What Kind of Functionings Matter for Global Justice for Children?

ABSTRACT In this paper I am interested in the ‘currency of justice’ question, which is important for a theory of global justice for children, based within the capability approach. My paper is structured in three parts. In the first part I will examine the distinction between capabilities and functionings and consider the role this plays in global justice for children. In the second part I will discuss the temporality of functionings as a key feature for understanding global justice for children. It is important to distinguish here between the timing (when children have certain functionings) and the duration of functionings (for how long children have those functionings). In the third and final part I will discuss the role of specific functionings which have been termed in other debates ‘kindergoods’ or the ‘intrinsic goods of childhood’. I will argue in this regard that child-specificity can have different meanings, that some functionings can only be had by children and that some functionings are particularly valuable only for children despite the fact that adults can also have them.


Introduction
The capability approach established itself as a major player in the global justice debate and it would also seem to be well-suited to provide a fruitful framework for discussing global justice for children. But although there is ever increasing empirical and theoretical literature on children and childhood applying and reforming the capability approach Hart, Babic, and Biggeri 2014;Dixon and Nussbaum 2012) there are still only very few works on its philosophical conception of (global) justice for children . This paper aims to fill certain gaps in this respect by questioning the role of functionings as a currency of global justice for children. This paper does not aim, however, to defend the plausibility of the capability approach or a specific capability-based take on global justice against competing approaches such as resourcism or utilitarianism. That has been covered elsewhere already (Kelleher 2015;Oosterlaken 2013;E. Anderson 2010). Furthermore, this paper cannot hope to develop and properly defend a capability/functioning-based theory of global justice for children in all its complexity. Rather I hope to clarify smaller but important questions which arise within such a broader project and which centre around the role of capabilities and functionings as the currency of global justice. I also do not engage in a close-reading of Martha Nussbaum or Amartya Sen's version of the capability approach and their writings on global justice or children for that matter but follow a systematic approach rather than a hagiographic one. The broadness of the capability approach, which is perhaps one of its great strengths and necessary for its appeal to so many different disciplines, also makes it quite hard if not impossible to speak for it in its entirety and this paper does not wish to do so. Finally, my arguments here do not say much, if anything, about two other important questions regarding global justice for children, namely how functionings or capabilities should be distributed on a global scale and who would be responsible for such a just distribution.
Nonetheless I want to say a few words about what I mean by global justice for children, which helps to frame the discussion in this paper. After that I will give a brief overview of the paper and its structure. There are many different understandings of global justice, and of what the two components 'global' and 'justice' should refer to. I acknowledge that there are no clear-cut definitions available but the following working definition seems helpful. A theory of global justice for children should be able to determine what all children, wherever they are, are entitled to as a matter of justice, which implies determining what kind of goods they are entitled to and to what extent they are entitled to them. We can call this the positive approach to global justice. Furthermore a concept of global justice should be able to determine what constitutes global injustice for children. Here, adapting a suggestion of Gillian Brock (Brock 2017), I propose that global injustice is an injustice that affects a large number of children in different countries and that its solution demands the coordination or cooperation of many different states (directly or through established global institutions). We can call this the negative approach to global justice, because it starts with global injustice. The positive and the negative are not fully overlapping. For example, some social injustices (such as relative poverty in a rich country) are not global injustices (because they only affect children in one country and do not demand cooperation between countries but only redistribution within this one country) but they might be covered by the positive approach to global justice (because it could also demand that no child should be poor compared to relative standards of poverty). The clarifications provided in this paper could prove to be helpful for both the positive and negative approaches, because they can help to design global justice for children and to detect global injustices.

Capabilities or functionings
Capabilities and functionings are two key terms of the capability approach and they have a particular role in understanding global justice for children. I will begin by clarifying how I will understand them in this paper -within the capability approach there are different interpretations of these two terms. (Nussbaum 2011;Robeyns 2005). Functionings, on the one hand, are the activities and states that make up a person's life; they are the different 'beings' and 'doings' living consists of. And since human existence consists of many different doings and beings, the category of functionings is a broad one. Examples include, being healthy and educated, having shelter, taking part in the life of the community, or being undernourished, as well as riding a bike or a car or feeling emotional distress. Capabilities on the other hand are defined as the functionings a person actually has access to and these reflect the person's freedom to realize different functionings. To give an example: eating is a functioning, while the real opportunity to eat is its respective capability. The capability approach values capabilities and views them as essential and more important than functionings, when it comes to evaluating a person's life. Also, for global justice it argues that we should promote capabilities rather than functionings and that a society is just if it gives people the opportunity to live a life they have reason to value, which means providing them with capabilities, which they can, if they wish, convert into functionings. Autonomy and personal freedom lie at the heart of the capability approach.
