Legal Issues for Pro-Choice Opinion - Abortion and Disability

On 25 November 2001, Pro-Choice Forum held a day of discussion at Queen's University Belfast, entitled 'Legal Issues for Pro Choice Opinion'. There were two main aims for the day; first to provide a forum for pro-choice opinion formers, from a range of cultural contexts, to discuss key issues about abortion law, and second, to disseminate main aspects of the day's discussion through printed and on-line media. There were three sessions held during the day, the first on strengths and weaknesses of different forms of abortion law; the second on abortion in Ireland; and the third on abortion and disability. A summary of the third session is below. To read a summary of the first discussion click here, and to read a summary of the second, click here.

Abortion and Disability

This session began with a presentation from Dr Sally Sheldon, of the Law Department, University of Keele. A paper based on the presentation appears below.

Termination of Pregnancy for Reason of Fetal Disability

Sally Sheldon and Stephen Wilkinson, Departments of Law and of Philosophy, University of Keele

Current abortion law discriminates against the impaired foetus ... The law should not discriminate between impaired and non-impaired foetuses: a common time limit should be adopted for all pregnancies (Shakespeare 1998: 671).

At a 1998 workshop on Abortion, Ethics and the Law, a rare and interesting point of consensus emerged between the proponents of two sides of a debate normally renowned for intractable difference. Ann Furedi, a well-known 'pro-choice' advocate currently employed by the British Pregnancy Advisory Service, and Helen Watt of the 'pro-life' Society for the Protection of the Unborn Child and Roman Catholic Linacre Centre for Health Care Ethics were debating termination of pregnancy in the presence of a likely fetal disability. Furedi argued that any abortion should be allowed for no more reason than a woman has requested one. Watt contended that (virtually) all abortions are wrong and should be prohibited, and that whether the fetus is disabled or not makes no moral difference. Accordingly, both speakers agreed that we should not regard aborting an 'able-bodied' fetus as morally different from aborting an 'impaired' fetus and that the law's treatment of the two should be the same.

This point of agreement between speakers from two sides of a notoriously polarised debate is particularly noteworthy in that the speakers find themselves joined together in taking a position which is starkly opposed to public opinion. Opinion polls have consistently found that people consider termination more acceptable in the presence of a disability (Lee and Davey, 1998, Lee 2000). Furthermore, their view is also out of line with contemporary abortion practice: between 1500 and 2000 terminations per year are performed for reason of fetal disability, with several hundred of them occurring after 20 weeks gestation.

Still more interesting is the fact that both speakers find themselves in broad disagreement with the current law in England, Wales and Scotland. The 1967 Abortion Act provides that terminations are only lawful where performed by a registered medical practitioner, and where two doctors agree that one of a number of conditions is met. One of these conditions, set out in s.1(1)(d) of the Act, is that abortion may be authorised by two doctors who agree that there is a substantial risk that if the child were born it would be 'seriously handicapped'. Since 1990, there have been no time limits for terminations performed on this ground. Likewise, the existence of a presumed fetal disability is also relevant in Northern Ireland, where the Abortion Act 1967 does not apply. Such terminations as are carried out there are done on the basis of R v Bourne [1938] 3 All ER 612, which holds that a doctor may lawfully perform a termination where 'the probable consequences of the continuance of the pregnancy will be to make the woman a physical and mental wreck'. Whilst it is not obvious that this should more readily support terminations on the basis of abnormality than on other grounds, nonetheless the majority of the terminations performed in Northern Ireland each year are performed for this reason (Lee 1995).

This paper will take its starting point in this disagreement between, on the one hand, Furedi and Watts, and on the other, public opinion and the law of England, Wales and Scotland. It will assess the three major arguments which have been advanced in support of s.1(1)(d) of the Abortion Act 1967: (1) the fetal interests argument; (2) the replacement argument; and (3) the parental interests argument (see Gillam 1999). The aim is to see whether any one of these three arguments is capable of justifying a more permissive attitude towards termination of presumed disabled fetuses in a context of a generally more restrictive policy towards abortion.

