Defeasibility in the Law

Defeasible reasoning characterises the law at different levels.
First, clues to the defeasibility of legal reasoning are embedded in the very language of legal sources. As we saw in the previous example, the legislator itself often suggests how to construct defeaters to certain arguments. For example, to indicate that liability in tort can be excluded by appealing to self-defence or a state of necessity, the legislator may use any of the following formulations:
According to all these formulations, to build an argument to the effect that one must make good some damage, it is normally sufficient to ascertain that one voluntarily caused that damage, but this argument is defeated by counterarguments appealing to the fact the person turns out to have acted either out of necessity or in self-defence.
Defeasibility is also an essential feature of conceptual constructions in the law. Legal concepts must be applied to such a diverse range of instances that they can at best offer a tentative and generic characterization of the objects to which they apply, a characterization that must be supplemented with exceptions. General legal concepts presuppose defeasibility: the requirement of absolute rigour in defining and applying concepts—the demand that all features which are included in, or entailed by, a concept apply to each of its instances—would paradoxically run counter to the very possibility of being “logical” in the sense of using general concepts. In fact, even the definitions of the legal concepts that can be found in statutes and codes reflect the stepwise defeasible process of establishing legal qualifications: first, a general discipline is established for a certain legal genus (e.g., the genus “contract”); special exceptions are then introduced for species within this genus (e.g., the species contract of sale); finally, further exceptions may be introduced for specific subspecies (e.g., the sale of real estate). Consequently, when using conceptual hierarchies, we must apply to a certain object the rules governing the category in which it is included, but only insofar as no exceptions emerge concerning a subcategory in which that object is also included.
Defeasibility can be deliberately established by the legislator, but it may also result from the evolution of legal knowledge: after a general rule has been established, exceptions are often provided for those cases where the rule appears to be inadequate.
This is typically the evolution of judge-made law, where general rationes decidendi are often limited by way of distinctions , that is, by way of exceptions introduced for specific contexts (on defeasibility and precedents, see Prakken and Sartor 1998; Horty 2012). In such cases, judges often leave the original default rule unchanged and add a new, prevailing rule that addresses the specific situations requiring a distinction. For instance, in the Monge case (US Supreme Court, 28 Feb. 1974, No. 6637), the judges introduced an exception to the idea that contracts of employment at will (lacking any set term) could be terminated by both parties regardless of the reason (“for any reason or no reason at all”). They stated that “a termination by the employer of a contract of employment at will which is motivated by bad faith or malice or based on retaliation […] constitutes a breach of the employment contract. Correspondingly, on the basis of this rule the dismissed employee Olga Monge could build an argument (her dismissal was a breach of contract, being based on malice and retaliation) that could defeat the employer’s argument that she could be legitimately dismissed on the ground that her contract was at will. Note that the judges could also have revised the original rule into a new rule: “a contract can be terminated by both parties for any reason unless the employer is terminating the contract motivated by bad faith or malice or based on retaliation.” The new rule would have triggered the same dialectical exchange, as long as the unless clause was interpreted as attributing to the employee the burden of proving bad faith, malice, or retaliation.
Finally, we need to also consider the procedural aspect of defeasibility. As noted, this aspect concerns the fact that defeasible reasoning activates a structured process of inquiry in which we draw prima facie conclusions, look for their (prima facie) defeaters, look for defeaters of defeaters, and so on, until stable results can be obtained. A process like this one reflects the natural way in which legal reasoning proceeds. This is especially the case in the law’s application to particular situations, when we have to consider the different, and possibly conflicting, legal rules that apply to such situations and must work out conflicts between these rules.
The defeasibility of legal reasoning also reflects the dialectics of judicial proceedings, where each party provides arguments supporting his or her position, and these arguments conflict with the arguments made by the other party. The debate of the parties is usually transferred to the judicial opinion that takes in the results of the dispute and determines its output. To convincingly justify a judicial decision in a case involving genuine issues, it is not sufficient to state a single argument; it is necessary to establish that the winning argument prevails over all arguments to the contrary, especially those that have been presented by the losing party, or that the latter arguments have to be rejected on other grounds.
Finally, doctrinal work cannot avoid being contaminated by the dialectics of legal proceedings, since its main function consists in providing general arguments and points of view to be used in judicial debates. From this perspective, doctrinal reasoning may be viewed as consisting in an exercise in unilateral dialectics , understood as a disputational model of inquiry in which “one develops a thesis against its rivals, with the aim of refining its formulation, uncovering its basis of rational support, and assessing its relative weight” (Rescher 1977, 47).
The significance of defeasibility in legal reasoning has been recently confirmed by the psychological experiments by Gazzo et al (2016), which show how both lawyers and laypersons reason defeasibly when applying legal norms. When presented with a legal conditional, in its usual formulation (If somebody kills another human, then this person should be punished for manslaughter), and with an instance of the antecedent condition (Bert killed another human), most participants in the experiment conclude for the conditional’s conclusion (Bert should be punished for manslaughter), but withdraw this conclusion when told that an exculpatory circumstance (because of a psychological disorder, Bert was not able to control his actions) also obtains. The experiments also show that lawyer are better than laypersons in withdrawing legal conclusion when faced with legally recognised exceptions, having a more precise knowledge of such exceptions and of their role in legal reasoning.