Seales v Attorney General: report from the court

Posted on by

Lecretia Seales’ quest to be permitted aid in dying is both tragic and important. It raises myriad legal and ethical questions. I spent the first two days of the case in Wellington High Court, listening to her legal team present their arguments. (Full disclosure: I also presented an expert witness affidavit in support of her case.)

Much of her claim centres on the NZ Bill of Rights Act, as explained by my colleague Andrew Geddis here. But the case also involves some detailed consideration of some core criminal law concepts like intent and causation.

Lecretia’s claim refers to “aid in dying” – it avoids the question-begging term “assisted suicide”; whether what she wants the right to do constitutes “suicide’ is one of the points in dispute. This “aid in dying”, though, can take two forms. As described in her claim, “administered” aid in dying (AAID) involves the direct administration of a lethal drug by a doctor. “Facilitated” aid in dying (FAID) involves the provision of the lethal drug by that doctor, but requires the final step of taking that drug to be by Lecretia herself.

The concern with AAID is that it may form the basis for a charge of culpable homicide or even murder. For that to happen, s 160 of the Crimes Act requires that the doctor must have caused death by “an unlawful act”, which is now defined as “a breach of any Act, regulation, rule, or bylaw.” 

What the “unlawful act” might be would depend on the form that the AAID took. If the doctor injected the patient with the lethal drug, the possibility has been discussed that this could constitute assault. But what if the doctor merely added the lethal agent to an IV drip that was already in place? It seems unlikely that this could be seen as the “application of force”. An alternative possibility would be “poisoning with intent to cause grievous bodily harm”.

Lecretia’s claim, though, is that, in circumstances like hers, the administration of a lethal drug does not constitute an “unlawful act” at all. This interpretation, she argues, is necessary to render the Crimes Act compliant with sections 8 and 9 of the BORA. If AAID in those circumstances does constitute an “unlawful act” for the purposes of s 160, then s 160 is inconsistent with the BORA.

What, then, of FAID? This would be dealt with under a different part of the Crimes Act. Although suicide is not a crime in New Zealand, s 179 makes it an offence to “aid or abet” someone to commit suicide. Providing someone with the means to commit suicide would violate that provision, if the assistant knew that suicide was being contemplated, and had the intent to assist.[2]

Would providing someone like Lecretia with a lethal prescription be an offence under s 179? At first glance, the answer seems obvious. But there are two reasons why things may be more complicated.

First, an act can only amount to assisting suicide if what the other party does or attempts to do amounts to “suicide.” I can’t assist you to do X if you are actually doing Y. Part of Lecretia’s argument is that ending her life in such circumstances would not amount to “suicide” for the purposes of s 179.

This may seem like a curious claim – surely we all know what “suicide” means, and we don’t need smart-alec lawyers to tell us. The Crimes Act, though, doesn’t offer a definition of “suicide”, which means that it has been left up to New Zealand courts to do so. So, last year, we had the High Court decide that a prisoner on hunger strike was not attempting suicide.

The manner in which Pankhurst J explained that decision is a bit ambiguous. At one point, he seemed to be basing it on the fact that All Means All (that was actually the prisoner’s name!) did not want or intend to die as a result of the hunger strike, but instead, to force the authorities to concede to his numerous demands. But the judge also cited with approval Lord Goff’s judgment from Bland, where he said that a refusal of treatment could never amount to suicide.

Whether it makes sense that no-one can ever commit suicide by “omission” is an interesting question. It doesn’t seem obvious that starving oneself to death would differ in any of the important ways (intention, causation, etc) from taking a lethal overdose. What’s more, if true, it would raise the potentially troubling possibility that inciting someone to do so – say, by stopping eating or foregoing their insulin – would not violate s 179. But that’s a thought to pursue another time.

For Lecretia and her team, what’s relevant is that it is already accepted that “suicide” is a term that’s open to a degree of judicial interpretation. The question is whether the interpretation she seeks would require more judicial flexibility than Collins J feels able to adopt.

There is also overseas precedent for distinguishing “suicide” as commonly understood from the sort of rational life-ending decision that Lecretia might want to take. In the USA, the 9th Circuit appeal court has recognised “a strong argument that a decision by a terminally ill patient to hasten by medical means a death that is already in process, should not be classified as suicide.” So if Collins J is persuaded to accept this distinction, he would not be the first to do so.

