Legally, identifying a child’s parents was important. Entitlement to property might depend on a claimant’s relationship; and only a parent was under a legal liability to maintain a child and legally entitled to exercise parental authority over a child. But at the beginning of the twentieth century only the most rudimentary scientific evidence was available to resolve a dispute about parentage; and it was often said (only half-jokingly) that paternity was a matter of opinion rather than fact. In contrast, there was rarely any dispute about maternity: someone would usually have seen a woman giving birth and (provided the child could be identified as the baby she had borne)1 there could be no real dispute. But even so there were still occasionally disputes.2
The Slings by family owned large settled estates in Yorkshire. The family did not approve of the life tenant’s marriage to a widow. Mrs Slingsby suffered a number of miscarriages and the marriage was childless for more than nine years; but in the spring of 1910 (when she was 41) Mrs Slingsby became pregnant. She claimed that a son was subsequently born to her in San Francisco, California (whither she had gone, unaccompanied by her husband). The family uncovered many inconsistencies in Mrs Slingsby’s accounts of what had happened and became convinced that Mrs Slingsby’s pregnancy had ended in a miscarriage, that she had never borne a(p.530) child at all, and that the child she claimed to have borne was in fact the illegitimate child of a chauffeur. The Court of Appeal and the House of Lords3 held that Mrs Slingsby had perjured herself, that the whole case she had put forward was one of fraud, and accepted the evidence of private detectives employed by the family that the child had been handed over to her in response to her newspaper advertisement seeking a child for ‘adoption’. The child was supposititious and not entitled to succeed to the family estates as a Slingsby.
Disputed Paternity: Affiliation Proceedings
In contrast, disputes about paternity were commonplace. Most of the litigation occurred in so-called affiliation proceedings, that is applications for maintenance made by unmarried mothers under the Bastardy Laws Amendment Act 1872.4 The legislation provided5 that a ‘single woman’6 could take out a summons against the man she alleged to be the father of her child, and that the magistrates (having heard the evidence) could adjudge the man to be the ‘putative father’ of the child and make an order7 against him for the child’s maintenance and education.
The mother was required to establish paternity only on the balance of probabilities rather than the stricter ‘beyond reasonable doubt’ basis required in the criminal law, but statute (unusually in civil proceedings)8imposed9 a requirement that the mother’s evidence be ‘corroborated in some material particular by other evidence to the satisfaction of the … justices’. The rule requiring corroboration was intended to ‘prevent men being at the mercy of profligate women’10 or at least to provide some protection against ‘wicked or unfounded charges which might so easily be made if the evidence of the woman without corroborative testimony were sufficient’.11
(p.531) Corroborative evidence is evidence, independent of the applicant’s own statement, which shows or tends to show that the applicant’s story is true:12 classic examples are letters written by the man admitting paternity expressly or impliedly or his having made financial provision for the child or that the defendant had given different accounts of his behaviour on different occasions.13 Even evidence that the parties had been seen to behave with a degree of familiarity consistent with an intimate relationship was capable of constituting corroborative evidence. As Lord Chief Justice Goddard put it in 1947:14
’ For many years … it has been the practice, and very often the only way of giving corroborative evidence in these cases, to prove that the two young people concerned were, perhaps, a courting couple or sweethearts or, at any rate, were associating together on terms … of intimacy … [In the case before the court, the magistrates] were satisfied on the evidence before them that these young people were, and had, for a long period of time, been associating together on the closest terms of intimacy and affection, visiting places of amusement and refreshment, and going to dances together, and being in the company of each other in the evenings and that the [defendant] visited the girl’s home, which means, I suppose, that her parents were allowing [him] to go there because … the parties were obviously on courting terms. It should be added that there was no suggestion that the respondent was having an association with any other man, a very important matter in these cases …’
What was required was evidence that tended to make it probable that the applicant’s story was true. Something more than evidence of opportunity was needed. Thus:
In Burbury v. Jackson15 a 15-year-old girl, employed to do light jobs at Crew Farm Kenilworth, claimed that she had been seduced by the farmer’s son and that he had frequently made love to her in a barn in the farmyard. She also said that she had never slept with anyone else. The justices held evidence from the girl’s mother that she had on several occasions seen the two together in the barn to be sufficient corroboration of her story, and made an order. The Divisional Court held that they had been wrong: all that the mother’s evidence did was to show that it was possible the couple had had intercourse at the relevant time, whereas corroborative evidence had to show that it was probable they had done so. If (said the Lord Chief Justice) the man and woman ‘were seen in the neighbourhood of a wood or other dark place where they had no occasion to be’ that might possibly be corroborative evidence.16 But in the present case the parties’ work required them to be in the barn; and they had a reason (other than (p.532) love-making) to be in the same place. There was thus no corroboration of the applicant’s claim, and the case should have been dismissed.
