Determining the quality of someone’s life is nigh impossible without first making some basic assumptions. First we must assume that another mind exists, and that this mind can exist in more than one mental state. Next, we must ascribe positive value to some ideal, attainable mental state and negative value to some unideal, attainable mental state. Finally, we must claim to be able to deduce whether another mind is in one of these aforementioned ideal or unideal mental states.
In the case of persons who have at least some basic capacity to communicate and understand their surroundings, we can have them communicate their mental state to us as being ideal or unideal. The newborn infant, however, has a very limited capacity for communication (through crying, smiling, laughter, etc.) and cannot fully understand their surroundings. This does not, however, mean we can know nothing about their mental state, and certainly does not warrant a default presumption of an ideal mental state and therefore an adherence to a sanctity-of-life principle in potential end-of-life decision-making.
To start with, as described by Eduard Verhagen’s 2005 article (p. 959) on the Groningen Protocol for Euthanasia in Severely Ill Newborns (which is actively used in the Netherlands and a basis for practice elsewhere), we can know a lot about a newborn’s probable mental state by observing “different types of crying, movements, and reactions to feeding. [Which allow us to create p]ain scales for newborns, based on changes in vital signs (blood pressure, heart rate, and breathing pattern) and observed behavior, […] to determine the degree of discomfort and pain.” Since persons who can communicate unideal or ideal mental states can have known vital signs, and these signs repeatedly correlate with these particular states, we can translate a great deal of these medical facts (alone) into what we have good reason to suspect as particular mental states in the infant.
We can also consider future medical facts and likely future quality of life using a similar method. For example, for a newborn with severe spina bifida we might be able to determine whether curvature of the spine is likely and if so how severe it would be. This would tell us whether the newborn would develop into a person who could make use of a wheelchair, would likely suffer from epilepsy, or would have normal brain development or develop a cognitive disability which might compound existing problems if they kept living, etc..
These sorts of facts were employed by medical and legal professionals in the case of ‘Baby M’, who was born in Melbourne, Australia, in July of 1989 and lived just 12 days (Kuhse, Quality of Life and the Death of ‘Baby M’, 1992, p. 234). Like many similar cases in which neonatal euthanasia and potential end-of-life decisions have arisen, Baby M was born with a very severe case of spina bifida. The treating doctors, in conjunction with the parents, decided that the newborn’s quality of life was so terrible that they should opt for a ‘conservative’ treatment regime whereby their primary focus was to treat only the symptoms and otherwise keep the newborn comfortable. However, sanctity-of-life activists attempted to personally intervene by appealing to the parents. When their intervention was unsuccessful they notified police, who subsequently launched an investigation but became satisfied that the treatment regime was appropriate. The state’s coroner agreed with the police assessment, and explicitly stated that quality of life was a relevant factor in her conclusion (Kuhse, p. 241).
This seems like a thoughtful and, for that reason, laudable conclusion given the terrible diagnosis of the infant and the demonstrably poor current and likely future quality of life. As Loane Skene argued (The Quality of Life and Disabled Infants, 1992, p. 998): “If, on the basis of the current diagnosis of an infant’s condition, the clear prognosis is that the infant’s quality of life is likely to be so ‘demonstrably awful’ that no reasonable person would choose to live it, then active treatment need not be given to prolong the infant’s life [although treatment to alleviate pain and distress should still be given].”
However, if an infant’s quality of life is indeed this ‘demonstrably awful’, it might not only be justifiable to withhold life-prolonging treatment, but in fact it could be a moral obligation to bring about its death as soon as possible. If we agree that such a quality of life is so unideal that “no reasonable person would choose to live it” (Skene, p. 998), then what would make a reasonable person choose to linger in that state if they were already in it? The only scenario in which such a reasonable person (if we agree with Skene’s ‘reasonable person’ argument) might linger in it is if they did not have the personal capacity to bring about their own death. This is the case for the newborn. That the newborn happens to lack the capacity to consciously choose to bring about its own death is a mere accident of circumstance, and we should not allow it to block a less egregious outcome.
Imagine that someone has just fallen into a large, active incinerator. Neither you nor the person who has fallen in can shut off the incinerator or get the person out. They will now certainly die an agonising death over the next few minutes. Next to you are two loaded guns – one which contains bullets and another which contains tranquiliser darts. Shooting a bullet will kill the person immediately whereas shooting the tranquiliser dart might ease the person’s suffering, however you cannot be completely sure. Do you shoot either of the guns or do you do nothing at all? Almost no one could morally defend not at least shooting the tranquiliser dart, I think, since it might at least ease the person’s pain. However, what if the tranquiliser dart is ineffective and the person continues to writhe in pain? Surely then we should shoot the gun with bullets. In this analogy, we should view terminally-ill newborns who are in extreme suffering like we view the person who falls into the incinerator. Further, we should view it as a moral obligation to, if the darts don’t sufficiently reduce the person’s pain, use the gun with bullets.
Given that it is possible to accurately infer facts pertinent to a newborn’s mental state, and this mental state is relevant to wellbeing and quality of life, there is no defensible reason to assume an ideal mental state exists and thus adhere to the principle of the sanctity-of-life (especially as a default position). Upon knowing something about the probable mental state, if it is deemed to inflict a terrible quality of life upon the newborn (now, in the foreseeable future, or both), ‘conservative’ treatment regimes like that used with Baby M are justified, if not obligatory.