Before I turn my attention to children it is useful to make another terminological clarification. Capabilities depend on conversion factors, which are necessary so that a person can actually achieve the respective functionings. Conversion factors can be of different natures. There are innate personal conversion factors, like one's body and brain, and acquired personal conversion factors, like one's education, knowledge and skills. There are natural environmental conversion factors, like the climate and there are social environmental conversion factors, like technologies or societal norms and practices. So, conversion factors can be material or immaterial, and they significantly vary across time and place. Resources also play a major role here. For the capability approach, they are not ends in themselves, but means to achieve ends. Put simply, conversion factors make it possible to convert existing resources into functionings. Let's look again at the example of eating. That a person has the capability to eat, or what I called above the real opportunity to eat, depends on many different things: that the person is able to eat or if not to have the appropriate support, that tolerable food is available and that the person has access to it. Depending on the social setting other conversion factors might also play an important role: is the available food banned by a food taboo because of religious or cultural reasons, is the appropriate cutlery available, if the food is not cooked it is necessary that the person has access to cooking tools, a kitchen, energy and knows how to cook. Furthermore, if the capability to eat is somewhat expanded to mean that the food is also sufficient for the person, other innate personal conversion factors might come to play like the health status and dietary needs and what kind of food is available. This example also shows that the different conversion factors (personal and environmental) are deeply entangled. The social environment shapes both the natural environment and how it operates and can be a barrier or an enabler. The same is true for personal conversion factors of which some are acquired in social processes and others are shaped by natural influences. The take home message is that without sufficient conversion factors there are no capabilities, and that conversion factors, and also the personal ones, are deeply social.
What does this all mean for a capability-based understanding of global justice for children? A capability-based approach to global justice for children needs to clarify whether capabilities or functionings are aspects to which children are entitled as a matter of justice. Only then can we clarify what kinds of capabilities or functionings matter in this respect. The mainstream of the capability approach, including that of Nussbaum and Sen, opts for capabilities in the case of adults and is very reluctant to value functionings for adults because that would be closely aligned to paternalism and anti-liberalism (Nussbaum 2011;Sen 2009;Robeyns 2005). Adults should be allowed to choose which functionings they want to realize out of a wide range of capabilities. But there are exceptions like Jonathan Wolff and Avner de-Shalit de-Shalit 2007, 2013), who argued for the importance of fertile functionings instead of fertile capabilities. The focus of the capability approach on capabilities and the decision to value them more than functionings rests on assumptions about what features a person has, what is important in a person's life, what should be the goals of justice and the relation of the agents of justice to those subjects of justice. By agents of justice I mean those persons, institutions or systems, which should enable justice and by subjects of justice I mean those for whom justice should be established. In a nutshell the subjects of justice are deemed as autonomous and rational agents, who have an interest in living a life they have reason to value and who reflect on those reasons and on those values. I will largely ignore here further assumptions, which are sometimes made, for example that the subjects of justice are members of a society which is characterized by a division of labour, where everyone is working not only for themselves but together they produce the common good. An alternative assumption might be that the subjects of justice share a sense of justice and have a mutual interest in each other's well-being. The question then is, if and to what extent such assumptions about the subjects of justice also apply to children and if not, what that would that imply for the question regarding the adequate currency of justice for them, in other words whether one should opt for capabilities or functionings in this regard.
Surprisingly, this question has received limited attention so far within the capability approach. Some continue to speak of children's capabilities without questioning the adequacy of the term . Nussbaum and Dixon (Dixon and Nussbaum 2012) argue in favour of functionings because children are not (sufficiently) autonomous and others have also followed that path . Here I want to try to give a more nuanced answer, which comes with the downside that I will not be able to provide a definite one but rather several criteria which are then in need of sensitive application to particular cases. Consequently, it is not as simple as saying that, in the case of an immature and non-autonomous person, we should always choose functionings and in the case where we are dealing with an autonomous person, we should choose capabilities. There are cases, where this would be unjust. This sole focus on the actual abilities of the persons in question can be misleading, and the case of children illustrates that.