One preliminary point: throughout the paper, our focus is on a law which explicitly distinguishes between the termination of (presumed) disabled fetuses and (presumed) non-disabled fetuses in the sense of providing that the former is permissible whilst the latter, in the absence of other contraindications, is not. We do not address the morality of termination decisions made by individual women. While the factors we discuss below may be relevant to such decisions, the countervailing arguments may be very different. As these countervailing arguments are not considered here, it would be wrong to attempt to read the arguments which we make with regard to the law as transferring in any straightforward way to the morality of individual abortion decisions.

The Fetal Interests Argument

The Fetal Interests Argument attempts to justify allowing termination for reason of fetal disability by claiming that termination actually benefits the disabled fetal, by saving it from a life of suffering. It claims that termination in these circumstances can thus be thought of as a kind of fetal euthanasia as death (or non-existence) is the only way to avoid later suffering. As Glover puts it:

... some kinds of life are perhaps worse than not being alive at all ... if it makes sense for people to see death as in their interests, there seems a parallel possibility of parents or doctors thinking that not being born may be in the interests of a potential child (Glover 1989: 129).

There are a number of objections to the Fetal Interests Argument. The first is that it only applies to a very narrow range of cases where the likely alternative to termination is a resultant child whose quality of life is not merely low, but negative: that is, she would, quite literally, be better off dead, or better off never having been born. Whilst we are happy to grant, for the sake of argument, that there are cases in which impairment is so severe that any resultant child will have a negative quality of life (for example, a child suffering from Tay-Sachs) many actual fetal impairments are indisputably not like this. For in most cases, the resultant child will have a quality of life which, although arguably less good than it would have been without impairment, is still positive overall and therefore a 'life worth living'.

The second problem with the Fetal Interests Argument is that this justification of s.1(1)(d) does not seem to fit easily with the jurisprudence which has developed to deal with issues arising from the selective treatment of impaired neonates. For example, in Re A (Children) (Conjoined Twins: Surgical Separation) [2000] 4 All ER 961, the Court of Appeal expressly rejected the finding of the Family Division of the High Court that Mary, a weaker conjoined twin who was only sustained by her sister's blood flow and who would almost certainly die in a matter of months if the twins were not surgically separated, had no continued interest in living. Ward LJ held that Mary's life did have a value to her, even though she had severely limited brain function and was incapable of crying or expressing pleasure or pain. On this issue, he concludes:

The question is whether this proposed operation is in Mary's best interests. It cannot be. It will bring her life to an end before it has run its natural span. It denies her inherent right to life. There is no countervailing advantage for her at all. It is contrary to her best interests.

In this case, then, the Court has judged even a few months of pitifully low quality life to be worth living and preferable to non-existence.

The gulf between the standards applied in these cases and in s.1(1)(d) of the Abortion Act can be clearly illustrated by taking the example of Down's Syndrome. Routine pre-natal testing now predicts a large number of likely cases of Down's, and termination is then offered under s.1(1)(d). However, in the case of Re B (A minor) (Wardship: Medical Treatment) [1981] 1 WLR 1421, the courts held that it would be unlawful to withhold treatment to remove an intestinal blockage from a child suffering from Down's as there was no evidence that it was in this child's best interests to die.

The Fetal Interests Argument, then, is fairly hopeless as a general justification of s.1(1)(d). There are two major problems with it. The first is that it patently fails to justify most of the terminations which are generally thought to satisfy that section's criterion of 'substantial risk' of 'serious handicap', since it only really applies to one small subset of such cases: those where termination amounts to prenatal euthanasia. Hence, even if prenatal euthanasia were justified, the Fetal Interests Argument would still fail to justify the current, widely accepted scope of s.1(1)(d) because most of the terminations currently performed under that section are not cases of prenatal euthanasia. The second problem is that the Fetal Interests Argument is not consistent with the jurisprudence which has developed in other areas, notably around the selective non-treatment of neonates.

The second argument for s.1(1)(d) is the Replacement Argument. This can be explained in the following terms:

The killing of the [disabled] fetal or baby is wrong insofar as its life would have contributed some happiness to the sum total; but if one kills it, and conceives and produces another with no disability, then the sum total will be increased by at least the same amount, and probably more (Hursthouse 1987: 149).

