Generalisations, causal relationships and moral responsibility
| Published date | 01 April 2016 |
| Author | Federico Picinali |
| DOI | 10.1177/1365712716628539 |
| Date | 01 April 2016 |
| Subject Matter | Articles |
Article
Generalisations, causal relationships
and moral responsibility
Federico Picinali
London School of Economics and Political Science, UK
Abstract
In Chapter 4 of his book Character in the Criminal Trial, Mike Redmayne addresses a series of
non-epistemic arguments against the use of bad character evidence. Redmayne is particularly
successful in debunking arguments focusing on the defendant’s autonomy. Inspired by the early
drafts of Chapter 4—which I had the honour to read—I published an article trying to con-
tribute to the debate. At the time of writing, however, I was not aware of a beguiling argument
concerning the use of behavioural generalisations, formulated and defended by Amit Pundik in a
recent research paper. According to this argument, we should not prove guilt using general-
isations that presuppose a causal relationship between an antecedent factor and individual
behaviour, if this relationship negates the freedom of the individual under any theory about
determinism and free will that one may adopt—be it a libertarian or a compatibilist theory. In
this article I critically assess this argument and attempt to show that it does not succeed.
Keywords
admissibility, causation, criminal trials, freedom, generalisations, moral responsibility
Introduction
In Chapter 4 of his book Character in the Criminal Trial, Mike Redmayne addresses a series of non-
epistemic arguments against the use of bad character evidence. Redmayne is particularly successful in
debunking arguments focusing on the defendant’s autonomy. According to these arguments, it is morally
wrong to prove the defendant’s guilt by relying on behavioural generalisations
1
about groups of which
Corresponding author:
Federico Picinali, London School of Economics and Political Science, New Academic Building, Houghton St., London, WC2A 2AE.
Email: f.picinali@lse.ac.uk
1. With the expression ‘behavioural generalisation’ I am referring to generalisations concerning the behaviour of individuals.
More precisely, behaviour must feature in the second term of the generalisation for it to be ‘behavioural’—although it may also
feature in the first term. Consider, for example, a generalisation stating that ‘cyclists in London are very likely to wear helmets’.
The first term is ‘being a London cyclist’. The second term is the behaviour consisting in ‘wearing an helmet’. Notably, the
second term of a generalisation is its focus, being the fact ‘in question’. The first term, instead, represents a ‘starting point’. On
the definition of ‘generalisation’ see Picinali (2012: 199–201).
The International Journalof
Evidence & Proof
2016, Vol. 20(2) 121–135
ªThe Author(s) 2016
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1365712716628539
epj.sagepub.com
To continue reading
Request your trial