Firstly, autonomy, as well as other capabilities, is not something that comes into existence at one time and is without degrees or variation. Childhood is the phase during which autonomy is developed, exercised and trained. Children -as well as adults for that matter -might show autonomy in some areas of their lives while they are only semi-or non-autonomous in other areas of their lives. That is part of what it means to go through childhood. The distinction between local and global autonomy is useful here, one which I borrow from Andrew Franklin-Hall (Franklin-Hall 2013). It means that autonomy can be viewed as the capability to make little choices for oneself, such as what to wear today or which kinds of hobbies one prefers. These little choices are important, because they influence who we are, and children are able to make them at an early stage. For such choices to be autonomous I mean that it is not sufficient for a child to simply say what they want to wear but that they also reflect on it and have developed their own sense of self and how this relates to their choice. Preferences in this regard are also in flux, and childhood is a phase during which most people experience changes in this regard and try different types of style that might fit for them. This is local autonomy, whereas global autonomy describes choices that affect one's life that are deeper and more lasting. Franklin-Hall calls this 'life-authorship' and we might say that they write the story of one's life. Such choices usually have higher demands in terms of self-reflection and maturity. They are, for example, the choice of one's occupation, to marry or to raise children. Children often lack the abilities needed to make such important choices, and many societies are designed in a way that they do not have to make them or that guardians make them for them. Needless to say, local and global autonomy are broad categories, with overlaps, and that adults also often make irrational and poorly reflected choices in both areas. So, if autonomy is not a monolithic bloc but rather develops and comes in various degrees and parts, at some stages during childhood and with respect to some areas of a child's life capabilities should matter whilst at other times functionings should take priority. But autonomy is not only an capabilitya meta capability that allows to have further capabilities -but also a value. Autonomy is considered intrinsically valuable because being autonomous is part of a good life. Autonomy is also instrumentally valuable because it allows to have abilities and to choose from options. The value of autonomy is often also closely tied to rationalityautonomous is the one who does not just make any decisions, but makes those decisions based on considerations that are right for him or herself. To say that autonomy is valuable leads to a moral appeal that people should be autonomous or that they have a right to be autonomous and exercise autonomous decisions (within certain limits). Especially for children, whose autonomy is often curtailed by many adults in a casual and unquestioning manner, the demand for respect for the value of autonomy has ethical and political consequences. Unfortunately, I do not have the space here to examine the many pitfalls associated with the concept of autonomy as a value, nor how the connection between autonomy and rationality is to be adequately understood for children, for example, the question of whether the articulation of desires is already sufficient to speak of autonomy or whether it is necessary to reflect on and justify these desires.
Secondly, abilities are not the only things that matter when it comes to determining if we should opt for capabilities or functionings in the case of children. For example, Joel Anderson and Rutger Claassen (J. Anderson and Claassen 2012) argued for a 'regime of childhood', which is based on other important considerations like the limited liability of children, parental responsibilities and the phase of childhood as one of special protection. This regime of childhood overrules a simplistic test of (autonomy) abilities as the only benchmark to decide what children should be allowed to do and to decide for themselves. Likewise Andrew Franklin-Hall (Franklin-Hall 2013) rejects the idea that children should be treated as similar to adults as soon as they show similar autonomy. Based on his differentiation of local and global autonomy he argues that childhood and life stages in general have a significant and distinct moral value, which should be respected and it is necessary to act paternalistically towards children, who are similar to autonomous adults, because a moratorium of youth allows them to acquire the skill, experience and knowledge to become authors of their lives. I believe that these arguments have some normative force to overrule the autonomous choices of children and that this provides some back-up to opt for functionings instead of capabilities. It is important, however, to note that within the protective state of childhood there is still plenty of room for allowing children some autonomous choices and thus to provide them with capabilities instead of looking only at functionings. Franklin-Hall makes room for autonomy during childhood by reference to local autonomy, although I believe that he somewhat underestimates the value of it. Here I borrow an argument from Monika Betzler (Betzler 2011), where she showed that the exercise of autonomy is crucial during childhood and that children should be supported to make some choices (although under the guidance of their guardians and with some protection from the consequences). This exercise of autonomy implies that children are provided with several capabilities as a matter of justice supposing that autonomy is a valid goal of justice. This argument can be further strengthened with regard to research which shows that making choices is not only valuable as a kind of training but is intrinsically for children's well-being (in particular for older children) (Tisdall et al. 2006).
