The General Comment on Children in Street Situations: Insights into the Institutionalisation of Children’s Rights

In: The International Journal of Children's Rights
View More View Less
  • 1 Geneva School of Social Sciences, Institute of Sociological Research; Centre for Children’s Rights Studies, University of Geneva, Switzerland,
Full Access

The drafting of the General Comment (gc) on Children in Street Situations (), whereby the sociological perspective () that informed this labelling becomes lost in translation, provides a convincing example of the ‘paradox of institutionalisation’ (). The vision of children’s “living rights” as the outcome of a structured process translating specific claims into an institutionalised set of norms () is thus specified. Analysis of the labels used for “street children” underlines the transformability of signification, domination and legitimisation in the theory of structuration ().


The drafting of the General Comment (gc) on Children in Street Situations (uncrc, 2016), whereby the sociological perspective (Lucchini, 1993, 1996, 2007; Stoecklin, 2000a, 2007) that informed this labelling becomes lost in translation, provides a convincing example of the ‘paradox of institutionalisation’ (Stammers, 2013). The vision of children’s “living rights” as the outcome of a structured process translating specific claims into an institutionalised set of norms (Hanson and Nieuwenhuys, 2013) is thus specified. Analysis of the labels used for “street children” underlines the transformability of signification, domination and legitimisation in the theory of structuration (Giddens, 1979, 1984).

1 Conceptual Outset

Where do you begin when you want to observe the institutionalisation of children’s rights? What should you look? at Where to start? There is actually no point from which you get the complete picture at once. As with any phenomenon, institutionalisation is an on-going process you simply observe when you are in it. You do not need to be officially recognised as being part of the process; there is no main entrance, but as many access doors as there are positions you may have, or take, in the discussion. The observer creates his own access to reality. I have chosen to study the institutionalisation of children’s rights by entering the door of the drafting of an important text about children in street situations.

The call of the un Committee on the Rights of the Child for the drafting of a General Comment on Children in Street Situations (un crc, 2016)1 is an opportunity to rethink central issues around children experiencing the street as a working, living and, sometimes, a survival space. On January 2016, the un Committee on the Rights of the Child started a consultation process and, consequently, issued an outline for the “General Comment on children in street situations”, based on the “Scoping Document for the General Comment and the Terms of Reference for the Consultant for drafting the General Comment, both agreed by the Committee in January/February 2015” (un crc, 2016).2 The outline comprises seven sections, namely:

  1. (1)Introduction (rationale for the present gc).
  2. (2)Objectives of the gc.
  3. (3)Being in public spaces (setting the scene; definitions and terminology; challenges; opportunities).
  4. (4)Key articles of the Convention in relation to children in street situations.
  5. (5)Broader context of the Convention (other relevant articles).
  6. (6)Developing rights-based, holistic, long-term strategies to prevent children developing strong street connections and to support children in street situations.
  7. (7)Resources for implementation (un crc Outline, 2016).3

Section 4 of the outline comprises two general principles and three articles which are seen as “key articles” for css. The two general principles are article 2.1 (“non-discrimination on the grounds of social origin, property, birth or other status; collective discrimination against children in street situations; individual discrimination against children in street situations”) and article 6 (“right to life; right to quality of life; right to survival and development”). One wonders why the other two general principles (art. 3 on the child’s best interests and article 12 on the right to be heard) are not also considered as key articles for css, but only as « other relevant articles » (they are mentioned in Section 5 of the gc’s outline).

The selection of some articles considered as “key articles” can be considered to be a reflection of an underlying view of css in which they are primarily considered as deprived of a family environment (art. 20), living below a minimum standard of living (art. 27) and deprived of the freedom of association and assembly (art. 15). The “other relevant articles” (Section 5 of the outline) seem to be considered secondary to these in the minds of the drafters of the gc.4 What does the selection of rights, foreseen as most important for css, tell us about the institutionalisation of children’s rights ? How is the agency of css evidenced and how is it accounted for in this process ? These are the two starting questions of the present paper.

It will highlight the central issue of the labels and definitions used in theories and practices. The alternative use of wordings such as “street-connected children”5 and css within the draft gc signals some semantic hesitations. The two expressions – css and scc – although relatively close, display differences in their focus. These very small details reveal how the institutionalisation of children’s rights occurs.

