Complicating the Duality: Reconceptualising the Construction of Children in Victorian Child Protection Law

This article seeks to challenge the prevailing view that child protection legislation introduced in the late Victorian and Edwardian periods was characterised by a victim/threat duality. An examination of the parliamentary debates leading to the passage of major reforming Acts leads to a more complex construction of children. While the classic duality was present in respect of some issues earlier in the period under examination, it gave way to a view which placed greater emphasis on the child as a victim. At the same time, the law began, in an admittedly limited sense, to recognise children as agents with individual stories capable of contributing to the legal process, especially through increased capability to participate at trials. This conclusion can add significantly to our understanding of how the law understood childhood before the development of the modern children's rights movement.

parents to safeguard their welfare, the heightened social regulation of parenting, changes in the sentencing rules applicable to offences against children and the rules for how children were to give evidence in criminal trials. They also allow us to focus on a relatively narrow range of concernsstatutes dealing with a wider range of children's issues such as the Children Act 1908, or statutes dealing with employment or education have not been included in this study.
Legislators were capable to seeing children in multiple, sometimes contradictory ways depending on what a particular section of legislation sought to achieve, and so adopting a narrow focus on the interaction between criminal law and child protection allows us to better understand the construction of children in this particular context, while acknowledging that other approaches may be useful when examining reform in other areas of law affecting children.
Literature in this area has sometimes utilised the idea of a victim/threat duality to understand how legal reformers approached children protection and welfare laws. This article critiques the idea of a duality through an examination of the parliamentary record itself. A close reading of legislation, proposed amendments, and the subsequent parliamentary debates demonstrates that, while some of the claims relating to the duality are partially verified, the use of the this approach can lead to a narrowing of our understanding of the views held at the time. This article, therefore, seeks to add another layer of complexity to the historical literature by arguing that we must add the concept of agency to our understanding of how and why child protection developed the way it did. These claims should not be overstated. I do not make the case that the duality is obsolete, or that legal reform was approached from something approaching a modern children's rights perspective, or that legal institutions treated children as equal citizens; merely that during this period, and in the field under examination, children were beginning to be seen as capable of having a more complex, and indeed active, role than previously thought.
The first part of the article deals with the underpinning concepts of the duality, and also outlines an approach to agency that can help us to understand how children as a group can, and did, influence the direction of the legal process. It will then examine the reform of age of consent rules in 1885 in order to ground the already well-established idea of the duality firmly in the legislative debates. It then examines legislation from 1889 and 1908 which sees a shift in the construction of children, the result of which was that they were now seen predominantely as victims.
The final section will then examine reforms of the law of evidence which demonstrate a nascent understanding of the agency of children. The idea of children as a threat did not disappear completely in these areas, as there is clear evidence of a belief in the supposed threat which stemmed from too ready a belief in the claims of children, but these debates gave far less emphasis to the idea of children as a threat than the debates leading to the 1885 Act.
In exploring these arguments, the article will attempt to shed light on the political discourse of the period by examining parliamentary debates and to the legal reforms to which they led. As such, the article is not about the wider social movements around child protection or moral hygiene, nor does it adopt framing analysis designed to examine public perceptions of the issue. Rather it will focus on the way in which lawmakers presented their ideas about children and the child protection process to each other, in the hope of persuading each other to adopt or vote down a particular measure. As a result, the article will not only examine the reforms which were ultimately adopted but will also highlight some which were proposed but ultimately failed to pass. Such proposals can help to significantly complicate our understanding of attitudes towards children and childhood because they can show ways of thinking about children's place in the legal system which were entirely novel in Victorian times.

| CORE CONCEPTS IN THE CONSTRUCTION OF CHILDHOOD
The Whig tradition hangs over any attempt to assess the historical development of law and policy towards children.
With its emphasis on social progress towards enlightenment and the potential for heroic narratives of the roles of individuals and organisations (Fraser, 2009), it provides an appealing framework for any examination of a history that does show undoubted improvements in the legal position of children. Pinchbeck and Hewitt's account, for example, (Pinchbeck & Hewitt, 1973) provides a useful cautionary tale, given how much emphasis it places on the stirring of national conscience and the growth of charity to the exclusion of more complex factors. Behlmer's work takes a more critical approach, attempting to remedy some of these defects by further engaging with issues such as class and the understanding of parental power but, in providing an account that focuses on the NPSCC, can be accused of drawing too clear causal link between the development of a more elaborate child protection system and the actions of a group of indefatigable reformers (Behlmer, 1982). This is not to deny the role of individual and organised advocates of reform. Rather, it highlights the need for more analytically complex accounts of reform that recognise the multiple ways in which children, and poor children in particular, were thought about by reform advocates, child protection groups, and legislators. One of the most important concepts, therefore, in providing an important corrective to the more optimistic narratives that preceded it, is the idea of the duality which underpins Hendrick's historical accounts of child welfare (Hendrick, 1994(Hendrick, , 1997(Hendrick, , 2003, which has also been applied in more recent scholarship (Jackson, 2000;Platt, 2005;Buckley, 2015). This posits that children have been constructed around seemingly opposite and contradictory images, such as victim and threat. While one aspect of the duality may be dominant at a particular period of time, the idea of dualism is ever-present. Crucially, despite the advantages stemming from protective legislation, Hendrick posits that 'victims were rarely allowed to reap the benefits of sympathy for their condition without the suspicion of what they might become' (Hendrick, 1994: 8). It is therefore important to realise that ostensibly protective legislation was often predicated on the idea that children and young people were victims, while also simultaneously presenting them as a threat to the established social and moral order. While over the course of the twentieth century, the concept of victimhood came to subsume the concept of the child as a threat, it is necessary when discussing social policy reform prior to World War I to understand just how much of the 'so-called protective legislation [was] concerned with their presence as threats rather than their suffering as victims. Indeed, more often than not, the image of young people as threats has undermined their reality as victim' (Hendrick, 1997: 7).
