Some Reflections on the Proof–based Theory of Legal Exceptions
| Author | Alessio Sardo |
| Published date | 01 March 2017 |
| DOI | http://doi.org/10.1111/1468-2230.12258 |
| Date | 01 March 2017 |
bs_bs_banner
REVIEW ARTICLE
Some Reflections on the Proof–based Theory of Legal
Exceptions
Alessio Sardo∗
Lu´
ıs Duarte d’Almeida,Allowing for Exceptions: A Theory of Defences
and Defeasibility in Law, Oxford: Oxford University Press, 2015, 293 pp,
hb £50.00.
In his 1949 essay, ‘The Ascription of Responsibility and Rights’, H. L. A.
Hart first introduced the concept of defeasibility, which would later become
the object of lively discussion in various domains: epistemology, logic, artificial
intelligence, and meta-ethics. In this essay, Hart tried to character ise the role of
exceptions in legal reasoning. In a nutshell, he considered that the conditions for
the application of a legal concept, such as ‘contract’, ‘although necessary are not
always sufficient’ for justifying a certain legal conclusion.1Our understanding
of the concept of a contract is incomplete if we only take into account the
positive conditions required for a valid contract to come into existence (such
as offer and acceptance): a complete understanding also requires knowledge
of what conditions can defeat the claim that there is a valid contract (such as
misrepresentation or undue influence). Hart extended this claim to non-legal,
ordinary ascriptions of responsibility, praise, and blameworthiness. Of course,
he was not the first author to deal with the problems raised by exceptions and
defences in normative reasoning. If we remain within the analytic tradition, we
can easily discover that, twenty years earlier, Sir W. D. Ross had stressed that
moral theories should allow the possibility of prima facie duties, which could
be defeasible.2If we do not confine ourselves to analytic philosophy, we might
even be willing to ascribe this thesis to Aristotle.3However, Hart elaborated
his arguments with a degree of precision that cannot be found in the works of
his predecessors.
The debate over defeasibility reached its peak in the 1990s, when the
approach of standard deontic logic to dealing with defeasibility was challenged
by those who used non-monotonic logics. The approach of standard deontic
logic rests on a basic principle called ‘antecedent strengthening’,4which states
∗Post-doctoral researcher, Bocconi University (Milan).
1 H. L. A. Hart, ‘The Ascr iption of Responsibility and Rights’ in Proceedings of the Aristotelian
Society New Series, 49 (1948-1949) 171; reprinted in E. Flew (ed), Logic and Language: First Series
(Oxford: Basil Blackwell, 1951) 145.
2 W.D.Ross,The Right and the Good (Oxford: Clarendon Press, 1930) 19.
3 R. C. Koons, ‘Defeasible Reasoning’ Stanford Encyclopaedia of Philosophy 2009 at http://plato.
stanford.edu/entries/reasoning-defeasible/ (last accessed 12 December 2016).
4 J. Rodriguez and P. Navarro, Deontic Logic and Legal Systems (Cambridge: CUP, 2014) 105.
C2017 The Author. The Modern Law Review C2017 The Modern Law Review Limited. (2017) 80(2) MLR 352–369
Published by John Wiley& Sons Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA
Alessio Sardo
that the proposition ‘if pthen q’ entails (evidences, validates) the proposition
‘if pand rthen q’.5So, for instance, in standard deontic logic, the nor mative
proposition ‘Dogs are not allowed in this area’ entails (in a strictly logical sense)
the normative proposition ‘Pet-therapy dogs are not allowed in this area’. Now,
as we can envisage, this principle brings about some counter-intuitive implica-
tions. Take the following scenario: a certain nor mative authority A, at time t1,
issues the normative proposition N1, ‘Dogs are not allowed in this area’; then,
the same authority A, at a later time t2, issues a second normative proposition
N2, ‘Pet-therapy animals, as an exception, are allowed in this area’. If we
endorse antecedent strengthening, accepting that the normative proposition
N1 entails N3 ‘Pet therapy dogs are not allowed in this area’, then we are
forced to consider this kind of situation as a normative conflict,6since dogs
clearly belong to the class of animals. Accordingly, we shall conclude that our
normative system ascribes incompatible normative solutions (‘allowed/not
allowed’) to an overlapping class of cases (‘pet-therapy dogs entering a certain
area’).7This means that if we accept antecedent strengthening, it becomes
impossible to explain exceptions and defences without introducing unwanted
contradictions in the normative system. In effect, the antecedent strengthening
principle compels us to accept the notion that every time the legislator intro-
duces a new exception, she produces an unwanted normative conflict in the
system.
Given such problems, non-monotonic logics appear to be more suitable
than standard deontic logic for dealing with defeasibility, being able to explain
with a powerful formalisation how, for instance, it is normally forbidden
to kill, but that such an action may be allowed under certain exceptional
circumstances (such as in the case of self-defence). This is because they are
grounded on a different consequence relation: in opposition to classical logic,
these new philosophical logics assume that some conclusions may be ‘blocked’
by the introduction of a new premise.8Going back to our example, a defender
of the non-monotonic approach would say that the normative proposition N1
‘Dogs are not allowed in this area’ is compatible with N2 ‘Pet therapy animals,
as an exception, are allowed in this area’, for N2 works as a barrier to the
implication which derives N3 ‘Pet-therapy dogs are not allowed in this area’
from N1. Non-monotonic logics were first developed to explain reasoning
with incomplete knowledge, having in mind those situations where certain
5R.Stalnaker,Inquiry (Cambridge, Mass: MIT Press, 1984) 123-124.
6 A. Ross, On Law and Justice (London: Stevens, 1958) ch 4.
7 In dealing with this kind of situation, standard deontic logic seems to face the same difficulties
that classical logic encounters in the explanation of puzzles such as the ‘Tweety Case’. In the
Tweety Case, we commit, at the same time, to a universal proposition such as ‘all birds fly’, and
to a singular proposition such as ‘Tweety, being a penguin, does not fly’. According to the rules
of classic propositional calculus, these two propositions bring about a contradiction, for their
conjunction entails that Tweety both flies and does not fly, since she is, at the same time, a bird
and a penguin. Therefore, one of them should be rejected.
8 This assumption satisfies also the ‘Requirement of Modularity’, intended as ‘the possibility of
formalizing an element or piece of information in the domain without having to consider each
time the rest of it.’ J. C. Bay´
on, ‘Why Is Legal Reasoning Defeasible?’ (2002) 2 Diritto e Questioni
Pubbliche 1, 8.
C2017 The Author. The Modern Law Review C2017 The Modern Law Review Limited.
(2017) 80(2) MLR 352–369 353
To continue reading
Request your trial