Who is Responsible for Remedying the Harm Caused to Children of Prisoners?

ABSTRACT It has been argued that the social circumstances of many children of prisoners goes against established principles of social justice. In this paper the proper allocation of responsibility for remedying this social injustice is discussed. Through a discussion of four principles for allocating remedial responsibility, it is argued that the responsibility for children of incarcerated parents is shared among several actors, including the incarcerated parent, remaining caregivers, prison officials, social work professionals, and, to some extent, members of the wider community. While incarcerated parents are partially responsible for remedying the harm caused to their children, prison officials have the responsibility to uphold the types of prison conditions under which incarcerated parents are able to fulfill their responsibilities to their children and socially contribute to their well-being. Similarly, whereas the main responsibility to care for the children of incarcerated parents falls on the caregivers (such as the remaining parent or other family relatives), states are at the same time responsible for implementing social welfare policies of the sort that can help caregivers fulfill their responsibilities for the well-being of these children. As for individual members of the wider community, they have an obligation not to contribute to the stigmatization and marginalization of the families of incarcerated individuals.


Introduction
Children of prisoners are often described as the hidden or forgotten victims of the criminal justice system (Minson 2020;Murray and Farrington 2008;Scharff Smith 2014). Both political discourses concerning crime and punishment and academic discussions on the ethics of punishment tend to focus primarily on criminal wrongdoers and their direct victims. Although a few scholars have examined the implications of philosophical theories of punishment for correctional practice (see, e.g. Deering and Smith 2016;Duff 2001;Lippke 2007;Ward and Salmon 2009;Wolff 2020), they unfortunately often remain silent on the impact of prison on the rights and well-being of children of prisoners.
The pan-European network Children of Prisoners Europe (COPE) estimates that 2.1 million children have a parent in a European prison on any given day (COPE 2021). Many of these children face a number of difficult challenges because of their parent's incarceration. Among other things, they might experience stigma, social exclusion, separation anxiety, disrupted attachments, depression, and strong feelings of loss, all which in turn might hinder their development and educational achievements (Arditti 2012;Condry, Kotova, and Minson 2016;Kotova 2020;Minson 2020;Murray 2007;Nesmith and Ruhland 2008;Scharff Smith 2014). The incarceration of a parent might also lead to the loss of a household income, which could either lead to or increase the severity of child poverty. This is a serious ethical concern in its own right, not least because child poverty can have a damaging impact on the well-being and well-becoming of the children affected by it (Schweiger 2019;Schweiger and Graf 2015). On the face of it, all of this is problematic from the point of view of social justice. After all, children of incarcerated parents have done nothing to deserve the harm they endure because of their parents' incarceration. Yet many of these children are at risk of suffering a range of different harms and are likely to face more adverse circumstances relative to those of other children (Bülow and Lindblom 2020;Condry 2018;Murray 2007).
To be sure, the harm suffered by the families of prisoners might sometimes be morally justified as a regrettable side effect of an otherwise morally justified criminal justice system. Hence, even if the harm suffered by children of incarcerated parents is bad, some might argue that it is the incarcerated parent, rather than the state, that is to be blamed in this case. The state's responsibility in this context, or so the argument goes, is first and foremost to mete out appropriate punishment. However, as a few philosophers and legal theorists have argued already, the question of whether a certain criminal justice system is morally justified or not does to at least some extent depend on its impact on the rights and interests of third parties. In particular, it has been argued that efforts should be made to either prevent or mitigate the harms caused to family members of incarcerated individuals, given the undeserved nature of these harms (Bülow and Lindblom 2020;Condry 2018;Lippke 2017;Manning 2011;Ristrouph 2016;Tadros 2011). One important issue in this context, one that has been somewhat overlooked in the rather limited philosophical literature on this topic, concerns the proper allocation of responsibility for remedying the harm caused to the children of prisoners. As is evident from the literature on other areas of applied philosophy, any philosophical theory of justice needs to specify not only the rights or goods that people are entitled to, but also what Onora O'Neill calls the 'agents of justice ' (O'Neill 2001; see also Hickey et al. 2021). Among other things, it needs to be settled which agents are responsible for realizing justice, how the responsibility for remedying a certain injustice ought to be distributed, and the different responsibilities of the various agents involved in a specific context of injustice.
My aim in this paper is to shed some light on these issues in relation to the social injustice suffered by the children of incarcerated parents. To this end, I examine four prominent principles for allocating remedial responsibility and their respective implications for who should bear the responsibility to remedy the injustice suffered by these children, as well as some of their more specific responsibilities in this context. To be clear, my aim is not to provide a comprehensive discussion of these admittedly very difficult questions.
Instead, my aim in this paper is merely to provide some tentative thoughts about this topic, hoping that this may stimulate further philosophical and ethical discussions on child welfare and criminal justice policies, as well as the intersection between these two areas. These are important discussions to be engaged in, not least since public policies especially designed to support children of incarcerated parents are often lacking (COPE 2021).