Yet the literature on these Acts has omitted two crucial points. First, the construction of children in the debates forming part of the legislative process has not been considered in a sustained way, thereby preventing an analysis of how parliamentarians spoke about children in a legislative forum from being undertaken. The literature cited above has touched on these exclusively legal sources occasionally, and the lack of attention to this issue is somewhat understandable, given that online versions of parliamentary debates are more readily available than before. While the approach of this article undoubtedly narrows the range of sources under examination, limiting the ability to rely on extraparliamentary speeches, memoirs, or correspondence, it allows a deeper understanding of how the laws affecting children were made by those tasked with a law-making function when actually exercising that duty. Secondly, existing literature has mostly focussed on the content of the final form the Acts took, or on the more colourful comments of parliamentarians, but has neglected the substance of the legislative processthe discussion of what the legal and social consequences of reform would be, the successful and unsuccessful proposed amendments suggested at various stages of the debates, and the margins by which such amendments were accepted or defeated.
As a result of these omissions, the idea that children could have some form of agency has been almost entirely absent from the literature examining the Victorian legislative reforms. Henrick himself has recognised the need to think about children as historical actors (Hendrick, 2008), while Gleason (Gleason, 2016) has critiqued Lancy's argument that a focus on agency creates problems for the study of childhood (Lancy, 2012). Taking one of Lancy's concerns seriously, it is important to clarify exactly what agency means and how it is being used. Agency has been traditionally understood as linked with concepts such as autonomy, rationality, and action in pursuit of a goal, (Emirbayer & Mische, 1998) and in opposition to the dominance of social structures as the determining forces in one's life. Yet children's agency must be understood in a more complex way. Beginning in the 1990s, new social constructionist theories of childhood began to regarded children as "social actors" (James, Jenks & Prout, 1998: 207).
While this may conjure the image of a child or children being heavily engaged participants in policy design, as happens in contemporary youth participation schemes, it is not the only way in which children can be social actors.
Importantly, there is a way to think about agency where children can influence structures without having to actively choose to do so, one which is particularly useful when discussing legal approaches to a problem which are built around making accommodations for children. David Oswell's work emphasises the importance of agency as a collectivity (Oswell, 2013). Children are commonly spoken of as belonging to a single sociological category or class based on their generation (Esser et al., 2016).
It is important to question whether this risks totalising the experiences of the different children within that collectivity -'even within one generation of children, in any one society, each child's experiences of 'childhood' will nonetheless be tempered by the particularities of their social circumstances' (James and James, 2004: 22). With this goes the idea that it may not make sense to speak of "children" as possessing agency, but rather of individual children being capable of doing do. Yet not all children will be in the position to exert this kind of influence on their surrounding social structures; indeed, precious few may be able to do so to even a limited extent.
The collectivity approach to agency does not reduce all children to a single sign or measure but accounts for the fact that children 'as a collectivity compromise a multitude of experiences and positionalities' (Oswell, 2013: 77). In this way of seeing agency as compromised of multiple experiences, children can have different effects on different parts of the infrastructure which surrounds them. He uses the analogy of a children's ophthalmic clinic to demonstrate the point. In such a clinic, children with no knowledge of each other or commonalities except for being children are continually subdivided and aggregated in order to design and administer tests and strategies for the rectification of eye problems. Using Foucault's language of conditions of possibility, Oswell makes the point that once such conditions are created by the clinic, children can then begin to act in and upon the world around them -'capacity to see is enhanced through the spectacles; their capacity to act in the world is enhanced by the facilitatory technology of lens and metal frame' (Oswell, 2013, 78). Agency is thought of relationally, not as a binary property that one has or does not have. By treating children collectively for certain analytical purposes, Oswell's clinic neither denies that individual children are capable of impacting on the world, nor does it treat children as being either an active participant in or passive recipient of social action in all circumstances; it recognises the individual experiences of children and their individual capacities while using an aggregate as shorthand for determining what kind of intervention is necessary to promote and strengthen a given child's capacities within a particular problem space.
As indicated above, the article will deploy this collectivist approach to agency when discussing how children's agency played a role in the child protection reforms of the Victorian era, not necessarily by direct participation, but by children being considered as having interests and capabilities which the legal system needed to recognise, and by creating conditions of possibility in which children could increasingly act through new facilitatory measures. This collectivist approach to agency is adopted within this article because it is the model which allows us to best make sense of the material under examination. While other more individually based approaches would be more suitable for the examination of, for example, the testimony of individual children in court records or child protection files (Jackson, 2000;Stevenson, 2017), when confronting parliamentary records, we need to understand how children were thought about in a forum from which they were absent. As no individual children were present, and the sources being examined were designed by adults, for adults, we need to find a collective approach capable of accommodating a multiplicity of individual experiences and capabilities. In this way, we can make use of the clinic analogy to ask whether the law began to recognise children as capable of impacting on the ways in which the law was designed and implemented.

| AGE OF CONSENT REFORM -THE HIGH POINT OF THE DUALITY
English laws have consistently demonstrated concern with ensuring that it was a criminal act to engage in sexual activity with children, and especially with prepubescent girls. The First Statute of Westminster of 1275 stated that it was an offence to ravish a 'maiden within age' (that is under 12 years of age) with or without her consent (3 Edw 1, stat 1 c 13). Several other legislative interventions clarified that carnal knowledge of girls under 10 was a felony (Benefit of Clergy Act 1575 (18 Eliz 1, c 7) s 4; Offences against the Person Act 1828 (9 Geo IV, c 31), ss 16 and 17).