In order to analyse the drafting of the gc on css as a case of institutionalisation of children’s rights, I will use the theory of structuration (Giddens, 1979, 1984) as well as the more recent notion of “translations” (Hanson and Nieuwenhuys, 2013). I will start with the latter, and then come back to the Giddensian notion of “structure” to show what each concept highlights and the benefit of linking them together. I will show that Giddens offers a relevant framework to deal with major questions that are raised with the concept of “translations”. Conversely, there is an added-value in the concept of translation for the understanding of “signification”, “domination” and “legitimation” in Giddens’ theory. Hence, bridging the theory of structuration (Giddens, 1979, 1984) with the notion of translation (Hanson and Nieuwenhuys, 2013) is mutually beneficial. This leads me to a reassessment of the Giddensian approach, based on the central notion of “structure” and its “duality”. The case of the gc on css is especially illuminating in this regard. This empirical example supports the reflection about theoretical frameworks in the field of child participation (Stoecklin, 2013) and, consequently, that which is outlined here is what I can now call a theory of situated agency.

The paper begins with a brief presentation of the notion of translations and asks what is being highlighted when institutionalisation is seen as a matter of translation (Section 2). Giddens’ theory of structuration is then presented in a quite a detailed way (Section 3). This is necessary to come to a clear understanding of “structure” and “structuration” before these notions can be used to highlight the institutionalisation of children’s rights. The example of the gc on css, considered as an instantiation of the institutional order, can then be scrutinised under the combined lights of these two approaches (Section 4). The importance of the labels used for “street children” will be evidenced. I will show that “signification” (Giddens, 1979), and specifically the capacity to make sense of discursive categories, is a central issue in the process of institutionalisation of children’s rights; this capacity is not just a personal competence, it is bound to dominant modes of action. These have a filtering impact, not only on children’s voices, but also on those academic perspectives considered as incompatible with dominant accounts. Hence, structural dynamics explain the paradox of a gc named along a sociological perspective (Lucchini, 1993, 1996, 2007; Stoecklin, 2000a, 2007) getting lost in translation (Section 5).

2 Translations

The concept of “translations” stems from the field of children’s rights (Hanson and Nieuwenhuys, 2013). It depicts processes of bottom-up meaning-making and top-down implementation of normative claims and standards, and is defined as follows:

The concept of translations is about what happens with rights in the encounter of children’s and other actors’ perspectives, movements for social justice and the elites, authorities and opponents.

hanson and nieuwenhuys, 2013: 16
Questions that are raised with the concept of translations are quite relevant in general, and especially with regard to the gc on css :

Whose interpretations, and whose priorities of children’s rights, are being defended? How do children’s living rights coalesce with top-down international child rights implementation strategies? What are the trajectories of both approaches to children’s rights? Where and how do bottom-up and top-down interpretations meet, if they meet, and what are the consequences of such an encounter?

hanson and nieuwenhuys, 2013: 20–21
If institutionalisation is a matter of translation, then we must scrutinise how this notion contributes to former conceptions of institutionalisation. The application of the notion of translation to the gc on css will give us some hints.

2.1 Institutionalisation as a Matter of Translation

A very important feature of translations is their multidirectionality, fostering ‘a multiple-way process that transforms the power relations of all actors involved’ (op. cit.: 21). The multiplicity and multidirectionality of translations in the case of the gc on css can be approached through diagram 1.

Diagram 1
Diagram 1

Translations in the General Comment on children in street situations

Citation: The International Journal of Children's Rights 25, 3-4 (2017) ; 10.1163/15718182-02503014

The notion of configuration (Elias, 1991) is helpful here to map out the different kinds of actors (ngos, Universities, uncrc, children, parents) involved within various channels of communication, participation or representation (Stammers, 2013). Some channels are formal and others informal. In the case of the gc on css, the formal channels of communication are, inter alia, the call for the drafting of the gc issued by the un Committee for the Rights of the Child, the messages to and from ngos and academics, the General Days of discussion, expert consultations, etc. Informal channels are those set up on a random basis, mainly consultations of street children (and sometimes their parents or caretakers) and exchanges between ngos and academics. Diagram 1 differentiates between formal and informal channels of communication, participation or representation; hence, “representational power” (Stammers, 2013) can be visualised. It appears that children in street situations (and their parents or caretakers) are the only actors that are deprived of formal channels of communication, participation or representation. Other interactions may also be rather informal, or casual, based on benevolence rather than obligations, like the exchanges taking place between ngos and academics.

