The Decriminalisation of Abortion: An Argument for Modernisation

While abortion is now offered as a routine part of modern NHS-funded reproductive healthcare, the legal framework regulating it remains rooted in the punitive, conservative values of the mid-Victorian era. This article argues that this framework is in need of fundamental reform to modernise it in line with the clinical science and moral values of the 21st century. It assesses the current statutory framework regulating abortion against the purposes that are typically claimed to motivate it: the protection of women; and the prevention and condemnation of the intentional destruction of fetal life. It argues that it fails to achieve either of these broad aims and that we should thus remove specific criminal penalties relating to abortion. This, it is suggested, would be likely to have very limited impact on the incidence of abortion but would, however, better recognise contemporary medical realities and moral thinking.


Introduction
Abortion is a criminal offence in England, Wales and Northern Ireland by virtue of a statute passed at a time when 'our society was only on the brink of the beginnings of the modern world'. 1 The Offences Against the Person Act 1861 (OAPA) was passed in the middle of the reign of Queen Victoria, some 20 years before married women were recognised as legal persons able to own property in their own right, almost 70 years before the achievement of women's right to vote on equal terms with men, 2 and at a time that 'in matters sexual was almost unimaginably different from ours'. 3 As such, it is unsurprising that the Law Commission has recognised that the OAPA is severely outdated and is consulting on how it might be modernised. 4 While far reaching in its extent, however, this consultation explicitly excludes offences relating to abortion on the basis that they 'are not included in the 1998 draft Bill or previous Law Commission projects on offences against the person, and raise issues going well beyond the law of offences against the person'. 5 While this is true, the refusal of successive governments to update the law governing abortion leaves intact an archaic legal framework that suffers from many of exactly the same problems that the Commission sees as providing a compelling case for general reform of the OAPA. Moreover, while the harshest punitive effects of the OAPA were mitigated by the therapeutic exception carved out by Abortion Act 1967, that too is now a badly outdated piece of law, with multiple inadequacies rendered ever more apparent in the face of evolutions in clinical practice. This article argues that this legal framework is now in need of fundamental reform to modernise it in line with the clinical science and moral values of the 21st century.
Moral consensus in this area is notoriously elusive and I do not aim to contribute to the very extensive literature regarding the ethics of abortion. 6 However, even within this polarised debate, typically dominated by vocal minorities, it seems to me that the following broad principles that provide the premises for my argument are capable of commanding widespread support in the British context. First, women should be enabled fully to participate in the public sphere on equal terms with men and, prima facie, control of one's own fertility is a fundamental prerequisite for such full participation. 7 Second, states have an important responsibility to support and promote the health, including the 2 Some women gained the vote via the Representation of the People Act 1918, with full female franchise achieved in the Equal Franchise Act 1928. The Married Women's Property Act 1882 changed the law to permit married women to own, buy and sell property in their own right.
3 Smeaton (n 1) [332] Munby J. 4 The Commissions Act 1965, s 3, sets out the Law Commission's duty with regard to 'the elimination of anomalies, the repeal of obsolete and unnecessary enactments, the reduction of the number of separate enactments and generally the simplification and modernisation of the law'. In its own words, the Commission is charged to ensure that the law is fair, modern, simple and effective, see Law Commission, <www.lawcom.gov. uk> (accessed 11 August 2015). 5 Law Commission, Reform of Offences Against the Person: A Scoping Consultation Paper (Law Com, CP No 217, 2014) 54. The draft Bill referenced here was published as part of an earlier Consultation exercise, entitled Violence: Reforming the Offences against the Person Act 1861. 6 For a very small taste of the voluminous literature, see J Finnis and others, The Rights and Wrongs of Abortion reproductive health, of their citizens. 8 Third, while its application in the abortion context is controversial, few would deny the importance of the general principle of respect for patient autonomy in medical practice. 9 In principle, this raises a strong argument in favour of supporting women to make their own, informed medical decisions about a pregnancy, unless there is very good reason to refuse this right. 10 Fourth, it flows from this that there should be robust provision for informed consent before an abortion, with measures to ensure that that the woman's decision is voluntary; that full, accurate, evidence-based information is given about all the options open to her; and that sufficient time is allowed for her to make a decision. Fifth, while not a full moral person with equal ethical status to someone who has been born, the human fetus is of moral value and holds a significance that increases as it grows throughout pregnancy. 11 Sixth, both for this reason and because of the greater risks to the woman at later gestations, other things being equal, it is better for abortions to take place early in pregnancy. Seventh, where abortions are performed, they should be done in accordance with the best available standards of medical practice. Eighth, debate with regard to law reform should be honest: religious values should be weighed as matters of religion, ethical issues should be debated as matters of ethics, and medical claims should be evidenced through a robust scientific base, with no toleration of political ideology masquerading as scientific fact. And, finally, the criminal law, which involves the most onerous and draconian of state powers, should be invoked only where it provides a necessary and proportionate response. 12 In this article, I suggest that taking these principles seriously requires fundamental, root and branch legal reform, serving to decriminalise abortion (which I take to mean the removal of specific criminal prohibitions relating to abortion, without intending that it should be taken out of the ambit of any 8 Reproductive health is recognised as a basic right by the World Health Organisation, which understands it to include the right 'to be informed of and to have access to safe, effective, affordable and acceptable methods of fertility regulation'. WHO, 'Reproductive Health' <www.who.int/topics/reproductive_health/en/> (accessed 11 August 2015). 9 This received a clear judicial articulation in the case of Re T (Adult) [1992] 4 All ER 649 (CA) and, more recently, in the Supreme Court decision in Montgomery v Lanarkshire Health Board [2015] UKSC 11, [2014] 2 All ER 1031. The latter is further discussed below. 10 Even Foster, in a sustained critique of what he sees as undue weight given to autonomy in modern medical ethics and law, does not deny that it has a vital role: C Foster, Choosing Life, Choosing Death: The Tyranny of Autonomy in Medical Ethics and Law (Hart 2009). 11 Neither is the fetus a legal person until it is born alive and separate from the body of its mother, Rance v Mid-Downs Health Authority [1991] 1 QB 587 (QB). I make no attempt to convince those who object to abortion on the basis of what Dworkin calls the 'Derivative Objection': that the fetus is a creature with interests of its own right from the start, including an interest in remaining alive, and that it therefore has the rights that all humans have. As Dworkin suggests, however, moral objections to abortion are, for most, not grounded in this view but rather stem from what he calls the 'Detached Objection': that human life has intrinsic innate value and is sacred in itself, detached from any particular rights or interests: Dworkin (n 6). This final empirical claim appears to be borne out by the polling data discussed at nn 100 and 113-16 and accompanying text. 12  general criminal law offences that apply to medical practice). 13 The guiding principle of such reform would be that where self-induced or requested by the pregnant woman, the destruction of an embryo or fetus would no longer form an independent ground for criminal sanction. This would not, of course, leave abortion in a legal vacuum. Rather, it would be treated as any other area of medical practice, remaining subject to the same range of criminal, civil, administrative and disciplinary regulations that apply to all clinical procedures. Specifically, this should mean that criminal sanction remains available where terminations involve a serious harm to the woman concerned, most obviously, where they are non-consensual.
My argument has three parts. First, I set out the relevant law, demonstrating that it is grounded in the medical and social realities of another era and briefly outlining some of the unjustifiable restrictions that it imposes on contemporary clinical practice. These unwarranted limitations, along with the stigmatising impact of criminal sanctions, 14 give cause to reject any suggestion that reform is unnecessary because abortion providers 'can work around' existing deficiencies in the law. 15 Second, and more fundamentally, I suggest that the OAPA reflects modern moral values as poorly as it reflects modern medical science. I consider the broad historical purposes that are said to underpin the law, arguing that, as currently enforced, our legal framework plays no useful role in fulfilling them in practice. I finish by discussing briefly what decriminalisation would mean in the UK.