During the Victorian era, several important reforms were undertaken, beginning with the 1861 Offences Against the Person Act. First, s 51 of the 1861 Act raised the age of consent to 12, while s 50 created a separate offence for the unlawful carnal knowledge of a girl under 10, with the difference between the two provisions being the length of sentence a convicted man was liable to receive. Additionally, s 52 of the Act also created a new offence of indecent assault, thereby extending the ambit of criminal law to cover sexualised conduct towards a child other than penetrative intercourse, and, until a House of Lords decision in 2004 (R v J [2004] UKHL 42), provided a means by which penetrative intercourse could be prosecuted where a complaint was not made soon after the incident. The age at which it became criminal to have sex with a girl was further raised by one year from 12 to 13 in 1875 (Offences against the Person Act 1875 (38 & 39 Vict, c 94) ss 3 and 4).
The Criminal Law Amendment Act 1885 sought, according to its long title, to be an 'Act to make further provision for the Protection of Women and Girls …', and grew out of the campaign against prostitution, especially juvenile prostitution. Among its most important reforms was the raising of the age of consent to 16 (s 5), as well as a variety of measures designed to tackle child prostitution. The Government initially suggested an age of consent of 15 to ensure the smooth passage of the legislation (HL Deb 13 April 1885), while Lord Bramwell thought that 14 may be a suitable age limit (HL Deb 28 April 1885). The debates leading up to the introduction of this legislation, as well as a full examination of the contents of the final legislation provide clear evidence of the presence of Hendrick's victim/ threat duality. Yet when individual contributions are examined in the context of the entirety of the legislative debates, it is arguable that there was a more heterogeneous view of children taken by members of the British parliament. Some Lords and MPs clearly viewed children as innocent objects of concern, some took an approach rooted in the view of children as social threats, while there were also faint glimmers of a more agency-based approach in some of the mooted reforms. Taken in totality, however, the evidence for a duality seems clear.
The campaigns leading up to the introduction to the 1885 Act are well known. W.T. Stead's articles for the Pall Mall Gazette tapped into a mood of public concern for children who were entering what was seen as a white slave trade of juvenile prostitution; moral purity campaigners focused on images of sexual innocence betrayed. The social desire to see the young prostitutes as sexually innocent girls who had individually been brought to sin was therefore central to the Act's final form. Both Jeffrey Weeks (1989) and Deborah Gorham (1978) have argued that this image of girlhood presented a more palatable vision for late Victorian society than that which would have been presented if it had been recognised that prostitution was seen as a way out of poverty by young women with no other real alternatives. It also ignored the reality that children were sexualised from a relatively young age due to factors such as living conditions, whereby children shared bedrooms with their parents, and that teenagers sometimes acted out of sexual desire (Gorham, 1978: 365). Scholarship examining this legislation has therefore tended to highlight the confluence of sexual discourses which came together in the late nineteenth century (Weeks, 1989), and the role that Christian moral economy played among reform advocates outside Parliament (Jackson, 2000). Hendrick (1994) has posited that the victim/threat duality was central to the 1885 Act. Earlier legislation, on which the 1885 Act built, was partly designed to protect a guardian's right to control a girl's sexuality, rather than prevent harm to the girl (Gorham,(363)(364)(365)(366). This was said to be particularly true of the offences relating the procuring of girls under 21 (s 49), and the abduction of girls under 16 (s 55). This mixture of regulatory impulses with some elements of protective intent may have arisen from 'an uncertainty about the nature of childhood and the proper relationship of children to the structure of the family and the wider society' (Gorham, 1978: 355;Keating, 2012), and continued to be present during the drafting of the1885 Act. While social reformers saw children as exploited, and so as victims, those same reformers' concern with a highly gendered understanding of moral purity meant that those same victims could be simultaneously constructed as social threats.
There is clear evidence that children, and those working in their interests, were seen as a threat to the social order. It was with respect to the age of consent that the victim/threat duality manifested itself most clearly. It was argued that the only objection that had been raised to the measure [raising the age of consent] was that men might be induced to yield to the temptations of these poor children, who might use the powers of this Bill for the purpose of obtaining money. Well, let men take that risk (HC Deb 9 July 1885).
This quote does indeed provide clear evidence of Hendrick's duality. The implication that children pose a real risk of temptation for men, and the supposition that this temptation would be exploited for monetary gain, was a common theme among those parliamentarians who saw girls and young women as objects for discipline and control, a theme that was present from the earliest stages of debate. Underpinning all of this was the belief that amid the 'strong language of abuse … a man might be more sinned against than sinning' (HC Deb 31 July 1885). On the Bill's introduction in the House of Lords, one of the first statements made was that '[t]he danger to be guarded against was … that of unfounded charge' (HL Deb 13 April 1885). Similarly, a defeated proposal to introduce a provision that would effectively provide a defence to a charge under the Act if a girl under 16 was a 'common prostitute or of known immoral character' was based on the fear that a man might be 'inveigled, or enticed, or decoyed by a woman who was a woman of the town' who happened to be under 16 (HC Deb 31 July 1885).