Diagram 1 serves as a basis to highlight some blind spots in the concept of translation. This is a configurational representation of actors linked to, and by, an object, here the gc on css. But the institutionalisation conveyed by the gc cannot be solely reduced to the kind of channels of communication that actors are using. We must come closer to what institutionalisation actually is by asking: institutionalisation of what? A simple and direct answer to this question could be: institutionalisation of measures most appropriate to css. However, a centration on the content being institutionalised, similarly to a centration on the channels of institutionalisation, does not give a complete picture of what institutionalisation actually is. It is necessary to examine the interplay of content and channels of institutionalisation.

A first and very broad approach to institutionalisation defines it as the relative stabilisation of prescriptions resulting from the complex set of positions and exchanges among different actors. However, in this process of embedding some conception within an organisation, a system or society at large, it is difficult to see clearly the distinctive role of “representational power” (Stammers, 2013) because the outcome of institutionalisation is closely linked to the very means used in the process. We end up with some sort of tautological thinking whereby the means of institutionalisation are considered the main, if not the whole, part of institutionalisation. The problem here lies in the fact that reducing institutionalisation to its processual components (formal and informal channels of communication) does not account for the “content” of institutions which is actually promoted. Outcomes are not reducible to processes. There is something more in outcomes, because they have consequences that were not foreseen by the actors involved in processes. It is therefore necessary to distinguish, on the one hand, the “power to institutionalise” as a process, and, on the other hand, the institutions as the manifestation or outcome of this power. And it is crucial to have a theory which explains the relationship between the power to institutionalise (the process) and the institutions (the outcome). Meanwhile, the concept of “translation” seems merely to depict the processual dynamics taking place among actors :

Translation of children’s rights into practice is never solely either a top-down or a bottom-up activity … but a circular process whereby source and target languages constantly engage with one another giving rise to unforeseen complexities that produce a state of constant indeterminacy.

hanson and nieuwenhuys, 2013: 19
The circularity, and alleged recursivity, of translations implies that the target population (children) may also, at some point, be heard as an inspirational source for the reframing of norms:

… translation must acknowledge difference and lend some recognition to children’s representations of their lives and situation. This provides children with room to negotiate meanings and influence practicioners’ interpretation of their rights that find their way, ultimately, into some kind of acknowledgement in the circles of legal experts.

hanson and nieuwenhuys, 2013: 20
But what exactly is this space for negotiation made of ? The authors refer to a rather undefined “third space”, or “betweenness”, suggested by others:

Sally Engel Merry (2006: 11) points at the double movement involved in translation, whereby international human rights are not only translated “down” into local systems, but local stories are also translated “up” into a common human rights idiom. Mark Goodale and Sally Engel Merry (2007) suggest that focusing on a third space, the “betweenness” in which the dialogue between individuals and groups on values and norms takes place, may help bypass the binary of the global and the local. It is in this in-between space, where the exchange of equally legitimate sets of values and norms take place, that new social practices emerge.

hanson and nieuwenhuys, 2013: 20
Gidden’s concept of structure certainly has something to do with the “betweenness” mentioned here. Actually, it provides a very convincing theoretical framework that enables one to come closer to a more comprehensive understanding of institutionalisation of children’s rights. In my view, the structuration theory casts more light on what is actually moving within the process of translation. Institutionalisation is neither just a procedural follow-up of children’s rights by the un Committee for the Rights of the Child (notably through gc 5 regarding general measures of implementation), nor just a dialogue (top-down and bottom-up) among individuals and groups. I consider that translation is part of institutionalisation, in the broader sense of the structuration theory (Giddens, 1979, 1984) and not, conversely, that institutionalisation is a part of translation, or that institutionalisation comes down to just another word for translation.