Current Law
The law governing abortion provides the oldest extant statutory framework governing any specific medical procedure in the UK. 16 A. The Offences Against the Person Act 1861 First, the OAPA applies in England, Wales and Northern Ireland (but not in Scotland, where abortion remains an offence at common law). It contains three offences that are relevant to the prosecution of abortions: s 58 Every woman being with child, who, with intent to procure her own miscarriage shall unlawfully administer to herself any poison or other noxious thing, 13 The applicability of such offences in the context of abortion is briefly explored in section 3. 14 R Cook, 'Stigmatized Meanings of Criminal Abortion Law' in R Cook, JN Erdman and BM Dickens (eds), Abortion Law in Transnational Perspective: Cases and Controversies (University of Pennsylvania Press 2014). 15 The head of Britain's largest charitable abortion service provider, the British Pregnancy Advisory Service (bpas), reports that ' [m]inisters and officials at the Department of Health have repeatedly said to us that they see no need to change the law because it is possible to ''work around'' its deficiencies. This is not good enough. The law as it stands undermines the delivery of safe, evidence-based abortion services.' A Furedi, 'A Shocking Betrayal of Women's Rights' (Spiked, 28 October 2008) <www.spiked-online.com/index.php?/site/article/5845/> (accessed 11 August 2015). 16 The statutory framework for abortion is contained within the first four pages of the chronologically ordered 270 pages of Blackstone's Statutes on Medical Law (5 th edn, OUP 2007), which begins with the OAPA 1861. or shall unlawfully use any instrument or other means whatsoever with the like intent, and whosoever, with intent to procure the miscarriage of any woman, whether she be or be not with child, shall unlawfully administer to her or cause to be taken by her any poison or other noxious thing, or shall unlawfully use any instrument or other means whatsoever with the like intent, shall be guilty of an offence and being convicted thereof shall be liable to be kept in penal servitude for life. s 59 Whosoever shall unlawfully supply or procure any poison or other noxious thing, or any instrument or thing whatsoever, knowing that the same is intended to be unlawfully used or employed with intent to procure the miscarriage of any woman, whether she be or be not with child, shall be guilty of an offence, and being convicted thereof shall be liable to imprisonment for a term not exceeding five years. s 60 If any woman shall be delivered of a child, every person who shall, by any secret disposition of the dead body of the said child, whether such child died before, at, or after its birth, endeavour to conceal the birth thereof, shall be guilty of a misdemeanor, and being convicted thereof shall be liable, at the discretion of the court, to be imprisoned for any term not exceeding two years.
These provisions were passed without any debate within Parliament or, indeed, outside it, with a remarkable silence in the editorial columns of the Lancet and British Medical Journal and an absence of 'letters to The Times from mid-Victorian clergymen'. 17 Potts, Diggory and Peel conclude that '[s]ociety in 1861 had not developed the machinery to discuss any aspect of sex openly or objectively' and that 'prior to the sort of vocabulary and insight which Darwin and Freud gave to the world, some problems were just not open to analysisand abortion was one of them.' 18 The provisions of the OAPA had largely carried forward those of an older statute, 19 which itself had framed abortion offences in broadly similar terms to those contained in its first (1803) legislative prohibition, when procurement of miscarriage attracted a potential death sentence if the woman was 'quick with child' or a 14-year prison term or transportation where she was not. 20 Since 1861, apart from some minor changes in the available sentences, 21 these ancient provisions have survived unaltered. They make no explicit exception for therapeutic abortion, 22 and provide no difference in available sentence between a woman who self-induces her own miscarriage and a third party abortionist. Further, the OAPA draws no distinction between abortions earlier and later in pregnancy, with any procedure that occurs after implantation (6-12 days after ovulation) potentially caught by the law, 23 creating serious impediments to the development and use of potentially beneficial treatments that operate very soon after intercourse. 24 In line with the punitive values of mid-Victorian Britain, s 58 provides one of the harshest penalties for unlawful abortion imposed by any country in Europe: only Ireland (with a maximum 14-year prison term) currently foresees a similarly onerous sanction. 25 While s 60 is seldom discussed in accounts of abortion, the offence is closely related to s 58, offering the possibility of prosecution for the lesser offence of concealment of birth when a more serious offence (unlawful procurement of miscarriage or murder of a newborn child) is suspected but cannot be proven. 26 While this aim is itself difficult to square with a presumption of innocence, the section might nonetheless be said to retain some modern justification as a public health measure, aiming to prevent the irregular disposal of human bodies. However, such justification would support classifying concealment of birth as an administrative and not a moral offence, implying a far lower penalty. 27 Indeed, any facts that would support a prosecution under s 60 would already also be punishable as such, given that they would ex hypothesi involve the failure to register a birth, the failure to notify the registrar of the place and date of disposal of a dead body 28 and, possibly, the common law offence of preventing the lawful and decent burial of a dead body. 29 As such, there appears no clear need to retain this offence on the statute books and, at the very least, there is a strong case for its continued purpose to be considered as part of the Law Commission's current review.
Sections 58-60 of the OAPA are infrequently charged. Police statistics record fewer than ten prosecutions per year under ss 58 and 59 combined in England and 23 Wales, 30 the great majority of which would appear to have been brought in the context of assaults on a pregnant woman or the non-consensual administration of abortifacients. 31 I have succeeded in finding accounts of just two convictions of women who have unlawfully procured miscarriages in the last ten years (each acting well after viability), 32 and no convictions of clinicians who have done so while acting in a professional role. 33 Cases involving concealment of birth are similarly rare. Only one conviction under s 60 has been legally reported over the last ten years, with a newspaper search revealing a small number of further cases, none of which resulted in custodial sentences. 34

B. The Infant Life Preservation Act 1929
While the OAPA makes no distinction between abortions early and late in pregnancy, a second statute, the Infant Life (Preservation) Act 1929 (ILPA), which applies in England and Wales, 35 prohibits the intentional destruction of 'the life of a child capable of being born alive. . . before it has an existence independent of its mother', unless this is done 'in good faith for the purpose only of preserving the life of the mother'. 36 The statute was not intended to regulate abortions but rather to close a legal loophole whereby someone who killed a baby during the process of spontaneous birth would commit neither the offence of unlawful procurement of miscarriage nor murder, if the child did not yet have an existence independent of the mother and was thus not yet 'a person in being'. 37 The Act contains a rebuttable presumption that capacity for life is acquired at 28 weeks of gestation, 38 reflecting the state of neonatal medicine in the 1920s: subsequent advances mean that today this capacity is generally accepted to be acquired around four weeks earlier. 39 The hypothetical possibility for which the ILPA was introduced is one for which it appears never to have been charged. Rather, the few prosecutions brought under the ILPA (numbering fewer than five per year) seem again to have involved assaults against pregnant women, resulting in miscarriage. 40 In this regard, the ILPA offers an example of the overlapping offences that the Law Commission notes as a matter of concern elsewhere, providing an alternative charge to (late) unlawful procurement of miscarriage under s 58 of the OAPA. Further, if it is accepted that terminations very late in pregnancy are more serious than those that occur earlier, the fact that conviction under the ILPA attracts the same upper sentence as that foreseen in s 58 provides an example of the problematic inconsistency in sentencing cited by the Law Commission as a reason for reform in other contexts. 41 The ILPA is also important for the significant role that it has played in judicial interpretation of the OAPA. It has been held that the word 'unlawfully' in s 58 presupposes that, on the contrary, in certain circumstances abortion must be lawful, with the interpretation of the term inferred from the exception contained in the ILPA: that a miscarriage was procured for the purpose of 'preserving [the woman's] life'. 42 This continues to form the basis for legal abortions performed in Northern Ireland each year, where it has been subject to an extremely restrictive interpretation. 43 Only a very small number of women in Northern Ireland who wish to end their pregnancies are thus able to do so within the jurisdiction, 44 with others either doing so clandestinely or travelling outside it to access legal services. This means that that while taxes in Northern Ireland contribute to NHS-funded services for other UK women, women in Northern Ireland must find the money for a termination and any associated travel and accommodation costs themselves. 45 Along with any problems caused by the need to arrange time off work or find childcare cover, 38  this may put legal abortion beyond the reach of many, leading to increasing reliance on the purchase of abortion drugs via the internet. This offers a potentially far cheaper option and one that avoids the need to arrange time off work or child care cover. However, it leaves women to negotiate the risk of encountering unscrupulous traders who supply pills that are harmful or contain no active ingredients, or simply send nothing at all, leaving them now facing more advanced pregnancies. 46 It is thus no surprise that women in Northern Ireland who seek legal abortions in other parts of Britain are treated at higher gestational ages than other resident women, 47 or that the status quo has been criticised as significantly in breach of human rights norms. 48

C. The Abortion Act 1967
Finally, the Abortion Act 1967 (AA), which applies in England, Wales and Scotland, carves out a detailed therapeutic exception to prosecution for offences relating to abortion. 49 In its current form, the Act provides that: 1.