Charles Henry Hopwood MP, who would go on to oppose many of the proposed reforms, stated bluntly when responding to the introduction of the Bill in the House of Commons that '[t]he provisions of this Bill … would place young men and boys in the power of designing girls, and would be a most powerful weapon for extortion.' (HC Deb 9 July 1885). Taking the theme further, he later argued that if the age of consent were raised there would be "practically be no protection for youths and the sons of persons of position. There were cases in which girls were steeped in depravity at the ages of 13 to 16 years" and he believed that if the Bill were allowed to pass there would be "great danger of young men and even boys being betrayed by designing creatures, whose object was to levy 'black mail'" (HC Deb 30 July 1885).
Objections to the legislation such as these were almost always expressed in class-based terms, underscoring the potential for young people to threaten the established order of social relations. The trope of working-class girls seducing wealthier men was repeated constantly in the debates. The quote above speaks of the need to protect young men of position from extortion, with no heed taken of the young girl in question. Another opponent of the legislation, while stating that proper protection must be afforded to young women, claimed the real purpose of the 'promoters of the Bill-ill-conditioned Democrats and Salvationist sentimentalists-was to set class against class.' (HC Deb 30 July 1885). This was not an isolated expression of the idea that those promoting the legislation were doing little more than causing trouble for class relations. Another opponent of the legislation directed his ire at the child protection societies, many of whom supported the raising of the age of consent to 16. He claimed that prosecutions undertaken by these groups were 'got up by Societies pretending to be formed for the protection of young girls' and that no man, 'however innocent and respectable, was safe from such organizations' (HC Deb, 9 July 1885). Child abuse was, therefore, not recognised as a real phenomenon but was seen as the product of a conspiracy bent on promoting class war. In these contributions, the construction of children as the source of threats to the social order could even be extended to those who advocated on their behalf, so that their attempts to reform and enforce the law would be resisted.
When the age of consent provisions of the legislation were debated, there was also clear evidence that children, and girls in particular, were sometimes constructed as innocent victims, and not just as threats as the previous paragraphs would indicate. One MP stated that 'it should be clearly understood that this was not a Bill for penal legislation directed against the poor helpless girls in question, but for their protection.' (HC Deb 9 July 1885). Offences against young girls were described as 'horrible' and 'the most brutal offences in the world' (HC Deb 31 July 1885), and as the 'most horrible and revolting barbarities' (HL Deb 28 April 1885). The debates demonstrate early, striking examples of the idea that legislation dealing with child protection matters ought to be cognisant of the risks posed by people who may seek to exploit a young person's vulnerability. This concern with the vulnerability of potential victims was in evidence elsewhere in the legislative debate, extending beyond a concern for childhood to encompass other vulnerable women.
A successful proposal was made to extend the zone of protection afforded to women who lacked the capacity to fully consent to sex. Section 5(2) of the Act states that it shall be an offence to have, or attempt to have, unlawful carnal knowledge of any 'female idiot or imbecile woman or girl, under circumstances which do not amount to rape, but which prove that the offender knew at the time of the commission of the offence that the woman or girl was an idiot or imbecile' . In the original draft of the legislation, the phrase 'idiot' alone was used. At the time, this meant a person who was deemed to lack all capacity, or at the very least those with profound or severe disabilities (Clemente, 2015). In proposing the amendment which extended the application of the provision to include 'imbeciles', Denzil Onslow MP highlighted that the law ought to protect many more women and girls than the original draft allowed for, showing a concern that it was the vulnerable condition of these women that promoted his desire to alter the draft (HC Deb 31 July 1885). Interestingly, there was no opposition to the introduction of this amendment. Additionally, no attempt was made to narrow its application through parsing the meaning of 'imbecile', or through the introduction of an upper age limit. While it cannot be said for certain why there was no opposition, it is possible that there was some cross-party consensus of a recognition to afford greater protection to women who were in a position of particular vulnerability.
One proposal relating to the age of consent demonstrates that some parliamentarians had a clear understanding of what we would now call sexual exploitation. An amendment tabled in the House of Lords proposed to insert a clause which stated that if any master (effectively an employer) or guardian assaulted a girl under the age of 18, they would be guilty of an offence and liable to imprisonment for up to two years with hard labour (HL Deb 28 April 1885). In putting forward this amendment, Lord Mount-Temple recognised that employers or guardians were in a 'position of trust' and that 'abuses of power by masters … were common and by guardians … were not rare' (HL Deb 28 April 1885). The amendment was rejected, but the fact that it was put forward, almost 120 years before modern abuse of trust provisions became commonplace in sexual offences legislation is striking (Sexual Offences (Amendment) Act 2000, s 3). A further noteworthy feature of this is that the age at which a woman would be classified as a victim for the purposes of this provision was substantially higher (18) that the age of consent found in the final version of the legislation. The realisation that sex in these kinds of circumstances was not only common but resulted from an abuse of a position of trust and power demonstrates a clear concern that only novel legal protections could address.
The material relating to the age of consent would appear to indicate that the existence of a victim/threat duality can be very readily substantiated. The mixture of protective impulse and the desire to control the sexuality of young people, and young women especially, provides clear evidence that the reform of consent rules in the 1885 Act was undertaken in the midst of a kind of legislative doublethink, whereby children's bodies could simultaneously be in need of protection and a site of discipline. What makes this Act so important for the construction of childhood however is not just the fact of the existence of such a duality, but the fact that it very quickly begins to slide out of view.
As the following section will demonstrate, the duality came to be largely replaced by a more complicated understanding of childhood, where the protective impulses grew stronger, the conditions of possibility for children's engagement with the legal system were created, and the belief that this was a threat to the social order was muted, but not absent.