Thirdly, the question whether capabilities or functionings matter more for children needs to take into account the fact that childhood is a particular phase but one that ends at some point in life and that most children enter adult life. This means that global justice for children can look at the actual state of children, and what they can or should do and have during their childhood, or it can look at them as adults-to-be or it can try to balance both views (Bagattini and Macleod 2014;Graf and Schweiger 2015). The balanced view, however complicated this might be to achieve in theory and practice, will acknowledge that although for children as children functionings matter more, capabilities certainly matter for them in terms of what they are entitled to as adults. I will not argue here for the value of autonomy for adults because others have done that extensively (Adams 2008;Mullin 2014;Betzler 2011) but from this latter, adults-to-beoriented perspective, capabilities are the right currency of global justice for children, although the protection of functionings during childhood is the right means to achieve justice in this regard.
Fourthly, one should take into account the different roles capabilities and functionings can play. While capabilities are important because they reflect autonomy, choice and agency they are not only possibilities. They are real in a different sense from functionings. This different ontological status of capabilities and functionings allows the latter to play a particular role: they can be the basis for other capabilities and functionings. This influence can go either way but depends on the functioning being an actual feature (being or doing) of the child or adult and not a mere possibility, something she might do or be. For example, while the functioning of illness negatively affects several capabilities and functionings, the functioning of good health has positive effects. Borrowing concepts from Jonathan Wolff and Avner De-Shalit (Wolff and de-Shalit 2007) I want to call those functionings with positive effects fruitful, and those with negative effects corrosive. For children some functionings are decisively important because they are necessary preconditions in their development path. I will say more about that in the next section though.
Based on these observations I would conclude that functionings should play a greater role in global justice for children than for adults. Children are, on average, not only less autonomous and their development depends on achieving important functionings, but they are also entitled to experience childhood as a protective phase with guidance from guardians. That said capabilities do play an important role for them, both as the ends and means of justice (what they should achieve and acquire during childhood and what can help them to do so). I would propose a few criteria that can help us to decide in particular cases whether capabilities and functionings should be given priority: the abilities of the child (is she able to decide for herself?), the embedding of the capability or functioning (would other people/children be affected? Are there social norms and practices involved in shaping the decision and the capability or functioning?), the status of the child within the development path (what does the child need for her healthy development), the capability or functioning in question and its effects (would a 'wrong' decision have severe effects? Are the effects short-or long-term? Would it be reversible?), the needed conversion factors (what is necessary for the child to achieve this functioning? Are there any risks involved?), the personal and social environment (is there support to reflect the decision and to deal with the achieved functioning? Is there support in case something does not work out?).

Temporality and global justice
The question of whether we should favour capabilities or functionings is only one among many that needs clarification. It is also a very broad one, because it does not go very deep into the issue of what kinds of capabilities or functionings should matter for children, let alone clarify which concrete capabilities or functionings are important for global justice or how to choose them. In this section I will focus on functionings, following the path laid out above. This does not mean that capabilities do not matter for children. They do as I argued for above. It is mainly for reasons of keeping my argument simple and fitting it in the space of a paper that I choose to speak from now on primarily of functionings. I believe though that they have a special place in global justice for children and I hope that my arguments in the previous section showed that sufficiently. So, what I am going to investigate in this section is the temporality of functionings, in particular the importance of their timing and duration in order to make sense of what children are owed as a matter of global justice.
The capability approach in general has quite a static view of the world (Robeyns 2005(Robeyns , 2006. It focuses, both in its theoretical and empirical-oriented versions, on what people have at a given moment. That shortcoming is well-known as are also the approaches to overcome it. For children it is of particular importance because children are, as I have already said more than once in this paper, developing beings and time matters for them significantly. Childhood is usually determined by being younger than a certain age, and development is itself a temporal concept. So, although time is certainly also important for adults and their lives, choices and actions are also happening within time (and space) and time and timing seem to play an even bigger role in children's lives and in a normative evaluation of them. It is for these reasons that researchers like Jérôme Ballet, Mario Biggeri and Flavio Comim  have tried to better adapt the capability approach for children in ways that incorporate concepts of development or evolvement, as they call it. They show how evolving capabilities -inspired by the concept of evolving capacities (Lansdown 2005)play a crucial role in determining children's well-being and development. Evolving capabilities have a trajectory and their evolvement influences the development of other capabilities. Here I want to push that agenda a bit further and introduce a more nuanced understanding of temporality in global justice for children.