(1) Subject to the provisions of this section, a person shall not be guilty of an offence under the law relating to abortion when a pregnancy is terminated by a registered medical practitioner if two registered medical practitioners are of the opinion, formed in good faith-(a) that the pregnancy has not exceeded its twenty-fourth week and that the continuance of the pregnancy would involve risk, greater than if the pregnancy were terminated, of injury to the physical or mental health of the pregnant woman or any existing children of her family; or (b) that the termination is necessary to prevent grave permanent injury to the physical or mental health of the pregnant woman; or (c) that the continuance of the pregnancy would involve risk to the life of the pregnant woman, greater than if the pregnancy were terminated; or (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.
In deciding whether the continuance of a pregnancy would involve 'risk of injury to health' for the purposes of s 1(a) or (b), doctors may take account of the pregnant woman's 'actual or reasonably foreseeable environment'. 50 46 Horgan (n 43). 47 73 per cent of NI women are treated at under ten weeks (compared to 80 per cent of resident English and Welsh women) and 87 per cent at under 13 weeks (compared to 92 per cent of English and Welsh women), Department of Health, 'Abortion Statistics, England and Wales: 2014' (June 2015). 48 A successful application for judicial review of the law has been brought by the NI Human Rights Commission, with the judgment still pending at the time of writing, H Macdonald, 'Northern Ireland High Court Grants Judicial Review of Abortion Law' Guardian (2 February 2015) <www.theguardian.com/world/ 2015/feb/02/northern-ireland-high-court-hears-abortion-challenge-rape-incest> (accessed 11 August 2015). 49 The Act's differential operation in Scotland, where it carves out a therapeutic exception to a common law offence rather than statute, is considered by McKnorrie (n 35). 50 AA 1967, s 1(2).
The overwhelming majority of legal terminations are performed on the basis of s 1(1)(a), which explicitly allows for a broad exercise of clinical discretion. 51 Modern abortion procedures are very considerably safer than carrying a pregnancy to term and thus in all cases there will be a basis for a doctor to reach a good faith determination that termination is indicated on the basis of the so-called 'statistical argument' that it would pose a lesser risk to a woman's health than would continuing the pregnancy. 52 There is likewise a clear basis for an abortion to be authorised where two doctors form a good faith view that continuing an unwanted pregnancy is likely to pose a risk to a woman's mental health.
The AA thus recognises an important role for doctors as gatekeepers to abortion services. 53 In addition to placing limitations on who may authorise and perform procedures, the AA restricts the locations in which they may be offered, 54 and sets out notification requirements. 55 Through such limitations, the UK parliament of the late 1960s aimed to address the problem of backstreet abortions, ensuring that henceforth terminations would be performed openly by an appropriately skilled doctor, in approved premises, following a second opinion. While the Act has been subject to repeated attempts at reform, 56 it has been amended just once. In 1990, along with some other minor amendments, 57 the upper time limit for abortion under s 1(1)(d) was extended, rendering abortion potentially lawful until term in the presence of a substantial risk of serious fetal anomaly.
Turning to how the AA has been applied in practice, it can be seen that reported numbers of lawful terminations steadily increased from 1968, before stabilising in the region of 200,000 procedures per year for women resident in England, Wales and Scotland, with 2014 seeing the lowest incidence of abortions for over a decade. 58 This represents an abortion rate that is broadly 57 A new s 3A provided a specific power to approve a 'class of places' for the performance of medical abortions; s 5(2) clarified that both the AA and OAPA are engaged in the context of selective reduction of a multiple pregnancy; s 5(1) extended the AA to offer protection from prosecution under the ILPA. In addition to the various other drafting problems with the AA discussed below, it is noteworthy that this section is worded so as apparently to offer protection only to the doctor and not to other healthcare professionals involved in the termination: I Kennedy and A Grubb, Medical Law in line with that seen in other Western countries. 59 One in three women will have an abortion at some point in her life, 60 making this the most common gynecological procedure performed in the UK and one that is sought by women of all ages and from all walks of life. The majority (and a steadily increasing proportion) of terminations take place early in pregnancy, with 92 per cent carried out within the first 12 weeks, only 2 per cent at over 20 weeks, and one tenth of one percent after 24 weeks. 61 While unsafe abortion remains one of the most significant causes of maternal mortality worldwide, 62 in line with the hopes of those who had advocated liberalisation of the law as a public health measure, mortality resulting from abortion is now virtually unknown in the UK, with termination very significantly safer than carrying a pregnancy to term. 63 With the notable exception of terminations for women in Northern Ireland, 64 almost all procedures are funded by the NHS. 65 In sum, since 1967, abortion has become entrenched as a normal part of routine healthcare, with the AA offering a platform for the provision of safe, high quality, state-funded services, typically provided in the first trimester of pregnancy.
However, 50 years is a long time in clinical practice and the multiple cracks in what is now a very dated statutory framework are clear. In 1967, the overwhelming majority of abortions were performed by risky, technically demanding surgical techniques, 66 whereas today abortions are generally performed by straightforward, highly effective, low-risk procedures in early pregnancy. 67 Further, a high level of deference to medical authority made it natural to entrust doctors with the kinds of social and ethical decisions that know precisely what impact the AA has had on the numbers of abortions performed because of the difficulty of obtaining data regarding illegal procedures. 59 15.9 per 1000 resident women in England and Wales aged 15-44; 11.0 resident women per 1000 resident women in Scotland. This is the lowest rate for 17 years in each country, Department of Health, 'Abortion Statistics' (n 47), ISD, ibid. Globally, the age standardised abortion rate stood at would today be seen as self-evidently belonging to patients. And the dangerous backstreet procedures that provided such compelling impetus for reform have all but disappeared in the face of the availability of safe, legal, state-funded services. Yet while these underlying concerns have largely evaporated, the legal infrastructure that was shaped around them continues to have a significant effect on the way that services can be offered.
First, the Act's requirement that two doctors must certify the need for an abortion is grounded in the assumption that doctors, rather than women, are best placed to decide whether an abortion is justified. In 1967, the understanding that medical control of abortion should go well beyond that which would accrue merely on the basis of a technical expertise reflected the belief that 'social conditions cannot be and ought not to be separated from medical considerations' and that the AA 'by its very drafting. . . [encourages] the concept of socio-medical care'. 68 While this captures well the values that characterised 1960s practice, modern medicine has shifted fundamentally away from 'doctor knows best' paternalism: today patients are routinely trusted, and indeed expected, to make medical decisions for themselves, with pregnant women not treated as an exception to this fundamental legal principle in other contexts. 69 Contemporary abortion practice reflects this same evolution in attitudes, with the broad wording of s 1(1)(a) having permitted doctors to exercise their discretion liberally in favour of authorising abortions. However, in the context of a consistently liberal interpretation, the requirement for two medical signatures becomes an entirely bureaucratic one, serving no obvious broader purpose. 70 Moreover, it has also been suggested that this requirement may, in some circumstances, breach the European Convention on Human Rights. 71 Second, the legal requirements that abortions should be performed only by a doctor and only on approved premises are likewise unsupported by any current medical evidence base. These provisions reflected a desire to eradicate dangerous, clandestine abortions and to recognise the best practice of the late 1960s when, as noted above, abortion was a far more technically demanding and risky procedure. Today, however, these requirements have become particularly nonsensical in the context of early medical abortion (EMA), which accounts for around half of the terminations performed in 68 D Steel, 'Foreword' in K Hindell and M Simms (eds), Abortion Law Reformed (Peter Owen 1971) 7. 69 Montgomery (n 9); Re MB (Adult, Medical Treatment) [1997] 38 BMLR 175 (CA). 70 The argument that it is this liberal interpretation of the law that is at fault is discussed at nn 108-12 and accompanying text. 71 R Scott, 'Risks, Reasons, and Rights: The European Convention on Human Rights and English Abortion Law' Med L Rev (forthcoming). Scott argues that to make access to lawful abortion within early pregnancy conditional on fulfilment of the terms of s 1(1)(a) is an unjustified interference with a woman's private life under article 8(2). She also raises concerns regarding the lack of a system of formal review in the event that doctors decide not to grant a termination.