| CHILDREN AS VICTIMS IN CRUELTY AND INCEST LEGISLATION
The attitudes to children evident in the parliamentary debates about 1885 Act changed radically within a few years; the passage of the Prevention of Cruelty to Children Act 1889 showed that lawmakers had come to see children almost exclusively as victims of circumstance, and the image of children as threats virtually disappeared from the parliamentary discourse. The 1889 Act introduced a new set of principles to the law and policy surrounding children whose welfare was endangered by parental behaviour, destitution, or other factors. Previously, children who lived in poverty were subjected to the Poor Law system of relief (Charlesworth, 2011). Further piecemeal reform had seen the introduction of legislation which imposed limitations on the employment of children and minimum educational attendance requirements. The 1889 Act, by contrast, saw the adoption of an expanded understanding of welfare and assigned responsibilities to different parts of the state's apparatus designed to safeguard welfare when it was endangered. The Act criminalised abuse and neglect for the first time and permitted children to be taken into care.
What is most remarkable about all of this is that the debates in Parliament were not primarily debates about principle, much as the debates about the 1885 Act were. As a result, we can detect a recognition of the fact that children had entitlements, and that there was a need to balance the interests of parents and children.
Care must be taken not to take the idea of legal entitlements too far; despite being referred to as the 'Children's Charter' (Masson, 2007), the 1889 Act was not the birth of a children's rights consciousness in the modern sense. It was made clear that care must be taken not to 'interfere with the legitimate conduct of parents' (HC Deb 19 June 1889). In one of the early comments on the Bill's passage through Parliament, it was stated that '[c]hildren have very few rights in England, and by this Bill I am really only anxious that we should give them almost the same protection that we give under the Cruelty to Animals Act and the Contagious Diseases Act for domestic animals' (HC Deb 19 June 1889). Nonetheless, it was recognised in the House of Lords that a parent who mistreated a child 'shows himself or herself to be unfit for guardianship, and it is in the public interest that the child should be removed from such a parent' (HL Deb 22 July 1889). The Lords also commented on, and commended, the speed at which the Bill passed (HL Deb 22 July 1889), highlighting that despite the legal novelty of some of its provisions, it commanded crossparty support across both Houses.
Section 1 of the 1889 Act represents a crucial development in child protection law. It provided that any person over the age of sixteen who had the custody, control or charge of a child who wilfully ill-treated, neglected or abandoned the child would be guilty of an offence. It was also an offence to expose a child to something that would cause unnecessary suffering or injury to its health. This for the first time criminalises not only child abuse in the sense of causing deliberate harm to a child, but also neglect, and did so on pain of significant penalties. If a person were convicted on indictment, the maximum penalty was imprisonment with or without hard labour, and/or a fine of £100.
Should the person be convicted summarily, the lesser maxima of three months' imprisonment and/or a fine of £25. S 2 stipulates that where the offender had a known financial interest in the death of a child, the punishment would increase to £200. The child need not have died for this provision to be triggered, merely that the offender was prosecuted under section 1 and had the relevant financial interest.
Despite the near unanimity of the recognition of the necessity of the legislation, the definition of a child to whom s 1 applies is worth noting, as it represents a practical compromise that helped to ensure the relatively smooth passage of the legislation when contrasted with the 1885 Act. The original version of the Bill stated that a child meant anyone under the age of 16. However, this was amended to cover boys under 14 and girls under 16. The rationale for the amendment was to protect parental rights (HC Deb 19 June 1889). It is evident from the speech of the Bill's sponsor, AJ Mundella MP, that significant compromises had to be made in order for the legislation to pass.
He stated bluntly that 'I do not think the House has the smallest conception of the appalling cruelties that are practiced upon children, and the hardships they have to suffer' but that nonetheless '[t]he Bill has been watered down' (HC Deb 19 June 1889). This alerts us to the fact that there may well have been some opposition to the Bill that is not present on the parliamentary record. However, this opposition was clearly not as strong as that found in the debate surrounding the 1885 Act and may have been muted in light of the supposition that the Bill would pass despite opposition. This does not allow us to claim that children were no longer perceived as threats, and were perceived solely as victims by lawmakers, but it does allow us to see that the idea of children as victims had significantly wider support than it did only four years previously.
A further consequence flowing from a conviction under s 1 is found in s 5. This provides that the courts were empowered to remove the child from the custody of that person and place the child with a relative or with another fit person in three situations: where a person had been convicted under s 1, had been sent forward for trial for an offence under that section, or had been bound over to keep the peace towards a child. While this can be seen as a further penalty imposed on a parent, it is also possible to read it as a child protection measure. This was, it seems, the actual intention of Parliament. An earlier draft of the provision was described by one MP as providing a penalty, and as an extreme breach of parental rights (HC Deb 19 June 1889). It was proposed that, given the extraordinary nature of the provisions it would only be possible to exercise the power where it was proven that the child had already suffered harm, as it was regarded as 'too much to say that the child is to be taken away from the control of the parents, and given to some stranger, merely because some other person … treated it in such a way as was likely to cause it unnecessary suffering.' (HC Deb 19 June 1889). Several MPs opposed this amendment on the basis that it would defeat the purpose of the Bill, as '[t]he Bill contemplates prevention of crime, and surely our desire should be to prevent the improper treatment of children without first waiting to see if the child will be injured by that treatment' (HC Deb 19 June 1889). This is a hugely significant shift in attitudes towards children, and indeed towards the idea that the state could intervene when children were at risk. The idea that children's interests in not suffering illtreatment, and that the state had the right to take preemptive action to avoid the child coming to harm, had for the first time been embedded in legislation.