Functionings, as states and doings, are temporal in at least two respects. They have a temporal extension, for example the time it takes to eat breakfast, and they occur at a particular time, for example a child might eat breakfast at seven in the morning. Both aspects matter for global justice, and this is true for both adults and children. For example, it makes a significant difference how long phases of hunger and non-hunger are. Leaving aside the question of voluntariness here it appears quite normal that every person experiences some short periods of hunger from time to time (because other things are more important to do or because one forgot to pack food for a short trip), but they become problematic if they go on for a longer period of time (and also if that happens more often during one's life). For some beings and doings a certain temporal extension is constituent because they cannot be shortened or extended freely. For other beings and doings their temporal extension is important in determining their value. Some beings and doings are more valuable if they are prolonged, others might lose their value if they are overly long. But I am here not only and not even primarily interested in the subjective satisfaction beings and doings bring to adults or children but in their objective value for global justice (of which the subjective dimension might be an important part). Health, nutrition, education, bodily integrity and other such important functionings all have a temporal extension, and also a particular timing about which I will say more later.
Taking the example of education, it seems plausible to argue that education is a matter of global justice for children (Unterhalter 2005;Walker and Unterhalter 2010). One way to make sense of this claim is to ask how long children should be educated as a matter of global justice (in comparison to the capability question of how long they should have the opportunity to be educated). Leaving aside questions about the nature, scope and aims of education, it would not be sufficient for global justice if children were to be educated only for a short time, say for a few months. Children need informal and formal education for a longer period of time, maybe for the majority or the whole of their childhood. This is simply because there are many things to learn and because they have to learn many things to become autonomous adults, with sufficient options in their lives and equipped with the skills and knowledge to lead the lives they have reason to value in a complex world such as today's. So, although just a short period of education is better than no education it would not be sufficient for global justice -and I believe this claim is true irrespective of the principle of global justice one adopts. So, the temporal extension is key to understanding, when it comes to education, what children are entitled to as a matter of global justice. The respective goal, that they have a justice-based claim to be educated (at some point in their lives, maybe when they become adults) does not say much about the crucial insight that this demands education for a certain period of time. This comes with many important implications, which are not unknown but which are sometimes overlooked in the philosophical global justice literature. Many functionings demand constant investment and engagement over a longer period of time by the agents of justice. That makes them harder to achieve. These demands in resources and time might trigger obligations on the part of the agents of justice not to discontinue their involvement. Linda Alstott concluded, based on a similar thought about the lengthy involvement needed to support a child to grow-up, that parents have a duty to stay and are morally bound not to leave their children or exit the parent-child-relationship (Alstott 2004).
The temporal extension of a functioning is also important in order to understand its development and its connection to other functionings and capabilities. As I understand it, Wolff and de-Shalit (Wolff and de-Shalit 2007) introduce the concepts of corrosiveness and fruitfulness without thinking much about them having a temporal dimension or what that would mean for them. They can be interpreted as static, which would mean that at a given point in time it is recorded if a child has certain corrosive or fruitful functionings and how they had diametrical or positive effects on other functionings. For example, a child is sick and she is socially excluded because of that. This static view hides a few important insights that lie within the concepts of corrosive and fruitful functionings. Corrosiveness and fruitfulness often do not happen instantly but they develop over time. The child gets socially excluded only if she is sick for a certain period of time; the adult, who is unemployed gets poorer over time and then at one point we can make the observation that unemployment proved corrosive for the functioning of not being poor. So, corrosive and fruitful functionings often unfold their effect over time, and for many of them it gets worse and worse or better and better over time. That has important implications for our understanding of global justice, in particular also the fragility of such functionings and the importance of resilience factors to protect fruitful functionings over time. If we know that corrosive functionings unfold their negative effects as long as they last, this also has implications for the timing of interventions and possibly the prioritization of resources.
The second aspect of temporality refers to the timing of functionings. Timing is equally crucial both for the subject and from an objective point of view. That the functioning of getting medical treatment needs to happen at the right time is obvious and also for a job application it is often crucial to actually have the skills and formal education needed and not just be in the process of acquiring them at some time in the future. For children and global justice I want to point out two things, which show how important timing is here.