The alternative claim regarding FAID is that, even if Lecretia’s final act would constitute “suicide” for the purposes of s 179, the doctor providing the lethal drug would lack the requisite intent to be guilty of “assisting” her.

Again, at first glance, this may seem unlikely. But on closer analysis, this seems to me the part of the claim that may have the highest likelihood of success. It’s already well established in many jurisdictions that a doctor who administers a palliative medication that she knows might hasten a patient’s death will not necessarily be deemed to have intended that death. Such a doctor will be able to rely on the so-called doctrine of double effect, the rule that seeks to distinguish outcomes that are truly intended from those which are merely foreseen side-effects.

Whether the DDE makes an important moral distinction, or constitutes a dishonest fudge, is a matter of great debate in bioethics and law. Nonetheless, it is a defence that has been accepted in English law (albeit less commonly and less clearly than is sometimes assumed). Its status in NZ law is a bit less clear, but if it is part of our law, then it may be thought to apply as much to the doctor in the FAID situation.

As we know from Oregon, where such assistance has been legal for many years, a substantial number (more than a third) of people provided with lethal prescriptions never actually go on to use them. Many of those patients speak of the reassurance and relief they gained from knowing that the option was available should their suffering ever become unbearable. A doctor providing such a prescription may sincerely hope that the patient would never have occasion to use it. She may be confident that the patient’s symptoms could be managed to a tolerable level. Her “primary purpose” may be to provide the patient with the relief of knowing that the option was there, should worst ever come to worst.

Following the rationale of double effect, such a doctor will not have “intended’ the patient’s suicide, even though she will surely have foreseen it as a real possibility. (though interestingly not, on these statistics, a “virtual certainty”, the standard usually required to infer intent from foresight.)

Suzanne Uniacke has suggested a “test of failure” to help distinguish truly intended outcomes from those that are merely foreseen.[1] On this approach, we would ask how the doctor would feel if, having administered the analgesia, the patient actually lived. If the doctor would be pleased with that outcome, then we can say that she did not in fact intend the patient’s death. If not, if that was a disappointing outcome, then it becomes hard to argue that death was not at least part of what was intended.

The same test, I have argued, could be applied to FAID. Imagine the doctor provides her patient with a lethal prescription, hoping against hope that the patient never has to take it. As it happens, the patient is one of that 35% who dies without ever using the drug, but derives comfort from knowing that it was there. If the doctor would regard that as a good outcome, then – on Uniacke’s ”test of failure” – the patient’s suicide was not part of the doctor’s intention.

Is this approach to “intention” a bit artificial, or likely to encourage a degree of self-deception? Maybe. Maybe it would be more honest to deem everyone to have intended all of the foreseen consequences of their deliberate actions. But that doesn’t seem to be the approach that the courts favour in other end-of-life contexts.

That being so, then it does look like a strong case might exist for regarding at least some instances of such provision to lack the required mens rea for “assisting suicide.” Intriguingly, the Solicitor General seemed to accept this possibility when Collins J put it to him on Tuesday.

For Lecretia, of course, the main concern is that she is allowed a peaceful and dignified death. But however Collins J decides, it seems inevitable that the reasons for his decision will bear very close analysis.

[1] Uniacke, S. ‘The Doctrine of Double Effect.’ In Ashcroft R, Dawson A, Draper H, McMillan J, eds. Principles of Health Care Ethics (2nd ed.) Hoboken: Wiley 2007

[2] R v Tamatea (2003) 20 CRNZ 363. This case dealt with inciting and counselling suicide, which is dealt with in a different subsection, but it seems certain that similar elements would have to be present for aiding and abetting.

This entry was posted in Uncategorized by Colin Gavaghan. Bookmark the permalink.