It is now routine for doctors to offer special tests to pregnant women who run an unusually high risk of having an abnormal baby ... If the test shows that the fetal does have Down's syndrome, the woman is able to have an abortion. The same happens with women who are shown to be carriers of the gene for haemophilia: the fetal can be checked to see if it has the disease. If it does, the woman can have an abortion, and then try again, so that she can have a normal baby. Why do we regard this as a reasonable thing to do, even when the handicap is one like haemophilia, which is quite compatible with a worthwhile life? ... [Because] we are offsetting the loss of one possible life against the creation of another life with better prospects (Kuhse and Singer 1985: 158).

Proponents of the Replacement Argument do not normally advocate applying it to adults with disabilities, killing them so that they can be 'replaced' by others without disabilities. This reveals one of the argument's premises: the view that (in comparison to competent adults) fetales have very low moral status (see Gillam 1999: 168):

... on any fair comparison of morally relevant characteristics, like rationality, self-consciousness, awareness, autonomy, pleasure and pain, and so on, the calf, the pig and the much derided chicken come out well ahead of the fetal at any stage of pregnancy - while if we make the comparison with a fetal of less than three months, a fish, or even a prawn would show more signs of consciousness (Singer 1979: 118).

On this view, it is acceptable to 'trade off' the life of one fetal against that of another in a utilitarian way, their status being such that killing one solely in order to generate an increase in the general good is permissible.

It should first be noted that the Replacement Argument relies on a (likely to be a contentious) empirical premise that non-disabled children would contribute more to general happiness than would disabled children. Even if we accept this premise for the sake of argument, there are still a number of problems with the Replacement Argument. The first, a practical one, is that it assumes something which cannot simply be assumed: that the women in question will at least try to become pregnant again, thereby 'replacing' aborted fetuses with new non-disabled ones. This is a problem for the Replacement Argument in two ways. First, it means that the proffered justification simply doesn't apply to many cases covered by s.(1)(1)(d) (those where there is in fact no 'replacement'). Second, and perhaps more seriously, it reduces the argument to absurdity - for what it now appears to justify is a policy under which women are only allowed to abort disabled fetuses if they agree to do their best to 'replace' the aborted fetal with a 'happier' one. For in cases where the woman decides that she doesn't want (or can't have) any further children, the idea of replacement offers no justification for termination.

This objection is clearly a huge problem for the Replacement Argument as applied to the morality of particular abortions. It is not so clear, however, that it poses the same problem when we move up to the level of law and policy. For what can be argued at this level is the following. Admittedly, we cannot guarantee that each particular aborted fetal will be 'replaced' by another non-disabled fetal from the very same mother. However, viewing the situation holistically, it is likely that, if we allow abortion for reason of fetal disability, a group of mostly non-disabled fetuses will come to exist which would not otherwise exist. So, although one-to-one 'replacement' may not always occur, 'replacement' at the group level does. And, provided that the 'replacement' group is substantially happier than and not substantially smaller than the group with disabilities would have been, this should be enough to meet the relevant utilitarian requirements. Hence, the Replacement Argument is able to survive this first objection.

However, the Replacement Argument faces a second and more serious objection: there is a clear tension between the premise on which it is based (the view that the relevant fetuses have very low moral status) and what it is trying to prove (the view that law should, at a given stage, be more permissive regarding the abortion of disabled fetuses than other fetuses). If the fetuses in question really do have very low moral status, such that they can be killed for purely utilitarian reasons, then it is not clear why special provisions covering disability are required. For eliminating disabled fetuses is, at most, only very indirectly related to increasing the general good and furthermore (insofar as it is a means of increasing the general good at all) it is certainly not the only relevant way of doing so. Compare, for example, parents who wish to abort their disabled fetus and then replace it with a non-disabled one with parents who wish to abort a 'normal' fetus and replace it with an 'enhanced' one. In both cases, given the low moral status of the fetus, the utilitarian considerations should win out and it would be at least permissible to abort.

Similar issues would arise if the parents decided on termination because of unforeseen job difficulties, or because they came to see that the world was overpopulated; again the utilitarian considerations should win out. So our point is this: the Replacement Argument relies on the assumption that fetuses at the developmental stage in question have a moral status which is so low that they can be permissibly destroyed for purely utilitarian reasons. With that assumption in place, though, we have a justification not for having a special exception for disabled fetuses - but rather for a much more permissive policy across the board. In other words, the Replacement Argument proves too much. For, if it works, what is warranted is not a special legal provision regarding disability (one like s.(1)(1)(d)) but rather a much more general relaxation of restrictions on abortion.