England and Wales and over 80 per cent of those in Scotland. 72 In EMA, there is no clinical need for drugs to be taken on approved premises nor for it to be a doctor who administers or prescribes them. Indeed, the same drugs are already taken at home in other contexts; 73 a woman undergoing an EMA is allowed to leave the approved premises immediately after taking them in order to arrive home before her miscarriage begins; the most commonly used EMA drugs are comparable to or safer than many drugs which are routinely prescribed by appropriately trained other providers; 74 and nurses are already permitted to prescribe mifepristone, one of the drugs used in an EMA, for other medical reasons. 75 Further, looking beyond EMA, while late surgical procedures are likely to require the training and skill of an experienced doctor, it seems plausible that earlier procedures might be performed equally well by other trained professionals. 76 These restrictions thus appear redundant in terms of safeguarding women's health and, moreover, their rigid enforcement risks impeding the efficient delivery of services so as to delay timely access to abortion. Given the greater risks involved in later terminations, this creates a clear potential for these provisions to increase the dangers to women seeking abortion services. 77 The questions of where and by whom abortion procedures can be safely provided are, of course, empirical ones raising important health concerns that should be answered through reference to a robust evidence base. The modest claim defended here is that the evidence base in question should be that offered by best current medical practice and knowledge, rather than that of the 1960s.
In sum, UK abortion law is characterised by archaic language, overlapping offences, inconsistencies in available sentences and clinically unwarranted restrictions on best practice. It has also been argued that it breaches international human rights obligations. 78 So far as possible, service providers have worked around the deficiencies in the law, resulting in a situation of good 72 Department of Health, 'Abortion Statistics' (n 47), and ISD Scotland (n 58). In 2014, for the first time, medical abortions accounted for over half (51 per cent) of the total number of abortions performed in England and Wales, Department of Health ibid. The term 'medical abortion' is used to refer to any termination of pregnancy that is provoked using drugs. 73  access to state-funded services in England, Wales and Scotland. This, in turn, has served to mitigate some of the worst consequences of the very restrictive provision in Northern Ireland and maternal mortality resulting from abortion has been close to eradicated in the UK. However, in addition to stigmatising women and service providers, this criminal law framework creates a number of clinically unwarranted impediments to the provision of high quality abortion services. If the need for good, modern abortion services is accepted, it is thus important to question whether these negative consequences can be justified with reference to any ongoing useful role played by the existing criminal law framework in policing its boundaries. I move now to consider this question.

The Historical Purposes of Criminalising Abortion
Criminal law represents the most onerous, intrusive and punitive of state powers and it is reasonable to assume that it should be invoked only where it offers a necessary and proportionate means to achieve an important objective, with the onus on those who would seek to deploy it to demonstrate that these criteria are met. 79 Moreover, any such demonstration should be subject to particularly robust scrutiny in the abortion context, given the significant considerations of gender equality, autonomy, and reproductive health that point powerfully in favour of liberal access to safe, legal services. I move now to consider the purposes served by ss 58-60 of the OAPA, read in the light of the therapeutic exception carved out by the AA, in order to assess whether they outweigh these other kinds of considerations.
First, it is necessary to identify what precise purpose is served by these sections. While the legislation clearly reflects archaic, highly conservative attitudes to gender norms, female sexuality and fertility control, 80 and has been read as part of the medical profession's fight to establish professional dominance over the management of pregnancy and childbirth, 81 it is commonly taken as representing an ongoing commitment to two specific purposes. It is said to be necessary, first, to prevent or to condemn the intentional destruction of fetal life; and, second, to prevent harm to women. 82 As noted above, the ILPA was introduced for very specific on the Elimination of Discrimination Against Women' (A/54/38/Rev 1, 1999) paras 309-10; CEDAW, 'Summary Record of the 844th Meeting' (CEDAW/C/SR 844, 2008) para 289. 79 Jareborg, Husak, Packer (all n 12). This principle might be extended to suggest that criminal laws should be occasionally reviewed and not allowed to stagnate on the statute books, Packer (n 12). 80 As a powerful illustration of late 19th century moral norms, Munby J describes the 1878 case of Annie Besant, whose daughter was removed from her custody with Besant judged unfit to raise her because she had written and published a treatise on contraceptive methods, Smeaton Today, the OAPA and ILPA must also be read in conjunction with the AA, which is equally said to be underpinned by two broad parliamentary purposes. First, the AA also reflects a concern with preventing harm to women, aiming to 'ensure that the abortion is carried out with all proper skill and in hygienic conditions'. Second, it was to extend access to abortion in a way that foresaw ongoing control over a controversial procedure, being intended 'to broaden the grounds upon which abortions may be lawfully obtained', permitting only those abortions that were deemed 'socially acceptable'. 83 Combined, then, we might say that the current criminal framework aims both to prevent harm to women and to prevent or condemn the intentional destruction of fetal life when this does not take place within tightly medically controlled circumstances. How well does current legislation further these goals?

A. Preventing Harm to Women
In 1861, abortion was a technically demanding, dangerous surgical procedure, offering clear medical grounds to support limiting its use to only the most compelling of cases. 84 By 1967, while termination procedures were far safer, they still carried significant risks. 85 Today, however, the claim that the restrictive, criminal prohibitions contained in the OAPA might be in any way justified by concerns for women's health is simply unsustainable. As noted above, in the UK abortion carries a far lower risk of maternal death than does carrying a pregnancy to term. 86 Claims that abortion causes breast cancer or infertility have been demonstrated to be scientifically baseless. 87 Likewise, in mental health terms, there is no good evidence to support suggestions that abortion injures women psychologically. 88 While women are not harmed by safe, legal abortion, however, they are significantly and demonstrably harmed where the enforcement of restrictive criminal laws obliges them to seek out illegal terminations. While maternal mortality resulting from abortion has now been close to eliminated in the UK, 89 unsafe abortion is estimated to result in chief evil of an abortion is no longer thought to be the loss of the unborn child, but the injury done to the mother by the unskilled abortionist'. 83  around 47,000 deaths each year worldwide, with these deaths overwhelmingly concentrated in countries with strictly enforced prohibitive legislation. 90 Further, there is some evidence to suggest that women with negative attitudes towards abortion are more likely to experience mental health problems following a termination, 91 and it therefore also seems probable that the stigmatising effect of criminal prohibitions on abortion may contribute to damaging women's psychological health.
In Britain, the prohibition on abortion must be read in connection with the therapeutic exception carved out by the AA. As noted above, the liberal interpretation of the latter has meant that the very few prosecutions for unlawful procurement of miscarriage or child destruction have tended to be reserved for cases involving non-consensual or very late terminations. However, the offences also potentially capture healthcare professionals who fail to comply with the bureaucratic requirements imposed by the AA and those women who are unable or unwilling to access legal services. Women on the Web, a feminist medical collective that prescribes and supplies abortion drugs via the internet, reports that drugs are frequently supplied to women in Northern Ireland and that regular requests are received from elsewhere in the UK, with the following cases typical: We had an Islamic girl forbidden from leaving the house without a chaperone. How is she going to get to an abortion clinic? She can't. For her, her only option might be that she could get the medicine sent to her by post. We have British women in abusive relationships whose boyfriend will beat the hell out of her if he finds out she is pregnant and wants an abortion. 92 While it would clearly be preferable for these women to have the possibility of accessing formal health services (and, where necessary, a referral to other support), the threat of potential life imprisonment does not obviously perform any useful role in protecting them. Further, insofar as the purpose of the abortion prohibition is to safeguard women's health, it seems odd to include women themselves within its scope: concerns regarding the dangers of women being injured in the course of elective cosmetic surgery, for example, have rightly led to calls for greater regulation of services rather than demands that women who put their health in the hands of unskilled providers should be punished for so doing. 93 These women are, of course, unlikely to be prosecuted. However the fact that a law is not likely to be enforced is the poorest of justifications for its retention.
