Turning a graphical method of evidential reasoning into an operational tool for judges? Empirical evidence
| Published date | 01 April 2022 |
| DOI | 10.1177/13657127221076172 |
| Date | 01 April 2022 |
| Subject Matter | Articles |
Turning a graphical method of
evidential reasoning into an
operational tool for judges?
Empirical evidence
Olivier Leclerc
CNRS, Centre de théorie et analyse du droit (CTAD), Université Paris Nanterre, France
Etienne Vergès
Université Grenoble Alpes, France
Géraldine Vial
Université Grenoble Alpes, France
Abstract
Research on graphical methods of reasoning has made enormous progress since the pioneering
workofWigmoreintheearly20thcenturyanditslater rediscovery in the 1980s. While the use-
fulness of graphical methods for student training and research is widely acknowledged, their use by
judges remains marginal, if not non-existent, even though this was Wigmore’s objective. This art-
icle explores the difficulties that graphical methods of reasoning must overcome if they are to be
integrated into the practice of the courts, at a time when courts are faced with ever more pressing
imperatives of efficiency. The research is based on a partnership with the French School of
Magistrates (Ecole Nationale de la Magistrature) and is informed by training courses given to magis-
trates on the basis of real cases, during which the authors proposed that they implement what we
have called the Orderly Method of Evidence Analysis. Although the research confirms the value of
graphical methods in promoting rigour in evidential reasoning, it also reaffirms the already clearly
identified limits related to their complexity and time-consuming nature. The article also points out
the difficulties that still need to be overcome in order to operationalise graphical methods of evi-
dential reasoning, and the difficulties encountered by these methods in avoiding judgment bias.
Keywords
chart method, diagramming schemes, evidential reasoning, French law of evidence,
J. H. Wigmore
Corresponding author:
Olivier Leclerc, CTAD UMR 7074, Université Paris Nanterre, 200 avenue de la République, 92001 Nanterre Cedex, France.
Email: olivier.leclerc@cnrs.fr
Article
The International Journal of
Evidence & Proof
2022, Vol. 26(2) 136–156
© The Author(s) 2022
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/13657127 221076172
journals.sagepub.com/home/epj
Introduction
Wigmore’s chart method is the first attempt to link a method of reasoning with masses of mixed evidence
to a theory of evidence. Formulated in 1913, it reflects Wigmore’s ambition to contribute to the develop-
ment of a science of proof which would bring the analysis of evidence out of a pre-scientific phase in
which it was not systematised but left to professional know-how. In order to ‘rationally determine the
net persuasive effect of a mixed mass of evidence’(Wigmore, 1913: 747), judges and lawyers need, he
asserted, a tool for the rational analysis of evidence. To this end, Wigmore proposed to list the available
evidential facts (Evidence List) and to identifythe subfinal facts, from which it would be possible to decide
whether or not the ultimate fact (or probandum) is proven. For this purpose, Wigmore drew up a graphic
representation of the evidentialfacts derived from testimonial and circumstantial evidence, the provisional
credit given to them by the person making the chart, the inferencesdrawn from each piece of evidence, the
conviction attached to them, and theirprobative effect. By means of a rich symbology, these charts ration-
ally order a large amount of evidence that the charter’s mind could not immediately grasp as a whole. The
method does not command what conclusion is to be reached, but should help to represent clearly the rea-
soning that is followed. Ambitious and innovative in its time, Wigmore’s aspiration to forge a novum
organum for the study of evidence in court contrasts sharply with William Twining’s sweeping observa-
tion that Wigmore’s proposal was merely a ‘lead balloon’(Twining, 1985: 164). This observation is
widely shared by scholars of evidence law. It is based on the fact that Wigmore’s chart method was
barely taken up by scholars of evidence law and has been taught only marginally in universities, even
during Wigmore’s lifetime. Worse still, there is no indication that the judges for whom the method was
intended ever relied on it to improve their professional practice.
However, Wigmore’s chart method has never ceased to attract interest, extending far beyond specia-
lists in the law of evidence. In the 1980s the intelligence community saw it as an effective tool for ana-
lysing complex situations, where the analyst must take into account and put into order a large number of
elements of assessment (Schum, 1987). Anderson and Twining in turn reinvested in Wigmore’s chart
method and became ‘the most active evangelists’(Vignaux and Robertson, 1992: 94). Anderson and
Twining, joined by Schum, extensively recast the method, simplifying its presentation and the
symbols used (Anderson and Twining, 1991; Anderson et al., 2005).
Wigmore’s chart method has also been explored by scholars of argumentation theory who have sought to
describe judicial reasoning by a diagramming scheme (Tillers, 2007; Walton, 2005; Walton et al., 2008).
Wigmore from this perspective appeared as a lesser known precursor of argument diagramming (Rowe
and Reed, 2006; Vignaux and Robertson, 1992: 96), relevant for analysing the particular context of legal
evidence where argumentation is conducted between two actors (lawyers, judges) who respond to each
other. Thus the description of the pattern of an argument proposed by Toulmin (2003 [1958]: 92 seq.)
has much in common with the fundamental operations identified by Wigmore and with the basic principles
of his graphical formalisation. Both symbolise by an arrow the link between a claim and the data that sup-
ports it (Nance, 2007). Similarly, Toulmin’s warrants and rebuttals echo Wigmore’s distinction between the
facts that can corroborate or explain away an inference. And although Wigmore was obviously unable to
imagine such a development, since it would come much later, his pioneering contribution to argument dia-
gramming and assessment has also been debated by informal logic and argumentation diagramming scholars.
Goodwin (2000) argued that Wigmore’s rhetorical orientation to argument diagramming ameliorates some of
the difficulties of evaluating arguments in tree diagrams. Wigmore’s graphical method has not been simply
repeated, but has been the subject of multiple reappropriations, simplifications and additions, so as to include
temporal analysis (Tillers and Schum, 1988), contribute to defeasible logic (Bex et al., 2003; Verheij, 2000)
and estimate the weight of the ultimate probandum (Chalamish et al., 2011). But Wigmore’s proposal has
also been progressively challenged by the development of probabilistic networks, whether in influence net-
works or, even more so, Bayesian networks (Dawid et al., 2011; Hepler et al., 2007; Kadane and Schum,
1996; Kjærulff and Madsen, 2008; Vignaux and Robertson, 1992).
Leclerc et al. 137
To continue reading
Request your trial