. 3
( 12)


Negligence and abortion

A duty to the mother
Very few cases of a simple failed abortion due to obstetric negligence are
reported. The majority that are will either fall into the categories of
wrongful pregnancy or of wrongful birth “ conditions which are described
and discussed in Chapters 4 and 3 respectively. Occasionally, however,
the fact that a pregnancy has not been terminated goes by unnoticed at
the time and these form something of a separate category of unwanted
pregnancies “ if for no other reason than that the fact of pregnancy itself is
appreciated only relatively late. This may mean that the deadline beyond
which a remedial termination that can be justified under section 1(1)(a)

Smeaton, n. 98 above, at [126].
It is clear that, if one transposes Munby J™s figures to Dhingra, the case for withdrawing
the charge in the latter case is far from obvious “ although, as argued above, a conviction
would have been unlikely. But it does illustrate the importance of the improvements in
medical science over a period as short as a decade.
[1969] 1 QB 541, [1968] 2 All ER 282.
Voluntary and involuntary termination of pregnancy 39

has passed; equally often, the pregnancy has reached such a stage that the
woman is no longer prepared to proceed to termination.
Just such a situation occurred in Scuriaga v. Powell109 which is an old
and poorly reported case but is, nevertheless, important in the present
context for two reasons. First, it laid the grounds, albeit tenuously, for
allowing recovery for the uncovenanted birth of a healthy child and,
second, it established that a doctor whose negligence has caused such a
pregnancy cannot rely on the woman retrieving the situation by way of
termination and we will see in Chapter 4 how significant this can be.
Failure to carry out an abortion within the parameters of good medical
practice will, of course, be judged on the normal criteria for establishing
negligence. Thus, in the County Court case of Chissel v. Poole Hospital
NHS Trust,110 the obstetrician failed to remove a twin whose presence
was unsuspected. The court held, however, that there was nothing to
suggest, or to lead Ms Chissell™s doctors to suspect, that she was bearing
twins and the chances of one twin fetus surviving lawful abortion were
slim; the case, however, failed mainly on the grounds that the experts
called disagreed as to whether or not the follow-up of the case satisfied a
˜reasonable degree of care™ “ it, thus fell foul of the Bolam test.111 The fact
that Ms Chissell refused a second termination on the grounds that it was
now too late to use the simple technique of endometrial suction did not
count against her. While one may regret the judge™s enforced position in
such a case “ and there is little doubt his hands would be tied by the
jurisprudence of the time112 “ the principles as to a duty to the mother are
reasonably clear.113 What is less clear is whether he has any duty to the
fetus. This secondary question was fully examined in the interesting
Canadian case of Cherry v. Borsman.114

[1979) 123 SJ 406. 110 [1998] Lloyd™s Rep Med 357.
Bolam v. Friern Hospital Management Committee [1957] 2 All ER 118, (1957) 1 BMLR 1.
The current significance of the case has already been discussed in Chapter 1. The
corresponding test in Scotland is to be found in Hunter v. Hanley 1955 SC 200, 1955
SLT 213.
Maynard v. West Midlands Regional Health Authority [1985] 1 All ER 635, [1984] 1
WLR 634.
The result can be compared with that in a recent Canadian case: Roe v. Dabbs 2004
BCSC 957. Here, the woman already had an IUD embedded in the uterine wall. The
Court considered that the obstetrician should have, as a result, been particularly careful.
In particular, the doctor failed to meet the necessary standard in failing to consider the
negative post-operative pathology report “ no products of conception were discovered in
the curettings. The duty of care owed to the patient was, thereby, breached. For further
illustration of the importance of the histology report, and the additional question of
negligence on the part of the pathologist, see Crouchman v. Burke (1998) 40 BMLR 163.
(1990) 75 DLR (4th) 668, BCSC; (1992) 94 DLR (4th) 487, BCCA.
40 The troubled pregnancy

Here, a negligently performed abortion, followed by a remarkably lax
follow-up, resulted in a severely disabled neonate. It was held, and con-
firmed on appeal, that, insofar as a negligently performed abortion can
cause foreseeable harm to the fetus, the practitioner owed a duty to the
fetus to prevent that harm; accordingly, he was liable to the neonate in
negligence. This case appears to be unique. No-one would deny that the
doctor caused the injuries but he did so in the process of trying to kill the
fetus. Given that the law considers the greatest injury to be death, it is a
little difficult to see how mitigating that injury can be tortious “ or,
indeed, how the traditional approach to compensation of restoring the
plaintiff to the position he or she would have been in but for the negli-
gence would be anything but absurd. An analysis of the case leads
strongly to the conclusion that the court in Cherry was anxious to recom-
pense the newborn child and was, therefore, concerned to avoid the case
being labelled as one of so-called ˜wrongful life™ when, as we will see later,
recovery would have been very difficult. The decision in Cherry may, thus,
be seen as an ingenious piece of legal legerdemain and we will return to
the case in detail in Chapter 6.
The idea of a duty to the fetus raises a conceptual problem insofar as a
dead fetus cannot bring an action; a doctor can only be liable to the fetus if
it survives as a neonate.115 If, therefore, we are speaking about legally
actionable fetal injury we should, in fact, be referring to some such entity
as the ˜neonate-to-be™. It is important that we bear in mind the well-
known fact that the fetus has no positive legal rights.116 The words of Sir
George Baker in Paton:
[t]here can be no doubt, in my view, that, in England and Wales, the foetus has no
right of action, no right at all, until birth117

have been followed ever since in virtually every Anglophone jurisdic-
tion.118 It is for this reason that I have couched the discussion of the
morality of abortion in terms of fetal values. This lack of rights depends on

And, in England and Wales, it must survive for a minimum of 48 hours: Congenital
Disabilities (Civil Liability) Act 1976, s.4(4).
As we have seen, the viable fetus has a negative right not to be killed intentionally: Infant
Life (Preservation) Act 1929.
Paton v. British Pregnancy Advisory Service Trustees [1979] QB 276 at 279, [1978] 2 All
ER 987 at 989. For Scotland, see Kelly v. Kelly 1997 SC 285, 1997 SCLR 749, per
Cullen LJ-C at SC 291, SCLR 755.
In England, the importance of avoiding maternal/fetal conflict was first emphasised in
Re F (in utero) [1988] Fam 122, [1988] 2 All ER 193. But, even so, circumstances alter
cases; Re F was, for example, specifically rejected in the remarkable New Zealand case
Re an Unborn Child [2003] 1 NZLR 115 where it was proposed that the birth of a child be
included in a pornographic film.
Voluntary and involuntary termination of pregnancy 41

a legal denial of personhood for the fetus which is, thereby, also denied
protection against criminal or negligent activity “ at least in the pre-viable
stage. There may, therefore, be times when the mother is deprived of a
child she wishes to keep and, while this is, strictly speaking, distinct from
the abortion debate, it is certainly an aspect of troubled pregnancy which
merits discussion at this point.

Involuntary termination of pregnancy
We have seen that the fetus that dies in utero as a result of criminality or
negligence has no right “ or, indeed, possibility “ of action per se; more-
over, since it cannot have either a guardian or an executor, no action can
be taken on its behalf. At the same time, most people would, it is imag-
ined, hope that the distress of the deprived mother would be adequately
In practice, this is remarkably difficult and the courts have to use all
their wiles to achieve a fair result. The case of Bagley119 is very much in
point. Here the pregnancy appears to have been complicated by Rhesus-
incompatibility but the mother was negligently denied an early induction
and was delivered of a stillborn child. Damages under the Fatal Accidents
Act 1976 were expressly disallowed because the child was not live-born
but Simon Brown J was able to recognise the loss of satisfaction in bring-
ing up a healthy child as a recoverable head of damages. Recompense
under the other heads that he identified “ namely the consequent reduced
ability or inability to have another child and a recognisable illness brought
about by the misfortune “ would, however, be contingent upon the
precise conditions of the case. Thus, maternal recovery for the ˜wrongful
death™ of her fetus will, at best, depend on something of a lottery.
The difficulty was exemplified by the New Zealand Court of Appeal in
the five-judge hearing of Harrild v. Director of Proceedings.120 The precise
nature of the case concerned the interplay of the New Zealand Health
and Disability Commissioner Act 1994 and the Injury Prevention,
Rehabilitation, and Compensation Act 2001 but, for present purposes,
the issue can be seen as whether or not the death of a fetus in utero due to
medical negligence could be construed as compensatable injury to the
mother who was, in fact, unharmed. Elias CJ concluded that severance of
the physical link between mother and fetus that occurs when the fetus dies

Bagley v. North Herts Health Authority [1986] NLJ Rep 1014. Bagley is, in fact, virtually
the only UK case available; Grieve v. Salford Health Authority [1991] 2 Med LR 295 is
similar but is less instructive.
[2003] 3 NZLR 289.
42 The troubled pregnancy

in utero constitutes a personal injury to the mother “ and this was the
majority view.121
There is no doubt that, underlying the opinion in Harrild was an
anxiety to preserve what was described as the ˜non-niggardly™ attitude
to compensation that has been established in New Zealand and which
most people would, one feels, applaud, particularly in the context of the
death of a baby due to negligence. The alternative logical conclusion that:
[t]o deny a stillborn recovery for fatal injuries during gestation . . . would make it
more profitable for the defendant to kill the plaintiff than to scratch him122
is hardly an attractive construct “ albeit posing a legal conundrum that
persists in the United Kingdom. The almost unique decision in Amadio
was made in very specific circumstances and, insofar as it is directly
in contrast to the House of Lords decision in Attorney-General™s
Reference (No. 3 of 1994),123 it is very unlikely that it could be maintained
in the United Kingdom. But the opinion in the latter case derived from
a consideration of the criminal law and it is that to which we must
now turn.