Firstly, childhood as a phase of development can be conceptualized as moving from one stage to another while acquiring new features (which can be conceptualized as functionings and capabilities). For example, the weight and height of a child at a particular age is used as an important indicator in paediatrics and developmental delay is a serious health issue (Yousefzadeh et al. 2018). Childhood is full of such milestones of physical and mental abilities, which signal healthy development and also that a child is getting what she needs and is not deprived (Gardner 2013). So, for global justice for children it is crucial that they achieve certain functionings within a particular time frame during childhood. Here, again, it is worth noting that often functionings are what matters and not capabilities. A child should actually be well-nourished (measured using height and weight and other indicators at a particular time in her life and comparing that with what is bio-statistically normal) and not only have the capability to be so. A child should actually be able to walk and talk and eat on her own and not only have the capability to do so.
Secondly, what functionings children are owed as a matter of justice depends on their maturity, abilities and age. So, timing is again crucial here, because for some functionings (as well as capabilities) the child might be too young, while for others they are too old -to put it simply. In the latter case the functionings have lost their value, in the former they have not gained it. In the first section I briefly said that autonomy develops and does so by degrees. This plays a major role here again and can be further differentiated. The development of autonomy not only implies that capabilities become more important but that different capabilities (and functionings) become important. The functioning of sexual health is important for children at any age but it has a distinctively different meaning based on a child's age and abilities. Sexual health for a teenager consists not only of being physically healthy but to have certain knowledge about sexuality and about one's own body and sexuality (Graf and Schweiger 2017). It is a particularly complicated concept, involving physical, mental and social dimensions and also relationships to other human beings (whether or not in the form of engaging in sexual relations but at least in the sense of reflecting about one's own desires towards other human beings). The message here is that the functioning of sexual health -which is certainly crucial for global justice for children and is so often violated and harmed (Bang et al. 2014;Barth et al. 2013) -is made-up of different functionings and capabilities which vary significantly based on the children's age, maturity and other abilities. Therefore the timing of these functionings and capabilities is crucial, although the children's age is only an indicator here (because children do not develop at the same rate).
Based on these observations I would want to argue that for a thorough understanding of global justice for children it is not sufficient to simply choose a few important functionings (and capabilities). Rather we need to examine which functionings are important and consider when and how children can be provided with them at the right time and for a sufficient amount of time. That said it might be the case that the specific temporality of functionings (when and how long they matter) is also dependent on other external social, cultural or environmental factors, which differ between societies and local contexts in which children grow-up. If such context-sensitivity exists and to determine its influence on certain functionings is up for debate and I cannot discuss it here.

Child-specificity, kindergoods and global justice
In the discussion about justice and childhood, arguments have been brought forward that different types of goods matter for children than matter for adults. Some authors have labelled them kindergoods or 'intrinsic goods of childhood' (Macleod 2010;Gheaus 2015;Brennan 2014;Hannan 2017). I believe there are several interpretations possible for this claim, but the most prominent is that such 'intrinsic goods of childhood' are valuable only to children and that they are entitled to them as a matter of justice because of this. Examples of such goods might be innocence or imagination. As far as I know this debate has not been revisited from a capability approach perspective although it would certainly be important for it. Capability theorists have long aimed to differentiate child-sensitive or child-specific capabilities and functionings in order to measure child well-being and poverty or deprivation . So, it would also be of interest for them to explore the question of whether and what goods would matter (only) for children. My contribution in this section is of limited scope with the aim of making a few differentiations, which could prove helpful for a concept of global justice for children based within the capability approach. So, I am not going to revisit the whole debate about the value of childhood, what role 'intrinsic goods of childhood' might play here and how they relate to justice for children in general. I believe that some distinctions are worth making in this regard. In particular it is necessary to clarify, what is meant by 'valuable', by 'only' and by 'functionings' (or capabilities or goods) in the claim that some functionings are only valuable for children. I interpret the debate about kindergoods as one that is concerned with the right 'currency of justice' for a particular stage of life -childhood -and it thus incorporates, albeit most often implicitly, considerations about the temporality of the goods owed to children as a matter of justice.
