In August 2022, Archie Battersbee, a twelve-year-old boy who had suffered severe brain damage in a domestic accident, died following the legally sanctioned removal of his life support at the Royal London Hospital. This case was notable for the five-month legal dispute between Archie's family and the hospital trust about whether his life support should be ended. The most striking feature of this dispute was the adjudication by the High Court in June 2022 that Archie Battersbee was dead, followed by the adjudication of the Appeal Court in July 2022 that he was not dead. One might wonder why the highest legal authorities in the UK are not able to agree whether someone is dead or not.
Some of this legal disagreement was partly based on the fact that there is no statutory definition of death in the UK. There is a medical Code, associated with the Royal Medical Colleges, which provides de facto criteria for determining death, but as was evidenced in the Archie Battersbee case, there remains uncertainty – at least in the courts – about the application of this Code. There have also been important disagreements among medical professionals and philosophers about the definition of human death and whether what is referred to as ‘brain death’ constitutes death. Most of this controversy was probably ignored in the decision that Archie was dead, but possibly reflected in the Appeal Court ruling that he was not dead. Nevertheless, that a High Court could mistakenly determine that someone was dead may strike some as cause for alarm.
Before the 1950s, it was fairly clear when someone had died. Doctors would basically listen for heartbeat and check for breathing. If there were neither, and there was an identifiable cause of death, the patient was declared dead, with time of death being given as the time of the doctor's pronouncement. Occasionally, of course, patients were declared dead when they weren't, and recovered in the mortuary. Anxiety about false positive announcements of death led in the nineteenth century to a macabre market in safety coffins, fitted with devices for alerting those above ground when you woke up. But generally, the dead were accurately identified by what became known as the cardio-pulmonary standard for death, irreversible absence of heartbeat and breathing.
All this changed with the development from the 1950s onwards of technologies for resuscitating people whose hearts had stopped and for maintaining the breathing and thus the heartbeat of patients who had stopped breathing. Most of these patients had brain or neurological damage. These technologies were supported by new ways of providing nutrition to such patients when they were unconscious, as well as techniques for supporting other biological functions that were compromised by brain injury such as blood pressure maintenance, temperature control, fluid level management, and so on.
The consequence was that if a person's heartbeat and breathing now stopped they were not necessarily dead. Heartbeat could be started again and breathing maintained by mechanical respiration. This was good news in a lot of cases, but it gave rise to some obvious problems. It meant that a growing number of patients emerged, mainly with severe brain damage, whose heartbeat and breathing were supported artificially but they were unconscious and, in many cases, completely unresponsive and not going to recover. What was the status of such patients? Were they, for example, dead? In the 1950s and early 1960s, the widespread view was that, by virtue of the cardio-pulmonary standard, they were not dead, although they were dying. They could be declared dead once their heartbeat and breathing ceased. But this took time. The number of such patients increased steadily, placing increasing demands on hospitals.
In 1968, in the US, a landmark attempt was made to address this ‘problem’ by a group at the Harvard Medical School. They proposed that any patient who was in what they called an ‘irreversible coma’ should be regarded as dead. They proposed a number of medical tests of irreversible coma, essentially showing a complete lack of responsiveness and complete absence of reflexes; an encephalograph could optionally be used to verify lack of brain activity. The aim of such tests was to pick out patients who were believed to have a ‘permanently non-functioning brain’, that is, their whole brain had ceased functioning. Such patients were considered ‘whole brain dead’ and thus dead. Therefore, life support could be removed, for a dead person does not need life support. Such patients could also be made available for organ harvesting for transplantation.
But this proposal had a number of conceptual flaws. To begin with there was a verbal infelicity. If a person is in a coma, even an irreversible one, they are not dead. In the 1950s, some French neurologists had referred to such patients as ‘coma depassé’ or ‘beyond coma’, which was more to the point. But perhaps this was only a verbal problem. A bigger problem was that there were many patients who did not have ‘whole brain death’ but who were in what seemed irreversible coma. These patients had lost higher brain function, that of the cerebrum, but had retained lower brain-stem function. Some of these patients could even breathe unassisted, as breathing is mediated by the brain-stem, but they were in irreversible coma, and would not recover. According to the Harvard Medical School criteria such parents were not dead, for their whole brain was not dead, even though they were in irreversible coma. So the Harvard criteria were not precisely a sufficient condition of death, as the committee had presumably intended it to be. From 1972, these patients would be usually be described as being in a persistent, vegetative state (PVS).