In my view, institutionalisation is more than what the concept of “translation” evokes, in a similar way, to many authors who use the concept of institution as referring to all political, administrative and legal practices that establish and institutionalise social relationships at any level whatsoever (Bonny, 2015: 34). In fact, the concept of translation, seen as practices establishing institutions, pushes the problem further: where do the evoked practices stem from ? The Durkheimian position, whereby the social explains the social, ending up with the tautological statement of social practices coming from social practices, presupposes an ontology of the social. This is not my position, and I will show that experience can come from experience as only humans have ontology and hence play the crucial mediating role in reproducing social practices. The latter do not reproduce themselves outside of human agency. I here intend to show that the notion of translation can only gain from a reassessment of the theory of structuration (Giddens, 1979, 1984) in the field of children’s rights, and that the notion of translation is an added value for this theory.

2.2 Differentiate Meanings and Their Manifestations

In order to answer the question: “Where do practices translated into institutions come from ?”, it is necessary to differentiate the development of meaning (we will see that Giddens calls this signification), which is at the heart of institutionalisation, from the organisations that instantiate its concrete manifestations. The daily and common-sense language speaks of institutions as organisations. As social scientists, we use the notion of institution in a different way, meaning the spirit that inspired specific social arrangements. It is in this sense that, for instance, we can consider private property as a major institution. Actually, private ownership is first of all an abstraction, a point of view, a categorisation, but it generates specific concrete organisations. As we know, the origin and the diffusion of the spirit of capitalism (its institutionalisation) are difficult to ascertain and the link with a specific ethos remains a controversial issue.6 This famous case is very interesting, as Weber has often been misunderstood: his thesis was not that protestantism was the origin of capitalism, but that the protestant ethic and the spirit of capitalism had elective affinities. Hence, there is no anteriority of one over the other but, rather, a reinforcing cycle among them.

2.2.1 The Linear Conception of Institutionalisation

This recursive aspect is very important, and I will come back to this crucial point regarding synergetic forces in the conclusion of this paper. For the moment, it can be underlined that contemporary perspectives very frequently fall into the same logic that made Weber misunderstood, namely a linear logic whereby causality is unidirectional:

In this perspective, institutions are first to be considered as meta-pragmatic elaborations (Boltanski, 2009) which are then translated into the pragmatic form of administrations, organisations, institutions, material assemblages, policies, devices, etc., but never reducible to these manifestations.7

bonny, 2015: 34
In this conception, there are two stages: first the (metapragmatic) elaboration, then its (pragmatic) translation into its various manifestations. This actually falls into the trap of the anteriorisation of thought with respect to action. This dissociation between a symbolic world and a material world is problematic in at least three ways, according to the Giddensian theory that I use here.

First, the “metapragmatic” level (Boltanski, 2009) depicts a symbolic world that would exist somewhere in itself, independently of actors. This conception reintroduces the dualism subject/object or individual/society that the Giddensian approach helped to overcome. The problem comes from the confusion, made by many, between “structure” and “system” (I will come back to this in Section 3). In Giddens’ approach, structure only exists as “structural properties” and hence does not have an existence of its own, independent of actors. Structure is defined as ‘rules and resources, organised as properties of social systems’ (Giddens, 1979: 66). Therefore any conception giving ontology to social systems or structures is problematic as it postulates that symbols stand alone, independently of human pragmatism, in a sort of cloud (that may be named “metapragmatic”).

Second, the manifestations are necessarily manifested by social actors differentially positioned in contexts made of various constraints and opportunities, hence it is unrealistic that they would just pick up elements in the metapragmatic level and translate them according to their own freely-elaborated fantasies. The Giddensian approach opposes this naive concept of actors by showing that the three dimensions of structure (signification, domination and legitimation) are interdependent. Hence, no meaning-making can be independent of power and sanction.