These provisions were built on by the Custody of Children Act 1891 which provided in s 1 that where a child was living with someone other than their parents, the courts could refuse to allow the parents to commence legal proceedings to recover the child. Permission would be refused if the court was of opinion that the parents had abandoned or deserted the child, or that they had conducted themselves in such a way that the court should refuse to enforce the right to custody. Where the court did allow the case to proceed, s 3 prevented the court from making an order for the child's return unless satisfied that the parent was a fit person to exercise custody rights, having regard to the welfare of the child. These provisions, although only applicable in certain situations and with the courts given quite wide discretion, again represent a shift towards greater recognition that children were entitled to have their welfare considered as of right, and that parental conduct causing harm to a child could not be disregarded. When the clause was discussed in the House of Lords, it was recognised that, despite the highly novel nature of the provisions, the evidence presented to the Lords and the 'general consensus' was that this was a necessary measure (HL Deb 6 February 1891). That consensus was that the necessity was borne out of the need to remove control of children's lives from 'drunken fathers and profligate mothers over children who have been saved from misery, degradation, and ruin' (HL Deb 6 February 1891). Opposition to the Bill in the Lords was described as fastidiousness (HL Deb 6 February 1891). Opposition in the Commons was virtually non-existent; its smooth passage summed up by the Attorney General's statement that '[j]udging from the representations received from all parts of the House, there appears to be a unanimous feeling that the Bill should be passed into law as soon as possible' (HC Deb 17 March 1891).
One area in which the victimhood of children was also eventually demonstrated in the wake of the 1885 Act was in the criminal law relating to incest. This had traditionally been dealt with by the ecclesiastical courts in the context of the law of marriage, but efforts had been made since the late nineteenth century to make incest a criminal matter. Incest Bills were therefore introduced in 1899, 1903 and 1907, but failed to pass each time. As a result, incest was not properly regulated by the criminal law until the passage 1908 Act (Wolfram, 1983). Much like the 1885 Act, the 1908 incest legislation owed much to the efforts of social reform organisations such as the National Vigilance Association and the National Society for the Prevention of Cruelty to Children, who engaged in sustained campaigning (Walker, 1964;Bailey & Blackburn, 1979). The Royal Commission on the Housing of the Working Classes (1885) also heard evidence from witnesses who testified that incest was common in one-room tenement accommodation and that action must be taken to remedy this, although the main thrust of their submissions was to reform housing provision.
The Punishment of Incest Act 1908 is a short, concise statute. The main provisions are outlined in s 1 which makes it an offence for any man to carnally know his grand-daughter, daughter, sister, or mother. On conviction for this offence, the man was liable to penal servitude of between 3 and 7 years, or imprisonment for a maximum of two years with or without hard labour; if the female victim was under thirteen years of age, a harsher sentencing regime would apply (s 1(1)). In all cases, the consent of the woman provided no defence (s 1(2)). Consent was only relevant if the woman was over 16, in which case she too would be guilty of an offence (s 2). All proceedings under the Act were to be held in camera (s 5) (Stevenson, 2017).
The discourse in years leading to its introduction show children occupying the sole role of victim, presenting a clear contrast to the debates of 1885. One of the striking features of the 1908 legislation is that children were never really presented as potential direct threats during its passage. It can be claimed that this is not surprising as, given that eugenic arguments played such a significant role in the lead up to the legislation (Wolfram, 1983), incest did not and by definition could not carry the same possibility for inter-class fear of blackmail found in the debates surrounding the 1885 Act. This would lead to the supposition that, because incest had to be an intra-class phenomenon, there was little reason for nineteenth and early twentieth century parliamentarians to construct child victims as threats to the social order. However, this view also supposes that class was the dominant motivating factor behind child protection reform; it would be more accurate to instead consider it as one factor among several influencing legislation which dealt with childrens issues. In the case of the 1908 Act, the parliamentary record indicates that the protection of children was among the preeminent motives ensuring its passage onto the statute book.
Opposition to the legislation instead centred on three grounds, the first two of which were the belief that offences against morals should not be made matters of concern for criminal law, and that there was little need for the legislation. Yet the Home Office was aware, according to Blackburn and Bailey (1979), of problems with the operation of the law as it stood; the existing child protection legislation did not protect victims over the age of 16, and the courts were circumspect in their attitudes to charges brought under the 1885 Act, highlighting a real need for legislative action. The idea that there was already enough criminal law, and that offences could be prosecuted under other legislation such as the 1885 Act has no real bearing on the changing social construction of children themselves, and so will not be assessed further.
The final ground of opposition, that blackmail by neighbours would be a possibility, does have some bearing on the matter. As already discussed, much of the opposition to the 1885 Act centred on the possibility that girls and their families would pose as victims, or entice a man to commit an offence, and use the legislation to extort money from wealthy families. However, blackmail was construed entirely differently in the debates about incest. Blackmail, it was feared, would not emanate from children telling 'false' abuse stories, but from neighbours seeking to cast aspersions on other families, such as where a brother and a half-sister lived together (HC Deb 26 June 1908). At no point in the debates was the spectre of false accusations emanating from children raised, a radical shift from the discourse surrounding earlier legislation, and important in recognising children as innocent victims of sexual offences, rather than as justifiable targets or as a threat in their own right.
The view of children as victims, and especially as victims the protection of whose interests required legislative action, was crystallised in the discussions about the frequency of incest. Some public figures argued that there was no real need for the legislation due to the rarity of the offence; Bailey and Blackburn cite the opposition Lord Chancellor in 1899 when he stated that 'the evil, though shocking, is of rare occurrence' (Bailey & Blackburn, 1979, 714). Even during the later debates, some MPs felt that the incidence of incest was on the decline; one particularly noteworthy contribution came from Charles Hemphill, MP for North Tyrone in 1903, who stated that 'this crime had never been heard of in Ireland … [i]n ancient times there were certain offences so bad that no law was made against them, because it was inconceivable that such offences could be committed' (HC Deb 26 June 1903). However, the arguments of this kind were consistently rebutted by those seeking to criminalise incest, by reference to personal experience derived from practice in the criminal courts (HC Deb 26 June 1908), information provided by the NSPCC, and information held by the Home Office (HL Deb 2 December 1908). Indeed, the Lord Chief Justice, unable to attend the debates in person, sent a telegram which was read out stating that incest had to be criminalised due to the frequency of assaults by fathers on their daughters (HL Deb 2 December 1908).