Before I proceed, I should make a few preliminary remarks. First, since my focus here is on responsibilities for remedying the social injustice that befall many of the children of prisoners, my focus is largely on what is commonly referred to as forward-looking rather than backward-looking responsibility. That is, my focus is on the extent to which a certain agent can be expected to do something in respect to either remedying or preventing a particular injustice rather than on allocating blame to whoever caused the injustice in the first place (although the latter issue does have important implications for the former (Hickey et al. 2021;Miller 2001Miller , 2007). Second, since my interest here is in the proper allocation of responsibility (i.e. how responsibility ought to be distributed from a moral perspective), I am first and foremost concerned with moral responsibility rather than any existing legal responsibilities for either remedying or preventing the harm caused to children of incarcerated parents. As is illustrated by the subsequent discussion of principles for allocating remedial responsibility that I provide in this paper, this responsibility is shared among an important number of agents, including the incarcerated parent, prison officials, social work professionals, policy-makers, and members of the wider community. Before turning to this issue, however, I will first say something more about the social circumstances of children of incarcerated parents, and the associated social injustice that should be remedied and prevented.

The impact of prison on children of incarcerated parents
It is widely acknowledged in the scholarly literature that children of incarcerated parents suffer in multiple ways because of their parents' incarceration. As previously mentioned, parental incarceration is associated with a range of problems, including separation anxiety, trauma and loss, social exclusion, stigma, financial stress, disrupted attachments, change of caregivers, anger and depression, sleeping problems and poor school performance (see e.g. Arditti 2012;Condry, Kotova, and Minson 2016;Kotova 2020;Minson 2020;Murray 2007;Nesmith and Ruhland 2008;Scharff Smith 2014, 2015. Parental incarceration is also associated with an increased risk of criminal behavior and subsequent criminal offenses by children of incarcerated parents (Dobbie et al. 2019;Lippke 2017;Murray 2007;Murray and Farrington 2008), although important differences among different countries have been observed (Dobbie et al. 2019;Scharff Smith 2015).
To what extent the problems listed above are due to the incarceration of the parent or some circumstantial background variables, such as lack of family stability, pre-existing poverty or lack of social support, is somewhat uncertain (Arditti 2012;Minson 2020). Even so, it is frequently recognized that children of incarcerated parents are vulnerable individuals and that at least part of the reason for this is their parent's incarceration. 1 According to Arditti (2012), parental incarceration can be understood as an ambiguous loss and enduring trauma, associated with feelings of uncertainty about the parent's well-being, whether the parent will come back, and whether things will return to the way they used to be. She further notes that the uncertainty and pain associated with parental incarceration 'are compounded by hostile, disapproving, or indifferent social attitudes pertaining to the loss of a family member through imprisonment ' (2012, 103). Consequently, while parental incarceration might cause strong feelings of grief, it is not always a loss that can be openly acknowledged or mourned. Nor is it a loss for which these children or their caregivers always get appropriate social and emotional support (see also Minson 2020). Here we should add that the financial stress that a parent's incarceration may cause for the remaining household members can create additional burdens for the caregivers who are expected to care for the children during their parent's incarceration (Arditti 2012;Minson 2020;Murray 2005;Nesmith and Ruhland 2008). In addition to the loss of a household income, having an incarcerated family member is sometimes associated with additional expenses such as travel costs for prison visits, loss of job opportunities, and the cost of telephone calls (Arditti et al. 2003;Arditti 2012;Comfort 2007;Condry, Kotova, and Minson 2016;Minson 2020;Murray 2005;Scharff Smith 2014). As Minson (2020) points out, these burdens are also experienced by close family relatives who, as a result of maternal incarceration, often take on the burden of caring for children whose mothers were the primary caregiver.
In sum, the incarceration of a parent is associated with a range of potential harms for the children and families of prisoners. It is also frequently observed that prison disproportionately affect already socially and economically disadvantaged families and, as a result, magnifies and compounds patterns of pre-existing disadvantages (see e.g. Condry 2018; Condry, Kotova, and Minson 2016;Murray 2007). That said, it is important to remember that the children of incarcerated parents represent a very heterogeneous group and that the effects of parental incarceration vary significantly from child to child depending on a range of factors such as age, gender, and past relationship with the incarcerated parent (Minson 2020). Moreover, the incarceration of a parent might sometimes have beneficial consequences for the child, such as in situations where the incarceration of a parent amounts to the removal of an aggressive and/or violent person from the child's immediate environment. Families of incarcerated individuals may also perceive prison as a 'safe haven' that can help the parent stay out of trouble. And for some families, the prison term might lead to the parents' rehabilitation and help improve family functioning and communication (Arditti et al. 2003;Tasca et al. 2016). However, even if a parent's prison sentence is not always entirely detrimental to the child, it is important to remember that it might still be very burdensome and challenging for the remaining family members. This is perhaps especially so for the large portion of cases where the child and their incarcerated parent had close contact before the parent's incarceration and when the remaining parent or foster parents have to struggle to fulfill their already demanding responsibilities in a situation of social and economic hardship 2 (Arditti 2012;Condry 2018;Minson 2020;Nesmith and Ruhland 2008;Scharff Smith 2014).