The Parental Interests Argument

According to the Parental Interests Argument, s.1(1)(d) can be justified by reference to the interests of the biological parents, especially the woman, since the strain of caring for a disabled child may be substantially greater than that of caring for a non-disabled one. This argument is accepted by many academic lawyers as the best rationale for s.1(1)(d) (see for example Grubb 1999, Williams 1978). As Williams claims:

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...(Williams 1978: 256).

This understanding of the section fits well with the general spirit of the Abortion Act 1967 which views abortion as generally undesirable but permissible in certain exceptional circumstances where the mother will be placed at special risk or under particular strain in carrying the pregnancy to term, with due account taken of her 'actual or reasonably foreseeable environment'. There are, however, problems with the Parental Interests Argument, even if one accepts for the sake of argument (as we will here) its empirical premise that caring for a child with a disability is typically more difficult and costly than caring for a child without a disability.

The most serious objection to the Parental Interests Argument has been vocally made by some advocates of 'disability rights'. We will call this the Disability Discrimination Objection. These writers argue that the Parental Interests Argument fails to take account of the fact that many of the problems faced by parents of children with disabilities are the result of social discrimination, rather than of impairment per se:

A disabled fetal represents for parents a problem that may have far more to do with society than with disability. Disabled children confront a hostile environment (Johnson 1990: 14).

At this point, a parallel is sometimes drawn between disability and categories such as skin-colour, sex or sexuality. Given that black and female children face social discrimination, it is asked, how would we respond to a legal provision authorising the abortion of fetuses on the basis that they are black or female and that, as such, their parents (and the children themselves) will be subjected to additional strain and worry? Surely we would want to say that enshrining such a provision in law would be outrageous, amounting to collusion with and (at least implicit) endorsement of discrimination or 'eugenics'? In this vein, Hubbard argues:

Most of us would be horrified if a scientist offered to develop a test to diagnose skin color prenatally so as to enable racially mixed people ... to have light-skinned children. And if the scientist explained that because it is difficult to grow up black in America, he or she wanted to spare people suffering because of the colour of their skin, we would counter that it is irresponsible to use scientific means to reinforce racial prejudices. Yet we see nothing wrong, and indeed hail as progress, tests that enable us to try to avoid having children who have disabilities .. (1997: 187-200).

Similar arguments are made by other commentators who cite the purported existence of the 'gay gene' and ask whether a woman who has decided that a gay child would be destined for a life of prejudice, discrimination and suffering should have her wish to abort respected (Fletcher 1998/9: 2). If science made such a thing feasible, what would we say about a proposed amendment to the Abortion Act, authorising termination on the grounds of 'serious risk' of 'substantial homosexuality'?

These parallels with colour, with sex and with sexuality have prima facie plausibility as well as substantial rhetorical force. The issues raised by them, though, are far from simple. For they include conceptual questions about what disability is and in what ways, if any, it is similar to these other characteristics. It seems to us that if traits like colour, sex, and sexuality share the same status as disability, then Hubbard's comparison (above) is decisive and the Parental Interests Argument must be rejected. However, it is not obvious that all these things do indeed share the same status. In order to explain why, we will focus specifically on the comparison with sex selection.

In what sense, then, is having a disability similar to being female? In order to answer this question, we need to address a dispute between those who advocate a 'social model' of disability and those who propound a more traditional 'individual' or 'medical' model, since disability and sex will only be relevantly similar if the social model is correct. According to the 'medical' model, at least some of the problems faced by people with disabilities are caused by their impairments, with 'impairment' being viewed as a negative health concept, along with diseases, disorders, maladies and so on. More specifically, impairment, on this view, is normally regarded as a harmful subnormal functioning of a bodily part, process, or system (including the brain, in cases of mental disability). The social model, on the other hand, claims that 'disability is a sociopolitical construction, a product of organisation and culture rather than a personal limitation due to a person's impairment' (Reindal 2000: 92). In other words, the social model holds that what makes people disabled is social discrimination, rather than impairment per se. On this view, a woman who is excluded from education or employment because of institutional sexism and a wheelchair-user who is excluded because of the inaccessible built environment are in fundamentally the same position. For both suffer disadvantage solely because of discrimination. If this is true, then it is also true that permitting the selective termination of disabled fetuses on the grounds of their disability is comparable with permitting the selective termination of female fetuses on the grounds of their sex. An abortion law under which a significant risk of being born female was explicitly singled out as an exception to a general ban on abortion would amount to colluding with sexism. So the same could and should be said about the former: permitting and practising the selective termination of disabled fetuses amounts to colluding with (and perhaps also encouraging) discrimination against people with disabilities.

