Legalistic Mistake

Legalistic Mistake

Marco Antonio Azevedo

On the occasion of Elizabeth Anscombe’s presentation of her paper “The Two Kinds of Error in Action” (1963) at the American Philosophical Association meeting in Columbus, Ohio, 1963, Joel Feinberg objected that murder was a “legal concept” so he might not have understood what Anscombe meant by the expression “morally speaking a murderer.” Anscombe reacted by saying that no one concerns oneself with questions of legality in calling, for example, the killings practiced by the Hitler regime murder. She further claimed that any view such that all unjust killings of people are purely a matter of legality consists of a piece of wicked positivism, comparing it to Thrasymachus’s claim in Plato’s Republic that justice is merely the advantage of the stronger (338c). This was a harsh criticism, and Feinberg’s response came one year later in a paper entitled “On Being ‘Morally Speaking a Murderer’” (1964). It was in this paper that Feinberg introduced the idea that, sometimes, using legal notions outside legal contexts may lead to a peculiar kind of fallacy he called the legalistic mistake.

In her original paper, Anscombe argued that sometimes “error destroys action,” meaning that ignorance of a circumstance can prevent an act from counting as the voluntary performance of a particular kind of action. Some of her examples concerned mens rea, or the mental element of an offense. In fact, her approach was mainly the same one presented in her classic book Intention (1958a). One problem she had in mind is that “when we consider error about whether doing S in circumstances C is doing A, it is very difficult to show the rationale of A’s not being imputable to the agent. For he did voluntarily, even intentionally, do S in circumstances C” (Anscombe 1963, 397–398). One example was the following:

A famous example is that of the public executioner who has private knowledge of a condemned man’s innocence. In some way he knows he cannot make use of it to get the man off; and he is to execute him. The man had a fair trial. The question is whether it is, morally speaking, an act of murder for the executioner, at the command of his superiors, to perform his office in these circumstances.

(Anscombe 1963, 398)

It was Anscombe’s mentioning of the possibility of one’s being called morally speaking a murderer that drew Feinberg’s keen attention. Even though it is possible that Feinberg had in fact misunderstood her point, his remark on the possibility of a fallacious use (or rather, a misuse) of the word ‘murder’ was spot on. In his response, he explained that:

In classifying ‘murder’ as a legal concept, [he] meant only that it finds its original, primary, and clearest application in legal contexts, and that therefore its use in those contexts is a convenient, even necessary, model for our understanding of its extended uses outside of the law.

(Feinberg 1964, 158)

Feinberg chose to call these concepts “legal‐like” instead of just “legal,” hoping that this remark could obviate Anscombe’s accusation of a wicked sort of positivism. On widening the use of such legal‐like terms, we must be cautious, for we might find ourselves guilty of making inferential mistakes or even proffering pure nonsense. The error, according to Feinberg, is committed by “one who, in stating a moral question using a legal‐like term, uncritically imports the precision of that term in its strict legal sense, while excluding appeal to the kinds of criteria which alone can decide its use” (Feinberg 1964, 161).

Feinberg’s examples of words that are subject to this kind of abuse are, in addition to murder, terms such as right, duty, responsibility, obligation, criminal, bankrupt, indebtedness, partnership, all of them legal‐like, that is, expressions whose original, primary, and clearest applications are legal or juridical. These terms can be distinguished from others that are widely employed in our normative vocabulary, such as ought, good, bad, better, which are not distinctively legal‐like. And if the word law itself is legal‐like, the same applies to moral law and the related concepts of morally obliged and morally responsible.

As Feinberg points out, Herbert Hart and Tony Honoré have suggested a similar kind of error committed by moral philosophers when uncritically using the term ‘responsibility’ in moral contexts, for, say they:

Where there is no precise system of punishment, compensation, or reward to administer, ordinary men will not often have faced such questions as whether the injuries suffered by a motorist who collides with another in swerving to avoid a child are consequences attributable to the neglect of the child’s parents in allowing it to wander on to the road.