Fetal death, fetal status and the criminal law
Attorney-General™s Reference No. 3 concerned the case of a man who
stabbed his pregnant girlfriend; at the same time the knife penetrated
the abdomen of her viable fetus.124 Following a premature delivery, the
baby died at the age of four months and the man was charged with its
murder although it was agreed that he had no intention to kill the fetus.
Inevitably, the trial judge ruled that there was no case to answer. When
the case was referred to the Court of Appeal, however, it was held that, in

The dissent, of course, maintained that the fetus and the mother must be regarded as
separate entities in both biological and practical terms.
Amadio v. Levin 501 A 2d 1085 (Pa., 1985). ˜Wrongful death™ statutes are in place in
most of the United States but their interpretation seems variable. In general, recovery
will be limited to the death of a viable fetus. Non-viable fetuses are protected by statute
in Georgia, Louisiana, Illinois, Missouri, West Virginia and South Dakota: see Santana
v. Zilog, Inc. (1996) 95 F 3d 780.
[1996] QB 581, [1996] 2 All ER 10, CA; [1998] AC 245, [1997] 3 All ER 936, HL.
A similar case occurring in the United States would now be covered specifically by the
Unborn Victims of Violence Act 2004.
I have discussed the case briefly in J. K. Mason, ˜A Lords™ Eye View of Fetal Status™
(1999) 3 Edinburgh Law Review 246“50. For discussion of the Court of Appeal decision,
see Mary Seneviratne, ˜Pre-natal Injury and Transferred Malice: The Invented Other™
(1996) 59 Modern Law Review 884“92 and, for the House of Lords, Sara Fovargue and
Jose Miola, ˜Policing Pregnancy: Implications of the Attorney General™s Reference (No. 3
of 1994)™ (1998) 6 Medical Law Review 265“96.
Voluntary and involuntary termination of pregnancy 43

the circumstances, it was possible to charge either murder or manslaugh-
ter. This decision was reached through a rather complex biphasic appli-
cation of the doctrine of transferred malice which depended, in the first
phase, on the understanding that the fetus was an integral part of the
mother. As a consequence, serious injury to the fetus constituted a serious
injury to the woman “ just as, as the Lord Chief Justice put it, ˜to injure
her arm or her leg would be so viewed™.125 In the second phase, the malice
towards the mother could be, again, transferred to the fetus on its birth.
As a result, the conditions for murder were established despite the fact
that, at the time of the assault, the fetus was not a legal ˜person™.
The result was widely interpreted as, at last, providing some protection
for the fetus in the criminal law. If it did, it was a Pyrrhic victory as the
protection offered depended upon denial of any autonomous fetal exis-
tence. Moreover, as was only to be expected, such advantage as was
conferred, applied only to the neonate “ the non-viable fetus in the
womb could still be killed with impunity.126
The theory of a double transfer of malice was, in any event, too much
for the House of Lords which rejected it out of hand and held, instead,
that the relationship between mother and fetus was one of bonding rather
than identity. Indeed, Lord Mustill, who wrote the most significant
opinion, pointed out the obvious “ that, at the very least, they have
distinct genetic profiles. The mother and the fetus were distinct human
beings “ albeit that the latter did not have the attributes of a person. Then
what was it? It was, said Lord Mustill, a unique organism and, ˜to apply to
such an organism the principles of a law evolved in relation to autono-
mous human beings is bound to mislead™.127
Looking back at the cases discussed in the last few pages, it is fascinat-
ing to see how this impasse of fetal identity can be manipulated according
to need. The court in Harrild used the concept of bonding to establish a
unity between mother and fetus; the House of Lords, on the contrary, saw
a bonding relationship as negating one of common identity. The Harrild
interpretation, albeit provided by the Chief Justice with obvious reluc-
tance is, in my view, to be preferred insofar as it provides something of a

[1996] 2 All ER 10 at 18. As has been noted, this view was rejected by the New Zealand
Court of Appeal in Harrild, n. 120 above, where the concept of a ˜union™ between two
saprophytic entities was preferred.
A charge of malicious wounding would be competent in the event of neonatal survival:
see Lord Mustill in Attorney-General™s Reference (No. 3 of 1994), n. 123 above, at AC 254,
All ER 942. A civil action might also be available to the child building on Burton v.
Islington Health Authority, de Martell v. Merton and Sutton Health Authority [1993] QB
204, [1992] 3 All ER 833.
[1998] AC 245 at 256, [1997] 3 All ER 936 at 943.
44 The troubled pregnancy

template from which to work. The House of Lords, however, leaves the
fetus in a legal limbo. It cannot be a person because it has no legal person-
ality, nor can it be protected as being part of its mother “ and we have no
clues to the answer because the House singularly failed to elaborate on the
status of this ˜unique being™.128 In fact, it may well be that the only way in
which the fetus can be protected as things stand is to regard it as a unified
part of the mother “ and this, again, may account for the generalised
preference for the approach of the Court of Appeal that is to be found in
Commonwealth decisions.129 In my view, the House of Lords ˜decision™
was, of necessity, based on legal precedents which most people would
regard as being too old to be relevant today; in the end, the conclusion that
the fetus has no rights remains grounded in rules established in the seven-
teenth century.130 Moreover, as we have seen, it was not well accepted in
other jurisdictions. A fresh consideration of the problem was needed, and
an opportunity seemed to present in the most apposite forum when Mrs Vo
took her case as far as the European Court of Human Rights.

The case of Vo v. France131
Mrs Vo was a Vietnamese woman living in France. She was some twenty
weeks™ pregnant when she went to hospital for a routine antenatal
appointment. Following a truly remarkable series of mishaps, including
mistaken identity, she was negligently treated for the removal of a non-
existent contraceptive coil and, during the process, her membranes were
ruptured.132 A week later, it was decided that the pregnancy could no

This is no place to discuss the intricacies of the criminal law but, for the record, the
House refused to consider the question of murder although a conviction for man-
slaughter was possible.
See, for example, R v. King [2003] NSWCCA 399: ˜The purpose of the law is best
served by acknowledging that, relevantly, the foetus is part of the mother™, per
Spigelman CJ at [97]. The position in Canada is similarly pragmatic: ˜It would not
have been illogical to find that bodily harm was done to [the mother] through the death
of the foetus which was inside of and connected to her body and, at the same time, to find
that the foetus was a person who could be the victim of criminal negligence causing
death™: R v. Sullivan and Lemay (1991) 1 SCR 489 per Lamer CJ at 506.
Sir Edward Coke, Institutes of the Law of England. Pt. III, Chap. 7 at 50. It is, however, to
be noted that the House was quite clear that this subsists: ˜It is established beyond doubt
for the criminal law as for the civil law that the child en ventre sa mere does not have a
distinct human personality, whose extinguishment give rise to any penalties or liabilities
at common law™ (per Lord Mustill in Attorney-General™s Reference (No. 3 of 1994), n. 123
above, AC 245 at 261, 3 All ER 936 at 948).
Vo v. France (Application No. 53924/00) (2004) 79 BMLR 71, (2005) 40 EHRR 12. The
paragraph numbers are those in the original transcript.
I have detailed the extraordinary story in J. K. Mason, ˜What™s in a Name? The Vagaries
of Vo v. France™ (2005) 17 Child and Family Law Quarterly 97“112.
Voluntary and involuntary termination of pregnancy 45

longer continue and it was terminated.133 Mrs Vo sued the hospital and
the doctor in both the civil and criminal jurisdictions alleging the uninten-
tional homicide of her child. The most important of the agreed findings at
autopsy were: the age of the fetus was between 20 and 21 weeks; there
were no indications of violence, there were no malformations nor any
evidence that death was due to structural disease or damage to the organs;
and, significantly, that the child had not breathed after delivery. Equally
significantly, pathological investigation indicated that the fetal lungs were
20 to 24 weeks old; the fetus was, therefore, on the fringe of viability.
The case had a turbulent passage through the courts. One of its main
interests lies in the various interpretations of fetal status that were made; it
is worth repeating these in some detail.
The court of first instance134 saw the issue as being that of whether or
not a fetus135 aged 20 to 21 weeks is a human person or ˜another™ within
the meaning of the Criminal Code.136 In reaching its conclusion, the
court made a number of assumptions and assertions which provided the
foundation of the arguments pursued throughout the jurisdictional hier-
archy. The main points were:
* The underlying statutory provisions are that the law guarantees the

respect of every human being from the beginning of its life;137
* There is no legal rule to determine the position of the fetus in law either

when it is formed or during its development;
* It has been established that a fetus is viable at six months and on no

account, on present knowledge, at 20 or 21 weeks;
* The court cannot create law on an issue which the legislators have not

yet succeeded in defining;
* Accordingly, a 20 to 21 week-old fetus is not viable and is not a ˜human

person™ or ˜another™ within the meaning of the Criminal Code.
As a result, it concluded that the offence of unintentional homicide or
unintentionally taking the life of a 20 to 21 week-old fetus had not been