Might the existence of legal abortion be harmful to women in a different way, leaving them open to coercion to terminate wanted pregnancies? This claim has been significant in discussions regarding sex selective abortion, where it was presented as the dominant concern motivating one recent reform attempt purporting to clarify the illegality of this practice. 94 The sponsor of the Bill, Fiona Bruce MP, described three cases, each of which focused on harm to pregnant women. Her first case was that of a vicious, unlawful assault by a husband on his pregnant wife, following his discovery that she was carrying a female fetus. 95 His actions would clearly be punishable under existing criminal law, whether or not he provoked a miscarriage, and while her situation highlights the need for services to support those suffering domestic abuse, the attacked woman would be assisted in no obvious way by a criminal prohibition on sex selective abortion. Bruce's second example concerned women who come under familial pressure to abort female fetuses, being forced to lie to abortion providers about their reasons for seeking to terminate a pregnancy. 96 Such cases underline the importance of robust procedures in clinics to ensure the voluntariness of a woman's consent: specifically, clinics should see each woman with no escort present, in order that she is given the fullest opportunity to discuss any pressures on her; and any translator used should be independent (rather than a partner or family member). 97 Again, however, it is unclear that a specific prohibition on sex selective abortion adds anything to the existing law (particularly in circumstances where women are coerced to lie about their reasons for terminating a pregnancy). Further, if the harm cited involves coercion, there is no obvious reason to single out sex selection for specific regulation: the voluntariness of consent is important in all cases. Bruce's third example is the story of Rupinder, who had chosen to terminate her pregnancy on the basis that she was expecting a girl: [Rupinder] was the eldest of six girls and she recalls that each time her mother went to hospital how disappointed everyone was when each time it was a girl. This experience traumatised and consumed her so much that the thought of giving birth to a girl meant disappointment, betrayal and lowered status within the family and the community. Rupinder made a painful decision to abort which she now regrets as she felt that she had no other choice. 98 This tragic case again underlines the need for robust informed consent provisions, the availability of high quality counselling and for women to be allowed adequate time fully to consider their choices. However, it is naïve to imagine that banning sex selective abortion would address the structural sexism identified here and it is these cultural pressures that are harmful to Rupinder rather than the existence of legal abortion, not least as refusing access to such services does not remove the possibility of terminating a pregnancy but only limits her ability to do so safely. It is also cruelly ironic to focus efforts on addressing such sexism through the state aligning itself with Rupinder's family and community in refusing her control over her own fertility. In sum, the structural sexism that leads to the practice of sex selective abortion is deplorable, a fortiori, when it manifests itself in violence and coercion. However, the examples above do not present a case for specific criminal prohibition but rather illustrate that fully respecting women's autonomy in this context requires not just robust consent procedures but also active commitment to securing the best possible conditions within which reproductive choice may be exercised. Moreover, there are strong practical reasons for being wary of seeking a response to these problems within the criminal law. A specific prohibition on sex selection is likely to be unworkable in practice and, if rigorously pursued, could not fail to be highly intrusive. Either screening out the very small number of cases where a termination might be sought for this reason would involve close questioning of all woman (not least given Bruce's concern with women persuaded to lie about their motivation) or, alternatively, it might potentially lead to a kind of racial profiling, with enhanced suspicion and scrutiny of women from particular ethnic communities.
Bruce's arguments reflect the currency of what Siegel has identified as a significant 'woman protective turn' in arguments for restricting access to abortion. 99 However, I have argued that the claim that liberal access to abortion harms women is as unconvincing in this specific context as it is more generally. If there is a continuing purpose for criminal prohibitions against abortion, then this can only lie in the claim that they are necessary to prevent or condemn the intentional destruction of fetal life.

B. Preventing or Condemning the Intentional Destruction of Fetal Life
As I noted earlier, in Britain there would appear to be significant support for the view that while not a full moral person with equal ethical status to someone who has been born, the human fetus is of moral value and holds a significance that increases as it grows throughout pregnancy. 100 However, this alone is insufficient to ground a criminal prohibition on abortion. First, the moral respect due to fetal life must be weighed against the significance of respect for 100 See the polling data considered below (at nn 113-1-6 and accompanying text). As I also note above also, my argument is not likely to convince those who believe that all human life is sacred or that abortion is morally equivalent to murder. However, I follow Dworkin (n 6), in assuming that moral objections to abortion are, for most, not grounded in this view, with this assumption supported by the fact that only 6% per cent of those questioned in a recent survey believed that abortion should be banned in all circumstances: see YouGov poll for the Sunday Times, http://cdn.yougov.com/cumulus_uploads/document/y4asheswh1/YG-Archives-Pol-ST-results-13-150112.pdf (accessed 4 June11 August 2015), discussed further below (at n 148). women's autonomy, gender equality and reproductive health. And, second, even if the scales are judged to come down on the side of protection of fetal life, it is also necessary to consider whether a criminal prohibition performs a useful function in achieving that end (either through preventing or condemning the destruction of the fetus).
Do restrictive abortion laws serve to prevent abortion? While the answer to this question is less straightforward than is often suggested, it is clear that even the most stringent attempts to enforce restrictive abortion laws will not succeed in all cases. At the extreme, Ceausescu's Romania saw abortion prohibitions rigorously enforced, with illegal abortion nonetheless remaining commonplace and maternal mortality rates soaring. 101 International data confirms that strictly enforced legal prohibitions are, at best, a poor indicator of low abortion rates and, indeed, tend to correlate negatively with them. 102 Rather, a lower incidence of abortion reflects a lower incidence of unplanned pregnancy which, in turn, reflects the availability and use of contraception. 103 However, it is also true that not all women who are refused a legal termination will seek out and secure an illegal one: sometimes, a refusal will result in a pregnancy continuing. While there is no good data regarding the relative incidence of illegal abortions and continuing pregnancies following a refusal, we can conclude that restrictive laws will prevent some, but by no means all, abortions. The cost of preventing some terminations and, thus, saving some fetal lives in this way must thus be measured not just in the moral, social and physical harms of enforced pregnancy, childbirth and child rearing; it must also be measured in the financial and emotional costs to women of needing to access terminations outside of their own jurisdiction and the maternal mortality and morbidity that typically accompany illegal abortions.
In many countries, a response to this moral calculation has been nonenforcement of the restrictive, punitive laws retained on the statute books, turning a blind eye to widespread disregard of them. Britain offered a clear example of this phenomenon even before the partial decriminalisation achieved by the AA, with few convictions for abortion offences and vanishingly small numbers of prosecutions of women who had undergone terminations or of doctors who had performed them in line with good medical practice. 104  102 Abortion rates tend to be lower in sub-regions with liberal abortion laws: the lowest sub-regional rates of abortion (12 per 1000 women) are in Western Europe, where laws are least restrictive, and some of the highest sub-regional rates (29-39 per 1000) are in Latin America, where laws are generally very restrictive: Sedgh and others (n 59). 103 ibid. See further, CF Westoff, A New Approach to Estimating Abortion Rates (DHS Analytical Studies No 13, Macro International Health 2008). 104 For discussion, see Sheldon (n 53) 21-24.
appears to coexist with a significant incidence of illegal terminations using drugs purchased on line. The absence of appetite for prosecuting the women who break the law in this way was recently demonstrated by the lack of any official response to an open letter, signed by over one hundred people, stating that they had either terminated pregnancies using this means or assisted others to do so. 105 This situation of practical access to abortion and non-enforcement of the criminal law can be viewed as a response to the moral calculation described above, suggesting that as a society we have already implicitly chosen to value women's autonomy and health over the attempt to protect fetal life through the criminal law. This is demonstrated through the weak enforcement of the law, the fact that even opponents of liberal abortion law now often frame their arguments in terms of women's health rather than the sanctity of fetal life, 106 and in surveys showing strong popular support for a woman's right to choose. 107 In other contexts, the fact that criminal prohibitions are so rarely and selectively enforced might be accepted as reason for their removal. Here, however, the stigma attached to abortion and the reluctance of politicians to confront the issue has led to legislative stagnation and the achievement of good access to legal abortion services through an implicit acceptance of liberal interpretation of the law, rather than the statutory reform that would be necessary to bring it into line with modern practice. Yet in the light of this current liberal interpretation, it is difficult to escape the conclusion that our current abortion legislation serves no ongoing purpose in preventing the destruction of fetal life.