But the fact that the supposedly corroborative evidence was consistent equally with action dictated by kind and humane considerations did not prevent the court from making a finding of paternity: provided there was evidence which was capable of constituting corroboration the question of the inference which should be drawn from the material before the court was one for the court to decide in the light of its assessment of the plausibility of the witnesses:
In Thomas v. Jones17 Miriam Jones went to work for a 43-year-old bachelor as servant and housekeeper at his farm in Llwynbeadd. On 11 May 1919 (the day before her period of service ended) she was ill and moaning. The farmer made a fire and gave her some brandy and tea. A doctor was called and a baby girl was born. The farmer allowed Miriam and her child to stay at the farm for some five weeks thereafter. On 14 July she wrote to him: ‘Dear sir, I just take the priv-iledge [sic] of writing these few lines to you hoping you are well as it leaves me at present. I should like to know what you intend doing in regard to the child, Do you intend paying or not… you know the child is yours …, Your’s truly, Miriam Jones’. The farmer (who denied ever having had intercourse with Miriam) did not reply. The magistrates held that the farmer’s actions on the morning of the birth, coupled with his actions in allowing her to stay on at the farm for a period (said by the Divisional Court to have far exceeded the ‘usual period of recovery in such cases’) and in not answering the letter were capable of corroborating her evidence. They made an order, and the Divisional Court (by a majority) accepted that the magistrates had been right to accept that the mother’s evidence had been corroborated. That being so, the decision was for the magistrates: did they believe the man or the woman? The magistrates had believed the woman, and that decision had to stand.
The most effective way of destroying the mother’s case was to establish that she had also had intercourse with another man at the relevant time18 but the percentage of applications which were successful was high.19 It is difficult to resist (p.533) the feeling that if the mother’s evidence that she had had intercourse with the man concerned at the relevant time remained unshaken and there was something which could be classified as corroboration an order would be made.20
Courts Reluctant To Bastardise a Married Woman’s Child
In affiliation proceedings the mother was by definition a ‘single woman’21 and her child therefore necessarily illegitimate. These factors were no doubt thought to justify the Justices’ comparatively undemanding approach to making a finding that a man was the child’s putative father.22 But things were very different in the superior courts where the outcome might be to ‘bastardise a child’ (as the expression was) who would otherwise be legitimate. Many23 of the more sensational cases involving parentage were in the Divorce Court: the fact that the husband was not the father of a child born to his wife was the clearest evidence of adultery.
The husband had to overcome two obstacles. First, the common law applied a strong presumption that a married24 woman’s husband was the father of any child born or conceived during the marriage. Thus:
(p.534) In Watson v. Watson25 the wife had been committing adultery with a lodger over a period of years and had herself claimed that the lodger was the father of her child. But she refused to allow scientific paternity tests26 to be carried out. The judge held that if he had been able to decide the case on the usual civil basis of ‘balance of probabilities’ he would have found the lodger to be the father. But the presumption of law meant that the judge had to be satisfied beyond all reasonable doubt; and for that reason the husband must be held to be the child’s father. As another judge had put it27 at the beginning of the century, the fact that a wife ‘had committed adultery with one, two, or twenty other men’ was insufficient to displace the presumption of the husband’s parentage.