On my reading of the case, the fetus could well have been dead at this time, making the
operation one of evacuation of the uterus. It was, however, considered throughout to
have been a therapeutic abortion. I have some, at least, tacit academic support:
Katherine O™Donovan, ˜Taking a Neutral Stance on the Legal Protection of the
Fetus™ (2006) 14 Medical Law Review 115“23.
Lyons Criminal Court, 3 June 1996.
The English translation of the judgment uses the spelling ˜foetus™. However, to avoid
confusion, I have used the preferred ˜fetus™ throughout.
The French Criminal Code, Articles 221“6 defines the offence of unintentional hom-
icide as ˜causing the death of another™ within the stated conditions.
Article 16 of the Civil Code; Law of 17 January 1975 on the Voluntary Termination of
46 The troubled pregnancy

made out since the fetus was not a ˜human person™ or ˜another™ “ and the
doctor was acquitted ˜without penalty or costs™.
The applicant appealed and was supported by the public prosecutor for
the following reasons:
By failing to carry out a clinical examination, the accused was guilty of negligence
that caused the death of the fetus, which at the time of the offence was between 20
and 24 weeks old and following, normally and inexorably, the path of life on which
it had embarked, there being no medical doubt over its future.
The Appeal was allowed138 and the doctor was fined and given a suspended
sentence of six months™ imprisonment. The Court of Appeal did, however,
in this writer™s opinion, start to muddy the waters by attempting to harmo-
nise the availability of voluntary termination of pregnancy in France before
10 weeks™ gestation with the provisions of, say, Article 2 of the European
Convention of Human Rights and Article 6 of the United Nations
Convention on the Rights of the Child;139 up to this point, abortion had
not been mentioned. Even so, it became central to the discussion later. The
views of the Court of Appeal on ˜viability™ are important in this context:
It follows that, subject to the provisions on the voluntary termination of pregnan-
cies and therapeutic abortions, the right to respect for every human being from the
beginning of life is guaranteed by law, without any requirement that the child is born
as a viable human being, provided it was alive when the injury occurred [my
emphasis] . . . Indeed, viability is a scientifically indefinite and uncertain concept . . .
devoid of all legal effect as the law makes no distinction on that basis. (At [21].)

The Court concluded:
It is not yet known with precision when the zygote becomes an embryo and the
embryo becomes a fetus, the only indisputable fact being that the life process
begins with impregnation
and that, since a previous judgment140 had decided that a non-fatal
wound would have been classified as an offence of unintentionally caus-
ing injuries, a fortiori, ˜an assault leading to a child™s death must be
classified as an unintentional homicide™:
Thus, the strict application of the legal principles, established scientific fact and
elementary common sense all dictate that a negligent act or omission causing the
death of a 20 to 24-week-old fetus in perfect health should be classified as
unintentional homicide. (At [21], emphasis added.)

Lyons Court of Appeal, 13 March 1997.
Signed in New York, 26 January 1990. It is to be noted that the former speaks of
˜everyone™s right to life™ and the latter to ˜a child™s inherent right to life™.
Uncited, Douai Court of Appeal, 2 June 1987.
Voluntary and involuntary termination of pregnancy 47

The judgment of the trial court was, accordingly, overturned.
The case then went to the Court of Cassation where the findings of the
Court of Appeal were summarily dismissed.141 The provisions of the
criminal law, it was said, must be strictly construed. The Court noted
that a 20 to 21 week-old fetus was not viable and, accordingly, was not
˜another person™ within the meaning of the Criminal Code; the Code
could not, therefore, be applied in the instant case.
Seldom can such a diversity of opinions have been expressed on a topic
of major significance and Mrs Vo then proceeded to the European Court of
Human Rights where it was hoped that such important issues might be
resolved once and for all. Unfortunately we were to be sadly disappointed.
It is important to remember that the action was now concerned with a
breach of human rights under the European Convention on Human
Rights and, of the contraventions raised, it is that against Article 2 that
is of main concern here. The following is the relevant extract:
1. Everyone™s right to life shall be protected by law. No one shall be
deprived of his life save in the execution of a sentence of a court . . . .
2. Deprivation of life shall not be regarded as inflicted in contravention of
this Article when it results from the use which is no more than abso-
lutely necessary:
a) in defence of any person from unlawful violence; . . .142
In short, the case raised the acid test of fetal status “ was fetal life
protected by law?
The Court was, of course, in a difficult position which it explained at
length,143 Attention was drawn to the international European jurispru-
dence, quoting in particular, the Convention on Human Rights and
Biomedicine144 “ in which the term ˜everyone™ is left undefined. It also
emphasised the differences in opinion as to the moral status of the embryo
and of the fetus that are held in Europe and it noted (at [41]) that the
offence of unintentional fetal homicide is unknown in the majority of
member states of the Council of Europe;145 legislation protective of the
fetus to some extent exists only in Italy, Spain and Turkey.

On 30 June 1999, see n. 131 above, [22].
For UK purposes, the Article is repeated in the Human Rights Act 1998, Schedule 1.
The Court also quoted an analysis which indicated that 28 out of 34 published com-
mentaries were critical of the Court of Cassation™s rigid attitude: Jean Pradel, ˜Violences
involontaires sur femme enceinte et delit d™homicide involontaire™ (2004) 7/7148 Recueil
Dalloz 449“50.
Opened for signature at Oviedo in 1997 and coming into force on 1 December 1999.
For the United Kingdom, see the authoritative article by Adrian Whitfield, ˜Common
Law Duties to Unborn Children™ (1993) 1 Medical Law Review 28“52. The early
analysis by Jennifer Temkin, ˜Pre-natal Injury, Homicide and the Draft Criminal Code™
(1986) 45 Cambridge Law Journal 414“29 is remarkably prophetic.
48 The troubled pregnancy

The applicant™s case was, therefore, wholly relevant to the nature of the
discussion in this chapter. In essence, she contended that it was now
˜scientifically proven™ that all life begins at fertilisation; a child that had
been conceived but was not yet born was neither a cluster of cells nor an
object, but a person. The term ˜everyone™ in Article 2 was to be taken as
referring to human beings rather than persons possessing legal personality
and that, subject to exceptions provided in the law on abortion, French
law guaranteed all human beings the right to life from conception; the
availability of abortion did not exonerate the state from its duty to protect
the unborn child under the terms of Article 2. She also contended that
unintentional homicide of a fetus necessarily merited a criminal sanction
because a civil remedy did not ˜satisfy the requirement of expressing
public disapproval of a serious offence, such as the taking of life™ (at
The Court, however, pointed out that the still unsettled question of
what constitutes the ˜beginning™ of ˜everyone™s right to life™ in the con-
text of Article 2 had, thus far, been raised within its jurisdiction only
in connection with laws on abortion,146 and it continued to look at
Mrs Vo™s case in this way even though it was of a quite separate nature.
By tying the problem of fetal homicide to national abortion laws, the
Court was able to say (at [82]) that the question of when the right to life
begins comes within the margin of appreciation which it generally con-
siders states should enjoy in this sphere “ in essence, it was undesirable,
and probably impossible, to answer in the abstract the question whether
or not the unborn child is a person for the purposes of Article 2 of the
Convention. Moreover, there was no call to decide whether or not ˜the
abrupt end to the applicant™s pregnancy™ fell within the scope of Article 2
insofar as there had been no failure in the particular case to comply
with the national laws on abortion (at [85]). Thus, in the end, what
had been hoped might become an ˜evolutive interpretation™ of a ˜living

It is, in fact, remarkable how few of the relevant applications have actually reached the
court. In X v. Austria (Application no. 7045/75, Commission Decision of 10 December
1976) the Commission avoided the issue. The close relationship between the mother
and her fetus was recognised in Bruggemann and Scheuten v. Federal Republic of Germany
(Application no. 6959/75, of 12 July 1977; (1981) 3 EHRR 244) but the Commission
did not find it ˜necessary to decide . . . whether the unborn child could be considered as
˜˜life™™ in the context of Article 2 of the Convention™. Similarly, in X v. United Kingdom
(Application no. 8416/79, Commission Decision of 13 May 1980) the Commission
denied the fetus an absolute right to life in the face of fetal/maternal conflict but avoided
a decision as to whether it was no rights or partial rights that were available. A further
evasive opinion was given in H v. Norway (Application no. 17004/90, Commission
Decision of 19 May 1992). See also Boso v. Italy no. 50490/99, ECHR 2002-VII, one
case which went to the Court itself.
Voluntary and involuntary termination of pregnancy 49

instrument™147 petered out into a non-decision.148 Due to a variety of
circumstances, the unfortunate Mrs Vo left Strasbourg empty handed;
she was even denied the opportunity to take action in the civil courts “
though, even then, she might have had considerable difficulty.