By 1981, a President's Commission on brain death in the US slightly modified the findings of the Harvard Committee, but still essentially held that death was either (a) the irreversible loss of cardio-pulmonary function or (b) the irreversible loss of all brain function. They were, however, careful not to refer to death as irreversible coma, but as death. They also made it clear that patients in a persistent, vegetative state were not dead, and could not have their life support turned off.
A notable and later controversial aspect of the approach of this Commission was to regard the brain as the primary integrator of the body's main systems, and to characterize whole brain death as leading, critically, to the loss of this integration. Whole brain death did not mean that every cell in the brain had ceased to function, only that its crucial function of managing the integration of bodily systems had now ceased. The patient was dead primarily because they could now be regarded as only a collection of unrelated systems: they had lost what became known as ‘somatic unity’. They might be breathing, they might have heartbeat and other biological functions might still be taking place, but these functions – in the absence of a working brain – were no longer joining together in a way that constituted a human life. So, they were dead.
In the UK, a similar approach emerged except that brain-stem death rather than whole brain death was taken as sufficient for death, given that the integrating functions of the whole brain were believed to depend on an intact brain-stem. Loss of brain-stem function seemed to lead to the most profound disintegration of somatic unity. Overall, the new view, which became widely shared throughout the world, was that death should now be equated with either permanent cardio-pulmonary cessation or permanent ‘brain death’ and the loss of somatic unity that followed from this.
But two philosophical concerns tugged at either end of the new idea that a person whose brain was dead was dead. One, notably advocated by the neurologist Alan Shewmon, as well as many others, felt the new approach went too far, and remained sceptical that whole brain death was tantamount to death. Shewmon wanted to say that someone whose brain had died was probably dying, but not yet dead. The other concern was pressed by, among others, the moral philosopher and campaigner Peter Singer, who initially felt the new approach did not go far enough. Singer, as a tough, atheist utilitarian, who judged moral positions only by their consequences, was sceptical of absolute moral truths and was opposed to what he saw as the old-fashioned view that human life was sacred no matter what form it took. Being notorious for once saying that a young foetus had the moral status of a prawn, he was prepared to take a hard look at patients with severe brain damage. He was at least initially attracted to the view that even patients with only higher brain death, and in a persistent vegetative state, could be treated as dead, for in them there was arguably the complete loss of the person, and what was death if it was not the complete loss of a person? He posed sharply the question: what was the moral difference between such patients and those with whole brain death?
Of these two alternative views to whole brain death as death, the higher brain death view had least traction. That higher brain death should be treated as death was widely discussed, but by the time of a second President's Council on brain death in 2008, it was clear that the implications of this idea were too uncomfortable for most. If absence of personhood constituted death then it seemed to follow that ‘diminished persons’, such as those suffering severe dementia and even newborns, could count as dead. Medical professionals were wary of such murky philosophical waters. Clinical doubts about the difficulty of diagnosing mere higher brain death also militated against acceptance of this as a criterion of death. And even Singer ultimately came to feel that there was something unappealing, for example, in agreeing to bury a person with a beating heart even with no higher brain. What all this anxiety revealed was an instinctive sense that in order to be dead there was a need for a human being to be showing as few signs of life as possible. Certainly, a dead person could not have a spontaneously beating heart.
This anxiety linked directly to the concerns that neurologists like Alan Shewmon were having with the view that whole brain death was sufficient for death. Shewmon's research had shown him that in patients with whole brain death (which he less tendentiously called whole brain infarction) a wide variety of important somatic functions might continue while the patient remained on life support. Activity in the spinal column, for instance, could be sufficient to maintain vascular tone and thus blood pressure. Food and fluid, if provided, would be digested and processed by the body and used to ensure cell growth and repair. The body removes waste. The liver detoxifies blood and maintains homeostatic balance. Kidneys manage fluid and electrolyte levels. The immune system fights infection. Wounds heal. Thus the ‘dead’ brain sat in what, in many ways, was a still living body. The 1981 Commission had argued that the key consequence of whole brain death was the loss of somatic unity. Shewmon argued, however, that the continuation of these many interrelated functions in a person with whole brain infarction pointed to something that could not be denied to be somatic unity. This level of somatic unity in a human being was also not actually dependent on the brain as an integrating organ. The brain did provide ‘fine-tuning’, but it was not the primary source of somatic unity. The source of this was something associated with the entire organism. For this reason, the view of the 1981 President's Commission, that death followed whole brain death primarily because of the loss of somatic unity, was mistaken. Shewmon's views were largely accepted in the 2008 President's Council report.