The Home Office's own data indicated that over one-quarter of the cases of carnal knowledge and rape of girls under 16 that it was aware of was incestuous, that incest was very common and that much of it went unpunished (Bailey and Blackburn, 1979, 713). The majority of parliamentarians accepted that the legislation was necessary, in part because of the significant number of cases encountered by both the courts and the child protection agencies.
This complicates the idea that legislators were reluctant to interfere in the domestic sphere, that class conflict dominated child protection reform, or that there was a consistently seen victim/threat dual construction of children.
Rather, children were beginning to be recognised as primarily, almost exclusively, as victims in this context, and legislation enacted to deal with concerns over their vulnerability.
Scholars of the Victorian and Edwardian periods have mostly examined reforms of the substantive criminal law, and examinations of the law of evidence and criminal trial procedure have not been undertaken to anywhere near the same extent. Some attempts have been made ot remedy this; Stevenson has undertaken valuable work on how courts approached the evidence of children prior to the 1860s (Stevenson, 2017), while Jackson's work focuses on the period up to 1910 (Jackson, 2000). Yet an examination of the parliamentary journey of the legal reforms allows us to see how the series of changes represent a different way of thinking about children, one which enables us to think about children as possessing some form of agency. Again, it is important to caveat this comment; as Jackson rightly highlights, child witnesses were not always viewed sympathetically and faced significant hurdles in giving their testimony (Jackson, 2000: 106). Nonetheless, the fact that statutory change was undertaken to address the problems highlighted in Stevenson's work is significant because it shows the system responding to children's needs and being, to some extent, shaped around them.
In the debates leading to the 1885 Act, significant parliamentary time was devoted to reform of the laws of evidence in cases involving young victims. Much of this revolved around proposals for young victims to give unsworn evidence. At common law, only evidence given under oath could be received in a criminal trial. In the case of children, this presented particular difficulties where the child was deemed incapable of understanding the nature of the oath, in terms of what it would mean in a religious sense to provide a false statement, and if they lacked an understanding of the importance of telling the truth more generally (Murphy, 2009). This meant that the ability of the child to give evidence depended on the 'sense and reason they entertain of the danger and impiety of falsehood' which was to be determined by a judge asking the child questions in the presence of the jury to determine their competence to give sworn evidence (R v Brasier (1779) 1 Leach 199, 1 East PC 443, 168 ER 202). The net effect of this would have been to deny many young children, especially those from poorer backgrounds who had not received significant formal religious instruction, the ability to participate in proceedings by giving evidence of the offences committed upon them (Stevenson, 2017).
The 1885 Act did contain some significant reforms which increased the ability of children to participate as witnesses. Where an offence was charged under s 4, a child who did not understand the nature of the oath was permitted to give unsworn evidence where they were deemed to be of 'sufficient intelligence to justify the reception of the evidence, and understands the duty of speaking the truth'. This was subject to the caveat that such evidence had to be corroborated, and that the child could be prosecuted for perjury in the same way as an adult giving false testimony on oath. This provision significantly altered the common law, allowing for the first time a child victim to give unsworn evidence and so to speak in court as a witness with limited restrictions.
More radical proposals were put forward but rejected. Samuel Smith MP tabled an amendment to the Bill proposing that a clause be inserted providing that where a girl was too young to understand the oath, she would be permitted to give unsworn evidence, provided that no one would be convicted in the absence of corroborating evidence (HC Deb 6 August 1885). Additionally, he proposed that an unsworn 'statement made by her before the committing justice or magistrate, and taken down in writing at the time' could be admitted in evidence at the trial, as if she were giving sworn, viva voce evidence (HC Deb 6 August 1885). This radical change in the laws of evidence was withdrawn without attracting support. If introduced, it would have removed the need for a child to meet any capacity based test of possessing "sufficient intelligence", and the need to even be present to give evidence, once a statement had been made to a court at an earlier stage in the criminal process. It was only with the passage of various measures in the 1990s that the law changed in a manner similar to the proposals made by Smith (Youth Justice and Criminal Evidence Act 1999).
Although this radical measure was withdrawn and not voted on, the debate on a narrowly defeated proposal gives greater insight into the extent to which children could be seen as possessing agency. This proposal sought to permit a girl who was too young to understand the nature of the oath to provide an unsworn statement 'in corroboration or explanation of any other evidence which may be given in support of' a charge being brought (HC Deb 31 July 1885). While this was criticised as a being too loose and informal a measure which ran the risk of putting words into a child's mouth (HC Deb 31 July 1885), the proposal attracted significant support given the nature of the crimes in question. One MP stated bluntly that he 'could not understand this squeamishness about children giving evidence in cases where that evidence was absolutely necessary' (HC Deb 31 July 1885). Others pointed out the necessity of including a measure of this sort as there would otherwise be no point in passing the legislation as far as young children were concerned; unless the law was reformed to allow young children top give evidence of in some manner, these crimes would continue to go unpunished (HC Deb 31 July 1885).