In summarising our discussion, it is worth reconsidering what Mrs Vo was
seeking. In the wide view, she was certainly hoping to establish a fetal
right to recognition as a form of human life that was entitled to legal
protection “ a right that was at least that of a human entity, if not that of a
human person. In the narrower, though perhaps the more easily appre-
ciated, context, however, she was a woman seeking redress for the negli-
gent loss of her child and, as such, she represented the, still, majority of
pregnant women who want to carry their pregnancies to full term and
parturition. I am, as a result, at a loss to understand how the European
Court of Human Rights, while acknowledging that the case raised a new
issue,149 came to argue it on the basis of legislation that was designed for
the benefit of those whose concern is diametrically opposed.150
I suspect that it stemmed from a fear that to address the question of
fetal status head-on might involve reopening questions of conflicting
maternal and fetal rights in pregnancy, which most would wish to be
seen as being largely settled. In fact, of course, there was no need to run
the risk. Insofar as there is a medico-legal definition, and absent disease in
the mother, abortion is generally understood to mean the termination of a
pregnancy that, for one reason or another, is unwanted. Thus, the major
question that could have been addressed, even if not answered, is whether
or not there is an offence of feticide that is independent of consensual
termination of pregnancy. As the analytical paper that was so critical of
the Court of Cassation judgment151 put it “ the freedom to procreate is

As the Convention on Human Rights was described. See n. 131 above, at [82].
It is only fair to say that this sense of disappointment was, in fact, shared by five of the
seventeen-judge panel. In brief, Costa J and Traja J thought that the fact that there was a
lack of uniformity among the legislatures of the member states provided no reason to
stop the court defining the meaning of a person with a right to life. The most important
aspect of their opinion for present purposes is that they considered the case to be clearly
distinguishable from those dealing with termination of pregnancy at the request of the
mother and could see no policy reason for not applying Article 2 in the circumstances.
The most trenchant criticism, however, is to be found in the overtly dissenting opinions
of Ress, Mularoni and Straznicka JJ who considered that Article 2 applied to the fetus.
´ ´
n. 131 above, at [81].
It is to be noted that all the dissenting opinions held firmly that Mrs Vo™s case was
˜wholly unrelated to laws on the voluntary termination of pregnancy™.
Pradel, n. 143 above.
50 The troubled pregnancy

less well protected than is the freedom to have an abortion, and this
cannot, surely, be the right approach. Irrespective of any question of
fetal rights, to deprive a woman of her future child cannot be seen as
anything less than a gross insult to her bodily integrity. Looked at in this
way, protection of the fetus in no way conflicts with women™s rights “
rather, it enhances their right to conceive and carry a child free from
external interference.
That being the case, the problem of viability, which is so much a feature
of the abortion debate at both moral and legal levels, loses its meaning
The woman who loses her fetus involuntarily is losing her prospective
child and the individual fetus is being deprived of his or her potential life.
The existing legal protection of the viable fetus is inadequate and one can
only question why the distinction should be made.152 Intuition surely tells
us that non-consensual feticide ought to be an offence rather than some-
thing on which the law turns its back.

Do we need an offence of feticide?
The offences under the Offences Against the Person Act 1861 are broadly
related to procuring the miscarriage of a woman and there is no doubt that
doing so will result in the death of the fetus she is carrying in the vast
majority of cases. But this is not to say that the two are necessarily the
same “ rather, they can be distinguished on several grounds. Historically,
the purpose of the original Offences Against the Person Act of 1837 was,
probably, to protect the woman against the negligence of the professional
abortionist rather than to secure the life of her fetus.153 Secondly, the
offences proscribed in the 1861 Act are offences of intent, and it is by no
means always the case that an intent to procure a miscarriage is the same as
an intent to destroy the fetus. And, finally, there are more ways of killing a
fetus than through the medium of abortion. A very good case can be made
out for seeking to fill what looks to be, at least, a lacuna in the criminal law.
Here, I suggest, the hiatus lies in the insistence that the fetus must have
progressed to the status of the neonate before it can be said legally to have
sustained injury, In the present context, the reductio ad absurdum lies in
the judgment of the French Criminal Division of the Court of Cassation
in a case heard on 25 June 2002.154 Here, the negligence of the health

As noted above, n. 20, Balcombe LJ in the Court of Appeal in Re F (in utero) [1988] Fam
122, [1988] 2 All ER 193, was unable to see any logical difference between the early and
the viable fetus as regards a need for protection.
See, for example, R v. Tait [1990] 1 QB 290.
Case 2, [29] of the majority opinion in Vo v. France.
Voluntary and involuntary termination of pregnancy 51

carers caused the death of a fetus that was not only full term but
was actually overdue. The court overturned the Court of Appeal™s verdict
that the midwife was guilty of unintentional homicide, stating that the
˜facts are not capable of coming within the definition of any criminal
At the same time, it is almost impossible to suppose that a United
Kingdom court could have upheld the criminal charges against Mrs Vo™s
doctor either, given the fact that the child was, I believe, stillborn “ and
this is clearly wrong. We tend to forget that the fetal ambience has
changed dramatically in the last thirty years or so.156 It is difficult, for
example, to be able to return a fetus to the uterus for maturation after
performing major surgery and still to maintain that it has no legal stand-
ing; common sense tells us that, if it merits individual medical treatment,
it also merits legal recognition.157 The conclusion that the legislators and
the lawyers have been so immersed in the abortion debate and the
associated problems of women™s rights that they cannot disengage from
them seems unavoidable. Remove the phrase ˜termination of pregnancy™
and substitute the word which describes what is in issue “ ˜feticide™ “ and
most of the jurisprudential problems disappear. In addition, and as I have
suggested previously,158 there is very little reason why a crime of feticide
could not be defined as being subject to the terms of the Abortion
Act 1967.
The facts of Vo v. France should, surely, encourage the medico-legal
forum to maintain its concern for fetal rights and, particularly, to consider
how these should, and could, be accommodated within Article 2 of the
European Convention on Human Rights. As to the former, it is suggested
that some such legal intervention is becoming increasingly inevitable as
the physical and social status of the fetus attracts new meanings and
understanding. As to the latter, it has been argued “ it is hoped,

And see also the Canadian case of R v. Sullivan, n. 129 above.
˜It no longer makes sense to retain the [˜born alive™] rule where its application woud be
perverse™ per Major J in Winnipeg Child and Family Services (Northwest Area) v. G(DF)
(1997) 152 DLR (4th) 193 (S Ct) at [110]. This was, admittedly, in a dissenting opinion
in a maternal fetal abuse case but it illustrates that others hold this view.
The majority in Vo reminded us that the French National Assembly voted to amend the
law and to create an offence of involuntary termination of pregnancy “ the so-called
˜Garraud amendment™. This amendment was rejected by the Senate largely on the
grounds that it ˜caused more problems than it solved™ (Assemblee Nationale, proposi-
tion du loi no. 837, 14 May 2003).
For previous discussion, see Mason, n. 132 above. A recent paper precipitated by
Mrs Vo™s case accepts the need to apply Article 2 to the fetus but only when the ˜bright
dividing line™ of viability has been passed “ which is roughly the present position in
England and Wales. See Aurora Plomer, ˜A foetal right to life? The case of Vo v France™
(2005) 5 Human Rights Law Review 311“36.
52 The troubled pregnancy

persuasively “ that we have no need to interfere with the abortion laws on
this score. In my view, the wholly negative approach of the European
Court of Human Rights perpetuates the fallacy that there is a necessary
connection between abortion and feticide and actually sets back any
reconsideration of both maternal and fetal interests in this area. I con-
clude this chapter by suggesting that we should seriously reconsider the
status of the fetus “ not with the intention of restricting women™s auto-
nomous choices but, rather, in the interests of all pregnant women.
3 Antenatal care and the action
for wrongful birth

