Skip to main content

Understanding Counterfactuals / Understanding Causation

A two-day workshop
15/16 December 2007
University of Warwick

Saturday, 15th December
9:00 David Mandel: Counterfactual and causal explanation: From early theoretical views to new frontiers
10:30 Coffee
11:00 Dorothy Edgington: Causation first: why causation is prior to counterfactuals
12:30 Lunch
13:30 Eva Rafetseder: Does children’s counterfactual thinking stick to the nearest possible world?
15:00 Christopher Hitchcock: Counterfactual Availability and Causal Judgment
16:30 Coffee
17:00 Sarah Beck: What needs to develop for children to relate counterfactual worlds to reality?


Sunday, 16th December
9:00 David Sobel: The importance of causal knowledge for children’s hypotheses, fantasy, and counterfactuals
10:30 Coffee
11:00 Teresa McCormack, Stephen Butterfill & Christoph Hoerl: Young children's causal and counterfactual judgments
12:30 Lunch
13:30 Johannes Roessler: Interventionism, agency, and causal understanding
15:00 Peter Menzies: The Role of Default Assumptions in Counterfactual and Causal Reasoning
16:30 Coffee
17:00 James Woodward: Causal and counterfactual reasoning: An interventionist perspective


Speakers and Abstracts

Sarah Beck (Psychology, Birmingham): ‘What needs to develop for children to relate counterfactual worlds to reality?’

One way to examine the relationship between counterfactual and causal thinking is to consider their development in early childhood. In this paper I will explore the development of counterfactual thinking, arguing that recent evidence indicates that a range of different abilities allow children to think counterfactually. These abilities develop over a relatively long time period ranging from at least 2 to 7 years. I will review empirical evidence from my own and others' studies that suggest that at least three distinct abilities must develop: ignoring the actual world to consider an alternative (Beck, Riggs, & Gorniak, under submission), holding multiple possible worlds in mind (Beck, Robinson, Carroll, & Apperly, 2006), and making comparisons between these possibilities (see e.g. Guttentag & Ferrell, 2004). I will explore whether domain general developments, particularly in executive function, might drive these developments. My aim is a more precise account of the development of children's counterfactual thinking which can then be related to developments in their causal understanding.

Dorothy Edgington (Philosophy, Oxford): ‘Causation first: why causation is prior to counterfactuals’

Christopher Hitchcock (Philosophy, Caltech): ‘Counterfactual Availability and Causal Judgment’

Although there are problems with efforts to analyze causation in terms of counterfactuals, mainly stemming from preemption, it is clear that causation and counterfactuals are closely related. In particular, a counterfactual test is a very good prima facie test for causation. While philosophers have primarily been interested in questions concerning the semantics of counterfactuals – which counterfactuals are true, and what are the logical relations among counterfactuals and other claims – psychologists have been interested in the question of which counterfactual possibilities we actually entertain. This is similar to the competence/performance distinction familiar from linguistics. Psychologists have enumerated a number of factors that render certain counterfactual possibilities readily ‘available’. I will explore the various ways in which differences in the availability of different counterfactual possibilities may affect our causal judgments.

David Mandel (Defence Research and Development Canada, Toronto): ‘Counterfactual and causal explanation: From early theoretical views to new frontiers’

I trace the evolution of thinking in psychology about the relationship between counterfactual and causal explanation. Early work emphasized the “cue-like” role of counterfactual thinking in determining “the cause” of a single-case outcome. Later work, however, highlighted a number of empirical and theoretical challenges to the early view. Notably, demonstrations that the contents of counterfactual and causal explanations of single-case outcomes predictably diverged in content led to a subsequent round of effort geared toward theoretical reformation. I examine the two main lines along which this reformation took place and address some of the challenges that they, in turn, have faced.

Peter Menzies (Philosophy, Macquarie): ‘The Role of Default Assumptions in Counterfactual and Causal Reasoning’

The traditional semantics employed in philosophical discussions of counterfactuals typically evaluate counterfactuals in terms of a similarity relation between possible worlds. The truth conditions for a counterfactual are stated in terms of whether the counterfactual's antecent and consequent are true in the possible worlds that are most similar to the actual world. This is sometimes expressed by saying that the similarity ordering of the possible worlds is centered on the actual world. This paper will question this centring assumption by arguing that it ignores or misconstrues the role of default assumptions in counterfactual reasoning. Philosophical arguments as well as experimental evidence from cognitive psychology will be presented to demonstrate the crucial role of default assumptions in counterfactual reasoning. The paper will also explore the implications of this feature of counterfactuals for theories of causation, such as interventionist theories, that make crucial use of counterfactuals.

Eva Rafetseder (Psychology, Salzburg): ‘Does children’s counterfactual thinking stick to the nearest possible world?’

Previous research suggests that children’s ability to entertain counterfactual speculations undergoes substantial development (Harris et al., 1996; Riggs et al. 1998; Amsel & Smalley, 2003). The various tasks in this research differ in their demand to relate counterfactual speculations to reality (akin to David Lewis’ criterion of “nearest possible world). We call reasoning that ignores this criterion, “hypothetical reasoning”, reasoning that does obey it, “counterfactual reasoning”. To check counterfactual reasoning abilities we designed tasks in which hypothetical default assumptions, e.g. based on typical regularities of the world, lead to different answers to a subjunctive question than counterfactual reasoning. For instance, a doctor sitting in the park to read his paper in the evening is called to an emergency at the swimming pool. The subjunctive test question, “if there had been no emergency, would the doctor be in the hospital or in the park?” should counterfactually be answered “in the park”. But ignoring the doctor’s intentions in the real world and just reasoning on hypothetical grounds one might answer: “in the hospital”, since doctors tend to be in the hospital and not in the park. Indeed, only by six years, did children give mostly correct answers. Much later than some of the existing data would suggest. But that literature did not control for correct answers based on hypothetical reasoning.

Johannes Roessler (Philosophy, Warwick): ‘Interventionism, agency, and causal understanding’

David Sobel (Cognitive and Linguistic Sciences, Brown): ‘The importance of causal knowledge for children’s hypotheses, fantasy, and counterfactuals’

I will present 3 lines of experiments, highlighting the importance of children's causal knowledge when forming hypotheses about possible causal relations, when making judgments about whether events are fantasy or reality (and how fiction is constructed in general), and when making inferences about counterfactual events. For each line of experiments, I will highlight the possibility of domain-general causal inference mechanisms interacting with domain-specific causal knowledge.

James Woodward (Philosophy, Caltech): ‘Causal and counterfactual reasoning: An interventionist perspective’


For further information about this workshop, please contact Christoph Hoerl