Conditionals Workshop

The meaning of conditionals is the topic of ongoing debate among philosophers, linguists and psychologists. Under which conditions are such sentences true? Can they even be true? The present workshop brings together researchers that seek to further the debate by submitting conditionals to empirical investigation.

Date: Friday November 9 2012
Time: 14:00-19:00 (approximately)
Location: Room Omega, Department of Philosophy.

Talks:

David Over, Durham University 
The psychology of uncertain reasoning

Abstract: There is a new Bayesian/probabilistic paradigm in the psychology of reasoning. In the old binary paradigm, inference was from premises taken as assumptions. The participants in experiments were to draw only conclusions that necessarily followed from the assumptions. If they failed to do this, but allowed their relevant and uncertain beliefs to influence their inferences, they committed fallacies or had biases. The new paradigm recognizes that almost all useful inference, in science and everyday affairs, is from uncertain premises, which often express uncertain beliefs. A new probabilistic definition of validity, p-validity, can be given in the new paradigm, and inferences that were seen in the old paradigm to be fallacies can be shown probabilistically strong in some contexts. However, there are questions about the relation between p-validity, basic psychological uncertainty, and standard logic and probability theory. There are also unanswered questions about the three-valued truth tables that people will produce when the components of compound propositions are uncertain. We will raise these questions in this talk and discuss ways of answering them.

Karolina Krzyzanowska, University of Groningen
Inferential conditionals

Abstract: Many indicative conditionals seem to convey the existence of an inferential connection between their antecedent and consequent. If this idea, which can be traced back to antiquity, is unpopular among contemporary philosophers, it is probably because its advocates failed to recognize that this inferential link does not always have to be understood in terms of classical deduction. Acknowledging the traditional distinction between deductive, inductive, and abductive reasoning can be a departure point for revisiting some old philosophical issues concerning the interpretation of conditional sentences. It can help to explain, for instance, why seemingly contradictory conditionals can both be true, or why some of them constitute a bad advice. Furthermore, it may shed new light on the meaning of certain linguistic markers of evidentiality, which are believed to indicate that a speaker bases his assertion on an inference. the aim of this talk is to provide both theoretical and empirical support for the claim that, for a large class of indicative conditionals, a conditional's consequent should be seen as a deductive, inductive or abductive consequence of that conditional's antecedent.

Janneke van Wijnbergen-Huitink, University of Groningen
Restrictions on left-nesting

Abstract: Conditionals may contain component sentences that are itself conditionals. These come in two forms: left-nested conditionals contain a conditional antecedent, and right-nested conditionals embed a conditional in their consequent. Folklore has it that some left-nested conditionals are difficult to make sense of, and this is sometimes thought to motivate the view that conditionals fail to express propositions. However, we also have examples of left-nested conditionals that appear to be perfectly interpretable. The goal of this talk is to come up with a classification of left-nested conditionals that not only gets the cuts in the right places, but that can also be subjected to empirical testing. 

Shira Elqayam, De Montfort University Leicester
Is-to-ought: Inferring deontic conclusion from utility conditionals

Abstract: Humans have a unique ability to generate novel norms. Faced with the knowledge that the globe is warming, we easily and naturally infer that we ought to reduce our carbon footprint. Although a contentious and lively issue in metaethics since Hume, such inference from ‘is’ to ‘ought’ has not been systematically studied in psychology of reasoning. We regard utility conditionals a paradigm case of inference from is to ought, combining agency, action, a(n un) desired goal, and a causal link between action and goal; e.g., ‘If you pull the dog’s tail, it will bite you’. A combination of these parameters invites the pragmatic inference that you must not pull the dog’s tail – a deontic conclusion. We draw on the metaethical notion of bridge constructs, which combine descriptive and evaluative aspects. In essence, people infer norms in the presence of values, and when they can be reasonably confident that the action will result in the desired outcome. When an action is causally linked to a valenced goal, valence transfers to the action and bridges into a deontic conclusion. This pragmatically rich ‘deontic introduction’ is mostly implicit. We report three experiments to support our theory. Participants were presented with utility conditionals in which an action results in a positive, negative, or neutral outcome (If Lisa buys the booklet, she will pass the exam), and asked to evaluate how strongly deontic conclusions (Lisa should buy the booklet) follow from the premises. Findings show that the direction of the conclusions matched outcome valence, whereas conclusion strength was determined by the strength of the causal link between action and outcome. The third experiment shows that deontic introduction is defeasible, and can be suppressed by conflicting goals or conflicting norms.

Igor Douven, University of Groningen
Conditionals and closure

Abstract:  It has long been held that high conditional degree of belief in an indicative conditional's consequent given its antecedent is necessary and sufficient for the acceptability of that conditional. Recently, it has been argued, both on theoretical and on empirical grounds, that the notion of evidential support plays a crucial role in the acceptability of indicative conditionals, next to that of high conditional degree of belief. It is still an open question under which principle acceptability is closed if we adopt this proposal. The present paper makes a beginning with answering that question.
Comments