British Academy: The UK's National Academy for the Humanities and Social Sciences
Enquiry, Evidence and Facts: An Interdisciplinary Conference
Common Sense Generalizations
Professor William Twining
University College London, Gower Street
London WC1E 6BT
An abstract presented to the conference
‘Enquiry, Evidence and Facts: An Interdisciplinary conference’
at the British Academy, London, on 13 December 2007
Biography
William Twining is Quain Professor Emeritus of Jurisprudence at University College London. He is the author of Rethinking Evidence (Cambridge UP 2006) and (with Anderson and Schum) Analysis of Evidence (Cambridge UP, 2005).
Abstract
This paper explores the thesis that generalisations are “necessary but dangerous” in inferential reasoning from evidence about particular past events. Generalisations are said to be the “glue” that warrants every inferential step from particular to particular (Schum 1994). Such generalisations are of different kinds, but most kinds are open to a series of challenges. For example, a single background generalisation - once articulated - may be vague, ambiguous, over-generalised, value laden, speculative, biased, prejudiced, irrelevant, and so on. But there are other hidden “dangers” relating to identity (which generalisation?), source, and form. Some of these are illuminated by considering the relationship of generalisations to stories in the context of decision-making.
Psychological research suggests that stories play a major role in decision-making about particular past events (e.g Hastie et al. 1983, 1993). But stories are also open to a range of challenges. “Stories” and “generalisations” are conceptually distinct, but they are intimately related in theory as well as in practice. Generality and particularity are relative matters, as is illustrated by such concepts as “type fact situations”, “story-types”, the “moral” of a parable, the point of a story, or the “holding” or “ratio decidendi” of a legal precedent. Commonsense generalizations and culturally shared stories are both said to be commonly derived from the same source - a community’s or culture’s allegedly shared “stock of knowledge” or “stock of beliefs”. They also are “dangerous” in different, but related ways. The relationship and some of the dangers are illustrated in this paper by reference to criminal investigation and the story of Cinderella.
Two chestnuts
(a) “If you want to win a prince, lose a shoe”. What other morals can be derived from standard versions of the Cinderella story?
(b) The following quotation is from an old training manual for the police entitled The Signs of Crime (Powis, 1977):
“Watch for
- Persons, especially those in the suspect age group, who appear unfamiliar in any way with the mechanism or controls of the vehicle they are using. Switching lights on in daylight or windscreen wipers on in dry weather, without realizing the switch has been used, is very suspicious indeed. In some vehicles inexpert shorting of the ignition to start the vehicle without a key can cause the windscreen wipers to operate and, more important refuse to stop! In dry weather, windscreen wipers operating for more than a few seconds are always worth prompt investigation.” (Powis. at p. 6).
On a fine day a police constable sees a red car 50 metres away moving slowly with its windscreen wipers operating. Is this sufficient on its own to constitute reasonable grounds for suspecting that the car has been stolen? What other generalisations suggest questions that the PC might answer merely by observing the car without stopping it? Could a strong case for reasonably suspecting that the car has been stolen be based on these observations? How many generalisations are involved in such a justification? Are all the generalisations of the same kind? From what source(s) are they derived?