There is a burgeoning literature that analyzes the situations of families of prisoners from a social justice perspective (see, e.g. Bülow and Lindblom 2020;Condry 2018;Condry, Kotova, and Minson 2016;Kotova 2020;Murray 2007;Wallis and Dennison 2015). Of course, this type of analysis will vary to some extent depending on which normative framework one endorses (whether one endorses a sufficientarian or an egalitarian account of distributive justice, for example). However, as Condry (2018) convincingly argues, the harm endured by many children of incarcerated parents and their families appears to be inconsistent with the principles of social justice, regardless of whether we understand them in terms of equality and desert, the elimination of institutional oppression and domination, or the capabilities approach of Martha Nussbaum. Families of prisoners, Condry argues, have done little to deserve the harm they suffer because of a parent's incarceration or the social and material inequalities that this might give rise to. She further notes that families of prisoners often experience marginalization and discrimination because of their relationship to the prison system. As a result, several of their substantive freedoms and core capabilities, as identified by Nussbaum (2000, 78-80), are often unduly compromised. Among other core capabilities from Nussbaum's well-known and influential list, these include the capabilities of bodily health (which includes mental health), emotions (which includes being able to grieve the loss of loved ones), affiliation (which includes having the social bases of self-respect and non-humiliation) and control over one's environment. 3 Insofar as the incarceration of a parent is associated with long-term negative effects to the child's well-being and opportunities later in life, this also runs afoul to the principle of fair equality of opportunity (Bülow and Lindblom 2020).
Condry does not provide that many details for how to either prevent or remedy the social injustice that she identifies. Instead, she merely notes that imprisonment should in general only be used as a punishment of last resort and that, where it is deemed necessary, the state might be seen to have an obligation to take certain measures in order to compensate for and ameliorate the harm caused to families and children of prisoners (Condry 2018, 37; see also Bülow 2014;Lippke 2017;Manning 2011;Minson 2020). However, Condry's account remains silent on what exactly these compensatory obligations should entail or how the responsibility for mitigating this harm should be allocated. Like many others she likewise never addresses to what extent, if any, the responsibility for the well-being of children of incarcerated parents might be shared among different actors, including the incarcerated parent and the wider society. In what follows, I hope that I will be able to add some further details to this picture through a discussion of various principles for the allocation of remedial responsibility.
Four principles for allocating remedial responsibility Following Miller (2001), I understand remedial responsibility as the responsibility to put a bad situation right. This, I take it, includes situations of social injustice. For the sake of clarity, being remedially responsible, or so I understand it, may not only entail having a responsibility to provide a remedy to an injustice, but may also include a responsibility to prevent this type of injustice in the future.
There are several proposals for the principles of how to allocate remedial responsibility in relation to a particular injustice. Here I will focus on four influential principles for allocating remedial responsibility commonly discussed in the philosophical literature: (1) The causation principle: The responsibility (as a whole or in part) for remedying a certain harm or injustice falls on those agents who have caused it.
(2) The capacity principle: The responsibility (as a whole or in part) for remedying a certain harm or injustice falls on those agents who, in virtue of their power or capacity, can do so.
(3) The contractual agreement principle: The responsibility (as a whole or in part) for remedying a certain harm or injustice falls on those agents who have previously agreed to prevent it. (4) The beneficiary pays principle: The responsibility (as a whole or in part) for remedying a certain harm or injustice falls on those agents who innocently benefit from it. 4 Each of these four principles is prima facie plausible and, when considered in combination, they may help identify both the agents of justice as well as their respective responsibilities in a specific context of injustice (Miller 2001). In what follows, I discuss each of them in relation to the injustices suffered by children of incarcerated parents. When doing so, I also draw from the scholarly literature on the impact of prison on children of incarcerated parents.

The causation principle
The causation principle, as it is here understood, is based on the largely intuitive idea that those causing a particular problem or injustice are the ones who should bear the primary responsibility of fixing the problem or compensating for the injustice in question. This is especially so if the problem or injustice could have reasonably been foreseen and avoided. Here it is worth noting that there are two versions of the causality principle that we should keep apart: a principle of causal responsibility and a principle of moral responsibility (Miller 2001). While the principle of moral responsibility involves an appraisal of the agent causing a certain state of affairs, causal responsibility only entails that the agent has had a causal role in the genesis of that state of affairs. However, as Miller (2001) points out in his discussion of allocation of remedial responsibility, the principle of causal responsibility usually only plays a limited role in the ascription of remedial responsibility. Thus, following Miller, I assume that the important aspect here is whether the agent causing a certain state of affairs is morally responsible for doing so.
At the face of it, the causation principle seems to imply that it is first and foremost the incarcerated parent who is responsible for remedying the harm caused to their children. After all, it is fair to assume that parents who commit imprisonable offenses could have reasonably foreseen the harm caused to their children (should they get convicted) and that they could have acted differently. In contrast, the state's duty in this case is first and foremost to punish those who commit crimes and mete out appropriate punishment in response to criminal wrongdoing (Metz 2000). 5 Consequently, it is the incarcerated parent that should bear the responsibility for remedying the harm caused to their children. However, while it is probably true that many incarcerated parents to an important extent are morally responsible for the harm caused to their own children, I am not certain that things are quite that simple.
First, we should bear in mind that not all incarcerated parents are equally responsible. For example, many incarcerated individuals have histories of mental health problems and substance abuse issues (Fazel et al. 2016), in which case it might not be appropriate to hold them fully responsible for their actions. 6 In cases where parents suffering from this type of issues have reasonable opportunities for rehabilitation while incarcerated, their incarceration may perhaps be beneficial both for them and for their children. However, it would not be fair to assume that the children in this case might not still suffer because of their parent's incarceration. To the contrary, the parent's involvement in the criminal justice system, although potentially beneficial for them, can still be very stigmatizing for their children and cause strong feelings of distress. Yet given that it is not appropriate to hold the incarcerated parent fully responsible for their actions in such a case, we should not hold them fully responsible for the harm caused to their children either.