Here, of course, it might be objected that the appropriate response is precisely not to accept this liberal exercise of discretion as a basis for viewing controls over abortion as redundant but rather to demand that they be more rigorously enforced. Such a view seems implicitly to inform the Department of Health's recent 'Guidance in Relation to Requirements of the Abortion Act 1967', which suggests that whilst not strictly legally required, it is 'good practice' for at least one of the doctors who authorises an abortion to see the pregnant woman in person (rather than relying on the assessment of other members of the healthcare team, as might be accepted medical practice in other contexts). 108 It further provides that the doctor must make an individual assessment of the woman, rather than simply relying on a general application of the 'statistical argument' described above as a justification for the legality of any early termination, or relying on the assessment of other members of the multi-disciplinary team treating the woman. 109 No justification for requiring a more robust level of medical scrutiny appears in the 'Guidance' and, notably, reference is given neither to legal authority nor to a medical evidence base in support of the Department's restrictive reading of the legislation, leaving the reader to speculate as to what it aims to achieve. It may be, for example, that the intention is that an enhanced level of medical scrutiny is intended to result in more women being dissuaded from terminating pregnancies or refused access to legal services. One powerful justification advanced in the 1960s in favour of entrenching the doctor as 'gatekeeper' was precisely that doctors might somehow take control of a woman's situation and offer the kind of persuasion and support that would convince her to continue with her pregnancy. 110 Yet whatever force this idea had in the 1960s, to modern eyes it appears troublingly coercive to suggest that the doctor's role should be one of active discouragement of abortion. Such conduct would constitute as clear a breach of the professional obligation to provide accurate information and non-directive counselling as would an attempt to persuade a woman to end a pregnancy (as the 'Guidance' appears to recognise). 111 Further, there would appear to be no strong contemporary support for requiring doctors to refuse more abortions: at least in early pregnancy (when a large majority of terminations occur), modern views tend towards seeing abortion as the woman's choice. 112 As noted above, this is implicitly accepted even in many political attacks on abortion services, which often argue not against liberal abortion access per se but rather suggest that existing consent provisions are insufficiently robust, leaving vulnerable women open to exploitation and abuse. Yet if it is not the intention that the doctor's role should be to refuse or to persuade against abortion, then demanding a tighter level of medical scrutiny appears, oddly, to be demanding that control happens entirely for its own sake.
Finally, it might be suggested that even if the criminal law is not enforced, it nonetheless offers the best available means for society to express its strong condemnation of the intentional destruction of fetal life as an important prima facie moral wrong or, at least, to underscore the moral gravitas of the abortion decision. However, even putting to one side more general concerns regarding the deployment of criminal law as a means of expressing moral disapproval, there are reasons for rejecting the specific application of this justification here. First, it seems strange to require our criminal law to express a moral message that is so poorly aligned with contemporary moral views on abortion. In a recent poll, just over half of those surveyed supported the view that 'a woman should not have to continue with her pregnancy if she wants an abortion'. 113 A second question, asked in the same survey, provided an even stronger response (with the difference between these two figures perhaps reflecting a restrictive view of the appropriate role of government in this context): when asked to select the statement that best reflected their views, only 17 per cent selected the statement that 'the Government has a responsibility to reduce the number of abortions', compared to the 70 per cent who chose the statement that 'it's a woman's right to choose whether or not to have an abortion and the Government should not interfere'. 114 While any polling data will be influenced by virtue of the exact question asked, no major poll in the last five years has identified the kind of substantive moral consensus against abortion that might justify its criminal prohibition, at least prior to viability. 115 This remains the case for polling data gathered from those who identify as Christian. 116 Second, the idea that the moral gravitas of a decision must be communicated by ensuring that it is made by doctors, rather than by the women who must live with the consequences of it, provides a clear reflection of the medical paternalism of the 1960s, which is poorly aligned with the values that inform modern British medical law. 117 The significance of the shift that has occurred over the last decades was powerfully recognised in the recent, unanimous, seven judge Supreme Court decision in Montgomery v Lanarkshire (2015), reflecting the extent to which social and legal developments over the last 113 15 per cent very strongly agreed, 12 per cent strongly agreed, 27 per cent agreed and 17 per cent disagreed. Ipsos MORI, 'Public Attitudes towards Abortion' <www.ipsos-mori.com/researchpublications/ researcharchive/2854/Public-Attitudes-towards-Abortion.aspx> (accessed 11 August 2015). A representative quota sample of 953 adults were interviewed face-to-face in their own homes in August 2011 at 156 sampling points across Great Britain, with data then weighted to match the profile of the population. 114 ibid. 115 Packer's authoritative account suggests that 'the criminal sanction should ordinarily be limited to conduct that is viewed, without significant social dissent, as immoral. The calendar of crimes should not be enlarged beyond that point and, as views about morality shift, should be contracted' (n 12) 264. For a counterview, see G Lamond, 'What is a Crime?' (2007) 27 OJLS 609, 617. The issue of later terminations is considered below. 116 Ipso MORI conducted 1136 face-to-face interviews with those who were recorded as Christian in the 2011 Census, or who would have recorded themselves as such. Of those interviewed, 63 per cent agreed that, within the legal time limit, an adult woman with an unwanted pregnancy should be able to have an abortion if she wants one, compared to 20 per cent against, with the remainder neither agreeing nor disagreeing, not knowing, or preferring not to say. See Ipsos MORI for Richard Dawkins Foundation for Reason and Science (UK), 'Religious and Social Attitudes of UK Christians in 2011' <www.ipsos-mori.com/researchpublications/researcharchive/ 2921/Religious-and-Social-Attitudes-of-UK-Christians-in-2011.aspx> (accessed 11 August 2015). 117 See Jackson for an early analysis of the tension between abortion legislation and the increasingly strong commitment to patient autonomy in medical law: E Jackson, 'Abortion, Autonomy and Prenatal Diagnosis' (2000) 9 Social & Legal Studies 467. decades 'point away from a model of the relationship between the doctor and patient based upon medical paternalism ': 118 What they point towards is an approach to the law which, instead of treating patients as placing themselves in the hands of their doctors . . . treats them so far as possible as adults who are capable of understanding that medical treatment is uncertain of success and may involve risks, accepting responsibility for the taking of risks affecting their own lives, and living with the consequences of their choices. 119 In the light of such a shift, which is equally visible in codes of professional ethics, 120 the fact that ending a pregnancy involves a morally serious decision appears to be all the more reason for formally recognising, and strongly communicating, that it is pregnant women themselves who must make it.
Thirdly, in any case, it is by no means obvious what message the OAPA succeeds in expressing. Given that people are far more likely to be aware of the widespread availability of abortion services in Britain than of the wording of the relevant statutes, they might reasonably believe that law communicates a rather permissive message. 121 For those who know the formal letter of the law, the criminal prohibitions of the OAPA taken alone might appear to express the idea that abortion is a serious moral wrong (potentially meriting the same life sentence as does murder), 122 at all gestations, and regardless of by whom it is performed. However, read in conjunction with the AA, the message is rather different: that abortion is seriously morally wrong when not carried out under medical orders and in line with the best medical practice of the 1960s. In this light, the clearest message expressed by the two statutes taken together might appear to be one of women's relative incapacity to make morally significant decisions and a refusal of the importance of updating laws in line with modern medical science, even when this serves to hinder clinical best practice.
In sum, our abortion law, as currently interpreted, fails to fulfil any demonstrable modern purpose. While its bearing on the incidence of abortion is unknown, it serves to stigmatise women seeking abortions and those who care for them, 123 and to impose clinically unwarranted, purely bureaucratic restrictions on medical practice. While it is likely that a more rigorous enforcement of the law would result in preventing some (but by no means all) abortions, this would come with inevitable costs measurable not just in terms of gender equality, reproductive health, and autonomy but also, potentially, in maternal mortality and morbidity. These are costs which modern British society appears unwilling to pay. Further, it would be contrary to accepted ethical practice for doctors to attempt to dissuade women from terminating pregnancies and there appears to be no general public appetite for Government to assume a more active role in seeking to reduce the number of abortions. 124 Yet if the law is not enforced in this more active, restrictive way, it appears to be redundant and this, in itself, offers a strong justification for root and branch reform.