Firstly, it seems plausible to assume that there are functionings (and capabilities), which only children can have. Functionings describe all different beings and doings a person can have and since children and adults differ in some particular respects there might be functionings which are based on these differences. This claim can be interpreted as having a natural or social meaning, depending on the underlying notion of childhood. The natural interpretation would argue that the difference between children and adults is not a social construction but rests, for example, on biological facts. If one understands childhood as a social construction, meaning that the differences between children and adults are not natural or only very remotely follow natural differences, some could still argue that some functionings are based on such 'constructed' differences but are only accessible to those persons who are 'constructed' in such a way. A strong interpretation of the 'social construction' thesis would then imply that in principle adults can be constructed as children, blocking them from having adultspecific functionings, and children can be constructed as adults, blocking them from having child-specific functionings. I do not wish to engage in clarifying the ontological status of childhood here but it seems plausible to assume that the differences between children and adults are not only socially constructed although this construction is highly important with regard to how natural differences are interpreted and shape social norms and practices towards children. Johannes Giesinger (Giesinger 2017) showed that the 'social construction' thesis, also in a lighter form, matters for the debate about 'intrinsic goods of childhood' because it rests on assumptions about what childhood and children are. Such claims about the 'nature' of childhood and children are important both with regard to claims that child-specific functionings exist and also for discussing whether they are valuable for children.
Secondly, functionings (and capabilities) can have quite different depth and breadth. Some aspects of it are relevant to distinguish. Functionings are often bundles of functionings, thus the combination of different beings and doings. Functionings can have context-specific variations. Functionings can be achieved in different degrees. This makes it complicated to answer the question of whether a specific functioning can only be had by children as well as whether it is only valuable for them. For example, no one would deny that education is important for children and for adults, but what is really meant by that greatly differs (Walker and Unterhalter 2010). The functioning of education consists for instance of the functionings to do things (reading, writing, calculating etc.) and to know different things. Which sub-functionings have to be achieved here to speak of education varies. In a highly developed knowledge society, different educational objectives are valued than in agrarian societies. The distinction between adults and children also plays a role here in order to indicate what is meant by the idea that children should acquire the functioning of education. If an adult only had the same level of education as a pre-school child this would not seem fair or sufficient from the point of global justice. So, while education is valuable for both adults and children and is not something only children or adults can have, certain important functionings within this bundle that are called 'education' might well only be available or valuable for children or adults. Other forms of knowledge, such as about some sexual practices, are kept away from children for moral reasons, as they are not yet mature enough to understand them. Certain specialized knowledge in professions is not yet valuable for children and cannot yet be learned by them. But there are also natural factors that favour children over adults. For example, children are made-up in a way that makes it easier for them to learn languages without textbooks -a functioning, which is almost entirely lost by most adults later in life (DeKeyser 2000). I would not say that it would not be valuable for adults to preserve that ability but that is impossible; it is a functioning which is reserved for children, but which is part of the larger functioning 'education'. Furthermore, it is worth noting again that the temporal dimension discussed in the previous section of this paper plays a crucial role here. The education of being able to read, write and to do basic maths has certainly a different value if it is acquired at the age of five or at the age of 25 or the age of 85. The fertility of education to enable to acquire other capabilities and functionings diminishes over time.
Thirdly, it is up for debate whether there are functionings (or capabilities), which are only valuable to children. One interpretation would be that there are certain functionings, which are only valuable to children because it is only during childhood that children can have them. Having a good childhood is only valuable to children because only they can experience it (and it is something valuable for them). This view is important if we look at a society at a specific moment in time and ignore the fact that these children will be adults later or that the adults have been children before. One might say that having a good childhood is still not 'only' important for children, because parents and others also value the fact that children experience it. I believe this is not what is meant here by 'only', but rather this should signify functionings which are not intrinsically important for adults as adults. So, for a world in which only adults exist or have ever existed a good childhood cannot be valuable. This means that the functioning of a good childhood (which is actually a bundle of functionings and capabilities) is only valuable to children because it is valuable for them and only children can have this functioning.
Fourthly, functionings can be valuable (only) for children in two different ways. We can look at them as developmental functionings, which are valuable (only) for children because they are needed during childhood to achieve something valuable as an adult, for example to become an autonomous adult or to have well-being. Or functionings can be valuable (only) for children because it makes their lives as children better. Both are important for global justice for children since justice should be interested in children as children (present-oriented global justice) and as adults-to-be (future-oriented global justice). This also means that some functionings, which adults can also have, are only valuable to children, and some functionings, which children can have, are only valuable to adults. A possible example would be labour (within a capitalistic system, one might want to add), which is deemed valuable for adults, and which can also be done by children, but which is deemed to be bad for them and child labour can be criticized as unjust if they have to engage in it (Satz 2010). This point is also related to the previous one I made about the right timing of functionings and how that matters for global justice.