Second, even if we assume that most incarcerated parents to an important extent are morally responsible for the harm caused to their children, it is debatable whether they alone are responsible for the entirety of the harm caused. To the contrary, we should note that the extent to which children of incarcerated parents suffer depends on a number of factors outside of the incarcerated parent's control. To take but one example, it is frequently acknowledged in the scholarly literature on parental incarceration that being able to visit, and thereby maintain regular and meaningful contact with one's parent, is beneficial for most children of incarcerated parent and important for helping these children cope with the trauma that this parent-child separation sometimes involves (Arditti 2012;Minson and Flynn 2021;Nesmith and Ruhland 2008;Scharff Smith 2014). However, as Minson (2020) points out in her study of maternal sentencing in the UK, opportunities for family visits are often at the discretion of the prison officials, and the frequency of visits can vary significantly due to such factors as staffing levels. Minson further notes that the quality of visits largely depends on the prison environment, which can sometimes be intimidating and hostile toward children (see also Aiello and McCorkel 2018;Nesmith and Ruhland 2008;Scharff Smith 2014), and that the frequency of visits also depends on available visiting hours and the traveling distance between the child's home and the prison where the parent is placed. While some might argue that these too are consequences that the incarcerated parent could have reasonably foreseen, I believe that a more reasonable conclusion to draw here is that prison officials and relevant policy-makers are at least partly responsible for the harm caused to the children of prisoners. After all, much of the harm caused by the state in this case is indeed foreseeable and could possibly be mitigated by increasing staff levels, making visiting rooms and procedures more child-friendly and, when possible, placing prisoners in facilities closer to their families (see also Condry 2018). I will return to a few of these points in the subsequent discussion of the capacity principle.
To be sure, the discussion so far does not suggest that incarcerated parents lack any responsibility for remedying the harm caused to their children. To the contrary, I believe that the causation principle rightly suggests that incarcerated parents should bear an important part of the responsibility in this case. At the very least, they owe it to their children to become better parents and not cause further distress or harm to them. To this end, the incarcerated parent might have an obligation to participate in rehabilitation programs, parenting programs, and vocational training during their prison sentence. Consequently, prison officials and relevant decision-makers are morally responsible for ensuring that incarcerated parents have reasonable opportunities to fulfill this part of their remedial obligations. However, it is worth noting that incarcerated individuals are often denied even minimally decent opportunities to be what Kimberley Brownlee calls social contributors, as they are frequently being denied opportunities to care for or provide social support to their families. By Brownlee's account, incarcerated parents and their children are both wronged, qua social beings, in those prison contexts where parents are denied decent opportunities to have regular and direct contact with their families and contribute socially to their well-being (Brownlee 2016(Brownlee , 2020. Similarly, Helen Brown Coverdale argues for the moral importance of not only recognizing criminal wrongdoers as subjects of state care but also enabling offenders to care for others, most notably their families. This, she argues, is crucial in order to support offenders as intrinsically valuable persons and might support desistance (Brown Coverdale 2018). Therefore, besides facilitating participation in rehabilitation programs and parenting programs, prison officials and policy-makers also have a responsibility to facilitate regular and meaningful family visits since this too might help mitigate the harm caused to children of incarcerated parents, maintain valuable social bonds between the child and the parent, and aid rehabilitation and desistance (Cochran and Mears 2013; see also Brownlee 2020, 187-191).
The discussion so far suggests that even if the incarcerated parent is to an important extent morally responsible for the harm caused to their child, this does not mean that prison officials or policy-makers can simply ignore the harm caused to children of prisoners, nor blame it completely on the incarcerated parent. To the contrary, these state actors too bear a remedial responsibility for the injustice suffered by these children. Even when parents are indeed morally responsible for committing a crime, and consequently also morally responsible for an important part of the harm caused to their children, we should note that the extent to which children of incarcerated parents are harmed and possibly wronged largely depends on the actions and omissions of prison officials and relevant policy-makers. Thus, in line with the causation principle, it is reasonable to hold them morally responsible for mitigating the harm caused by, for example, making it possible for the parent to participate in rehabilitation programs and to contribute socially to their child's well-being. However, we should also recognize that policy-makers, as well as other relevant state authorities, might have further responsibilities in this context, especially since the incarcerated parent might not always be able to compensate for the harm caused. In what follows, I will take a closer look at two additional principles of responsibility allocation that explain why state actors hold a particular responsibility in this context: the capacity principle and the contractual agreement principle.

The capacity principle
One way of determining who is morally responsible for remedying a particular injustice is by considering those agents who have the relevant power and capacity to remedy it (Hickey et al. 2021;Miller 2001). As noted above, incarcerated parents might not always be able to remedy the harm caused to their children, even if they are sometimes blameworthy for the harm caused. In such cases, it is reasonable to suggest that the responsibility falls on some other agent who has the relevant capacity to either rectify or mitigate this harm, insofar as this mitigation is not too demanding for the agent in question.
The idea that one's capacity to mitigate harm can give rise to special moral obligations is an influential idea in contemporary moral and political philosophy. In rescue cases, for example, it is largely intuitive that a person, A, might gain special obligations to rescue another person, B, from grave harm, especially under the condition that A is the only person capable of rescuing B and if doing so involves no risk or too large costs for A (Butt 2007). Similarly, it is sometimes argued in the literature on global ethics that part of the responsibility for remedying global injustices should fall on those agents most capable of doing so. This is perhaps especially so when the agent who otherwise has the primary responsibility for promoting the well-being of certain individuals is either unable or fails to do so (Brülde 2011;Miller 2001).