Secondly (and specifically) although this presumption that the husband was the father of his wife’s child could be rebutted by evidence28 establishing beyond reasonable doubt29 that the husband could not be so, the law for many years30 refused in the interests of public decency to allow either party to give evidence that they had not had intercourse at the relevant time.31 Thus:
In Russell v. Russell32, the husband convinced a jury that he had never had intercourse with his wife. She had borne a child, and must therefore have committed adultery. The House of Lords held that the husband’s evidence about the (p.535) parties’sexual relationship should not have been admitted. There was no other evidence sufficient to justify a finding of adultery against the wife. The decree nisi obtained by the husband was accordingly rescinded, and in 1926 the High Court made a declaration that the child was the legitimate child of the marriage.33
The law thus prevented the courts from hearing evidence from those best qualified to say what had actually happened; but it apparently allowed judges to decide that the circumstances were such as to make it a moral impossibility that husband and wife should have had intercourse at the relevant time:
In The Aylesford Peerage34 the question was whether a child, born five years after his parents’separation, was entitled to succeed to the husband’s peerage. Lord Bramwell explained the facts and the inference he drew from them as follows:
’ The wife, under the influence of a strong passion … for Lord Blandford, had left her husband’s house to live in a state of adultery with Lord Blandford. The husband, knowing of that, had taken his children from her, had given up all intercourse with her, had entered into a deed of separation by which he and she were to live separate, and had endeavoured to free himself from the marriage that bound them together … [F]or your Lordships to entertain a doubt as to the case would be to suppose it possible that this unhappy lady while living in a state of adultery with Lord Blandford, for whom she had this passion, could permit her husband, what one may call her nominal husband, to have sexual intercourse with her, and that he himself, insulted as he must have felt himself to be … could have intercourse with her which he knew he was sharing with Lord Blandford. People will commit immoral actions, but they would never, I think, do anything so nauseous as what I have suggested as possible; and the very feeling, the very passion that caused the immoral action to be done, almost precludes the possibility of such a course of conduct as I have supposed.’
It is true (as explained elsewhere in this book) that during the two World Wars35 the fact that the wife of a soldier serving overseas had given birth to a child could and often did form the basis of many divorce petitions. But that was because the husband’s service records provided virtually irrefutable evidence that he could not possibly be the father of the wife’s child and it was not necessary to rely on evidence from the parties that they had not had intercourse at the relevant time. But in other cases the courts, lacking any reliable scientific evidence, were faced with an impossibly difficult task36 and were driven to rely (p.536) on such doubtful material as the resemblance37 between the child and the alleged father.38
The Development of Scientific Testing
(i) Blood groups
In 1900 a crucial scientific breakthrough was made: the Austrian scientist Karl Landsteiner39 demonstrated that the red blood cells of some individuals contained different chemical substances from the blood cells of others; and all blood could be classified into a small number of groups. These characteristics are transmitted from one generation to another in accordance with recognised principles of genetics:
‘[A] comparison of the characteristics of a child’s blood with that of his mother and a particular man may show that the man cannot be the father … [and] if it is known that at the material times the mother had had intercourse only with Η (her husband) and X. and the blood test excludes Η but not X, then X must be the father.’40
The courts began to accept such evidence in cases where paternity was relevant. The first reported case appears to be Wilson v. Wilson otherwise Jennings in 1942:41
A husband petitioned to have his marriage annulled on the ground that the wife was, at the time of the marriage, pregnant by another man; and an expert witness gave evidence that the parties’ blood groups established that the husband could not be the father of the child the wife had born. The evidence was not disputed, and the judge granted a decree of nullity.
(p.537) But there were many problems in making wide use of blood tests as a means of determining parentage in legal proceedings. Of these the most obvious lay in the fact that for a satisfactory result it was essential to have samples not only from the child but also from the adults involved. What was to happen if an adult refused to provide a sample, or if a parent refused to allow samples to be taken from the child? These problems began to be widely appreciated in the 1960s;42 and in 1966 the judges of the Probate Divorce and Admiralty Division suggested that the Law Commission undertake a review. The recommendations in the Commission’s Report,Blood Tests and Proof of Paternity in Civil Proceedings43, were given statutory effect by the Family Law Reform Act 1969.