Legal reform resulting in abortion being available on request as a part of mainstream healthcare services would serve to update our laws in line with current medical practice and modern moral values. Moreover, such a move would provide a far better reflection of the broad principles set out at the beginning of this paper. It would recognise the importance of fertility control as a key part of ensuring women's reproductive health and full participation in society, and it would accord with respect for patient autonomy, removing a significant anomaly in the current law. The elimination of the current unnecessary bureaucracy built into the existing framework might make a modest contribution to abortions taking place earlier in pregnancy and in ways that are safer, more effective and more acceptable to women. 125 Further, there is no reason to believe that decriminalisation would have any negative impact on provision for informed consent: this would remain, as now, subject both to the standards of general medical practice and specific professional guidance. Finally, as noted throughout, a substantial evidence base supports the clinical safety and acceptability of these changes.
The most significant objection to my argument lies, of course, in the claim that decriminalisation of abortion would offer less appropriate recognition of the moral respect due to the human fetus. I conceded at the outset that I am unlikely to convince those who hold that the fetus is a full moral person and that ending a pregnancy is morally equivalent to murder. 126  preventing the intentional destruction of fetal life: were the law to be modernised in the way that I suggest, there is no reason to believe that this would have a significant impact on the incidence of abortions. 127 Second, liberalising the law is likely to have a modest effect in further improving the proportion of abortions that take place very early in pregnancy, 128 with this a welcome outcome for those who take a gradualist view of the moral value of fetal life. Most notably, if decriminalisation paves the way to the opening up of abortion services within Northern Ireland, then early abortion rates for Northern Irish women might potentially improve in line with those elsewhere in the UK. Third, there is a wealth of evidence that suggests that a concern for protecting fetal life can be more effectively pursued through policies that attack the incidence of unwanted pregnancy (for example, through improving the quality of sex education and contraceptive provision, and making motherhood a more realistic possibility for women struggling to balance childcare alongside other commitments). 129 Fourth, for those who accept the merits of an expressivist role for the law, I have suggested above that the message communicated by our current legislation is, at best, ambiguous. Decriminalisation would, however, mean that it would be women who would carry the responsibility for decisions regarding abortion, including weighing the ethical significance of ending the life of the embryo or fetus. Reform might, therefore, be seen as expressing the view that women are as capable as their doctors of making morally serious decisions. Finally, it should be acknowledged that a minority of people are likely to continue to believe that abortion constitutes a significant moral wrong. In a plural democracy, it is important that nothing would require those who hold this view to make use of abortion services. Further, of course, they would retain the right to make known their views and to attempt to convince others through legitimate forms of political protest. Finally, as I suggest below, a right of conscientious objection could offer continued protection to those healthcare professionals who do not wish to be involved in the provision of abortion services. However it is equally important that the views of a vocal minority should not be able to impede access to services or to stigmatise the many who take an equally sincerely held different moral view.

The Extent and Effect of Decriminalisation
The removal of specific criminal penalties relating to abortion in the UK would, of necessity, require a process of parliamentary reform, 130 involving a 127 The example of Victoria is instructive here (n 131 and accompanying text). I address later terminations and Northern Ireland as two possible exceptions to this claim at nn 143-50 and accompanying text. 128 Science and Technology Committee (n 73). 129 See Sedgh and others (n 59) and Westoff (n 103). 130 There is no mechanism whereby statutory provisions might simply be swept away by decision of a UK court, as in the Canadian Supreme Court decision in R v Morgentaler [1988] 1 SCR 30. Even if (aspects of) the radical revision of the law, yet one that would be likely to have anything but a radical impact on practice in England, Wales and Scotland. I have space here to do no more than to suggest some broad principles that should inform such a process. Most fundamentally, under the reform proposed, the destruction of fetal life would no longer provide an independent justification for criminal sanction, though such sanction should remain available to recognise the important harm done to a woman who is subjected to a non-consensual abortion. Below, I briefly consider the general impact of the removal of specific criminal penalties relating to abortion, before noting two particular 'hard cases', which would require careful consideration within any reform process.

A. The Broad Impact of Decriminalisation
In 2008, the Australian state of Victoria followed Western Australia and the Australian Capital Territory in decriminalising abortion, removing prohibitions that had been closely modelled on those contained in the OAPA. 131 The reform was designed to modernise the law, bringing it into line with current clinical practice and making terminations neither more freely available nor more difficult to access. 132 In the words of one commentator, the resulting legislation represented: a profound shift in the relationship between the state and its female citizens. It changes both nothing and everything. Nothing, because the number, rate and incidence of abortion will not change. And everything, because for the first time women will be recognised as the authors of our own lives. With that comes our full citizenship. 133 Given current, liberal access to abortion services within the existing law, there is good reason to believe that this claim would hold generally true in the UK.
Further, there seems little reason to fear that sweeping away specific criminal prohibitions might lead to the re-emergence of the problems that provided the impetus for the introduction of the AA, with profit driven, sometimes poorly qualified providers left free to prey on vulnerable women. 134 In the same way that a specific criminal law provision prohibiting amateur dentistry is unnecessary to discourage patients from seeking out unqualified providers, women are highly unlikely to frequent backstreet abortionists in a context current law were to be found incompatible with the European Convention on Human Rights, a UK court is empowered only to issue a declaration of incompatibility, leaving Parliament to address the offending provisions: s 4, Human Rights Act 1998. 131  where free, safe, confidential services are available within the NHS. 135 And were recourse to the backstreets to occur, surgical terminations would fall within common law provisions governing all invasive procedures where consent does not offer a defence to the infliction of actual or grievous bodily harm. 136 Whether this prohibition would capture abortions performed by unqualified providers using less invasive techniques is not clear. However, if the guiding concern is not with the ending of a pregnancy per se but rather, as for other procedures, with ensuring fully informed, voluntary consent and safeguarding women's health, then this question would appropriately turn on the intention of the abortionist, the woman's consent, the seriousness of the invasion and the level of harm caused, with these factors relevant to the determination of the existence and severity of the general criminal offences of common assault, or assault causing actual or grievous bodily harm. 137 Where the safety of patients is negligently or wilfully jeopardised, professionals (like unqualified abortionists) can likewise face potential action in the civil or criminal courts, 138 with deviations from appropriate practice also potentially provoking disciplinary sanction or action by the Care Quality Commission. In practice, abortion doctors who act outside accepted medical practice may already be more likely to find themselves sanctioned by disciplinary bodies rather than by courts, with the General Medical Council sometimes seen as better equipped to provide a thorough exploration of the boundaries of acceptable medical practice. 139 In any fundamental reform of abortion law, it would be necessary for legislators to pay close attention to the existence of specific circumstances that would merit the imposition of a criminal sanction. It was noted above that the majority of prosecutions under both the OAPA and ILPA have been brought against men who assault pregnant women in order to provoke miscarriages. Given the harm to the women involved, such actions should continue to be chargeable and would be so under general offences relating to the causing of actual and grievous bodily harm. It would be necessary, however, to consider whether any amendment to 135 The analogy with dentistry also illustrates, however, that what might well provoke a growth in attempts to secure treatment outside of mainstream health services would be the removal of NHS funding: S Armstrong and M Ruiz del Arbol, 'The Rise of DIY Dentistry' Guardian (3 April 2015) <www.theguardian.com/society/2015/ apr/03/rise-of-diy-dentistry-britons-doing-own-fillings-to-avoid-nhs-bill> (accessed 11 August 2015). 136  the scope of these offences is required to ensure that the extent of harm caused to a woman by the loss of a desired pregnancy is fully recognised in law. 140 Close attention should also be paid to the small group of cases where men have been prosecuted for attempts to procure the miscarriage of pregnancies that were wanted by the women concerned, through administering abortion drugs to them without their knowledge. 141 While such conduct might fall within existing criminal prohibitions on the administration of 'a poison or other destructive or noxious thing', 142 again it would be necessary to clarify that the scope of this offence covers these factual circumstances and that the available sentence captures the full harm caused by the loss of a desired pregnancy.