(In Feinberg 1964, 161)

To this, Feinberg (1964) adds that his point is even stronger than that: “[i]n ordinary life, in abstraction from all practical questions about punishment, compensation, and the like, there is no rational way of answering such questions” (161). Anyone may use a legal‐like term, but suppose one reacts to something you did, saying that you are “morally responsible for the harm.” If this was done as an attempt at keeping you from doing the same in the future (maybe making you feel bad about yourself), the statement, so says Feinberg, was well made and meaningful. This is different if “we consider the question as one calling for cool and exact judgment, a ‘verdict’ issued on past events.” In this case, he stresses, the question is ill formed, “for it asks us to sit on a kind of moral jury in a court in which normal juridical rules and reasons can have no relevance, but which has no alternative rules and reasons of a remotely legal‐like kind” (Feinberg 1964, 161).

Surprisingly, this remark is rather similar to Anscombe’s notorious observation in “Modern Moral Philosophy” (Anscombe 1958b). According to her, concepts such as moral obligation and moral duty and what is “morally right” or “wrong” are “survivals or derivatives from survivals from an earlier conception of ethics which no longer generally survives” (1). In this earlier and renowned paper, Anscombe (1958b) claims that the ordinary (and originally non‐legal‐like) terms as ‘should,’ ‘need,’ ‘ought,’ and ‘must’ had the special sense of ‘being obliged,’ ‘being bound,’ or ‘required to,’ in the sense of being obliged or bound “by law” (or as something required by law), as a result of the domination for centuries of the Christian law conception of ethics. In consequence of this cultural dominance, the concepts of being bound, permitted, or excused became “deeply embedded in our language and thought” (5). She claims that in the times of Christian dominance, this usage was perfectly in order; in fact, if you believe in God as a law‐giver, as do Jews, Stoics, and Christians, the legal‐like effects of those terms are natural and meaningful. So her critical assessment was rather of the usage made by the modern philosophers. At any rate, it is very likely, so she claims, that a legal‐like notion could exert its effect even after losing its social and cultural roots; but in this case, it is only a psychological, almost delusionary effect. “It is as if the notion ‘criminal’ were to remain when criminal law and criminal courts had been abolished and forgotten” (6). Obviously, we are not obliged to agree with Anscombe, nor to support her harsh criticism of her contemporary colleagues; anyway, her criticism is very similar to Feinberg’s, that is, when using legal‐like terms, we should pay special attention to the criteria that make these terms meaningful in their legal or legislative original contexts. Now, the disagreement between Anscombe and Feinberg is explainable. Anscombe would agree that ‘murder’ is a legal‐like term with an original or primary legislative origin; in her rebuke of the executioner’s case, what she probably had in mind was natural or divine law. For Feinberg (1964), instead, the case seems puzzling because there is a difficult problem of moral appraisal “that has nothing to do with the law or anything legal‐like”:

In acting on his mistaken moral judgment, the executioner was at fault, and the fault accurately reveals a defect of character – poor moral judgment in some respects – which counts against a man in a final appraisal. Conscientiousness, on the other hand, is clearly a virtue. […] Given that the executioner was at fault in killing his victim, and perhaps even therefore to blame for the death, it still does not follow that he is, in some strange moral sense, guilty of murder, or indeed that he is guilty of anything at all. (170)

Feinberg’s objection does not rule out legal‐like terms in moral discourse. It asks philosophers to explain what warrants applying them when the legal rules that normally settle their use are absent.

The use of “legal‐like” terms outside law is common. In the abortion debate, for example, some advocates claim that “abortion is murder”. To assess this claim, first ask what “murder” means in the argument. One proposed inference defines “abortion” as killing an embryo or fetus, treats embryos and fetuses as innocent human beings, and defines the intentional killing of an innocent human being as murder. Whether those premises establish the conclusion depends on the definitions and moral claims being defended. A legal classification of “murder” would require the rules of a specified legal system; a use of “murder” in a special “moral” sense would require moral criteria. Neither can borrow precision from the other without explanation. An analogy between abortion and murder can also be argued for, but calling an act “murder” does not by itself establish the analogy or resolve the underlying moral disagreement.