In the discussion of abortion that has gone before, I have deliberately
avoided the fourth ground on which a legal termination can be founded.
This is because the motivation of and the moral issues involved in that sub-
section are of a very different type from those addressed in Chapter 2. The
now relevant text of the 1967 Act, section 1(1), reads that a person shall not
be guilty of an offence under the law relating to abortion if, again, two
registered medical practitioners are of the opinion, formed in good faith:
(d) That there is a substantial risk that if the child were born it would suffer from
such physical or mental abnormalities as to be seriously handicapped.
I have to admit that, at one time, I was prepared to suggest that, second
only to the saving of the mother™s life, this was the ground which was most
likely to be widely acceptable. It will become clear in the following pages
that I now consider that to have been an unjustifiable view.
As a start to the analysis, it is to be noted that no time limit is now
imposed under the subsection.1 The practical reason underlying this
relaxation is that the diagnosis of genetic or chromosomal disease is
time consuming; given that the test may also have failed at the first
attempt, it was more than possible that the conditions justifying law-
ful termination would be discovered only after remedial action in the
form of abortion was rendered potentially illegal by virtue of the Infant
Life (Preservation) Act 1929, section 1. A classic case is that of
Mrs Gregory2 to which we will return later. The doctor who performs a
termination within the terms of the 1967 Act is, however, now no longer
subject to the 1929 Act3 “ and, because misfortunes comparable to that

Consequent on amendments to the Abortion Act 1967 introduced by the Human
Fertilisation and Embryology Act 1990, s.37(1). Note that discussion of cases under
this heading is, therefore, of a different quality depending on whether or not the amend-
ment was in force at the time of the hearing.
Gregory v. Pembrokeshire Health Authority [1989] 1 Med LR 81.
Human Fertilisation and Embryology Act 1990, s.37(4).

54 The troubled pregnancy

which befell Mrs Gregory were not uncommon,4 many would regard this
as an improvement on the original legislation. Nonetheless, the possibility
of a termination being lawful up to full term raises ethical issues of
considerable importance on its own account.

Late termination of pregnancy
First, it will be appreciated that the maternal environment has changed.
Whereas the woman seeking a termination under section 1(1)(a) of the
Act does not want a child, she who takes advice in respect of section
1(1)(d) wants a child “ and, for reasons given above is likely to have been
carrying her child for an appreciable time “ but she does not want a
disabled child. As a consequence, the ethical, emotional and legal con-
siderations surrounding the process are, now, of a different dimension.
Second, the doctor performing the operation must also reappraise his
motives, remembering that it was said in the debate preceding the 1990
If the fetus is mature enough to have a reasonable chance of survival with intensive
care, all possible steps are taken to optimise the recovery of both mother and
Which is a lofty view, taken from an impeccable moral position “ but is it
meaningful in practice? The obstetrician has contracted to relieve the
woman of a child who, at the time of the operation, was unwanted; he or
she cannot personally guarantee a successful adoption6 and it might be
considered that, in achieving a live birth, the obstetrician is abrogating his
duty of care to the pregnant woman. The legal and technical conditions
are, in fact, such that a living normal abortus must be a very rare event.7
The possibility of a live birth is far greater in the case of abortions

The previous effect of the 1929 Act is very well demonstrated and widely discussed in
Rance v. Mid-Downs Health Authority [1991] 1 QB 587, [1991] 1 All ER 801. However,
not all such instances resulted from genetic investigation “ Rance was a case of missed
neural tube defect.
Official Reports (Lords) vol. 522, col. 1043, 18 October 1990 per Lord Ennals at 1052.
Though, given the present situation, a successful outcome might well be anticipated. It
could, indeed, be suggested that, subject to the woman™s right to choose, adoption of a
viable fetus might well be the preferred objective in any abortion. See Sheila A. M.
McLean, ˜Abortion Law: Is Consensual Reform Possible?™ (1990) 17 Journal of Law and
Society 106“23.
It could occur when the expectation as to the risk of abnormality proves unfounded. The
mother might also have an unwarranted fear of a disabled child though, in such circum-
stances, the termination would probably be under s.1(1)(a) and subject to the 24-week
restriction. Otherwise, the situation could arise from terminations under s.1(1)(b) and (c) “
in which case a living child could be the desired outcome.
Antenatal care and the action for wrongful birth 55

performed under section 1(1)(d) and it is likely to be harder “ albeit,
mercifully, not always impossible “ to find adopting parents for a seriously
handicapped child.8
The doctor™s dilemma could, therefore, be acute9 and, as a result, the
current professional guidelines recommend that abortions performed
after the twenty-second week of pregnancy should be preceded by
feticide10 “ apparently a remarkable volte-face since the days of Lord
Ennals. This is clearly a pragmatic approach but one that is virtually
dictated by the uncertainty as to the doctor™s duty of care. We have,
moreover, to remember that the criminal law is also involved. The
Abortion Act 1967, section 5(1)11 protects the doctor from the charge
of child destruction and domestic law as to death in utero has been
reinforced by the European decision in Vo v. France.12 No overt statutory
protection is, however, offered once the child has been delivered and,
since it can be forcefully argued that the doctor now has a duty of care to
the neonate, charges of both manslaughter and murder could be brought
against one who abandons a living abortus. Empirical reports indicate
that the opportunity arises fairly often13 yet, the fact that there is so little
reported legal action14 suggests that the authorities, in addition to dis-
playing a clear reluctance to reopen the abortion debate at any time,
appreciate that this is a medical problem that is best resolved by the
medical profession “ indeed, it is both a logical consequence of and a
logical reason for the medicalisation of abortion.
All of which presupposes that the fetus wants to be born alive and, in
practice, the dilemma is aggravated by the possibility of a neonatal action

It is reported that there were 104 such live births in the West Midlands Region between
1995 and 2002 (which suggests that there could be some 130 cases annually on a national
scale): Elizabeth Wicks, Michael Wyldes and Mark Kilby, ˜Late Termination of
Pregnancy for Fetal Abnormality: Medical and Legal Perspectives™ (2004) 12 Medical
Law Review 285“305. This is an outstanding review of the situation.
1,900 abortions (1 per cent of the total) were performed by reason of suspected neonatal
handicap in England and Wales in 2004 (Abortion Statistics, England and Wales: 2004).
Lois Rogers, ˜Fifty Babies a Year are Alive after Abortion™ (2005) Sunday Times,
27 November.
As amended by the Human Fertilisation and Embryology Act 1990, s.37(4).
(2004) 79 BMLR 71, ECtHR, discussed in detail at p. 44 above.
For anecdotal evidence, which is probably all that can be obtained, see Sarah-Kate
Templeton and Lois Rogers, ˜Babies that Live after Abortion are Left to Die™ (2004)
Sunday Times, 20 June, p. 1.3.
In over 20 years, I have still only uncovered one apposite case “ R v. Hamilton (unre-
ported) (1983) The Times, 16 September, p. 1 “ where the magistrates decided there was
no case to answer in a prosecution led by the DPP. There have been a number of
coroners™ decisions sporadically reported in the news media in which no action has
been taken. These cases are discussed, inter alia, in J. K. Mason and G. T. Laurie,
Mason & McCall Smith™s Law and Medical Ethics (7th edn 2005), 5.99“100.
56 The troubled pregnancy

for ˜wrongful life™ “ a concept that is discussed in detail in Chapter 6. The
doctor terminating a late pregnancy is, therefore, in a most difficult
situation whether measured on the practical or the moral plane. The
latter is further complicated by consideration of the purpose behind
section 1(1)(d) which is colloquially known as the ˜fetal ground™. But is
it? Many years ago, Glanville Williams wrote:
The argument for abortion on the fetal indication relates to the welfare of the
parents whose lives may well be blighted by having to rear a grossly defective child,
and perhaps secondarily by consideration for the public purse . . . The fetus is
destroyed not necessarily in its own interest . . . but in the interests either of the
parents or of society at large, though of course only on the request of the mother15

and, to the best of my knowledge, this has remained the accepted legal
interpretation of the subsection.
The ethical implications are considerable. The conclusion that the
fetus is being destroyed simply because it is imperfect is hard to resist “
as is the extrapolation that section 1(1)(d) is an invitation to discrimina-
tion on the basis of disability;16 many who campaign for the rights of the
disabled will, additionally, hold that the section strikes at the status of all
disabled persons.17 We must, at the same time, appreciate the uneasy
position of the pregnant woman whose choice cannot be made in a
vacuum. It has, for example, been stated that more than 10 per cent of
obstetricians agree that:

The state should not be expected to pay for the specialised care of a child with a
severe handicap where the parents had declined the offer of prenatal diagnosis
[and, presumably, the resulting abortion “ Auth.].18

The profound dilemma for all concerned in a late termination, which
can be legally performed under subsection 1(1)(d) up to full term, is

