Explanations of Counterfactual Inferences

Pacific Philosophical Quarterly, 1991

Siegal, Elitzur and Nora Boneh (eds.), Perspectives on Causation, Springer., 2020

Counterfactual conditionals are used extensively in causal reasoning. This observation has motivated a philosophical tradition that aims to provide a counterfactual analysis of causation. However, such analyses have come under pressure from a proliferation of counterexamples and from evidence that suggests that the truth-conditions of counterfactuals are themselves causal. I offer an alternative account of the role of counterfactuals in causal thought that is consistent with these data: counterfactuals are used in a common method of causal reasoning related to John Stuart Mill's method of difference. The method uses background beliefs about causal relationships, history, and the natural laws to establish a new causal claim. Counterfactuals serve as a convenient tool for stating certain intermediate conclusions in this reasoning procedure, and that is part of what makes counterfactuals useful. This account yields a functional explanation of why our language contains a construction with the truth-conditions of counterfactuals. 3 For an interesting alternative explanation of the connection between causation and counterfactuals, see Maudlin (2004). 4 I do not claim that that is the only function of counterfactual conditionals. They clearly also serve other purposes, e.g. in making practical decisions.

International Journal of Philosophical Studies, 2007

The counterfactual analysis of causation has focused on one particular counterfactual conditional, taking as its starting point the suggestion that C causes E iff (~C □→ ~E). In this paper, some consequences are explored of reversing this counterfactual, and developing an account starting with the idea that C causes E iff (~E □→ ~C). This suggestion is discussed in relation to the problem of preemption. It is found that the “reversed” counterfactual analysis can handle even the most difficult cases of preemption with only minimal complications. The paper closes with a discussion of the wider philosophical implications of developing a reversed counterfactual analysis, especially concerning the differentiation of causes from causal conditions, causation by absences, and the extent to which causes suffice for their effects.

Among the many philosophers who hold that causal facts 1 are to be explained in terms of-or more ambitiously, shown to reduce to-facts about what happens, together with facts about the fundamental laws that govern what happens, the clear favorite is an approach that sees counterfactual dependence as the key to such explanation or reduction. The paradigm examples of causation, so advocates of this approach tell us, are examples in which events c and e-the cause and its effect-both occur, but: had c not occurred, e would not have occurred either. From this starting point ideas proliferate in a vast profusion. But the remarkable disparity among these ideas should not obscure their common foundation. Neither should the diversity of opinion about the prospects for a philosophical analysis of causation obscure their importance. For even those philosophers who see these prospects as dim-perhaps because they suffer post-Quinean queasiness at the thought of any analysis of any concept of interest-can often be heard to say such things as that causal relations among events are somehow "a matter of" the patterns of counterfactual dependence to be found in them.

Cornell University - arXiv, 2022

Counterfactual reasoning-envisioning hypothetical scenarios, or possible worlds, where some circumstances are different from what (f)actually occurred (counter-to-fact)-is ubiquitous in human cognition. Conventionally, counterfactually-altered circumstances have been treated as "small miracles" that locally violate the laws of nature while sharing the same initial conditions. In Pearl's structural causal model (SCM) framework this is made mathematically rigorous via interventions that modify the causal laws while the values of exogenous variables are shared. In recent years, however, this purely interventionist account of counterfactuals has increasingly come under scrutiny from both philosophers and psychologists. Instead, they suggest a backtracking account of counterfactuals, according to which the causal laws remain unchanged in the counterfactual world; differences to the factual world are instead "backtracked" to altered initial conditions (exogenous variables). In the present work, we explore and formalise this alternative mode of counterfactual reasoning within the SCM framework. Despite ample evidence that humans backtrack, the present work constitutes, to the best of our knowledge, the first general account and algorithmisation of backtracking counterfactuals. We discuss our backtracking semantics in the context of related literature and draw connections to recent developments in explainable artificial intelligence (XAI).

Cognitive Science, 2010

Bayes nets are formal representations of causal systems that many psychologists have claimed as plausible mental representations. One purported advantage of Bayes nets is that they may provide a theory of counterfactual conditionals, such as If Calvin had been at the party, Miriam would have left early. This article compares two proposed Bayes net theories as models of people's understanding of counterfactuals. Experiments 1-3 show that neither theory makes correct predictions about backtracking counterfactuals (in which the event of the if-clause occurs after the event of the then-clause), and Experiment 4 shows the same is true of forward counterfactuals. An amended version of one of the approaches, however, can provide a more accurate account of these data.

Abstracta, 2020

Mind, 1997

On David Lewis's original analysis of causation, c causes e only if c is linked to e by a chain of distinct events such that each event in the chain (counterfactually) depends on the former one. But, this requirement precludes the possibility of late pre-emptive causation, of causation by fragile events, and of indeterministic causation. Lewis proposes three different strategies for accommodating these three kinds of cases, but none of these turn out to be satisfactory. I offer a single analysis of causation that resolves these problems in one go but which respects Lewis's initial insights. One distinctive feature of my account is that it accommodates indeterministic causation without resorting to probabilities.

Artificial Intelligence, 1999

A stratified view of causal reasoning is set forth; one in which the identification of counterfactual dependencies plays an important role in determining what sort of causal connection, if any, exists between two events named by a given pair of partial descriptions. A semantics for temporal counterfactuals in which events are represented at the object level is then formalized based on a syntactic form of belief updating. Counterfactuals are evaluated relative to an agent's information state, taken to include a set of initial beliefs together with additional assumptions to handle the frame problem. Inertial inferences emerge as a side-effect of requiring minimal information change between states of the world in some chronicle. A chronicle is, in addition, assumed minimal with respect to an explanatory preference that minimizes the set of beliefs that are not part of an agent's initial set of beliefs or are not supported by some body of law-like knowledge. A number of epistemic preferences that underly the choice of alternative worlds in accommodating a counterfactual supposition are then examined ranging over types of knowledge, locality of action, and time. This leads to a semantics for causation.