Third, the translation itself is a structured process, involving struggles and power relations and, as such, it is not explained by this definition of institutions which in fact tells more about the linear causal thinking of its conceptors than it explains about what institutionalisation actually is. This approach also presupposes that all social actors elaborate their thinking and understand the elaborations made by others before acting. They would be like little planners or architects, thinking or drawing in their minds the things they intend to do and then, at some point, realising these things. This mode of action is, of course, possible but it is by far not the only one and, in many contexts, not the most current one. It can be seen as the ‘instrumental-rational social action’ (Weber, 1978). However, this is an ideal-type, it does not exist in reality, as Weber insisted:

An ideal type is formed by the one-sided accentuation of one or more points of view and by the synthesis of a great many diffuse, discrete, more or less present and occasionally absent concrete individual phenomena, which are arranged according to those onesidedly emphasized viewpoints into a unified analytical construct.

shils and finch, 1997: 90
That is to say that no empirical case can correspond exactly to a pure ideal-type as the latter is a methodological device to highlight reality, and not an instantiation of reality.

Applied to the issue of institutionalisation of the gc on css, the ideal-type of “instrumental-rational social action” is useful to understand what would be the process if it would exactly correspond to this ideal-type: all actors involved in the elaboration of the General Comment would act most rationally with the aim to achieve most efficiently the best interests of children in street situations. This is actually not the case as it is impossible to identify clearly what are the best interests of children in a general way, which is precisely the focus of a “general comment”. In reality, the best interests of the child (art. 3, uncrc) is an undefined and undefinable concept, as it cannot have a specific expression but must be adapted from case to case: what fits a specific child in a given context does not necessarily fit another child in another context. Moreover, different actors have different agendas and their respective contributions to the gc bear traces of these often conflicting views. So we must consider that actors are hardly acting in ways where the symbolic thinking precedes its materialisation in unconstrained ways. Rather, they are referring to already-existing instantiations of others’ points of view, using the same or new material realisations and devices, in order to act upon them according to the different kinds (economical, social, cultural, symbolic) of capital (Bourdieu, 1992) they may have as a reflection of their various social positionings.

Consequently, a more dynamic, less linear, concept of institution is needed. We shall ask how the institutionalisation of children’s rights is structured. And the notion of translation will certainly be helpful. But it is necessary to assess the dimensions of “structure”, in Giddens’ theory, before we can more clearly see the added value of the concept of translation.

3 The Theory of Structuration (Anthony Giddens)

Giddens’ structuration theory offers some very inspirational insights to understand the institutionalisation of the gc on css. I will extensively rely on excerpts from Giddens’ writings, as is required for a good understanding of his theory. His most quoted work, The Constitution of Society (1984), is, of course, central. However, it synthesises some essential developments which sometimes makes it difficult to understand the theory. Most of these elements are presented, in a more detailed and thus accessible way, in his former publication, Central Problems in Social Theory (1979). This is why I will use also this former book to expose some very important issues that will then be used for the analysis of the gc on css as an instantiation of the institutionalisation of children’s rights.

3.1 Distinction of Structure and System

Giddens criticises the substantivist approach that considers institutions in their concrete manifestations, whereby the existence of political institutions would be confined to contexts with distinctive forms of State apparatus, and economic institutions would correspond to societies where the “economy” has become a distinct part (Giddens, 1984: 34). He proposes to avoid the confusion between the elaboration of institutions and their manifestation by concentrating on what he calls “structure”, defined as ‘Rules and resources, or sets of transformation relations, organized as properties of social systems’ (Giddens, 1984: 25). It is this central concept of “structure”, distinct from “system”, that gives Giddens’ theory of structuration its originality:

The concept of structuration I wish to develop depends upon making distinctions between structure and system (without questioning that these have to be closely connected); but it involves understanding each of the terms differently from the characteristic usages of both structuralism and functionalism.

giddens, 1979: 62
Hence, Giddens insists on the difference between system and structure, a difference that is not made by functionalist sociologies equating the structure to a ‘kind of “patterning” of social relations or social phenomena’ (Giddens, 1984: 16). Giddens criticises the approach of functionalists:

This is often naively conceived of in terms of visual imagery, akin to the skeleton or morphology of an organism or to the girders of a building. Such conceptions are closely connected to the dualism of subject and social object: “structure” here appears as “external” to human action, as a source of constraint on the free initiative of the independently constituted subject.

giddens, 1984: 16
He is less critical of structuralists and post-structuralists :

As conceptualized in structuralist and post-structuralist thought, on the other hand, the notion of structure is more interesting. Here it is characteristically thought of not as a patterning of presences but as an intersection of presence and absence; underlying codes have to be inferred from surface manifestations.