Fifthly, what makes a functioning valuable for children can be claimed from a variety of normative perspectives. The capability approach is only one among others, and within it there are plenty of options as to why some functionings are valued while others are not. Nussbaum and Dixon (Dixon and Nussbaum 2012) refer to dignity, while others prefer to stick closer to a framework of children's rights laid out in the convention on the Rights of the Child (Stoecklin and Bonvin 2014) or to child wellbeing and well-becoming . This variety is well-known. The point made by Giesinger (Giesinger 2017), that the social construction of the understanding of what a child is and what the 'nature of childhood' is also plays a role here, although philosophers often claim to be aware of such limitations. Radically different concepts of childhood and child well-being, both contemporary and historic (Gittins 2009), are still hard to grasp and pose serious questions about how much objectivity is possible in the determination of what is valuable to children and what they are entitled to as a matter of global justice. This is related to the point of view of children in determining the "intrinsic goods of childhood". The capability approach is the framework for many empirical studies with children, and for how to provide them with proper space and time to voice their thoughts and preferences (Domínguez-Serrano, Del Moral-Espín, and Muñoz 2018). On the one hand the normative weight of children's views on these matters is still in question. Some have argued that they are not authoritative but only consultative (Brighouse 2003;Archard and Skivenes 2009), others are concerned with the epistemic injustices where children's views are unjustly ignored or downplayed by adults (Carel and Györffy 2014). On the other hand, children's views themselves can differ based on place and time, their gender, socialization or abilities. That is an issue especially for the project of global justice for children, which has a universalist reach and has to balance this universalism with the need for localism without becoming relativistic. This question arises not only regarding the question of which functionings (or capabilities) are the currency of justice but also how they should be distributed among children living in such diverse contexts.
Based on these considerations I would like to distinguish between child-specific, child-sensitive, child-like, child-oriented and child-neutral functionings. This categorization is neutral and does not prescribe a particular value to any of them for global justice for children. Child-specific functionings are such that only children can have. Child-sensitive functionings are such that both children and adults can have but their embodiment, shape and content is greatly different for children and adults. Child-like functionings are such that children and adults can have but which are more common in children and which are associated with what can be described as the 'nature of childhood' (either based on a naturalist or social constructivist understanding of it). Childoriented functionings are such that both children and adults can have but which are particularly important (either good or bad) for children and not so for adults. Child-neutral functionings are such that both children and adults can have and which are either equally common or equally important for them both.
Obviously the boundaries between them will be blurry and it depends on empirical and normative considerations as to which functionings belong in which category. Normative considerations play a role because assignment to a category must also be based on moral and social norms about childhood and what is good or bad for children. The above categorization can then play an important role in developing a theory of global justice for children. On the one hand it lets us distinguish a bit better between the value ascribed to certain functionings and how they relate to the 'nature of childhood'. For example, one cannot and should not assume that a functioning is valuable only because it is child-specific. Still it is plausible to assume that if some child-specific functionings are particularly valuable to children this has some important consequences for the proper understanding of global justice for children, because it then demands to be clearly separate from global justice for adults (at least when it comes to the currency of justice). On the other hand, it seems highly plausible that the categorization of functionings varies across children based on their age, maturity or abilities. We need to look at the temporal dimension of these functionings again. That is important for the question of their particular value for particular children and how that translates into policies which aim to target the right children with the right measures. Here it plays a crucial role if the functionings which will be developed and fostered can actually be developed and fostered (and if that development is valuable and proper for these children).

Conclusions
This paper examined some questions about functionings and capabilities as the currency of global justice for children. I argued that it is plausible to prioritize functionings over capabilities for children, but that this answer hinges on several conditions which need to be addressed when we want to say what is just for particular children in particular situations. That does not tell us, however, which functionings or capabilities should matter for global justice for children or how to choose them. Ingrid Robeyns (Robeyns 2003) made a suggestion about how to choose capabilities that matter particularly for women, and maybe her methodology could be transferred to the case of children. Mario Biggeri and Santosh Mehrotra (Biggeri and Mehrotra 2011) and Gottfried Schweiger and Gunter Graf  made suggestions about how to choose capabilities (Biggeri) and functionings (Graf and Schweiger) for children. These can function as starting points but a better and more comprehensive understanding of which functionings matter for global justice for children should take into account the considerations in this paper. This needs to happen so that we can move on from making overly broad assumptions about global justice for children to a refined and 'thicker' approach.
I would like to thank Nicolás Brando and the two anonymous reviewers for valuable comments.

Disclosure statement
No potential conflict of interest was reported by the author.