engagement with narrative fiction. Much of the psychological research on counterfactual reasoning depends on this analogy'; it studies the cognitive processing of hypotheticals by' asking experimental subjects to read miniature fictional narratives, then draws more general conclusions about the cognitive handling of possible (unwritten-about) states of affairs from the subjects’ behavior and self-reports.³⁵

This understanding of the relationships between fiction and fact challenges the familiar fact-fiction dichotomy relied on by McCulloch and many of his critics. From the perspective described here, the key distinction is not that between fact and fiction, but that between the actual world and possible worlds. In assessing the propriety or truth of a statement, the important question to ask is not whether it is a factual or fictional statement, but whether it is referring to the actual world or to an appropriate or inappropriate possible world. An appropriate possible world would, perhaps, be one whose relationship to the actual world is such that multiple people can agree on the truth of statements about that possible world. Although such a possible world is imaginary, it can in this way also be a basis for objective assessments. A statement or a story may thus refer to a possible world and yet be objective and legitimate. As the next section will explain, lawyers routinely depend on this possibility.