For the purpose of this article, we should note that prison officials, policy-makers, and social work professionals are all capable of alleviating the injustice suffered by children of incarcerated parents. In particular, each of these state agents can, given the power and capacity that follows from their professional roles and social positions, be reasonably expected to implement a number of child-friendly policies of the sort that may assist children of incarcerated parents and ease the burdens of their families, none of which are very demanding.
First, prison officials and policy-makers are, in virtue of their professional roles, especially capable of improving and directly influence prison conditions to better accommodate children's needs and thus avoid causing further distress among children of prisoners. As discussed in relation to the causation principle, one way of doing so would be to improve the opportunities for and frequency of family visits, but also to improve the physical environment in which family visits take place. As is illustrated by recent penal reforms in the Scandinavian countries, visiting rooms and visiting areas can often be made more child-friendly by having designated rooms with more welcoming and playful furnishing. Nowadays, many Scandinavian prisons also have specially assigned facilities for overnight visits for families and children, some of which are furnished as guest apartments. As Scharff Smith (2015) points out, one important vehicle for these and other child-friendly improvements within Scandinavian prisons has been the introduction of 'children's officers'. These are members of the prison staff who are responsible for anchoring a child's rights perspective in their respective institutions with regard to prisoner's children. This can include working to improve visiting conditions and procedures, implementing parental study circles and initiating special child-centered events in the prison (Scharff Smith 2015).
Second, we should note that states (at least well-functioning democratic ones) are often capable of helping children of incarcerated parents by providing additional financial and social support to families of incarcerated individuals. As previously noted, a parent's incarceration can lead to financial stress for the remaining household members, and because of the stigma involved in having an incarcerated family member, families are also at risk of social isolation when facing the many challenges that a parent's incarceration might create. It has been noted that these burdens may cause further distress among children of incarcerated parents as many of them worry about their stressed caregivers, who might be overly burdened because of a co-parent or close family members incarceration (Nesmith and Ruhland 2008). Consequently, as Nesmith and Ruhland (2008) points out, one important way of mitigating the harm caused to the children of incarcerated parents is by supporting their caregivers by providing more generous social welfare support. 7 I suspect that some might resist this proposal and argue that states are not necessarily morally obliged to provide this type of support, at least not to partners of prisoners. For example, as Tadros (2011) points out, one might argue that partners of incarcerated individuals have less of a complaint against being harmed, especially compared to victims of crimes that would be committed should the state refrain from punishing criminal wrongdoers. One reason for this, or so Tadros suggests, is 'that at least some people who have voluntarily associated with and now rely on criminal wrongdoers could have avoided being harmed in this way' (359). He further suggests that even if people do not always have much of a choice, given the ways of the heart, 'we might think that accepting harms of this kind [i.e. emotional, social and financial harms] is what relationships are all about; perhaps that is what the 'for richer or poorer' part of the marriage vow really means' (Tadros 2011, 359). If these observations are correct, the state might perhaps be morally justified in harming partners of incarcerated individuals without necessarily owing them any compensation for doing so. But then again, we should bear in mind that the situation that individuals may face because of their partners incarceration can be very burdensome and sometimes overwhelming (Minson 2020;Nesmith and Ruhland 2008). Therefore, even if the remaining parent might perhaps lack a legitimate complaint against being harmed by their partners incarceration, the resulting situation might still have a detrimental impact on their children who, unlike the remaining parent, cannot avoid relying on their incarcerated parent. 8 In addition to social welfare support, it is frequently observed in the scholarly literature that it is beneficial for children of incarcerated parents that they receive accurate and ageappropriate information about their incarcerated parent and what prison is like. As Nesmith and Ruhland (2008) points out, children who receive false or inaccurate information about their parent's incarceration are often left to their own imaginations, which in turn might result in further distress, anxiety and feelings of insecurity. Similarly, Jones et al. (2013) stress that disclosure of their parent's incarceration, when provided in an age-appropriate way, may help children of incarcerated parents adjust to the situation and reduce feelings of anxiety and guilt. While the responsibility to inform the child about their parents' incarceration arguably falls on their caregivers, it is crucial that the caregivers are informed about the importance of providing accurate information to their child, avoiding secret-keeping and talking to the child about their experience. In line with the capacity principle, I would suggest that both state officials and local authorities, most notably prison officials and social work professionals, have a particular responsibility in this context. In virtue of their professional roles they are especially capable of making sure that caregivers receive this information as well as how and where they can receive further social and emotional support. On an institutional level, relevant authorities also have a responsibility to develop appropriate training for police officers, prison staff, schools, and social work professionals to be sensitive to the needs of children with incarcerated parents and learn how to best provide both social and emotional support to the families of incarcerated individuals. 9 This should include education about how a parent's incarceration may affect the child at all stages of the criminal justice system, including the time of the arrest, during the pretrial phase, during the sentencing phase, during the parent's incarceration, and when the parent is released back into society. 10 The discussion above suggests that an important part of the responsibility for safeguarding the welfare of the children of incarcerated parents should fall on agents other than their direct caregivers. Given how they are especially capable of influencing the prison environment, prison officials and policy-makers can plausibly be held responsible for implementing several child-friendly policies within their prison settings. More so, implementing policies of this sort does not have to be particularly burdensome, especially not in otherwise already well-functioning criminal justice systems. On the same basis we should expect states to provide financial support to families of prisoners, social work professionals to inform caregivers about the importance of providing children with ageappropriate information about their parent's incarceration, and that the police, schools and social services, are properly educated about how a parent's incarceration may affect the child at all stages of the criminal justice system. As in my discussion of the causation principle, the fulfillment of these responsibilities amounts to the implementation of welfare policies of the sort that might help mitigate harm caused to children of prisoners and assist their caregivers, including the incarcerated parent, in fulfilling their obligations to care for the affected children.