Glanville Williams, Textbook of Criminal Law (London: Stevens & Sons, 2nd edn 1983)
at 297.
The moral position is explored in depth by Sally Sheldon and Stephen Wilkinson,
˜Termination of Pregnancy for Reasons of Foetal Disability: Are There Grounds for a
Special Exception in Law?™ (2001) 9 Medical Law Review 85“109. After extensive argu-
ment, however, the authors conclude that the essential discrimination lies in not allowing
women a choice of abortion whatever the circumstances.
Lynn Gillam, ˜Prenatal Diagnosis and Discrimination against the Disabled™ (1999) 25
Journal of Medical Ethics 163“71. For rejection of this thesis, see Raanan Gillon, ˜Is there a
New Ethics of Abortion?™ (2001) 27 Journal of Medical Ethics supp II: ii5“9. That author™s
argument, however, depends upon denying a full moral status to the fetus.
J. M. Green, ˜Obstetricians™ Views on Prenatal Diagnosis and Termination of Pregnancy:
1980 compared with 1993™ (1995) 102 British Journal of Obstetrics and Gynaecology
228“32. Quoted in a very sympathetic article by John Wyatt, ˜Medical Paternalism and
the Fetus™ (2001) 27 Journal of Medical Ethics suppl II: ii15“20.
Antenatal care and the action for wrongful birth 57

amplified by what I see as a third major moral consequence “ that the
distinction between feticide and neonaticide is irrevocably blurred, and
this, of itself, provokes further theoretical and practical quandaries.
The background difficulties are several. First, a genetic history may
alert us to the likelihood of a disabled fetus; it cannot tell us whether or
not a genetic defect is present. Second, while numerous antenatal tests,
involving both mother and child and of varying degrees of invasiveness,
are available by which to identify the presence of abnormality, they tell us
the degree of disablement with far less definition. And, finally, it is well-
nigh impossible, in terms of the overall loss of human life, to distinguish
between the killing of a fetus in the last week of pregnancy and ending
the life of the same child in the first week of extra-uterine existence.19
Why, then, do we accept feticide when we are in a state of uncertainty
and prohibit neonaticide when the precise state of the infant can be
It is difficult to find a satisfactory reply and we return to the subject in
Chapter 7. Meantime, the honest, albeit philosophically unsatisfactory,
conclusion is probably to be derived from the intuitive belief that, logic or
no logic, neither the law, nor the profession nor public opinion would
tolerate causing the deliberate death of a child unless it was achieved by
way of omission and, by consensus, that it was in the child™s best interests
not to be sustained. Such is the attitude to disability engendered by
section 1(1)(d), however, that one can only hope that it will be possible
to say the same a decade from now.

The uncertainties of subsection 1(d)
As a coda to this discussion, it has to be noted that the subsection is
singularly unhelpful in the resolution of these dilemmas. It speaks of a
substantial ˜risk™ “ and it means just that. The significance lies in the risk “
not the fulfilment of that risk; it follows that a termination under this head
can be lawful despite the fact that the fetus is normal. Moreover, ˜sub-
stantial™ is a subjective concept which is to be defined by those making the

This, of course, would be accepted by those who follow the Tooley school of ˜person-
hood™ “ for which see p. 16 above. That construct is, however, of no significance in the
present context as it holds that neither the fetus nor the neonate qualify as ˜persons™.
The case of Dr Arthur (R v. Arthur (1981) 12 BMLR 1), which is discussed in detail in
Chapter 7, stands out in the memory as an exception. It is, however, now 25 years old and
it is hard to believe that either the law or the medical profession would support Dr Arthur
58 The troubled pregnancy

choice “ that is, the woman, assisted by her medical adviser.21 The same
criticisms apply to the interpretation of ˜serious handicap™ which, again,
is a phrase which can be invoked according to choice. For example,
Kleinfelter™s syndrome (for which, see below at p. 61) is a genetic disorder
which results in some feminisation of the physique and the probability of
infertility in the male. The majority would probably place this very low in
their classification of physical handicap; but the author has yet to find an
obstetrician who would not certify a termination on the grounds that the
fetus was so affected if asked to do so.
A recently publicised example arose in Jepson v. Chief Constable of West
Mercia.22 Here, a minister of religion was granted judicial review of the
police decision not to prosecute doctors who performed a late termina-
tion on a fetus with a cleft palate “ a condition that may or may not
˜seriously handicap™ a child depending on its severity and on the skill of
the remedial surgeon. But, as has already been inferred, it is impossible to
demonstrate ˜bad faith™ on the part of the professionals involved granted
that the risk is there. The Crown Prosecution Service declined to
prosecute “ and the cynic might reasonably suggest that its hesitation was
yet another example of the politico-legal determination to avoid reopen-
ing the abortion debate. As a result, the case stalled at the first hurdle
and, even if it is reopened,23 it is unlikely to recover “ particularly as any
argument based on fetal ˜human rights™ has been virtually eliminated by
the decision in Vo v. France.24 The fundamental question is, however, of a
wider nature “ is fetal abnormality too complex a subject to be governed
by a section that is as loosely drafted as is section 1(1)(d) or is the section
deliberately so phrased in order to preserve its subjectivity? One suspects
that the latter represents the political view and the subsection is unlikely
to be changed in the foreseeable future; nevertheless, a test prosecution
could be helpful to all concerned and, for this reason at least, Ms Jepson™s
lack of success is to be regretted.

In this connection, the Royal College of Obstetricians and Gynaecologists believed that,
to be substantial, a risk should be recognisable as such by ˜informed persons with no
personal involvement in the pregnancy or its outcome™: Termination of Pregnancy for Fetal
Abnormality in England, Wales and Scotland (1996), para. 3.2.2. But, since only the
antenatal care team can realistically assess the risk, this is doing little more than applying
the Bolam test (see p. 10 above).
[2003] EWHC 3318, (2003) WL 23145287.
Press statement http://www.jjepson.org of 15 April 2005. My information is that
Ms Jepson has, in fact, abandoned her action. Currently, the case has been reported
only in the news media: Ruth Gledhill ˜Curate loses legal challenge over ˜cleft-palate™
abortion™ (2005) The Times, 17 March, p. 14.
n. 12 above.
Antenatal care and the action for wrongful birth 59

Antenatal care and fetal abnormality
Thus far, however, the road may have been difficult but it has been
relatively even. A pregnant woman is aware that her fetus is less than
perfect. She does not want such a child and she is aware that the law will
support her rejection of it in the fetal stage of development. But how has
this equilibrium been reached? Although she may be perfectly capable of
reaching her own conclusion, a woman cannot be expected to make a
reasoned decision as to the disposal of a fetus at risk of congenital
abnormality unless she is supplied with adequate information. This is
the function of the antenatal health carers and these comprise men and
women of varying, yet integrated, expertise and of proximity to the
pregnant woman. In order to appreciate this diversity of responsibility,
it will be appropriate, at this point, to consider briefly those congenital
abnormalities that are likely to give rise to an action for wrongful birth
when they occur in a child.25

Congenital disease
Congenital disease “ that is, disease in the fetus that is present at birth or
for which the seeds are present at birth “ can be looked on as being due
either to:
* genetic and chromosomal abnormalities “ which it is convenient,

though, strictly speaking, incorrect, to combine;26 or to
* environmental factors; or to

* a combination of the two.

Each of these can either cause or predispose to disease. Thus, as we have
already observed, the mere discovery “ or inference “ of an abnormal gene
in a fetus, or of a predisposing environment, does not necessarily inform
as to either diagnosis or prognosis in the resulting child.