Accordingly, if we accept the social model of disability, then it looks as if the Disability Discrimination Objection to the Parental Interests Argument does hold. Even if terminating particular disabled fetuses would benefit the parents, there is good reason not to have an especially permissive regulation for such terminations, since this would amount to endorsement of serious discrimination against people with disabilities. The fundamental question with which we are left, then, is whether the social model of disability is correct. In what follows, we suggest that there is good reason for thinking that it is not, that disability is not entirely 'social' and, therefore, that the Disability Discrimination Objection to the Parental Interests Argument is at best only partially successful.

The main reason for rejecting the social model is that it ignores the obvious fact that some of the disadvantages associated with disability are not caused by society, but are intrinsic parts of the impairment itself. Harris makes the point well:

I don't believe that it is social factors that make blindness and lameness and deafness into a disability. Social factors may exacerbate the problem of having such disabilities but they are disabilities because there are important options and experiences that are foreclosed by lameness, blindness and deafness. There are things to be seen, heard and done, which cannot be seen, heard, or done by the blind, the deaf and the lame whatever the social conditions (Harris 2000: 98).

Disabilities can involve pain, or reduced life-span, or important inabilities of various kinds: such as the loss of a sense modality or reduced mobility. Where present, these things are intrinsic 'evils': harms which are not caused by society and which, arguably, no amount of social intervention, short of removing the impairment, can take away. So while, clearly, many people with disabilities are the victims of harmful (and wrongful) social discrimination and exclusion, these social factors are not the whole story (Hull 1998: 203). For impairment is necessary for disability and impairments are, by definition, intrinsically harmful to varying extents.

This observation explains why disability is not just like sex. Disability necessarily involves impairment, 'an inability to do something which is characteristic of the species' (Culver and Gert 1982: 77) whereas being male or female does not. Given this, it is plausible to suppose that whilst all (or virtually all) of the special problems faced by women are caused by social discrimination, only some of the difficulties faced by disabled people are caused by such discrimination - the others being caused directly by impairment. Hence, the attempt to discredit the Parental Interests Argument by comparing selectively terminating disabled fetuses with selectively terminating female fetuses is somewhat misleading - failing, as it does, to take account of this important difference between disability and sex.

That said, we do believe that this objection to the Parental Interests Argument still has some force. For one of the reasons why selectively terminating disabled fetuses is judged to be in parents' interests is social discrimination, one aspect of which is a lack of support services for the parents of disabled children or prejudice against those children. In these circumstances, we recommend caution when invoking the Parental Interests Argument to justify s.(1)(1)(d), bearing in mind the danger that in so doing one may be colluding with social discrimination against people with disabilities.

In this paper, we have explored three attempts to defend s.1(1)(d)'s exception to a general prohibition on abortion. The Fetal Interests Argument was unsuccessful because, insofar as it works at all, it only covers a small minority of those cases which fall within s.1(1)(d)'s more general test of 'seriously handicapped'. The Replacement Argument failed for a number of reasons, most notably that it appears to justify too much: not a special exception for disability but, rather, a more generally permissive abortion policy. The position of the Parental Interests Argument is rather more complex. We have argued that it does go some way towards justifying s.1(1)(d) but (and this is an important caveat) the Disability Discrimination Objection shows us that only certain kinds of parental interest should be allowed to count for this purpose. More specifically, parental interests should not be taken to be decisive in relation to s.1(1)(d) where it is in the parents' interests to abort only or mainly because they or their child will be harmed by social discrimination (which may include lack of support services) - because this would amount to colluding with such social discrimination. While we accept, then, a limited version of the Parental Interests Argument, we would argue that the use of such arguments in practice gives cause for concern. For there is a danger that the special acceptability of terminations will be assumed also in cases where all or most of the harm is caused by social discrimination. If parents would feel able to care for a disabled child with greater social assistance and in the absence of prejudice and discrimination, then legally to endorse termination because prejudice exists and assistance will not be forthcoming is worrying. Whilst any society will have to make a choice about where resources are to be allocated, in this instance the availability of abortion may conceal the pressure placed on potential parents by the lack of financial and other support. A more honest and open evaluation of the rationale for s.1(1)(d) should at least contribute to public debate of this issue.