The contractual agreement principle
One important way of establishing moral responsibilities in any context is through contractual agreement. By signing a contract, the contracting parties gain certain responsibilities towards one another, just as someone who makes a promise has a moral obligation to keep it. Similarly, in the context of global justice it has been argued that part of the main responsibility to promote people's health falls on their respective national governments due to the signing, or contractual agreement, of international human rights conventions (Brülde 2011). For the purpose of this paper we should note that 196 national governments have explicitly agreed to protect children's rights by signing the UN Convention on the Rights of the Child (UNCRC). Therefore, aside from the fact that many of these states are often capable of mitigating the harm caused to children of incarcerated parents, they also have particular responsibilities in this context because they have formally agreed to respect children's rights.
In a way, the UNCRC does not merely establish who is responsible for protecting the rights of children. It also provides an account of social justice in its own right that has direct implications for the situation of children of incarcerated parents. To take but one example, the UNCRC holds that states 'shall respect the right of the child who is separated from one or both parents to maintain personal relations and direct contact with both parents on a regular basis, except if it is contrary to the child's best interests' (Article 9.3). As Scharff Smith (2014) points out, this is a far-reaching right that is followed by a recommendation for state parties to ensure the right to regular prison visits for children of incarcerated parents and for visits to take place in child-friendly environments whenever possible. The UNCRC further states that when a parent is separated from the child through imprisonment or detention, state parties 'shall, upon request, provide the parents, the child or, if appropriate, another member of the family with the essential information concerning the whereabouts of the absent member(s) of the family unless the provision of the information would be detrimental to the well-being of the child' (Article 9.4). This too reinforces one of the points made in our discussion of the capacity principle, namely that states and relevant authorities have a moral responsibility to ensure that children are provided with accurate and age-appropriate information about their incarcerated parent and their whereabouts. This is crucial, not least since a lack of information might be detrimental to the well-being of the child (Nesmith and Ruhland 2008).
While a large majority of states have signed the UNCRC, we should note that this is not true of all states. The US, for example, has not signed it. Moreover, even for those countries that have signed the convention, it is not necessarily legally binding. 11 Different states might also interpret the convention differently, not least in terms of what lies in 'the child's best interests'. However, for the purpose of my discussion here it is worth noticing that this lack of clarity, or the lack of legal rigidity, does not discharge states of their moral responsibility. With or without formal agreements, states are still to an important extent morally responsible for remedying the harm caused to children of incarcerated parents for reasons discussed in relation to both the causation principle and the capacity principle.

The beneficiary pays principle
The discussion above suggests that relevant state authorities and state officials have an important moral responsibility for the children of incarcerated parents. In particular, I have argued that policy-makers and relevant state officials are responsible for implementing child-friendly policies within prisons, providing rehabilitation opportunities for incarcerated parents, and assisting caregivers who are taking care of children with incarcerated parents. That said, I have not yet said much about whether fellow citizens too have particular responsibilities in this context. One possible reason for thinking that they do is that fellow citizens are the beneficiaries of the type of institution that contributes to the injustice suffered by the children of incarcerated parents, given that prisons are generally regarded as being vital for upholding law and order and deterring criminal activity.
This type of reasoning is in line with the beneficiary pays principle, which states that agents who innocently benefit from an injustice have rectificatory obligations toward the victims of the injustice or wrongdoing in question (Butt 2007;Duus-Otterström 2017;Hickey et al. 2021). However, while influentially evoked in a range of different fields of applied philosophy, 12 I am not certain to what extent this principle adds anything to my analysis of the proper allocation of remedial responsibility for the harm caused to children of incarcerated parents. Although most individuals benefit from having an institution of criminal punishment, it is not clear to what degree individual members of the wider society benefitif at allfrom the injustice suffered by children of incarcerated parents, at least not in otherwise reasonably just societies. Yet this does not mean that individual members of the wider society don't have important responsibilities towards these children even in this case. As previously mentioned, the trauma associated with losing a parent to the prison system is often 'compounded by hostile, disapproving, or indifferent social attitudes pertaining to the loss of a family member through imprisonment' (Arditti 2012, 103). Hence, I think it is fair to say that members of the wider society too are partly responsible for the harm and injustice caused to children of prisoners and their families, since they might often contribute to the social exclusion of families of prisoners. At the very least, individual members of the wider society therefore have a responsibility not to contribute to the social exclusions and marginalization of families of prisoners. However, this responsibility stems from the more general moral obligation not to harm others rather than the fact that the members of wider society benefit from the injustice suffered by children of incarcerated parents.

Shared responsibility for remedying the injustices caused by parental incarceration
So far, I have examined four principles for allocating remedial responsibility and their respective implications for who should bear the responsibility for remedying the injustice suffered by children of incarcerated parents. Putting these things together, I believe that my discussion of the aforementioned principles clearly shows that this responsibility is shared among several agents, including the incarcerated parent, remaining caregivers, prison officials, social work professionals, and, to somewhat lesser extent, fellow citizens.
