In late March 2020, I became a member of a Covid-19 ethics group, tasked with providing advice to local NHS hospitals. It has been an eye-opening experience, giving me some sense of the range of complexities involved in coping with the disease. There cannot be a simple focus on the patient, since the risk of infection of others is so great. A central challenge is to respond to this, while maintaining standards of patient-centred care and keeping inequality to a minimum. Much of the work of the group has been concerned with myriad details of how to strike this balance in day-to-day care.
When the group started though there was an issue that rather overshadowed these concerns. Fatalities were rising alarmingly in parts of Italy, as the hospitals there became overwhelmed. Much of the discussion in the UK was of how we should respond if a similar situation arose here. How we should allocate access to ventilators if there were not enough? Should only medical criteria be used in making a decision? Was age a medical criterion? What of other factors that might reduce one’s chance of benefitting from ventilation? Should NHS staff be given priority? Would it be acceptable to take someone off a ventilator if there were none to spare and another patient’s need were greater?
Although there was some discussion of these issues among members of the group, they were way beyond our remit. We were expecting some guidance from those with more authority for what would be, ultimately, a political decision. For a while there was talk of the British Medical Association — the doctors’ union — saying something. Then expectations became focused on the four Chief Medical Officers of the UK, who were supposedly going to provide some detailed guidelines. But the dates passed, and nothing came. In the end the BMA offered a very vague statement, but by then it was already becoming clear that the ventilators were unlikely to be overwhelmed.
The UK government has not handled the Covid-19 pandemic very well. As I write, the UK has the total highest per capita death rate of any country except Belgium. The rate is six times higher than that of Germany. Part of the problem has been a lack of clear and timely guidelines: care homes, for instance, got precious little advice in the early days on how they could protect their residents. Nevertheless, I think that, in not offering guidelines for who should get priority in access to ventilators, the government did the right thing.
What would have happened if they had decreed that some groups should be given lower priority? Many members of those groups would surely have complained; there may well have been, as in Italy, legal challenges. Perhaps these considerations loomed large in the Government’s thinking, but they are not at the heart of the issue.
To get closer to the real issue, consider an analogy. Imagine a couple watching the film Sophie’s Choice, a drama set in Auschwitz in which the lead character has to choose which of her two children to save, and which to allow to be killed. Afterwards, our couple reflect on how they would have acted in such a situation with their own children. Discussing it, they conclude that such decisions should not be left to the last minute. Instead they decide to create a ranking of which child they would prioritise in various situations. In the name of openness, they suggest, they should share it with the children; perhaps the children could be involved in the decision. And, to keep things fair, there should also be a ranking across other family members too, grandparents, and themselves.
Nobody, or virtually nobody, actually behaves in such a way. These are taboo questions, concerning what psychologists have called “protected values”, and what Bernard Williams called “the unthinkable”. As Williams’ term makes clear, it is not just that people are reluctant to make decisions about these issues; we are highly resistant even to think about them. The unsurprising psychological findings are that if they are forced to do so they feel contaminated, and seek to make recompense in some way. This is not just superstition. There is a reason for not seriously thinking about such horrible things. Thinking over which of your children, or your parents, or your friends, you would save carries risks. It risks damaging your attitudes to them. Involving them in the deliberations will almost certainly make things worse, damaging love, respect, and trust.
Such damage may not be quite so obvious when it comes to ventilator policy, but it is there. Those allocated lower priority would be likely to see this as a statement that they are dispensable, that their lives are less valuable. More broadly, a certain boundary would have been crossed, a precedent set. One group would have been officially ranked below another for purposes of treatment. Next time there were talk of how to allocate scarce resources, perhaps in a very different setting, this policy would resonate. Better not to provide plans in advance for such eventualities.
Suppose, though, that the country had in fact run out of ventilators. Would it still have been right to have had no government guidelines on who would have priority? Imagine what would happen then. The issue would not go away because of the Government’s refusal to think about it. It would fall instead on the hospital staff. Clinicians and administrators would have to decide on a policy, even if were only “first come first served”. They would have no mandate to make such a choice. And they would already be stressed and overworked, dealing with huge emotional distress and grief from patients, families and co-workers. Unless we were very lucky, the result would be an inconsistent set of different policies in different hospitals, and a morass of guilt, uncertainty and resentment.