Before finishing, two more general observations should be made. First, the realisation that it is the best interests of parent(s) rather than the quality of life of the eventual child which is the best way of providing a convincing argument for the exception made in s.1(1)(d) adds new weight to a rather old argument: that the fundamental basis of the current regulation of abortion in the England, Wales and Scotland is seriously flawed. How is it that two doctors can be deemed better placed to judge the strain put on parents by the birth of a (disabled) child, than are the parents themselves? What special expertise do doctors have in this regard? It is arguable that the medical profession are well placed to offer some guidance on the nature and extent of a likely disability, the child's life expectancy, and the degree of any pain and discomfort which she is likely to suffer. However, it is unreasonable to expect a doctor to have special expert insight into particular parents' ability to raise a disabled child and the positive and negative impact which such caring responsibilities will have on them. Whilst it might be suggested that, in practice, women will be allowed access to termination in the presence of serious fetal disability and indeed that it may well be unlawful to refuse such access in these circumstances, the fact that technically the legal right to make this decision rests with two doctors is an indefensible anachronism in the context of a legal system which has moved firmly in the direction of according greater recognition to patient autonomy (see Jackson 2000: 467).

Secondly, and finally, it is important to emphasise again that our focus in this paper has been a law which posits certain reasons, and those alone, as acceptable ones for seeking abortion. Our analysis has not considered the issue of the ethics of abortion decisions taken by individual women and cannot be extended to it in any straightforward way. Whilst there is no space here to develop an argument for this here, our own position is that no woman should be forced to carry to term a disabled, or any other, fetal. In other words, termination should be legally justified by the mere fact that a woman does not wish to continue with a pregnancy. So, while ultimately we agree with our opening citation from Tom Shakespeare in believing that the same principles should apply to all abortions regardless of any fetal disability, we believe that such principles should be very liberal.

C. Culver & B. Gert, Philosophy in Medicine (Oxford University Press 1982).

A. Fletcher, 'Testing Times' (1998/9) 16 National Abortion Campaign Newsletter 2.

L. Gillam, 'Prenatal Diagnosis and Discrimination against the Disabled' (1999) 25 Journal of Medical Ethics 163.

J. Glover, Fertility and the Family: the Glover Report on Reproductive Technologies to the European Commission (Fourth Estate 1989).

A. Grubb in I. Kennedy and A. Grubb, Principles of Medical Law (Oxford University Press 1999).

J. Harris, 'Is there a Coherent Social Conception of Disability?' (2000) 26 Journal of Medical Ethics 95.

R. Hubbard, 'Abortion and Disability' in L. Davis (ed.), The Disability Studies Reader (Routledge 1997).

R. Hull, 'Defining Disability - a philosophical approach' (1998) 4 Res Publica 199.

E. Jackson, 'Abortion, Autonomy and Prenatal Diagnosis' (2000) 9(4) S.L.S. 467.

M. Johnson, 'Aborting Defective Fetales - What Will it Do?' (1990) Link Disability Journal, August/September 14.

H. Kuhse & P. Singer, Should the Baby Live? (Oxford University Press 1985).

E. Lee and J. Davey, Attitudes to Abortion for Fetal Abnormality (Pro-Choice Forum 1998).

E. Lee 'Young Peoples' Attitudes to Abortion for Fetal Abnormality' (2000) 10(3) Feminism and Psychology 396.

S. Lee, 'Abortion in the Twilight Zone' in A. Furedi (ed.), The Abortion Law in Northern Ireland: Human Rights and Reproductive Choice (Family Planning Association Northern Ireland 1995).

S. Reindal, 'Disability, Gene Therapy and Eugenics: A Challenge to John Harris' (2000) 26 Journal of Medical Ethics 89.

T. Shakespeare, 'Choice and Rights: Eugenics, Genetics and Disability Equality' (1998) 13 Disability & Society 665.

P. Singer, Animal Liberation: A New Ethics for Our Treatment of Animals (Jonathon Cape 1976).

G. Williams, Textbook of Criminal Law (first edition, Stevens 1978).