As noted in relation to the principle of causation, it is reasonable to think that incarcerated parents are to a significant extent morally responsible for the harm caused to their children. After all, parents who commit imprisonable offenses can often be expected to have reasonably foreseen the harm that they would cause to their children, should they get convicted, and it is reasonable to think that they could have acted differently. Consequently, they are also morally responsible for remedying the injustice that they have played a part in causing. However, as I have argued here, it is important to remember that it is not always appropriate to hold incarcerated parents fully responsible for their actions (given how many prisoners have histories of mental health issues and substance abuse). Moreover, the extent to which children of incarcerated parents suffer is largely determined by the decisions and omissions of prison officials and policy-makers. Thus, it is reasonable to hold them morally responsible for at least part of the harm caused to these children. But even if prison officials and policy-makers are partially responsible for the harm caused to children of incarcerated parents, their responsibility in this case mainly consists of creating and maintaining the types of conditions under which incarcerated parents are recognized as social contributors and can maintain meaningful social contact with their children. This, or so I have argued, includes a responsibility to provide reasonable opportunities for parents to participate in rehabilitation programs and facilitate regular and meaningful family visits, which in turn includes responsibilities to make visiting conditions and procedures more child-friendly. In addition, prison officials might also have further responsibilities to facilitate parental study circles and special child-centered events in the prison. These recommendations are also supported by the capacity principle and the contractual agreement principle.
As argued in relation to the capacity principle, policy makers and state officials also have a responsibility of developing and implementing a number of policies that may help mitigate the harm caused to the children of incarcerated parents by assisting their caregivers in fulfilling their obligations of care. For example, it is reasonable to expect that states are capable of supporting caregiversincluding partners of prisonersby providing additional social and financial support. Social work professionals can be expected to inform caregivers about how to help children cope with a parent's incarceration, such as the importance of providing accurate and age-appropriate information about the incarcerated parent and what prison is like and avoid secret-keeping. Policy-makers and relevant authorities can also be expected to develop appropriate training for police officers, schools, social work professionals, and prison staff geared toward being sensitive to children's needs in this context.
Last but not least, individual members of the wider society are also responsible to some extent for the harm caused to children of incarcerated parents in virtue of the fact that they might contribute to the social exclusion of families of incarcerated individuals through the perpetuation of social stigma and discrimination. Consequently, they have, at the very least, a responsibility not to contribute to the social exclusions and marginalization of families of prisoners. As argued in relation to the beneficiary pays principle, this responsibility is grounded in the general moral obligation not to harm others.
The analysis provided in this paper overlaps with previous philosophical analysis of the situation of families of incarcerated individuals. Writing specifically from the context of the US criminal justice system, Lippke (2017) argues that our concern for how parents' punishment might affect their families provides a 'distinctive reason to urge more humane and helpful forms of imprisonment, along with shorter sentences ' (2017, 653). Further, he suggests the undeserved harm caused to families of prisoners can be mitigated by making prisons more accessible to families of prisoners and with more generous welfare support (Lippke 2017; see also Manning 2011). These suggestions are very much in line with the conclusions drawn from the overall analysis of this paper. What this paper adds, besides offering a novel theoretical approach to this issue, is a muchneeded discussion to the effect that the responsibility for remedying the social injustice that parental incarceration might give rise to is shared among a range of different agents, including the incarcerated parent, prison officials, social work professionals and, to a somewhat lesser extent, members of the wider community. As is evident from the above analysis, the state's responsibility in this case cannot be dismissed on the basis that the parent, rather than the state, is to blame for the harm caused to their children as a result of their parent's incarceration. Nor can the burdens endured by families of incarcerated individuals be dismissed on the basis that the remaining parent (or other caregivers) has voluntarily associated with the incarcerated family member. By bringing the question of remedial responsibility to the fore, my hope is also that what needs to be done to address the social injustice that befalls many children of incarcerated parents will become clearer, not only in the US (where the prison system is already widely considered to be deeply unjust) but also elsewhere. Focusing on the issue of remedial responsibility shows that policy-makers, prison officials, social work professionals and individual members of the wider community all have an important role to play. It also helps specify and clarify the obligations that these various agents owe to the children of incarcerated parents, thus moving beyond somewhat more general descriptions of what should be done. More generally, I believe that my focus on principles of allocating remedial responsibility points toward a number of wider societal conditions that should be fulfilled in order for a criminal justice system to be at least reasonably just. After all, we should expect that a minimum requirement of any criminal justice system that seeks to enact justice should not only be to punish criminal wrongdoers but also to avoid causing new injustices.

