Author: John O’Dowd, School of Law, University College Dublin
Finlay CJ formulated the crux of the Supreme Court’s decision in Attorney General v X as follows:
‘[I]f it is established as a matter of probability that there is a real and substantial risk to the life, as distinct from the health, of the mother, which can only be avoided by the termination of her pregnancy, such termination is permissible.’ ([1992] 1 IR 1, 53-54).
An obvious question is: what does the word ‘substantial’ mean? (‘real’ appears to exclude risks that are theoretical because they do not actually arise on the facts of the case, even though sometimes associated with a condition that is present.)
The Expert Group on the Judgment in A, B and C v Ireland reached the following view about whether legislation should further clarify this or other aspects of the test:
‘Although the medical decisions may be difficult in particular cases, the complexities will not arise from the words of the test but from diagnostic and treatment issues. Implementing the decision does not, therefore, require another definition of the test. Neither is it necessary or desirable to seek to explain it with synonymous terms.’ (para 6.2)
Dr Rhona Mahony, master of the National Maternity Hospital, made this submission to the Joint Committee on Health and Children:
‘[On] the highly difficult issue of defining what is a substantial risk to life during pregnancy … [is] it a 10%, a 50%, an 80% or a 1% risk of dying? The interpretation of risk is not the same for all people. A woman herself will have a view as to what is an acceptable risk of her dying during pregnancy. Her opinion deserves to be afforded consideration. Clinical flexibility, supported by law, is required. Doctors must be able to make sound common-sense medical decisions based on medical conditions and circumstances and not ever on ideology or philosophy.’ (8 January 2013)
From a different point of view, Deputy Peadar Tóibín also raised the issue of ‘quantification’ of risk—would a 3% risk of suicide be ‘substantial’, for instance? (9 January 2013)
Several lawyers making submissions to the Committee rejected the idea that ‘substantial’ can be reduced to percentages; doctors routinely perform assessments without doing so.
Although any legislation will likely not try to elaborate on what ‘substantial’ means, one should consider some relevant questions that remain open.
Does ‘substantial’ imply the same level of risk, expressed in percentage terms or not, in all circumstances?
Firstly, ‘the matter of probability’ required concerns the existence of the risk of death, not death as outcome. Furthermore, the majority in AG v X deliberately contrasted ‘real and substantial risk’ with ‘an inevitable or immediate risk to the life of the mother’, a criterion firmly rejected. (Although, as the Expert Group recognised, in cases of an immediate risk to the mother’s, a different legislative approach is required.)
To distinguish suicide risks from other situations, Finlay CJ observed in AG v X—
If a physical condition emanating from a pregnancy occurs in a mother, it may be that a decision to terminate the pregnancy in order to save her life can be postponed for a significant period in order to monitor the progress of the physical condition, and that there are diagnostic warning signs which can readily be relied upon during such postponement. (p 55)
It would be wrong to infer that all life-threatening physical conditions fall into this category. As recent events have highlighted, in several situations a termination could not prudently be postponed while more information is gathered and assessed about the actual level of risk. In these cases, Finlay CJ’s reasoning implies that the decision to terminate can be made at a point such intervention is likely to remain effective in obviating the risk and that ‘substantiality’ of the risk can be assessed on a precautionary basis; a lower level of risk will thus suffice than is required in cases with a possibility of ‘wait and see’.
Can the woman’s opinion of what is an acceptable risk be given much weight in favouring a decision to terminate?
Although occurring in the dissenting judgment in AG v X, there is little reason to doubt that Hederman J’s characterisation of the basic effect of the Eighth Amendment remains sound: ‘no recognition of a mother’s right of self-determination can be given priority over the protection of the unborn life. … [It is] the mother’s duty to carry out the pregnancy …’ (p 72). Despite the ‘freedoms’ conceded by the Thirteenth and Fourteenth Amendments, it is hard to see what else the ‘right to life of the unborn’ can mean. While the mother’s views are, of course, a vital factor for doctors to consider, perhaps the Eighth Amendment does still intrude its ‘ideology or philosophy’, preventing a ‘common-sense’ consideration of her view of the acceptable level of risk of her own death becoming a back-door form of the self-determination that this Amendment fundamentally repudiates.
Two other brief concluding remarks. Firstly, although ‘real and substantial risk’ to the mother’s life is the touchstone for the lawfulness of a termination, there is also the requirement that, even where that risk exists, it is only if there is no other means of avoiding that risk that the termination will be lawful. Secondly, as to both that requirement and the concept of substantial risk, not everything said in AG v X should be treated like a statutory provision or even a rule of law rather than a factual generalisation forced on the court by the paucity of evidence. One subsequent High Court decision, Cosma v Minister for Justice, Equality and Law Reform [2006] IEHC 36 (15 February 2006) was more sceptical both of the nature of the suicide threat (there made in response to the prospect of deportation) and the practicability of guarding against it. This should remind us that if legislation is able to place these decisions effectively in the hands of doctors, it is likely to be because of the courts’ reluctance to act as an ‘abortion licensing agency’ rather than a direct endorsement of the decisions taken.