The 1969 Act was skilfully drafted to make blood testing available to establish parentage44 whilst at the same time recognising concerns about the unde-sirability of compelling people to submit to the invasive procedure of taking a blood sample. The Act45 provided that in any civil proceedings46 in which the paternity of any person has to be decided the court could give a direction for the use of tests.47 But failure to comply with such a direction (unlike failure to comply with a court order) does not constitute a contempt of court punishable by fine or imprisonment. Indeed, the Act specifically provides that in general a sample shall not be taken without the consent of the person concerned.48 But the (p.538) court is empowered to draw such inferences from a refusal as it thinks fit;49 and in Re A (A Minor) (Paternity: Refusal of Blood Test)50 the Court of Appeal held that if a claim were made against someone who could possibly be the father, and that person chose to exercise his right not to submit to be tested, the inference that he was the father would (in the absence of very clear and cogent reasons for the refusal) be ‘virtually inescapable’. In that case:
A prosperous businessman refused to undergo a test because the mother had been working as a prostitute and there were known to be two other men who might on the facts equally be the father. He claimed it would be unreasonable and unjust to put him alone at risk of having paternity conclusively established against him.51 But Waite LJ said that any man who is unsure of his own paternity and harboured the least doubt as to whether the child he is alleged to have fathered may be that of another man now has it within his power to set all doubt at rest by submitting to a test. The accuracy of scientific testing made it impossible for any man in such circumstances to be forced against his will to accept paternity of a child whom he does not believe to be his.52
The 1969 Act also adopted the philosophy that it is generally in a child’s interests to have the truth determined:
In Re Η (Paternity: Blood Test)53 the child’s mother was adamantly opposed to samples being taken because she regretted the association she had had with another man after her husband’s vasectomy. It was said that the mother wanted to expunge and treat the relationship almost as if it did not exist. But the Court considered that it would be better for the child to know that he had two ‘fathers’ rather than to ‘leave a time-bomb ticking away’.
(p.539) The 1969 Act provided54 not only that samples could be taken from a consenting person aged 16 or over55 but also from a child under that age provided that ‘the person who has the care and control of him consents’. But this seemed to give a parent with care and control the right to refuse to permit a sample being taken from the child; and some judges were tempted to use procedures of doubtful validity56 in the interests of settling parentage issues. The matter was resolved by legislation:57 a child may be tested wherever the court considered it to be in the child’s interests for this to be done.
(ii) DNA Testing
The 1969 Act was passed at a time when blood group testing worked negatively and could only prove directly that the child could not possibly be the son or daughter of a particular person (although increasingly sophisticated analyses of blood samples would often in fact provide positive evidence that it was possible and sometimes indeed highly probable that the two were parent and child).58 Moreover, the Act only allowed directions to be given in respect of blood samples. At much the same time research59 into deoxyribonucleic acid or DNA (which carries genetic information from generation to generation) enabled tests on any human sample containing DNA (and not only blood) to establish a person’s parentage positively in many cases and, provided that the tests have been carried out under proper conditions, without doubt. In 1987, the legislation was amended60 to empower the courts61 to give appropriate directions for the taking of bodily samples in order to obtain reports of the scientific evidence about an individual’s parentage; and the function of testing was redefined. No longer were the tests to be directed primarily62 to ascertain whether a man was or was (p.540)not excluded from being the child’s father. Instead, the question was to be whether the tests showed that the person concerned was or was not the child’s father or mother. In short, the legislation now recognises that, where the necessary samples are available, genetic testing will usually enable parentage to be established with positive certainty.63 This is a profound revolution in forensic practice.
Artificial Insemination And Other Techniques Of Human Assisted Reproduction
So far, it has been assumed that a child’s legal parents are his or her genetic parents, that is, the man and woman who provided the genetic material (the egg and sperm, or ‘gametes’) which led to the child’s conception and birth. It is true that the procedure of legal adoption has, since 1926, recognised that the legal parents of a child may in certain circumstances not be the genetic parents, but that recognition was both hesitant and exceptional. Lawyers knew perfectly well that men or women might well think of the foster parents who had cared for them from infancy as their ‘real’ parents; but for the common law it remained true that parentage concerned genetics. The law would not recognise social or psychological parentage as sufficient to constitute the legal relationship of parent and child: genetic factors alone determined the identity of a child’s legal parents.64
But what was the position if the child’s conception had occurred, not as a result of sexual intercourse, but by artificial insemination, that is by the manual introduction of sperm into the mother’s cervix?65 Did the common law rule apply (in which case the donor of the sperm—occasionally the husband but more often an anonymous third party donor—would be the child’s legal father)? If so, the child born as the result of donor insemination (AID)66