However, that report still pressed a view that what is now more carefully called ‘whole brain failure’ was death. While the loss of somatic unity was now agreed not to be the key consequence of whole brain failure, the ceasing of what was referred to as the body's ‘vital work’ was seen as crucial. The body's vital work was its ability to respond to the external environment and its ‘appetite’ to breathe spontaneously. While a high degree of somatic unity was now accepted as present in many patients with whole brain failure, it was proposed that those who showed no response at all to external stimuli and could no longer breathe spontaneously were dead. In the UK, the 2008 Code of the Royal Colleges effectively embodied these criteria of death in its formal tests for brain-stem death. Tests for responsiveness should be done and finally an apnoea test should be performed which would test the patient's capacity to breathe without support. If patients with brain-stem damage failed these tests they were officially dead.
Doctors caring for Archie Battersbee were convinced quite soon after he was admitted to hospital that he was probably brain-stem dead, based on scans of his brain. Archie's family were sceptical that brain death was death. They took the view that only cardio-pulmonary death was death. They refused permission to withdraw his life support. They shared Alan Shewmon's scepticism that whole brain death was necessarily death and, interestingly, Shewmon was called as a witness to the court hearing considering whether Archie was dead. Under the UK Code, however, if it could be shown that Archie was brain-stem dead then this would have permitted the hospital to remove his life support. Shewmon's scepticism would have no legal weight in the UK.
The hospital proposed to do the formal test for brain-stem death, in line with the Code. However, Archie's family declined permission for this test on the grounds that the part that tested Archie's capacity to breathe spontaneously, which would have required temporary removal from life support, posed a risk. The hospital therefore applied to the court for permission to do this test, which was granted.
Bearing in mind the philosophical curve which Peter Singer had travelled in relation to this kind of case, one might wonder why the hospital was so preoccupied with the question of whether Archie was dead or not. Singer's view, backed by the decision in the Anthony Bland case, was that whether a patient was dead or not was not actually relevant to the question of whether their life support should be withdrawn. The only question was what was in the best interests of the patient. Why did the hospital not simply apply to the courts for a best interests hearing at which they could argue for permission to remove Archie's life support?
The team caring for Archie had little doubt he was brain-stem dead. Their view was that the brain-stem death test would show this and that this would lead to a prompt resolution of the case. The question of whether a patient like Archie was dead could therefore have a practical relevance, making it reasonable to defer the question of what was in his best interests. Interestingly, the question of Archie's status and his best interests were combined in the view of his court guardian – the person appointed by the court to protect the legal interests of a minor – that it was in his best interests for the hospital to know whether he was dead or not.
However, while they obtained permission to do the formal brain-stem death test, the hospital could not complete the test for technical reasons. This meant that according to the accepted medical Code, Archie could not be declared dead. At this point, the hospital submitted a curious application to the court. They asked the court, in the absence of the Code test, to declare Archie dead on the basis of diagnostic scans already obtained. This was unprecedented. Curiously, they also asked the court, if it could not declare Archie dead on the basis of these scans, to judge that it was in Archie's best interests to remove his life support. The question of best interests had now been raised, although it still remained embroiled in the question of Archie's status.
As was evidenced by the judgment of the Appeal Court, what the High Court should have done was reject the application to determine that Archie was dead in the absence of the formal Code test. Following Singer's philosophical curve – although the Appeal Court did not refer to Singer – they might have realized that while it had not been formally possible to declare Archie dead, this was not actually necessary in order to determine what was in his best interests. Nonetheless, the first court went ahead and did declare that Archie was dead on the basis of the diagnostic scans. More curiously, they also determined that even if they had not reached this judgment, they would have decided it was not in Archie's best interests to remain on life support.
What might be learned from all this? The most striking lesson is that it remains possible for an English court to declare you dead when you are not. In addition, it remains possible for an English court to do this by ignoring the formal medical Code that says that it should not. This is partly because there is no statutory definition of death in the UK. Another lesson is that if you have whole brain failure it remains possible for an English court to declare you dead, even while philosophers and medical professionals continue to debate whether whole brain failure is tantamount to death. Perhaps this last lesson is the least surprising. Legal authorities will stress that they are not philosophers, or neurologists, and there are pressures on them to make practical decisions while theoretical debate rages about them.
After the Appeal Court hearing determined that Archie was not, after all, dead, there was a best-interests hearing at which it was determined that it was in his best interests to remove life support. It is quite likely that even Alan Shewmon may have agreed with this judgment given the evidence that Archie's somatic unity was rapidly deteriorating. Shewmon might even have agreed that by this criterion – of very severely reduced somatic unity – Archie had already died. Peter Singer, of course, might well wonder why the hospital and the courts became so concerned about proving that Archie was dead. Life support had been removed from Anthony Bland even though he was alive, and Archie's brain damage was much more severe than that of Anthony Bland. A best-interests hearing when this was known would have presumably come to the same conclusion.