Conclusion
Children of incarcerated parents are often described as the hidden or forgotten victims of criminal justice. Moreover, it has been noted that the circumstances of many children of incarcerated parents goes against established principles of social justice. Thus, even though some might argue that the harm suffered by these children is merely a regrettable side effect of an otherwise morally justified criminal justice system, it is also difficult to dismiss the claim that something should be done to alleviate the harm caused. In this paper, I have discussed four principles of responsibility allocation (here referred to as the causation principle, the capacity principle, the contractual agreement principle and the beneficiary pays principle) and their respective implications in response to the question of who should bear the responsibility for remedying and mitigating the social injustice often suffered by the children of incarcerated parents. In doing so, I have argued that the responsibility is shared among different actors, including the incarcerated parents, remaining caregivers, prison officials, other state actors and, to somewhat lesser extent, individual members of the wider society. In particular, I have argued that even if incarcerated parents to an important extent are morally responsible for the harm caused to their children, this does not mean that prison officials and other relevant state actors do not have important moral responsibilities in this context. To the contrary, prison officials are morally responsible for creating and upholding the type of conditions under which incarcerated parents can fulfill their respective parental and remedial responsibilities and contribute socially to their children's well-being. This, or so I have argued, includes a responsibility to provide reasonable opportunities for parents to participate in parenting and rehabilitation programs and to facilitate meaningful family visits, which in turn might include further responsibilities to make visiting conditions and procedures more child-friendly. Similarly, while an important part of the responsibility in question falls on the children's caregivers (such as their remaining parent or another family relative), state officials have the responsibility to develop and implement policies that might help mitigate the harm caused and assist these caregivers in fulfilling their obligations to properly care for the children of incarcerated parents. Among other things, they should support caregivers by providing information about how to help children cope with a parents' incarceration, provide welfare support, and develop appropriate training for police officers, schools, social work professionals, and prison staff to be more sensitive to children's contextual needs. As for members of the wider society, they have moral obligations not to contribute to the marginalization and stigmatization of prisoners' families. Recognizing these different roles and the responsibilities attached to them is crucial for any society that aspires towards treating all of their citizens, including the children of incarcerated parents, justly. Notes 1. 'Vulnerability' is an ambiguous term used in a range of different ways in the philosophical literature. One helpful distinction here is between 'inherent' and 'situational' vulnerability (see, e.g.Bagattini 2019; Schweiger 2019). Children are often portrayed as inherently vulnerable in the sense that they have particular developmental needs and because of how they, qua children, lack certain cognitive, emotional, or physical abilities. This is different from situational vulnerability, which has to do with social practices, norms and other contextual factors, such as poverty, discrimination, and various forms of marginalization. As the discussion in this paper illustrates, I believe that parental incarceration can be understood as a specific form of situational vulnerability. 2. Scharff Smith (2014, 45-47) estimates that in more than half of all cases of parental incarceration, the children had significant contact and close connection with their parent before their incarceration. In a large portion of cases the parent and child were also living together. This, Scharff Smith argues, suggests that the number of children that suffer due to a parent's incarceration is not negligible. 3. There is an ongoing philosophical debate as to whether the standard candidates for the metric of justice, including the capability approach and the Rawlsian notion of primary goods, are sufficient for addressing the issue of justice for children. One criticism is that these metrics focus too much on what is valuable for adult human beings (Lindblom 2016;MacLeod 2010). I will not dwell on this issue here though. Instead, I assume that the capability approach as well as the Rawls's notion of primary goods provide plausible starting points for thinking about justice for children. For a more detailed discussion of the capability approach in this context, see Schweiger and Graf (2015). 4. This list of principles draws from Butt (2007). For a similar, yet somewhat different list of principles, see Hickey et al. (2021). For a careful discussion of (1) and (2), see Miller (2001). Miller (2001) also mentions a communitarian principle according to which people might gain special obligations toward those whom they have special ties, such as those arising from shared activities and commitments or a shared history or identity (see also Hickey et al. 2021). However, since my interest in this paper concern the proper allocation of responsibility within a particular political community, it seems to me that this principle becomes somewhat irrelevant in this context, even though it might perhaps be relevant in the context of global injustice in which Miller introduced it. In later work, Miller also endorses the beneficiary pays principle (see Miller 2007). 5. What constitutes an imprisonable offense is of course a matter of philosophical debate. For a discussion of this issue in terms of both crime-preventive and retributivist theories of punishment, see Lippke (2007, 63-79). 6. For an illuminating discussion of the criminal responsibility of people who commit crimes under the influence of mental health disorders, see Hartvigsson (forthcoming) 7. It should be observed that the large majority of caregivers in this case are women, regardless of whether it is the child's mother or father that is incarcerated (Minson 2020). Thus, this also raises a number of issues related to gender equality. 8. To be sure, Tadros (2011) too recognizes this point and suggests that states should perhaps ameliorate the harm caused to the children of incarcerated parents by improving opportunities for prison visits and provide financial compensation. See also Lippke (2017) for a careful discussion of whether it is appropriate to hold families liable to the harms they suffer due to the incarceration of a family member. 9. A fairly recent report from the Swedish National Board of Health and Welfare suggests that this type of competence is often lacking among those professionals who meet children with incarcerated parents (Socialstyrelsen 2018). Even if this report only focuses on the situation of children of incarcerated parents in Sweden, I suspect that the situation is similar in many other European countries. 10. For a comprehensive discussion of how the rights and interests of children are affected during the different stages of the criminal justice system, see Scharff Smith (2014). 11. That said, a few states, such as Sweden and Norway, have made the UNCRC legally binding. 12. For example, the beneficiary pays principle is commonly defended in the context of global justice, where it has been argued that present beneficiaries of injustices committed in the past have a duty to compensate the descendants of victims of such past injustices (see, e.g. Butt 2007;Collste 2015). It is also frequently discussed in the philosophical literature on climate justice and how the burdens of mitigating and adapting to climate change should be allocated. In domestic affairs it has been evoked in defense of preferential hiring (Duus-Otterström 2017).

Funding
This work was supported by FORTE [grant number 2018-01116].

Notes on contributor
William Bülow is a senior researcher at the Centre for Research Ethics & Bioethics at the department of Public Health and Caring Sciences at Uppsala University. His research interests are in legal philosophy and applied ethics.