• children

The four thoughtful commentaries on our feature article draw out interesting empirical and normative questions. The aim of our study was to examine the views of a sample of the general public about a set of cases of disputed treatment for severely impaired infants.1 We compared those views with legal determinations that treatment was or was not in the infants’ best interests, and with some published ethical frameworks for decisions. We deliberately did not draw explicit ethical conclusions from our survey findings, both because of the acknowledged limitations of survey methodology, and because survey conclusions cannot, in themselves, yield answers about what the right threshold should be for providing or withholding treatment.2

In this brief response, we are going to address head-on the important ethical question raised within our survey – when life is worth living for an infant. We follow-up on the suggestion of two commentators that the presence or absence of “relational potential” might be ethically important to report in studies of the outcome of severely impaired infants,3 and to whether parental requests for treatment should be supported.4

The notion of “relational potential” was introduced by John Arras in a 1984 commentary.5 Arras was responding to the Baby Doe Regulations and a report by the President’s Commission on Bioethics that strongly endorsed the ‘best interests standard’.6 Arras’s view was that in applying a best interests standard to ‘extreme cases’ of children with profound cognitive and sensory impairment, “the best interests standard ha[d] been pushed beyond the pale of its capabilities”.5 Instead of asking whether it would be best for a child like Charlie Gard or Alfie Evans to live or die, Arras contended that clinicians and families might ask whether the child has at least the minimum capacity to form human relations.

But what level of function is sufficient to ground relational potential? Wightman et al in their commentary4 cite a recently published account of an “expanded relational potential standard”. This expanded account, drawing on care ethics, recognises that parents may establish caring relationships even if that relationship is largely or entirely one-sided, and the “the child has limited or no capacity to reciprocate”.7

If relational potential is synonymous with the capacity to be cherished by care-givers, to be related to, then all of the cases in our survey would appear to have relational potential. Indeed, even children who meet neurological criteria for death (like teenager Jahi McMath), would appear to have relational potential in this sense.

However, we would suggest, with respect, that such an account confuses the capacity to relate with the capacity to be related to, it confuses the question of value to the individual with the question of value to others. Arras had in mind that the capacities for “self-consciousness and relating to other people” were distinctly human capacities that gave biological human life value. We concur. Although we do not have space here to fully articulate or defend such an account, we propose the following necessary and sufficient criteria for a Relational threshold.

Relational threshold: For biological human life to have significant value1 to the individual, they must have the following capacities

1. Self-consciousness - The minimal capacity to be aware of themselves2

2. Other-consciousness - the minimal capacity to be aware of others

3. Relatability – The desire and capacity to reciprocate and interact with others

4. Memory – the minimal ability to maintain and sustain interactions

There is more to be said about when this threshold would apply, but here we suggest that the cases from our paper of ‘unaware’, ‘possible awareness’, ‘minimal cognition’ and ‘significant burden’ would potentially fall below the Relational threshold. A child with complete locked-in syndrome might lack the ability to reciprocate and hence relate to others. However, they still arguably have the capacity to relate, and given the possibility of communicating by eye movement/finger movement (or even via neurotechnology), we would argue that many children with locked-in-syndrome would be above the threshold.

Where a child is so profoundly impaired that their life is or will be below the Relational Threshold, it does not necessarily mean that their life should end. There may be reasons, in terms of the interests of others, to sustain that life. For example, we fully accept that the child in a persistent vegetative state may remain a cherished member of a family. The family may continue to relate to the child, even if he or she cannot relate back. The parents may have a strong desire for the child’s life to continue, and that desire, itself, gives us a prima facie ethical reason to keep them alive. However, we should be clear that this reason – in terms of the interests of others - is different from a reason for continuing treatment for the sake of the child. What is more, this ethical reason may be outweighed.

First, where the child’s life is below the relational threshold and there is evidence that they are experiencing (and will continue to experience) pain, we suggest that continuing to sustain life may cross the harm threshold.8 In such a situation, parents’ wishes should potentially be overruled.

Second, there are very difficult questions about how to allocate resources within publicly funded healthcare systems. However, we suggest that there is no plausible justification for expending scarce or limited public healthcare resources to sustain biological human life in the absence of relational potential.