4 thoughts on “Seales v Attorney General: report from the court

  1. Hi Colin, great blog, thanks. I think there is a moral difference between an clinician assisting suicide by administering a lethal dose and providing a patient with the means that they might use at a time of their choosing. It seems to me that the position adopted in many American States whereby an action on the part of the patient is required for death is preferable because it helps to ensure that this is a genuine and deeply held autonomous wish. I’m attracted by the ‘test of failure’ strategy for determining what’s intended as opposed to foreseen and I’m hoping Collins J does too. However, I do wonder whether how well or how often it would be passed in assistance in dying cases. In the case of hair loss as a side effect to chemotherapy, a clinician might say that hair loss is a side effect and while it is foreseen it is certainly not intended. Moreover, if the therapeutic effect of the chemotherapy could be achieved without causing the side effect, a clinician could, hand on heart, say I would have preferred that did not happen. Now, it might be that a clinician who writes a script for a lethal dose could say, ‘my intention is to ease her suffering by empowering to take control of the manner of her death’, however in the event that someone does take their life, then, because the clinician has provided them with the means to do so, and that’s the therapeutic end that a lethal dose is very likely to bring about, it would be odd to put hand on heart and say, ‘I did not positively will that would come about.’ It seems to me more like a case where something is done which will lead to one of two intended effects. So, ‘double’ effects, but because a disjunction is willed, not a case of the doctrine of double effect. I’ll send Suzanne the link to this blog and see if she would like to comment.

    Cheers,

    John

  2. Thanks for that John. The scenario I was imagining would have to be one where:

    1. the patient (P) is not contemplating using the lethal drug immediately;
    2. the doctor (D) provides the lethal drug primarily to afford comfort and assurance to P, who is worried that his/her suffering might become unbearable in future;
    3. D is aware that P might use the drug at some point in future, but hopes that this never proves necessary;
    4. P also hopes never to have to take the drug, but wants to prepare for worst case scenarios.

    If D’s primary purpose would be thwarted by the P’s non-consumption of the drug – if s/he would view that as a bad outcome – then D would (if I’ve understood it correctly) have failed the “test of failure”; there would in that case be no denying that D’s primary purpose was for P to take the drug. But if P died without taking the drug, but reassured by having it, and if D regarded that as a good and happy outcome, then it seems more plausible to argue that D’s actual intent was to reassure P, rather than to help P die. P’s taking the drug is a foreseeable outcome that D hopes is avoidable.

    Of course, all of this is open to disingenuous claims about what D actually hopes and expects, but so much is also true of other applications of the DDE. (Please do pass it on to Suzanne, i’d love to know what she thinks about this.)

  3. Great summary and analysis, thanks, Colin. I have a couple of quick thoughts.

    If the courts end up favouring the doctrine of double effect as you’ve suggested, I wonder whether they might see the intervention in this case as lying outside the doctrine’s usual scope for medical purposes, or at least introducing a complexity that causes it to deviate from the normal case. This might not be a problem for the doctrine directly, but it could be a problem for medical practice (as I hope to explain).

    In this case, it seems that the effect intended by the prescribing doctor arises from non-use of the prescription. If non-use helps the patient in a medically relevant way (by making them feel a lot better), then it seems like a good medical outcome. However, the practice of prescribing for intended effects arising from non-use seems like it might cause problems if it was applied more generally. What if a worried patient would feel a lot better knowing that he could have x-rays, blood tests, or access to performance enhancing drugs in the event that they feel they need them. I’ll just stipulate that each of these would, because of features of the particular case, not be things that the Dr has a medically acceptable reason for intending the patient to use. Double effect seems to give the Dr a free pass to give patients access to a range of medical interventions if non-use is the action that gives rise to the intended effect.

    Just responding to myself here, but perhaps this range is constrained by the need for some independent justification of a right to access (the other aspect of the Seales case) though. If so, this would seem to weaken the objection.

    The second thought is just to emphasise the point made by one of the lawyers for Seales, which is that Parliament cannot be relied upon to act on this issue independently. We have seen this in the case of use of embryos in research, which is effectively prohibited due to government failing to approve new, more permissive, guidelines on this from ACART. There is a political vacuum when it comes to considering contentious issues such as these. It will be interesting to see whether the court will take the opportunity to fill that vacuum in this case, and perhaps others if the situation doesn’t change.

  4. Thanks for the opportunity to contribute to this discussion.

    We should distinguish the following questions:

    (1) Whether there is a defensible distinction between an intended as opposed to a (merely) foreseen effect of one’s action

    (2) Whether and under what conditions such a distinction is relevant to moral or legal evaluation.

    I’m not much of a blogger, so I’ll focus my initial contribution to this discussion on (1). Some comments:

    We need to clarify what is meant by an intended, as opposed to a (merely) foreseen effect. This has a significant bearing on the so called ‘test of failure’.

    An intended effect of my action is something I aim at, either as an end or as a chosen means, in acting as I do. A (merely) foreseen effect is something that I anticipate my action will cause but which is incidental to what I aim to achieve. (This is not a distinction between an actor’s primary and secondary purpose, since an actor’s purpose is intended.)

