Looking into oral argument about causation

Oral argument on appeals is quite an interesting thing, as you can hear and read in relation to Burrage v United States (Docket No 12-7515, 12 November 2013).

At issue here is what distribution of heroin “resulting in” death means. Here is discussion of the case, and here is the relevant legislation; you need to read down to the penalty part to get the phrase “if death or serious bodily injury results from the use of such substance.”

Was it necessary for liability that the heroin was the only cause of death? Or that it was a contributing cause of death? Or that it was not merely a contributing, but a significant cause of death? Another question was whether the death had to be a foreseeable consequence of the distribution of the drug. This turns on the extent to which the common law relating to the causation aspect of the attribution of responsibility applies to the interpretation of the words “resulting in” in this statutory context.

The appellant (called the “petitioner” in American terminology) argued that a “but for” causal connection is required: the prosecutor must prove that the death would not have occurred but for the distribution of the drug.

The victim had consumed other drugs too, and these had not been supplied by the appellant. Expert evidence was that it could not be said that the victim would have lived if the heroin had not been used. The heroin contributed to the death but could not be said to be its sole cause. To what extent, if any, is the defendant to be held responsible for the acts of strangers, including acts of the victim?

Standing back and looking at the problem in the conventional terms of common law causation, there is no great difficulty. As to the actus reus, liability requires proof that the defendant’s acts were an operative contributing cause of death. This can sometimes be put as a “substantial” contributing cause. The evidence in the case must be assessed against this requirement. Liability will be negatived by an intervening act that replaces the defendant’s acts as an operating cause of death. For an intervening act in this case the defence would have to point to evidence that raised a reasonable possibility that death would have occurred without the defendant’s act of supplying heroin. See, for example, the discussion of Hughes v R [2013] UKSC 56, here on 9 August 2013, Burns v The Queen [2012] HCA 35, here on 15 September 2012, Maybin v R, 2012 SCC 24, here on 22 May 2012, and R v Kennedy [2007] UKHL 38, here on 19 October 2007. In Kennedy the victim’s voluntary choice to use the drug supplied by the defendant was a novus actus interveniens. These decisions were not cited in the brief filed for the petitioner (which was confined to citation of American cases; not that there is anything necessarily wrong with that).

As to mens rea, liability for the victim’s death usually requires gross negligence at least.

These conventional considerations are merely a background for interpreting the relevant legislation. The legislature can be taken to be aware of the existing law, and to intend to change it only by clear provision: Hughes, above. The policy supporting the creation of the statutory offence must be identified if the plain words of the enactment are ambiguous.

The opening remarks of counsel for the appellant in Burrage show a departure from the ordinary legal meaning of cause, claiming that “but for” cause is the usual requirement. Would it have been better to refer to the evidence first, arguing that it supported intervening act? Justice Ginsburg began the questions from the bench by referring to the hypothetical that invites consideration of intervening acts. Justice Scalia makes this clear with his first comments, emphasising that the expert evidence was that the victim may have died without using the heroin.

The argument proceeds with discussion of hypotheticals and with counsel insisting on a “but for” interpretation of causation. Hypotheticals are used to raise policy considerations. What policy supported the appellant’s argument and made it more acceptable than the policy that supported the opposing argument? The three-drops-of-poison hypothetical raises the question of the appropriate boundaries – in this legislative context – of the danger to the victim that is attributable to the defendant’s conduct. This was alluded to by Scalia J in his reference to “the scope of the risk”.

The trouble with a “but for” requirement is that it easily reduces to a situation where no-one is guilty, as where poisoners independently administer a sub-lethal dose which cumulatively kills the victim. And the trouble with a substantial contributing cause requirement is that others may have already given a lethal dose when the defendant administers what is also a lethal dose; as the victim would die anyway, has the defendant caused the death, assuming he has not accelerated it? Is it acceptable that in this latter example the defendant might only be liable for the attempt to kill?

Sometimes judges play with numbers when the law prefers words, as happened with the questioning of counsel for the United States (the “respondent”) by Kagan J: how much more likely than other causes does the defendant’s acts have to make the victim’s death – 50%, 30%? Scalia J had the same difficulty with what is a “substantial” contributing factor to the death: “10 percent, 20 percent … 5 percent, what?” That is like demanding that the expression “beyond reasonable doubt” be put into numbers, something that the law does not currently do, as Kagan J pointed out.

Anyway, this transcript and recording is an interesting illustration of how argument on appeal can develop. Both counsel demonstrated significant expertise in handling questions from the bench.

Guidance for oral arguments is given in “Guide for counsel in cases to be argued before the Supreme Court of the United States” (Clerk of the Court, Supreme Court of the United States, October term 2013). Much of the advice on oral argument in that document will be of assistance to counsel in other courts.

Advertisements
%d bloggers like this: