Error Correction of Restrictive Measures: Appeals Made by Young People in Care

The article examines appeals against restrictive measures in child protection submitted to and decided upon by administrative courts in Finland. Restrictive measures used in care restrict children’s and young people’s fundamental and human rights considerably, meaning ‘confinement in fractions’. Therefore, young people’s access to justice is an important issue in situations in which they consider an error to have been made. We study appeals as a form of legal safety and as a means of error correction. Although young people, 12 years of age and older, have the right to appeal, the appeal system is mainly used by parents to appeal against restriction of contact. As such, the appeal system poorly protects young people’s individual access to justice, if at all. As only administrative courts can overturn the decision to apply restrictive measures, the adult-centeredness of the system and young people’s access to justice should be critically assessed and rethought.


Introduction
Unlike in many other countries, in Finland, the child protection system does not include any secure institutions for children, of whatever age. Presently, all child protection residential institutions should be open in the sense that young people should be able to unlock the doors from the inside and leave the institution. Prisons, Article 1 Tampere University, Finland, 2 University of Helsinki, Finland YOUNG 29 (1) on the other hand, accommodate annually only a handful of young people under the age of 18 (Huhtanen & Pösö, 2018;Pitts & Kuula, 2005). The Finnish ambition to avoid locked residential institutions in child protection is fundamentally different from 'secure care' practised in several countries in Europe and North America. Sometimes referred to as youth treatment centres or secure children's home, being in secure care means that young people are not free to come and go as they wish. As part of child protection or juvenile justice, young people are placed in secure care for months or years because they are deemed to be 'out of control' (Enell et al., 2018;Goldson, 2007;Harder et al., 2011;Roesch-Marsh, 2014). The literature regarding secure care includes concerns about children's rights to liberty and the lack of due process in decision-making systems (Brummelaar, 2016;Roesch-Marsh, 2014).
Although secure care institutions do not exist in Finnish child protection, legislation allows practitioners to introduce restrictive measures, which similarly restrict young people's right to liberty. Restrictive measures may target only one element in young people's life in care, for example, being in contact with a certain person as will be described in more detail later. Confinement through restrictive measures thus takes place in fractions of space, time and relations, unlike secure care, which primarily functions by locking up people in a closed space. Restriction measures are based on the assessment of the behaviour of a young person, or, in contact restrictions, on the impact of contact on the young person, and they are decided by individual practitioners or multiprofessional teamsnot by the courts. As the restrictive measures include the use of public power and the decisions are made in a case-based manner as part of the sometimes conflictladen practice of out-of-home care, the legal safeguards of young people become an important issue.
Very little is known about these error-sensitive restrictive measures; in particular, how young people act when they feel that their rights might be violated. Although the more or less informal practices that young people undertake to resist unfair decisions (e.g. running away from care) have previously been highlighted (Hoikkala & Kemppainen, 2015;Isoniemi, 2019;Taylor et al., 2013), there are also formal mechanisms made available to express disagreement. Making appeals to the court is one form of such a formal mechanism, which connects the lived experiences of restrictions of freedom, unfairness and disagreement, the child protection system and the legal court. Young people who are 12 years or older have the independent right to appeal in Finland. This article, written by one social work and two legal scholars, focuses on appeals against restrictive measures submitted to and decided upon by administrative courts: how and to what extent young people express their disagreements regarding restrictive measures in child protection by using the formal appeal system and how the appeal system responds to their appeals. Our analysis is guided by the view that making appeals is a way to express disagreement about perceived errors of restrictive measures, and that the appeal system is an essential part of access to justice. The forms of legal remedies concerning restrictive measures in child protection are not fundamentally different from those regarding young people sentenced to prison or treated in mental health care or in residential disability services (e.g. Pollari & Murto, 2016): appeals provide a route for overturning decisions and young people themselves have the right to appeal. However, no solid body of research exists to inform how young people use the opportunity to appeal in any of the abovementioned fields.
The article is structured as follows: we first discuss the appeal system and present our theoretical standpoint. Then we move to presenting the context of Finnish child protection and the restrictive measures therein. The next section presents the empirical findings of the analysis of appeals, followed by conclusions.

Legal Protection and Appeals as a Means of Error Correction
The concept of legal safety (or legal protection) has no precise meaning (de Godzinsky, 2013). A narrow definition of legal safety often refers to the right to a fair trial (due process). In a broader sense, legal safety is related to how our social system is structured (the rule of law). At the heart of the rule of law is the recognition of individual liberty and limitations on the use of public power. In this study, legal protection is approached in its broader as well as narrow sense, as both are important in restrictive measures.
Regarding young people in public care, legal safeguards are important as the complexities and controversies of measures restricting young people make it inevitable that errors occur in decision-making. For example, the process of defining and identifying 'out of control' behaviour of young people is not an objective or value-free process (Roesch-Marsh, 2014). When young people are restricted, it is likely that interests clash, and the decision-makers' view is disputed. Decisions of restriction are made in tense situations and sometimes in a hurry due to the urgency of the situation. Consequently, the decisions are error sensitive. Surprisingly enough, very little is known about the use of restrictive measures and the errors they may contain in Finland. There is no information available about the number of restrictions on the country level, and only a handful of research studies exist on this topic (the review of research: Huhtanen & Pösö, 2018).
Appeals and appeal processes are known in virtually all legal systems (Shavell, 1995). Courts, administrative agencies and other organizations have formalized appeal processes in which a person who is disappointed with the first decision of the first-hand decision maker can seek reconsideration by a second-higherdecision-making body (ibid.). The right to appeal is an essential part of a fair trial that is secured as a human right, for example, by the European Convention on Human Rights (EHCR art 6.1., see also article 13), and also by the Convention on the Rights of the Child (CRC/C/GC/12, s. 46-47, CRC/C/GC/14, s. 98). In recent years, especially children's and young people's right to a fair trial has been highlighted and specified in international instruments (see for example European Convention on the Exercise of Children's Rights and the Guidelines on Child-friendly Justice, 2012), as well as in the research literature (Berrick et al., 2018;Goodgame, 2016;Leviner, 2015).
The appeal processes serve several functions. One function is in fact to serve as a means of error correction, which, consequently, could lead to the making of better decisions as suggested by Shavell (2006). In Finland, error correction is seen as one of the main functions of the appeal procedure: the administrative courts' duty is to ensure that the authorities do not make illegitimate decisions or act illegally (see Mäenpää, 2007, pp. 25-43). Appeals as means of error correction are also the very core of access to justice (A/HRC/25/35, s. 4). In addition, appeals provide a forum for accountability for decisions legitimated by legislation, which, according to Lipsky (1980, pp. 160-162), is a link between bureaucracy and democracy. The link is, however, complex as public services such as child protection are expected to be accountable to legislation, agency policies and the related bureaucracy as well as to service users' claims. How the different functions of appeals are accomplished in child protection is apparently an under-researched topic. The existing research focuses on the procedural rights of children and young people, the right to appeal, the right to have a representative and, most of all, the right to participate (Goodgame, 2016;Leviner, 2015;Masson, 2006).
Appeals as a form of legal remedy also include enhancing service users' access to justice. The right to appeal usually belongs to the custodians of the children and young people. In Finland, the young people, aged 12 or older, also have an independent right to appeal.

The Types of Restrictive Measures
Although Finnish child protection focuses on providing supportive in-home services to children and families, an approach known in international comparisons as the family service approach (Gilbert et al., 2011), children and young people are also taken into public care and placed into foster homes and residential institutions when their health and development is seriously endangered. In addition, care orders may become topical if children or young people 'seriously endanger their health or development by abuse of intoxicants, by committing an illegal act other than a minor offence or by any other comparable behaviour' (Child Welfare Act 417/2007, CWA Section 40). Restrictive measures can be used if the child is placed in substitute care as a result of the decision of a care order or an emergency placement, both decisions concerning children and young people under the age of 18.
The objectives of restrictive measures are to ensure the aims of the placement or to secure the children's or another person's health, safety or some other interest laid down in the Child Welfare Act. Unlike rules and limits in upbringing provided in substitute care, restrictive measures interfere with fundamental rights that are guaranteed to children and young people, including civil and human rights such as self-determination, liberty, property and privacy (Constitution of Finland ss. 7, 10 and 15; see also CRC art. 16 and art. 9, para. 3; ECHR art. 8 and art. 5). In order to interfere with fundamental human rights such as these, the law must stipulate such interference clearly (Saastamoinen, 2018, pp. 9-11). However, the interference is justified because it is considered to protect the children's and young persons' more fundamental rights-the right to have essential care (Constitution of Finland s.19) and the right to special protection (CRC 3 art. 2; CWA s. 1).
The extent to which it is acceptable to interfere with a child's or young person's fundamental rights by the use of restrictions must be assessed on a case-by-case basis. Age is a consideration only for decisions about 'special care', defined below, as it can be introduced only for young people 12 years of age or older. Since restrictive measures can be applied against the will of the child or that of the child's custodians, the use of restrictive measures is also about using forced administration. The Child Welfare Act defines how restrictive measures should be carried out: for example, restrictive measures must always be implemented in a manner that is as safe as possible and respects the person's human dignity. The use of restriction must be discontinued after it is no longer necessary. Some restrictions are time-limited (e.g. a maximum of 48 h in isolation and a maximum of 30 days-in some situations 90 days-in special care). All forms of restrictive measures may be used in residential care, whereas only the restriction of contact is allowed in foster homes.
The restrictive measures laid down in Chapter 11 of the Child Welfare Act are the following: 1. The restriction of the contact between the child and a person close to him or her, for example, a parent, grandparent or a friend (the restriction of contact). 2. The confiscation of intoxicants and equipment suitable for using such substances, substances and objects that are meant to harm the child or another person or substances and objects that are likely to cause serious harm either to the child's substitute care or other children's substitute care or to public order (the confiscation of substances and objects). 3. A search for intoxicants or other harmful substances or objects in the child's clothing or otherwise on them, or to take a breath test or blood, urine or saliva sample (bodily search and physical examination). 4. Inspection of the child's room or the child's possessions in search of intoxicants or other harmful substances or objects (the inspection of possessions and deliveries); under certain conditions, it is possible to withhold a letter or other similar confidential message totally or partly from the child (withholding deliveries). 5. Restraining the child physically in order to calm him or her down (the physical restraint of a child). 6. The prohibition of the child leaving an institution's grounds, the institution itself or the premises of a certain residential unit for a fixed period (restrictions on the freedom of movement). 7. Isolation of the child from other children in the institution for a fixed period (isolation). 8. Arranging for so-called special care for the child that is multiprofessional care (with expertise in upbringing, social work, psychology and medicine) and intensive care, during which the child's freedom of movement may also be restricted.
The regulation of restrictive measures has varied considerably during the years (Hoikkala, 2011) and poor practices have been reported to exist, especially in the studies of historical abuse (Hytönen et al., 2016;Laitala & Puuronen, 2016) as well as in the inspection reports by the Parliamentary Ombudsman (Eduskunnan oikeusasiamies -tarkastukset, 2018). The recent regulation originates from 2006 with some very recent changes in 2019 that emphasize the importance of informing children and young people about their rights.

Legal Safeguards of Restrictive Measures
Since using restrictive measures is about exercising public power in an intense way, several efforts have been made at the legislative level to safeguard the legal protection of the child, young person and other people involved in the use of restrictive measures. First, the use of restrictions, the decision-making authority and protocol are closely regulated in the Child Welfare Act. The requirements include the transparency of the use of restrictive measures as each decision should be documented and the documents should be shared between the decision-maker, the child and his or her custodians and substitute carers as well as the social worker of the child or young person. Part of the legal protection is also who makes the decision about a restrictive measure. The decision is made either in an institution or in a child protection agency. The more the restriction interferes with the child's or young person's rights, the higher the level at which the decision is made. For example, the director of child protection makes the decisions on long-term restrictions on contact (over 30 days) and on starting special care, whereas the child's and young person's social worker makes decision on short-term restrictions on contact and on the termination of special care. Nevertheless, the court does not ever become involved in the first-hand restriction decision. Secondly, the residential institutions have a duty to keep a record of the use of restrictive measures (CWA, s. 74). The records are important in monitoring and supervising the use of the restrictive measures. The records must include a description of the restrictive measure concerned, the justification for the measure, the duration of the measure, the names of the persons who made the decision about the measure, who put the measure into practice and the persons present during the measure. The records must also specify how the child or young person was heard and his/her views on the measure. The content of the records must be sent to the child's social worker, who has a duty to supervise the use of restrictive measures.
Thirdly, for most restrictive measures, there is an option to appeal against the decision in the administrative court. Judicial disputes are heard in general and administrative courts in Finland. Administrative courts deal with matters concerning the exercise of official authority and they are mandated to make all child protection decisions according to the Child Welfare Act. Only bodily search and physical examination, the physical restraint of a child and the inspection of possessions and deliveries are not appealable. The courts, in a panel of three (two legal judges and one expert member), examine the appeals. The administrative courts mainly use a written procedure (Nylund, 2017). However, in child protection cases, oral hearings are quite common and also young people 12 years of age or older have an opportunity to participate (de Godzinsky, 2012).
All restrictive measures are also open to an administrative complaint to local, regional and national supervising authorities. However, only the administrative court can annul the decision about a restrictive measure based an appeal, and that is why the appeals are especially important for the legal safety of young people. The complainant may choose the authority to which he or she makes the complaint. Among the supervising authorities, the Parliamentary Ombudsman has a special duty to monitor children's rights, and in this task, the Ombudsman pays special attention to the fundamental and human rights of children. Children and young people may complain to the Ombudsman, but that is quite rare. Yet, the restrictive measures have been the most common subject of complaint for years (Parliamentary Ombudsman, 2017, pp. 113-114). Even if the Ombudsman cannot annul the restrictive decisions, the Ombudsman may oblige the authorities to change unlawful practices and give instructions on how to take better account of fundamental and human rights and to act in a more child-friendly manner. The Ombudsman can also make initiatives to amend the law; legislation on restrictive measures, for example, has been revised several times on the initiative of the Ombudsman (Nieminen, 2018, pp. 164-166).

Research Design
The appeals against restrictive measures in child protection are examined as expressions of errors as seen by young people and parents, that is to say those who have the right to appeal, and the administrative courts' decisions to reject or to accept the appeal are seen as their views on the perceived errors. In this study, the interest is in particular in the appeals submitted by young people. As all appeals and related judgements take a textual form, we examine the errors and judgements in writing. The data consist of the appeal judgements in three administrative courts (out of six in Finland) during the period of 1 January to 30 June 2016. There were 85 appeals. The data consist of the court judgements, which are written by the court referendaries. They include the summary of the appeal as expressed by appellants, the examination of the case and the reasoning for the court's decision. In addition, in 25 appeal cases, we also have the written appeal submitted by the appellants. This data come from one court.
The analysis is by its nature descriptive and maps out the profiles of perceived errors and the court responses to them. This kind of analysis is relevant as the existing research on appeals in child protection cases is very limited both in Finland (e.g. de Godzinsky, 2012) and, to our knowledge, elsewhere.
The analysis is based on inductive thematic coding of the perceived error descriptions and the reasoning of the judgements (Atkinson & Coffey, 1996). We also count certain topics such as the number of accepted appeals. The analysis shifts from reading the full data (85 court judgements of the appeals) to a closer analysis of the appeals written by young people and/or parents (25 appeals). The findings are presented under two headings: we first look at the errors as viewed by parents and young people, and then at the responses by the courts.

Limitations
The sample of the judgements covers a period of 6 months and represents the country well geographically although three courts are not included in the data. As we lack information about the overall number of restrictive measures, it is not possible to estimate the appeals' proportion of all restrictive decisions. However, the sample itself represents the appeals during that particular period well, and there are no legislative, policy or cultural factors, which would make that period different from other periods in the 2010s. Nevertheless, it is important to acknowledge that appeals do not exhaust all disagreement regarding restrictive measures, as not every young person or parent is willing to or capable of expressing their claims regarding errors in a written and formal form.
The analysis of the textual documents is bound to language. In this case, the analysis is based on the use of Finnish language. The findings are translated into English. The legal and social work terminology may not travel well into English as the legal and child protection systems differ between Finland and Anglophone countries (Gilbert et al., 2011;Koch et al., 2017). In a similar manner, the findings might not travel as such from Finland to other countries with different systems for the confinement of young people and related legal remedies of perceived errors. As we could not find literature on legal safeguards for young people in confinement, we hope that despite the fragmentary generalizability of these findings the study would inspire more interest in this topic also elsewhere.

Overview
The majority of the appeals concern the restriction of contact (81 out of 85), four are about restrictions of the freedom of movement and one appeal included several restrictive measures. In the data set, there was no appeal against special care or isolation, both of which strongly physically restrain young people in space and resemble, to some extent, the forms of confinement in secure care. The contact was typically restricted between the parents, the mother in particular, and the (young) child during the placement. In these cases, the child was not allowed to visit the parents during the weekends or holidays, the number of visits was restricted to be of a frequency which the parents disagreed with or the parent (the mother) was not allowed to call or visit the child in her substitute home. The arguments for the appeals as presented in the court judgements (85) are grouped into two main categories: first, false or impartial evidence (the major category) had been used to make the restrictions and, second, the failures of the process of making restrictions or related decisions. We will look at these arguments in more detail later.
Only three appeals had been submitted by young people on their own and one young person appealed with the mother. The number of persons appealing is even less as one young person had submitted two appeals. Two appeals of young people were against the restriction of movement and one against contact, and they argued that their behaviour had been misunderstood. This is to say that the appeals were especially submitted by parents: 89 per cent (n = 76) of the appeals were submitted by parents, of which 45 were submitted by the mother only, 20 by the father only and 11 of them by the parents jointly. The average age of the children was 9.3 years old in the appeal cases, the age of children varying from a couple of months to 17.5 years. The children appealing on their own belonged to the oldest (17-17.5 years). There were 28 young persons in these data, who were 12 years old or older and who could have submitted or co-submitted an appeal, but only four young persons had done it. A handful of appeals were made by siblings and other relatives, and by a combination of them. Half of the children (43) were placed in foster homes and the other half in residential institutions.
Legal aid was included in 56 appeals, whereas 29 were made without legal aid.

Appeals as Worded by Young Persons and Parents
A more detailed examination of the errors as viewed by the parents and/or young persons is done based on the full case files from one court, including the appeals as written by young persons, parents and their legal representatives (N = 25). In this data set, no appeal was submitted or co-submitted by a young person. All appeals were against the decision of restriction of contact. Six parents had made 16 appeals as access for contact between the parent(s) and two to four siblings had been restricted. Most of them also appealed against other child protection decisions. This is to say that disagreement against child protection services accumulated in some families. The nature of the restriction of contact, which was disputed, ranged from a total cut of contact to the timing of the contact. The total cutting of contact included one case in which the father had not been given any access to his young child or any information about the child's whereabouts due to the history of the father's violence towards the child and the family. On the other end of the spectrum, there was a dispute about the time of the day when the parents were allowed to meet the child: the parents wished to meet the child in the morning, whereas the meeting was scheduled to be in the afternoon. Between these extremes on the spectrum, the majority of the appeals focused on appealing against the frequency and nature of contact. The restriction decisions ordered a certain frequency for the visits (e.g. for the child or young person to spend every second or third weekend at his or her parents' home), which in these cases was appealed against (a parent or parents wished to have the child or young person more frequently at home or to start the weekend visit on Friday instead of on Saturday, for example). The disagreements were also about the nature of the contact. The contact was, for example, ordered to take place in a certain location, under supervision, whereas the parents wished to have the child or young person at their home. Or else the child or young person was only allowed to visit the grandmother's home (with her mother) and not her mother's home. Some parents disagreed with some of the conditions required, for example, the tests for drug use before the meetings with the child. The parents argued that such a test went against their personal rights.
The arguments for the appeals are thematically grouped in three themes. They reflect parents' views on errors and represent the parents' interest, excluding very much the child's or young person's views or interests in the matter. The first type of argument was about the type of evidence used in the decisions (theme 1). The parents claimed that the decision was based on old, false and biased evidence. The parents did not, for example, use drugs anymore or the violent partner had moved away from home and therefore the home was not dangerous for the child to visit. They also claimed that the contacts should not be restricted because of the behaviour of the child or young person seen in the substitute homes after the contact; in their view, it was natural that the children were upset and restless, as they had had to leave their home and parents to return to the substitute home. Secondly, the arguments against the restrictions claimed that the restrictions went against the purpose of the placement (theme 2). If the child and parent could not meet, family reunification would not be possible. That claim included references to legislation, as did the third type of argument as well: the restrictions went against human rights as they restricted family life (theme 3). The arguments of the third theme were common in the appeals written by legal representatives.

The Courts' Responses
The court judgements included the examination of the arguments of the appeal and of the evidence given by social workers and the experts involved in the case, and they weighed evidence against the appeal's arguments. The majority of the decisions (N = 71) were made based on written material only and, in 14 cases, the courts had organized oral hearings.
In 57 cases, the appeal had been rejected. The appeals submitted by young people on their own (3) were among those rejected: the restriction decisions had fulfilled the legal criteria. In addition, in three cases, the appeal itself had been rejected but a detail (such as a request for a greater number of hours of contact) had been accepted. At the end of its judgement, the court typically formulated its view of the rejection in a contact issue in the following way: When considering all the relevant information about the child's situation, the administrative court sees that the contact between the child and his/her parent would be harmful to the child's health and development if materialised according to the parent's wishes. The contact would not be in the best interest of the child. The administrative court sees that the contact has not been restricted more than it should be.
This extract demonstrates well how most court judgements approached the overarching principle of the child's best interest: they referred to the existing principle by mentioning it. Only in 16 judgements out of 85 did the courts present in detail how they viewed the principle in that particular situation.
Out of 85 appeals, 12 appeals (14%) were accepted. One accepted appeal was about the restriction of the freedom of movement, whereas the others were about restrictions of contact. The reasons for the acceptance of the appeal were related to the procedural elements of the decision-making process (e.g. the wrong authority had made the decisions or the timing of the decision) and in six cases, the justifications for restrictions had insufficient evidence (e.g. there was no evidence of harm to the child's well-being being caused by the contact). The accepted cases accumulated in the same courts: all procedural issues were pointed out by one court and five evidence issues (out of six) were pointed out by another court. The courts differed considerably in their acceptance rates: the first court accepted five cases out of 38 (13%), the second court accepted one out of 25 (4%) and the third one accepted six out of 22 (27 %).
Twelve cases had lapsed or had been removed from the docket. The panels of the administrative courts had made 84 decisions in agreement. In one judgement, one legal judge largely agreed with the rejection of an appeal but felt that the requirement for a drug test before contact with the child was unjustified.
The decision by the court was made on average 6.8 months after the original decision about the restriction in substitute care.

Discussion
As a means to correct errors, appeals are not productive from the point of view of applicants, as they rarely amend the errors that are appealed against (see also Simmons, 2016). The courts do not, in general, accept young people's and their parents' (and their legal representatives') views of errors and the related justifications. In the few cases in which an appeal was won, the decision was made so late that the actual restriction measure had already expired.
Restrictions on contact, that is the relational element of confinement, are the most error-sensitive restrictive measure in this data set. This is an important finding for two reasons: firstly, because there are no appeals against isolation or special care, which are the strongest restrictions on personal liberty and which could be especially error-sensitive from the point of view of young people (Pösö et al., 2010), and secondly, because Finnish child protection in general encourages constant contact between children and their families. The right to keep in contact with one's family is included in the child's rights in the CWA (CWA s. 62, CRC art. 9). Both the child's social worker and substitute carers have a duty to support and promote contact between the child and people close to him or her (HE 252/2006, CWA s. 54). The intensity of the contact is defined in joint client plans, often resulting in weekly visits at the birth parents' homes and regular phone contact (Pösö, 2016). However, despite the importance attached to maintaining contact, most appeals were lodged against restrictions on contact, especially by parents. The disputes were often about the alleged harm to the child or young person caused by maintaining contact and the evidence used to support this. Similar contact disputes have been observed in previous literature. According to Ainsworth and Hansen (2017), a child's behaviour before and after parental contact is often scrutinized by substitute carers in child protection. The substitute carers' feedback is frequently negative, suggesting that contact should be reduced or completely stopped to avoid the child becoming upset. A restriction on contact is not, however, in the authors' view, necessarily the best solution; instead, better management of the child's upset and anxiety is required as it is 'natural' to be upset in such artificial situations (Ainsworth & Hansen, 2017). The data used for this study do not allow us to speculate how children and young people viewed the 'harm' as evidence to support restricting contact. Yet the parent-centeredness of the arguments in the appeals suggests that it would be important to learn more about children's and young people's views on this matter. It would also be important to learn why the appeals against contact restriction were only about parental contact, excluding young people's appeals against restrictions on contact with their peers.
Regarding the legal protection of young people, the analysis of the appeals gives rise to the following three key critical remarks. As a first remark, the absence of young people as agents submitting appeals threatens the protection that is provided to them by law. This suggests that young people are highly dependent on adult help in order to gain access to justice: young people might not even know that they have rights; if they do, they might not know what kinds of rights they have; and above all, if they know they have rights, they might not know how to use those rights to seek corrections of errors (CRC/C/GC/2003/5, s. 24, CRC/C/GC/2005/2, s. 5). This is highly important in the context of substitute care where children and young people are strongly dependent on the support of their substitute carers and social workers and the information these prove them.
The second remark, the use of legal aid in more than half of the cases, suggests that the appeal system is not easily manageable without legal representation although the Finnish administrative court system is based on the assumption that legal aid should not be needed (Paso et al., 2015). It has been demonstrated elsewhere that the proportion of legal aid in child protection cases tends to be growing in administrative courts (de Godzinsky, 2012), which may support individuals' legal rights. Nevertheless, it is usually the parents who use legal aid. A lawyer has the duty to promote his or her client's-in practice, the parents'-rights, which are not always identical to the young person's rights. Children rarely have a lawyer or a guardian to assist them (de Godzinsky, 2012;Enroos et al., 2017). In addition to the better availability of legal aid, young people might need other forms of support in order to have their views about errors heard. Practitioners employing human rights-based approaches in child protection and peer group support could equally be important.
The third remark, the slowness of the decision-making process, highlights that a successful appeal is not efficient as a remedy. The slowness may also threaten an individual's sense of procedural justice (Thibaut & Walker, 1975;Tyler, 1990), that is to say the way one feels one has been treated in the procedure: fairly or unfairly. In a young person's timeframe, 6.8 months (the average the decisions take) is a long time. When the court decision is eventually available, the restriction that was appealed against is often already over, which makes it questionable whether an appeal is an efficient means to achieve the correction of an error. Moreover, the slow process violates a young person's right to decision-making without delay (CRC/C/ CG/2014/14, s. 93, CWA, s. 88).
Since both the restrictive and legal protection measures in child protection are very much the same as in mental health care, disability care and even in prisons, there is reason to assume that young people's access to error correction via appeals is similar, if not worse, in those other fields. If the appeal system is to guarantee legal safety, major changes are required to make it accessible to young people in vulnerable situations: information about their rights as well as legal assistance and other forms of support are needed. Furthermore, it should be critically examined to what extent appeal outcomes have an impact on young people's experiences of error correction and whether young people need other forms of recognition of their experiences of errors. The adult-centeredness of the appeal system is indeed a challenge when it comes to young people having their views heard and having access to justice.

Conclusions
The analysis shows that the appeal system is not very sensitive or effective: young people do not use this form of legal safety although it is precisely they who are in the first place affected by the restrictive measures and the full spectrum of the Finnish practice of confinement, whether regarding space, relations or time. The appeal system is mainly used by parents to challenge restrictions on contact between themselves and their young children, that is that element of confinement, which restricts relations as defined by parents. Young people are very much left to express their disagreement with restrictions in other ways and they may consequently experience a loss of trust in the child protection system and its formal legal mechanisms. This is an essential critical finding regarding the access to justice and children's rights. It is especially important because investigations into the historical abuse of children and young people in child protection institutions as well as educational and religious facilities have, in many countries, highlighted the lack of mechanisms for children and young people to express their experiences of abusive practices (Hytönen et al., 2016;Sköld, 2016;Wright, 2017).
The important issue for further studies is to include all forms of young persons' expressions of disagreement, both formal and informal, in a study and to explore how well they meet the principles of legal safety, access to justice and ultimately good care. It is imperative to include young people and their views about restrictive practices and error correction. In an explorative study about young people missing from care, for example, the inclusion of young people as research partners challenged the administrative view on the topic; at the same time, it highlighted the importance of such principles as autonomy, being heard and treated with respect and feeling that someone cares when tackling the problem of running away from care (Taylor et al., 2013). In a similar way, a fundamentally different system to correct errors in restrictive measures might be proposed by young people if they and their views were included. From a comparative point of view, it would be important to learn more about the nature of the legal safety of young people in child protection systems which involve secure care institutions, and how it differs from the Finnish approach of just using fractions of confinement instead of 'locked institutions'.