Genetic disease.27 The reasons for these shortcomings in respect
of genetic disease are several. As to the first, it is well-known that similar

So-called late onset genetic disease “ in which the condition may not be manifested until
adult life “ is a highly significant factor in genetic counselling but is unlikely to lead to an
action for wrongful birth (see p. 60).
The most important practical difference is that the former are hereditary conditions
whereas the great majority of the latter arise spontaneously and are environmentally
It is invidious to pick out any of the many authoritative texts on the subject, the majority
of which are probably too detailed for the lawyer™s purposes. This author happens to find
Robert F. Mueller and Ian D. Young, Emery™s Elements of Medical Genetics (Edinburgh:
Churchill Livingstone, 11th edn 2001) a useful vade mecum “ but there are many others.
60 The troubled pregnancy

genes are present in the cell nuclei in pairs or as alleles, one being derived
from each parent. The gene responsible for an abnormal condition may
be dominant over its normal counterpart “ in which case it will always be
free to express itself “ or it may be recessive, in which case it will be active
only when it is not suppressed by a normal partner “ in other words, when
both alleles are abnormal. The correlative importance is, of course, that,
although he or she does not manifest the abnormality, the person with a
single recessive gene becomes a carrier for the condition.
The second reason for the diversity of genetic disease is that the
penetrance of a gene is determined to large extent by the company it
keeps. In practice, almost all genetic disease is of this multifactorial type;
even if the origin of the disease process can be traced to a single, identi-
fiable genetic mutation, its expression will be influenced by other factors
and these may be related not only to associated genes but may also be
Two specific aspects of the transmission of genetic disorder must be
remembered. First, we have the occurrence of X-linked (or sex-linked)
disease in which the recessive mutation is carried on the X sex chro-
mosome. The presence of a normal X chromosome in the XX female
configuration means that, while the affected female remains a carrier,
the abnormal gene can only express itself in the XY male situation.
Second, we should note the condition of late-onset genetic disease.
This is of practical importance in that it explains the persistence of
severe disease caused by a dominant injurious gene. In the usual
circumstance, such disease should be self-limiting in that those in the
affected population will tend to die before the gene is passed on; if,
however, the condition does not manifest itself until late in life, they
may have procreated before the danger is appreciated. Huntington™s
disease is the well-known classic example which demonstrates some of
the complexities added to reproductive counselling in this special
situation. It is clear, for example, that termination of pregnancy because
of the presence of the Huntington gene in the fetus poses an excep-
tion to the general rule, discussed above, that termination under sub-
section 1(d) is for the benefit of the mother; rather, it is designed to
eradicate a deleterious gene which will express itself in the individual
with uncertain severity and at an uncertain time if it does so at all.

Recent work in so-called epigenetics suggests that current theories as to genetic expres-
sion are too simple and it may be that the whole nuclear mass is involved. A whole issue of
Science (2001), vol. 293, pp. 1001“208 is devoted to the subject but the papers are
extremely technical.
Antenatal care and the action for wrongful birth 61

The dilemma for the mother of the affected fetus must be well-nigh

Chromosomal disorder. Chromosomal disorder is of a different
nature and is, essentially, a matter of an excess of chromosomes due to
the fact that the pair do not split properly on the formation of gametes; as a
result, three chromosomes will be left at a given locus when the zygote is
formed.30 This is known as a trisomy of which there are many variants “
including trisomies of the sex chromosomes such as Kleinfelter™s syn-
drome (XXY) which has already been noted; however, trisomy-21 which
results in Down™s syndrome is by far the most important. As is well-
known, this “ and other trisomies “ can result in a number of fairly specific
anatomical abnormalities together with an unpredictable degree of learn-
ing disability.31 The significant difference from genetic disorder is that, in
general, trisomies arise spontaneously, the risk depending almost entirely
on the environmental factor of the mother™s age.32 The socially, and
legally, important exception to this generalisation lies in the so-called
Robertsonian translocation abnormality in which part of one chromosome
becomes attached to another. This occurs spontaneously and is of no
consequence to the original individual but, once established, the abnor-
mality can be passed on by way of standard genetic inheritance principles.
The situation, then, is that the child of the carrier may have a normal num-
ber of chromosomes but an increased chromosomal mass, and this will
result in Down™s syndrome if the extra mass derives from chromosome 21.

The problem of the late-onset genetic dsease is raised again in Chapter 6. There is an
associated minor legal quibble in that s.1(1)(d) of the 1967 Act legalises abortion if there
is a substantial risk that ˜if the child were born it would suffer™ from serious handicap.
Whether this means at the time of birth has not been decided in the courts. That being the
case, one would think that, as a matter of common sense, the additional words ˜now or
in the future™ can be read into the Act. It is not, however, a very satisfactory assumption.
I. Kennedy and A. Grubb in Medical Law (London: Butterworths, 2000) come to much
the same conclusion at p. 1427. For the interpretation of parliamentary intention, see R
(on the application of Quintavalle) v. Secretary of State for Health [2003] 2 AC 687, [2003] 2
All ER 113.
Clearly, the alternative is that the zygote contains only one chromosome but this is
virtually incompatible with life other than when it occurs in the sex chromosomes
(Turner™s syndrome, in which women have only one X chromosome, is a major exam-
ple). It is to be remembered that chromosome abnormalities are common and probably
are to be found in some 20 per cent of conceptions (Mueller and Young, n. 27 above at
p. 215). The great majority of affected fetuses are, however, lost naturally so that the
incidence at birth is less than 1 per cent.
Note, however, that the Fragile-X syndrome, which is probably the commonest cause of
such disorder, is an X-linked genetic condition.
Mueller and Young quote an incidence of Down™s syndrome of 1:1500 at the age of 20,
through 1:400 at age 35 to 1:30 at age 45.
62 The troubled pregnancy

The chances of a carrier of such a translocation producing a child with
Down™s syndrome are about 10 per cent in the case of females and 4 per cent
of males and this, of course, is independent of age. The case of Al Hamwi v.
Johnston33 is instructive and is discussed further at p. 77.

Environmental factors. Environmental factors may constitute the
only cause of congenital disease. The extent of such disability will become
apparent in the following pages. For the present, I suggest that viral
infection in the mother is by far the most important of these as a source of
actions in negligence. Rubella, or German measles, is the most common in
this context “ possibly because it can occur in such benign form that the
diagnosis may well be missed. Other cases have been caused by varicella
(chicken-pox) infection and there is no reason why viraemia of any type
should not affect the fetus in some way.34 The effects may be devastating
and include blindness, deafness and severe neurological disorder.
Drug induced congenital disease is a further significant category and, if
one includes the fetal alcohol syndrome and intrauterine opiate poisoning
in the classification, it may be one of the commonest causes of neonatal
morbidity.35 We will see in Chapter 6, however, that fetal injury from
prescription drugs raises particular jurisprudential issues and, although
intense research on drugs prior to their release has now decreased the
incidence of such injuries, the use of therapeutic drugs in pregnancy is still
a matter of major importance.
Finally, haemolytic disease of the newborn, which is due to rhesus blood
group incompatibility between mother and fetus, is often quoted as an
example of pure environmental disease although, since it results from
genetic incompatibility, that may be a semantic misallocation. Once again,
the condition is now rare but it provides a good example of the distinctive
pre-conception tort; we consider it further at p. 196.

Combined genetic and environmental factors. The purist might say,
with some justification, that all disease is, ultimately, the result of such a
combination. Be that as it may, the classic examples lie in the neural tube

[2005] Lloyd™s Rep Med 309, QBD. For an earlier presumed example, see Gregory v.
Pembrokeshire Health Authority, n. 2 above. Interestingly, both cases failed for much the
same reasons.
I am not discussing transmitted HIV infection here. Not only has it widespread and
irrelevant ramifications but actions for wrongful birth or wrongful life associated with the
condition would, I think, be untenable.
Again, neither is likely to precipitate actions of the type considered in this book but both
illustrate the great difficulty in allocating fetal ˜rights™ which have been touched on in
Chapter 2.
Antenatal care and the action for wrongful birth 63

defects represented by anencephaly and spina bifida “ there is a strong
genetic susceptibility but also a strong association with folic acid defi-
ciency in the maternal diet. Neural tube defects illustrate the importance
of the underlying cause in the calculation of risk of a child being affected.
In a pure single gene disorder the chances of the gene being transmitted
can, at least in theory, be accurately determined with relative accuracy.
But, because of the varying penetrance of genes and of the effect of the
associated mutifactorial elements, the chances of the mutation expressing
itself can, in the absence of total dominance, generally be predicted only
on the basis of probability. The problem is amplified as the importance of
environmental factors increases; in general, then, the counsellor can only
base his or her opinion on an empirical analysis of the history of the family
and of the population.

Investigations available in the face of potential fetal abnormality
Whether the antenatal care team can discharge its particular duty to the
pregnant woman as to the condition of her fetus depends on the tests and
techniques available for this purpose; whether or not they have satisfied
their duty depends, to a large extent, on how they have used and inter-
preted those tests and, since the great majority of this chapter “ and of
Chapter 6 “ is concerned with failure in this respect, a short aide memoire
as to their availability must be included at this pont.
Antenatal investigations designed to ensure the normality of or, more
importantly in the medico-legal context, to identify an abnormal fetus
must start with a family history “ while this is common to all medical
diagnosis, it is, of course, particularly imperative when one is specifically
seeking evidence of hereditary disease.36 Beyond that, investigations may
be maternal or fetal and, at the same time, relatively invasive or non-
invasive. Invasive testing introduces the hazards of fetal injury or disim-
plantation and all scientific tests cost money. The primary concern is,
therefore, whether or not the offer of a test is warranted, either in relation
to the risk to the fetus or to resource allocation “ is it, for example,
effective to test women under thirty years old for the presence of a
Down™s syndrome fetus and, if it is not, at what age should we introduce
such an investigation as a routine?