    There are clear examples in which a distinction between an intended, as opposed to a (merely) foreseen effect makes complete sense. (If I fail a student’s poor essay and I foresee that he will probably be upset, I don’t thereby fail his essay with the intention of upsetting him.) Difficult examples for this distinction can arise where an unintended effect is foreseen as certain or highly probable. Nonetheless, there are clear such examples in which the distinction is defensible (as in the incidental effects of chemotherapy – hair loss, nausea).

    The so called ‘test of failure’ (sometimes referred to as the ‘counterfactual test’) is invoked in the philosophical literature on the DDE as a suggested way of differentiating effects that an actor intends as opposed to those that s/he (merely) foresees. (The ‘test of failure’ is not original to my contribution to this literature, so it’s not “Uniacke’s test”.)

    It’s crucial that the ‘test of failure’ is framed in terms of whether the actor would achieve what he aims to achieve if, contrary to her belief at the time of action, the (merely) foreseen effect does not occur. The ‘test of failure’ applies at the time of action (not in retrospect). It’s also a test of the actor’s intention, and not a test of how she feels or of what she might be pleased about achieving or would be disappointed not to achieve. (I can intend something that I’m not pleased about having to do; and I can be pleased about an effect that I did not intend.)

    Some philosophers maintain that the ‘test of failure’ isn’t (always) satisfactory in distinguishing an intended effect from one that is (merely) foreseen. However, they tend to invoke highly fanciful or gimmicky examples to this end. I think it is a mistake to infer that because the ‘test of failure’ seems unsatisfactory as applied to such examples, that it is unsatisfactory as applied to other, more realistic or standard examples.

    My comments above refer to intention and foresight as ordinary language terms, as opposed to how these terms might be specified in law, for instance. In criminal law a person can be deemed to have intended effects of his actions that he foresaw as ‘virtually certain’ to occur, even if they were strictly incidental to his aim. For instance, a person who plants a bomb in a railway carriage is said to intend the deaths of all of those who happen to be in the carriage and are killed by the blast, even if he was aiming to kill only one particular person (and the presence and deaths of all of the others were incidental to this aim). The reason for the inclusion of ‘virtually certain’ effects in what is (legally) intended is, I assume, the view that in a case such as this the actor should be guilty of murdering all of those who are killed by the blast. So, the ‘test of failure’ doesn’t distinguish intention from (mere) foresight in such a case and this is because in this legal context ‘intention’ has been given a more inclusive sense than it has in ordinary language.

    However, in other areas of the law, the ordinary language distinction between intention and (mere) foresight is invoked. Medical law seems an obvious case. For instance, a doctor who does very risky surgery in a desperate attempt to save a patient’s life, and where this is the patient’s only hope of survival, might foresee that the surgery will probably kill the patient. In this case, the surgeon’s intention (to save the patient’s life) is actually incompatible with the foreseen effect (the patient dies on the operating table), which is an excellent reason for distinguishing between intention, as opposed to foresight in such a case. (This is the type of case in which it seems wrong to call the foreseen effect a side-effect, as opposed to an incidental effect.) More difficult cases are ones in which the actor’s intention is compatible with the foreseen effect. But here again, the legal permissibility of so called pyramid painkilling invokes the ordinary language distinction between intention and foresight.

    As regards the ‘test of failure’ as applied to FAID (where the patient is provided with the means of ending her own life): Here it seems to me plausible for a doctor to say that at the time of action her intention is to provide the patient with the means of taking his own life if he chooses so to do. This means that if the patient takes the drug intending to die and for some reason the drug doesn’t kill him, then the doctor’s intention (of giving the patient the means to kill himself if he chooses so to do) has failed. However, the doctor’s intention at the time of action is consistent with the patient choosing not take the drug. For these reasons it seems to me that while the doctor intends to supply the patient with a fatal dose of a drug, she does not intend that the patient takes it (provided the patient’s choosing not to take it would be consistent with the doctor’s intention).

    In FAID cases, the effect (whether it’s intended or (merely) foreseen) is indirect, in that it comes about through another person’s agency. Indirect effects can be intended. But the conditions under which a person can intend an indirect effect that comes about through another person’s agency are bound to be more complex than in a case where an effect is directly brought about by that person’s action.

Leave a Reply

Your email address will not be published. Required fields are marked *

* *

*