Finally, within the process of reform, it would also be necessary to consider whether it would be valuable to retain some aspects of the AA. While I have no space to expand here, in my view, it would be appropriate to maintain a right of conscientious objection for healthcare professionals who choose to opt out of participating in abortion procedures. Notification requirements might also continue to perform a useful role, not least in allowing for the rigorous testing of the claim made above: that decriminalisation would be likely to have little impact on the incidence of legal abortion.

B. Two Hard Cases
There are, however, two cases where decriminalisation would make a significant difference to legal access to abortion, potentially impacting on abortion rates. Further, to the extent that my argument is grounded in permissive public opinion regarding abortion, these are also cases that would require particularly close attention in any reform process.
First, it is impossible to know what impact there would be on the incidence of abortion if decriminalisation were also to extend to Northern Ireland, as there is no reliable means of estimating current numbers of terminations each year. Beyond the few dozen women who terminate pregnancies within the jurisdiction and the several hundred who give addresses in Northern Ireland when accessing services in Britain each year, there are undoubtedly many more who access legal services without using their real addresses and others who procure illegal abortions. 143 However, while decriminalisation would have an unknown impact on the absolute incidence of abortion among women in Northern Ireland, it would nonetheless have the very significant effect of opening the door to far greater provision of legal services within the jurisdiction. While assessing the legal and political fallout of such a move is beyond the scope of this paper, it is certain to be significant. While some data exists to suggest that public opinion in Northern Ireland would favour modest moves towards a less restrictive law, 144 and steps are underway to assess the merits of some very limited legal changes, 145 Northern Ireland MPs have consistently raised vocal opposition to liberalising reform. 146 However, if, faced with decriminalisation, the Northern Ireland Assembly chose to make use of its powers to regulate on abortion that would at least mean that women in Northern Ireland would gain a law that is the product of a modern, local, democratic debate. Moves towards decriminalisation might also provoke the kind of public consultation on the reform of the law in Northern Ireland that has been repeatedly demanded by CEDAW. 147 Second, removal of the specific criminal prohibitions regarding unlawful procurement of miscarriage in the OAPA, child destruction in the ILPA and corresponding offences in Scots common law, would have the effect also of decriminalising post-viability abortions. It should be acknowledged that later terminations raise particularly acute moral concerns for many and that the retention of criminal law restrictions would be likely to command more popular support in this context. 148 While this issue would thus require particularly close deliberation, such consideration should take seriously the question of what is to be gained by criminalising women at any stage of gestation and, further, the importance of removing barriers that discourage women from accessing professional advice and support. The difficulty and risks involved in later procedures also offer some reason to suppose that the removal of criminal penalties would not lead to a dramatic escalation in later terminations. In addressing this issue, Victoria chose to take women and healthcare professionals out of the criminal law altogether, while retaining criminal penalties against professionally unqualified abortionists at all gestations, and foreseeing the threat of professional sanctions as an appropriate way of building in safeguards against healthcare professionals who act outside of recognised practice around access to later terminations. 149 While there is no hard data on this point, this appears not to have resulted in increased numbers of later terminations in Victoria, as healthcare professionals have set their own limits on services. 150 If notification requirements were retained in the UK, the better data thus available would, of course, allow for this aspect of reform to be monitored closely.

Conclusion
The fact that a statute is old is not a problem in and of itself. However, any law fossilises the values and assumptions of the era in which it was introduced and the statutory framework regulating abortion is embedded within particularly deep historical strata. The OAPA provides a fascinating snapshot of the anxieties and realities of Victorian Britain, entrenching a motley collection of specific offences, including those of impeding a person endeavouring to save himself from shipwreck, 151 failure to provide 'apprentices or servants with food, &c. whereby life is endangered' 152 and 'assaults with intent to obstruct the sale of grain, or its free passage'. 153 It is no surprise that legislation grounded in those concerns appears anachronistic to modern eyes. 154 While they raise additional 'broad policy considerations', the offences relating to abortion suffer just as seriously from this problem as do those other parts of the Act that the Law Commission has identified as requiring reform.
In general, the danger of such legal 'fossilisation' is guarded against by a range of strategies. First, any statute is subject to interpretation by those who apply it day to day: here, service providers and doctors. This interpretation has evolved over time, in the obiter opinion of one judge, leading to a situation where the law is now 'wrongly, liberally construed in practice so as to make abortion available essentially on demand prior to 24 weeks with the approval of registered medical practitioners'. 155 Yet while this may reflect the judge's own moral or religious view, with respect, there is no basis for describing a liberal construal of the AA as 'wrongful' in legal terms. On the contrary, the Act is 'built on the premise of non-interference with clinical freedom': 156 in 1967 Parliament fully intended that the 'great social responsibility' for regulating access to abortion should be placed on the shoulders of doctors. 157 In exercising this responsibility and using their discretion liberally, doctors have done no more than to develop abortion services in line with evolutions in broader popular morality and best medical practice, interpreting the law in a way that is fully supported by concerns for women's reproductive health and patient autonomy. Further, this interpretation is one that respects the original purpose of the Act: to ensure that 'socially acceptable abortions should be carried out under the safest conditions attainable.' 158 Second, laws are subject to interpretation by the judiciary, who are tasked to read them as 'constantly speaking' 'living statutes', with considerable judicial creativity sometimes deployed to limit the extent to which our aged statutory framework impedes the modern provision of high quality services. 159 There are, however, limits to the elasticity of legal language. The judge's primary duty is to give effect to the ordinary (or, where appropriate, technical) meaning of words, yet this task becomes difficult when legislation needs to be applied in the context of medical realities unimaginable to its architects. 160 This is most graphically illustrated by the challenge of applying laws developed during an era of now rarely used surgical techniques in the context of a widespread reliance on medical abortion. 161 Where the gulf between the plain language of a statute and a sensible interpretation of it becomes too great, a third strategy becomes necessary: for lawmakers to step in to remove or revise offending provisions. Here, the Law Commission plays an important role, aiming 'to ensure that the law is fair, modern, simple [and] effective'. 162 Yet this work is blocked where issues are perceived as raising policy considerations that render them unsuitable to be considered by a law reform body and, a fortiori, if they are perceived as too controversial to be tackled by government. Abortion is, perhaps, the paradigmatic example of this problem, with the Commission's exclusion of offences relating to abortion from the scope of its current consultation providing only the most recent example of a longstanding official reluctance to put abortion law reform before Parliament. Many domestic abortion laws (including the AA) were introduced by way of private members' bills, often denying them the benefit of the skilled drafting that would be provided by parliamentary draftspersons. 163 Many reforms (again, including revisions to the AA) have taken the form of amendments tagged onto the vehicle of other statutes, of necessity thus offering a tinkering at the edges of existing statutory frameworks rather than providing the coherent, root and branch measures that might be envisaged in a specific reform bill. The costs of such political expediency are felt in the form of uncorrected poor drafting, archaic terminology that fits uneasily with modern reproductive healthcare practice, and-most fundamentally-underpinning values and assumptions that remain grounded in the moral mores and medical practices of a long distant era.
Writing some 30 years ago, the veteran pro-choice campaigner, Madeleine Simms, argued that 'the 1967 Abortion Act was a half-way house. It handed the abortion decision to the medical profession. The next stage is to hand this very personal decision to the woman herself.' 164 In practice, this second step has already been taken: doctors have used the broad discretion accorded to them under the AA to respect patient autonomy in this as in other contexts. What remains is to update the law to bring it into line with modern medical practice, leaving abortion services subject to the same complex web of regulation that governs other aspects of healthcare provision. Such a change would not remove social contestation around abortion. It would, however, recognise that a law is overdue reform when there is no appetite for enforcing it in the context for which it was intended, where it has no impact on abortion rates, where it imposes clinically unnecessary impediments that restrict the provision of a high quality, safe and compassionate service, and where it stigmatises one third of British women and the healthcare professionals who care for them. In 2018, we will mark not just the fiftieth anniversary of the coming into effect of the Abortion Act 1967 but also the one hundredth anniversary of the first British women achieving the right to vote. 165 It would be a fitting commemoration of each of these anniversaries were the MPs, who as women we are now formally empowered to share in electing, to recognise our formal legal right to control our own fertility.