Crucially, young children were spoken of as 'the most truthful, oftentimes, of all witnesses, if only they were examined in a kindly manner' (HC Deb 31 July 1885). This last point, about the manner in which children were examined by barristers, had been prompted by a discussion of how children were to be cross-examined, which highlighted that the ages of children needed to be taken into account so that their stories could be told (HC Deb 13 April 1885).
The amendment was not, however, carried, being defeated by a margin of only three votes. That so significant a change was defeated by so slim a majority is highly significant, as the inclusion of such a measure would have forced us to rethink the idea that the law did not accord any meaningful recognition to the idea that children's status as children ought to be accommodated within the legal system, rather than demanding that they adjust to the demands of that system. Some of the defeated proposals were eventually to find their way onto the statute book. In 1894, the Prevention of Cruelty to Children (Amendment) Act and the Prevention of Cruelty to Children Act were passed. i By the end of that year, judges were empowered to take a written deposition of a child's evidence on oath, if satisfied by the evidence of a registered medical practitioner that the attendance of the child in court would involve serious danger to its life or health (s 14 Amendment Act; s 13 Cruelty Act). This would apply in cases where the accused was charged with child cruelty, or any offence involving injury to a child. ii Additionally, this deposition could be admitted in evidence in court against an accused person, provided that the defendant was given the opportunity to crossexamine the child at the time the deposition was taken (s 15 Amendment Act; s 14 Cruelty Act). While it appears that this power was rarely, if ever, used (Henderson, 2012), it marks a significant advance in the realisation of the ability of child victims to participate as witnesses in ways designed to accommodate their vulnerabilities, but also their agency.
These measures provoked some limited debate when introduced in Parliament. It seems that the main aim of the power to take and admit a written deposition was to ensure that the evidence of a dying child should form part of the prosecution case (HC Deb 23 May 1894). When clarification was sought on whether the clause would apply only to dying children, none was provided, save to say that dying depositions could not currently be taken (HC Deb 23 May 1894). This was introduced as part of the earlier Amendment Act when the power only extended to prosecutions for cruelty. The subsequent extension of the provisions to cover any offence involving bodily injury so shortly after their initial introduction would indicate that the principle of accommodating children as witnesses was widely accepted.
Further important reform is found in s 15 of the Prevention of Cruelty Act. This provides that where a child is deemed not to understand the nature of an oath, the evidence of such child may nevertheless be introduced if the child is regarded as 'possessed of sufficient intelligence to justify the reception of the evidence, and understands the duty of speaking the truth'. Further, the evidence of child 'though not given on oath but otherwise taken and reduced into writing' was to be deemed to be a deposition and admissible in a trial on that basis. Where evidence was admitted under this section, the accused was provided with certain safeguards. First, a person could not be convicted of the offence unless the child's unsworn testimony was 'corroborated by some other material evidence in support thereof implicating the accused'. Additionally, any child who wilfully gave false evidence was themselves liable to prosecution. During the limited debate on these matters, it was simply stated that '[i]t had been the experience of the Judges that evidence given by children who were not sworn was perfectly trustworthy' (HC Deb 23 May 1894).
There appears to be no detailed consideration of these provisions, thereby indicating a fairly general acceptance of them as necessary measures to ensure the participation of child victims.
What is most striking about these changes is the extent to which they reflect the ophthalmic clinic analogy found in Oswell's work on agency. It could be argued that as children may well be the only witnesses to these crimes, some reform was needed. Yet if a strict victim/threat duality was indeed the operational construction underpinning the thought of these legislators, it would not necessarily follow that evidential reform would be the solution they reached for. Instead, attention could have been focused on provisions increasing police powers (which were indeed found in these statutes), with no facilitatory measures relating to children's testimony, or on offences relating to false reporting. Instead, these statues created the conditions of possibility whereby children who had been victimised could be facilitated in giving evidence against their abusers, thereby impacting on the legal system and, more widely, the world around them.

| CONCLUSION
Can we legitimately say that we must rethink our attitude towards Victorian legislators dealing with child protection concerns? Must we move beyond the well-established concept of a victim/threat duality to a more nuanced concept, whereby threat and victimhood sit side by side with the idea that children were thought of as being possessed of some form of agency? If a collectivist, relational approach to agency modelled on Oswell's work is adopted, then the reforms outlined in the article indicate that a positive answer must be given. While the debates of this period still indicate a strong desire to regard children as not the equal of adults, and indeed as sometimes presenting special threats, these concerns should not obscure the importance of the reforms undertaken in two specific respects. First, the changes ushered in during this period radically changed the legal landscape for children suffering abuse and neglect, as there was now a statutory code providing them with greater protection than ever before. Secondly, it is important to recognise what this meantit was recognised that children had particular interests which needed to be catered for, leading to the recognition that children were a class apart from adults requiring specialised legislative intervention. When it came to the implementation of those new legal provisions, children were no longer to be excluded merely because they were a class apart; rather their differences could lead to greater inclusion in the legal system rather than justify their exclusion from it. In this way, the conditions of possibility for children to help bring about change in how the law operated were created, if not fully realised. In doing so, the legal system recognised that children had a story to tell, that the story was worth hearing, and that it should shape itself to their abilities to tell that story.

ENDNOTES
i Rather confusingly, the Amendment Act (57 & 58 Vict, c 27) was passed in July, while the Prevention of Cruelty to Children Act (57 & 58 Vict, c 41), which consolidated child protection law, was passed in August and repealed the Amendment Act. The provisions of the Amendment Act with respect to the laws of evidence were effectively re-enacted, but in a slightly expanded form.
ii The earlier Amendment Act only applied in cases where the accused with charged with child cruelty, or any offence involving injury to a child.