We seem to be caught between two unacceptable approaches. On the one hand, establish a prior policy, and bring damage in its wake; on the other enter the crisis without a policy, and inflict a different set of costs. Certainly there is no easy solution here. But I think there is something of a path between the two. There is a distinction between a settled government policy, a considered position laid down ahead of time, and an emergency policy. In an emergency, decisions have to be made; the unthinkable has to be thought. Decisions need to be clear, well communicated, and, as far as possible, fair. But exactly because it is an emergency, they are likely to lack features that we would normally expect political decisions to have. For that very reason, they carry less weight as indicators of our commitments and values.
Much recent contractualist thinking has embraced the idea of reasonable rejection: a political decision is defensible just in case none of the parties affected by it could reasonably reject it. This may be idealistic, but it is an ideal worth aspiring to. Arriving at such an outcome takes time and effort: positions need to be acknowledged, compromises negotiated, compensations offered. Emergency policies sidestep such a process. If it had become clear that ventilators were going to run out, it would have been incumbent on the Government to issue a set of emergency guidelines on their allocation. I very much doubt that any guidelines would have passed the contractualist test. Many patients who would have been given low priority could reasonably have complained that there was no good reason why this should have fallen to them, rather than some other group. It’s not that there would have been no reasons for the policy; unless it were a terrible one, there would have been. But the reasons wouldn’t have been compelling; there would have remained reasonable grounds to reject them, and to have chosen another group instead, or to have embraced some other approach, a “first come, first served” policy, or some even more explicit policy of random allocation of treatment.
It is because the decision would have been an emergency one in this sense — one that was never expected to pass the contractualist test — that it would have provided no precedent. Just because, at this time, a certain group was given low priority, there would be no assumption that they would be given it again. The decision would have passed a minimal standard of justifiability, but no more. It would have been an expedient, to be treated with regret, and perhaps, after the fact, with compensation.
The main lesson to be learned from an emergency policy is different. It is that we should do our utmost to ensure that we do not get into a situation of needing it again. Indeed, that is the lesson to be learned from the fact that we so nearly needed one, and it remains to be seen if the government has learned it. Modern industry might work on a “just in time” model, where there is no slack in the system, no excess capacity in case of emergency. But this is no way to run a health system. The issue does not just concern the provision of supplies. One key contrast with Germany is that they benefitted from a system that was not stretched beyond capacity because it was only designed to cope with normal times — they did not, for instance, need to move elderly people out of hospitals into nursing homes to free up beds, people who, it transpired, were already infected with Covid-19.
As the philosopher on the ethics group, all this gave me pause. One thing that characterises contemporary normative ethics is its use of thought experiments about disasters. Explorers are stuck in cave entrances and must be detonated if those trapped behind are to be freed. Lifeboats are overloaded and will sink unless some passengers are thrown overboard. Falling figures will crush bystanders unless they are vaporised with ray guns. Most famously of all, in a dazzling array of varieties, runaway trolleys bear down on hapless victims, who might be saved in increasingly baroque ways. All of these are emergency cases, and we think we can learn something from considering them. Yet, when faced with a genuine emergency, the best advice I could offer was not to make a decision unless it was absolutely necessary.
So are we as philosophers wasting our time, and that of our students, in considering such cases? Worse, are we encouraging them in ways of thinking that are positively harmful? Some of the time I fear that we are. Most of these are emergency cases that we will never encounter, and we will learn little about the actual policies that we should follow by trying to work out what we might do in them. And the glib attitude to human disaster encouraged by an unremitting diet of such cases is very jarring in the face of real disaster. Still, such cases can tell us something. Much of the problem is that they have come to be used for things they were never designed to do. When Judith Jarvis Thomson gave the classic formulation to the trolley case, she wasn’t trying to devise rules that should regulate accident policy in the transport industry. She was pointing out that our reaction to pushing someone off a bridge to stop a trolley is very different to our reaction to diverting it to a place where it will kill fewer people. The former is not acceptable even as an emergency policy; the latter is. There is a role for considering emergency cases in moral philosophy; but we need to do much more to get clear on its limits.