Much is currently written about the ethics of genetic testing in general; it is often
forgotten that taking a family history as an aid to diagnosis has been the norm since
time immemorial “ yet it is no more or less than a primitive form of ˜genetic testing™.
Moreover, given the uncertainties surrounding the accuracy of newer forms of genetic
testing and screening, some commentators still view the family history as the most
important indicator of disease.
64 The troubled pregnancy

Maternal testing
Maternal serum tests. The most common maternal tests involve estimates
of various protein constituents of the blood that have been found to vary in
the presence of a fetus suffering from Down™s syndrome and also of one
with an open neurological abnormality. The simplest, but now replaced,
is to measure only the maternal serum levels of a fetoprotein and human
chorionic gonadotrophin (HCG) but the best that this can do is to
indicate the need for more specific “ and invasive “ investigations. It is
now claimed that the addition of two further analyses together with
repeated ultrasonography can provide a detection rate of 85 per cent
with a false positive rate of only 1 per cent “ which is virtually a diagnostic
test.37 It is noteworthy that the government at the time of writing is
seeking to introduce routine testing for Down™s syndrome in all pregnant
women.38 Insofar as Down™s children, for practical purposes, make no
contribution to the gene pool, that there is no treatment of the condition
and that its gravity cannot be determined by diagnostic tests, the only
significant effect of such a programme can be to encourage terminations
on the grounds of disability.39 Again, however, the ethics of screening
programmes are beyond the remit of this book.
Ultrasonography. Within the parameters of antenatal counselling, the
main function of fetal ultrasonography is to detect anatomical anomalies.
The simplest will be manifest abnormalities of the limbs or other parts of
the skeleton but the most important are those related to the central
nervous system in the form of neural tube defects, of which the common-
est is spina bifida. Anatomical abnormalities of the internal organs are
increasingly open to discovery. Changes in the soft tissues of the neck “ an
increase in the zone of so-called nuchal translucency “ is a useful indicator
of Down™s syndrome at ten to thirteen weeks™ gestation, particularly when
combined with serum testing as above.40

Invasion of the fetal environment. The primary purpose of non-
invasive investigation is to estimate the risk of the fetus being abnormal
and to provide firm ground on which to, either, reassure the pregnant

N. J. Wald, H. C. Watt and A. K. Hackshaw, ˜Integrated Screening for Down™s
Syndrome on the Basis of Tests Performed during the First and Second Trimesters™
(1999) 341 New England Journal of Medicine 461“7.
Department of Health Our Inheritance, Our Future (2003), Cm. 5791“II.
We have raised this problem elsewhere: Sheila A. M. McLean and J. Kenyon Mason,
˜Our Inheritance, Our Future: Their Rights?™ (2005) 13 International Journal of Children™s
Rights 255“72.
For a simple, up-to-date analysis, see Zarko Alfirevic and James P. Neilson, ˜Antenatal
Screening for Down™s Syndrome™ (2004) 329 British Medical Journal 811“12.
Antenatal care and the action for wrongful birth 65

woman or to recommend a specifically diagnostic invasive test. Current
practice is to recommend the latter when the risk is 1:250 or greater, but
this is, of course, modified by numerous subjective features of the indi-
vidual case. As it has been wisely put: ˜any ˜˜one size fits all™™ policy sits
uncomfortably with pregnant women and clinicians™.41 Effectively, there
are two major procedures available.42
Amniocentesis. In this process, a specimen of amniotic fluid is with-
drawn under ultrasound observation. The fluid can be analysed bio-
chemically, especially for the diagnosis of neural tube defects. The
contained fetal cells can also be cultured for chromosome and DNA
analysis and to demonstrate the presence of the rare, but important, inborn
errors of metabolism. The difficulties are considerable. Most importantly,
an induced miscarriage occurs in up to 1 per cent of investigations.
Occasionally, no fluid is obtained or the culture fails. Amniocentesis is
best performed at 16 weeks™ gestation; satisfactory culture generally occu-
pies a further two weeks. Given, also, that the progress of the investigation
may not be ideal, it will be appreciated that a termination, if indicated, will
certainly be late and, as we will see when considering the relevant cases,
many women who would have accepted an early abortion will not do so in
the second trimester and beyond.
Chorionic villus sampling. For these reasons, chorionic villus sampling,
which is performed at 10 to 12 weeks™ gestation, has obvious attractions.
Here, cells of fetal origin are taken from the early placenta. Chromosomal
analysis can be undertaken directly and the availability of large amounts
of fetal DNA means that many single gene disorders can also be diag-
nosed without the need for culture. The main contraindications are that
the technique is complicated; it is not everywhere available and it may be
associated with later limb deformities. Most significantly, it carries a
miscarriage rate of some 2 to 3% “ at least twice that of amniocentesis.
It may also harvest maternal rather than fetal cells “ occasionally with
disastrous results.43

Alfirevic, n. 40 above. At the same time, failure to offer counselling and testing may well
be regarded as a failure in the duty of care: Enright v. Kwun (2003) EWHC 1000 (QB),
The Times, 20 May.
Fetal or umbilical cord blood sampling, fetoscopy and fetal biopsy can be undertaken but
their usefulness is very specific and, currently, they are out of favour, largely by reason of
possible fetal loss.
A remarkable example is provided in the US case Schirmer v. Mt Auburn Obstetrics and
Gynecologic Associates, Inc. 802 NE 2d 723 (Ohio, 2003) in which maternal cells con-
taining a balanced translocation trisomy were mistaken for fetal cells which, in fact,
contained an unbalanced abnormality. For a very recent and significant UK case involv-
ing the b-thalassaemia gene, in which culture was needed, see Farraj v. King™s Healthcare
NHS Trust [2006] EWHC 1228, QB.
66 The troubled pregnancy

It is to be noted that, attractive though chromosomal and DNA analysis
may be, it is not without its own ethical problems. It is emphasised once
again that the mere discovery of the cause of an abnormality does not
necessarily indicate how serious will be the result. Moreover, many dis-
coverable single gene abnormalities do not indicate any more than the
propensity to disease.44 More important from the practical aspect is the
number and choice of tests. The patient may seek advice on the likelihood
of one particular abnormality being present. Yet routine testing may
involve several conditions and the mere process of drawing the chromo-
somal map (or idiogram) dictates that any abnormality is likely to show
up. Is it, then, ethical to load the woman with information she has not
sought?45 If not, then is it ethical to withhold information that may be of
major importance to the woman herself and/or to the future child she is
carrying? The dilemma is acute, and emphasises that antenatal care must
be carefully predetermined “ counselling begins before, not after, diag-
nostic tests have been performed.

The principles of antenatal care
In practice, the obstetrician has very little contact with the pregnant
woman between diagnosis and delivery; antenatal care in the United
Kingdom is the work of a team for which the relevant NHS Trust is
vicariously responsible. The interpretation of the various tests performed,
thus, devolves on a number of individuals.
Clearly there must be an overall coordinator, whether this be the
obstetrician or a specialist employed for the task, who will be responsible,
first, for taking a family history which is likely to indicate the scope of
further investigations and, second, for integrating these into a composite
of information that can be provided to the patient. From what has been
said, it will be apparent that a number of professionals will contribute to
information gathering “ among which one might include the radiologist,
the immunologist/microbiologist and the geneticist. Ultimately, however,
one person will be responsible for distribution of that information which
will enable the patient to decide on whether or not to continue with
her pregnancy; it is convenient to refer to that person as the antenatal

E.g. the breast cancer risk genes BRCA 1 and BRCA 2.
For discussion of the role of ignorance in genetics in general, see Graeme T. Laurie, ˜In
Defence of Ignorance: Genetic Information and the Right not to Know™ (1999) 6
European Journal of Health Law 119“32.
The title of ˜genetic counsellor™ is often given but is a misnomer insofar as much of the
data used is non-genetic.
Antenatal care and the action for wrongful birth 67

Antenatal counselling
The theory and practice of antenatal counselling lies at the heart of this
chapter yet it is almost impossible to provide an outline which will cover
all individuals and all circumstances. At base, the extent and nature of the
information provided must rest on the choice, or autonomy, of the
recipient rather than on that of the provider and should, in theory, be
non-directive. Yet, in satisfying the terms of the doctrine of ˜informed
consent™,47 the counsellor cannot help but indicate, at least, that there is
some information to give “ the neutrality of the ground is undermined
from the start.
Once having started, the problems accumulate.48 It is widely stated
that antenatal counselling must be non-inductive “ the parents should
make up their own minds according to their circumstances. But, at the
same time, one must ask whether this is ever possible “ and, even, is it
desirable?49 To begin with, the answer is seldom clear cut and the coun-
sellor is performing a mental balancing act ab initio. The parents cannot
do this alone “ they do not want a check list of facts, they want advice as to
the interpretation of those facts. And, if the counsellor has come to a
reasoned decision, is it ethically correct for him or her not to express that
in positive terms? The duty of the counsellor is to give positive advice in


. 3
( 12)