giddens, 1984: 16
Nevertheless, the problem with these two sociological traditions, in Giddens’ view, is the fact that, beyond their differences, neither differentiate between the concepts of “structure” and “system”. Hence, they treat the “structure” or “system” (the terms being equivalent) as an ontology with “girders” or “underlying codes”. In the Giddensian sociology, there is no ontology of the social: ‘According to the theory of structuration, social systems have no purposes, reasons or needs whatsoever; only human individuals do so’ (Giddens, 1979: 7). Therefore, the view of a structure (or system) that would be placed “over” the individuals and constraining them from “above”, or “under” them and making them stand on “underlying codes”, is misleading. Whereas structuralists are ambiguous, because they sometimes consider structure as ‘a matrix of admissible transformations within a set’ and sometimes as ‘rules of transformation governing the matrix’, Giddens clearly states that he speaks of ‘rules of transformation’ (Giddens, 1984: 17). This makes Giddens specify the difference between “structure” and “system”.

3.2 Structuration

This distinction between “system” and “structure” is reflected in diagram 2, which can be used to visualise Giddens’ whole process of structuration.

Diagram 2
Diagram 2

The Giddensian process of structuration8

Citation: The International Journal of Children's Rights 25, 3-4 (2017) ; 10.1163/15718182-02503014

The centrality of “structure”, separated from “system”, is of essential importance. Structures link actors and social systems. Social systems are ‘Reproduced relations between actors or collectivities, organised as regular social practices’ (Giddens, 1984: 25). Hence, “social systems” and “actors” are bound together by “structures”, which ‘can be identified as sets or matrices of rule-resource properties … recursively implicated in the reproduction of social systems’ (Giddens, 1979: 64). Hence, structure is both constraining and habilitating: this is why the theory is sometimes called a theory of “double structuration”, as social systems and actors are simultaneously structured by one another and structuring one another. As a consequence, structure has a “duality”: it not only goes in two directions (from actors to social systems and from social systems to actors) but it also has the peculiar characteristic of being simultaneously a medium and an outcome.

‘The recursivity of the model is obvious and central to Giddens’ theory of structuration, which refers to ‘Conditions governing the transmutation or continuity of structures, and therefore the reproduction of social systems’ (Giddens, 1984: 25). This recursivity is an essential feature of Giddens’ theory, as it justifies the rejection of ‘… the dualism of the individual and society, or subject and object’, as well as ‘… the dualism of conscious/unconscious modes of cognition. In place of each of these dualisms, as a single conceptual move, the theory of structuration substitutes the central notion of the duality of structure’ (Giddens, 1979: 4–5).9 Giddens argues that the “duality of structure” is precisely to be found in its recursiveness:

By the duality of structure, I mean the essential recursiveness of social life, as constituted in social practices: structure is both medium and outcome of the reproduction of practices. Structure enters simultaneously into the constitution of the agent and social practices, and “exists” in the generating moments of this constitution.

giddens, 1979: 5
We see here that Giddens’ answer to the question, ‘where do practices come from?’ is his notion of “structure”, and its dual character. By this duality he underlines the recursiveness of practices and the central role played by structure in their generation. Applying the Giddensian notion of “structure” to the present topic – the gc on css – will therefore highlight its centrality in the institutionalisation of children’s rights.

In order to make the somehow mysterious notion of duality understood in a more concrete way, and therefore show its relevance for the topic, I will use a metaphor. I think the metaphor of coin and hammer, and its limits (as any metaphor is limitative), helps to understand “structure” in a more concrete way. Imagine the times when coins were hammered, during the period lasting from the first millenium bc until the early modern period of ca. 15th–17th centuries. The coining was the process of manufacturing coins using a kind of stamping. Hammered coinage was made with a coin die (one for each side of the coin) containing an inverse image of the one that will be seen on the coin. The images on both sides of the coin were struck with hammers that would strike the coin, pressing the images into the blank metal disc (planchet). So we have a blank (a flat metal disk ready for stamping as a coin) and a flan or a die (a piece of metal shaped ready to form a coin but not yet stamped by the die), one for each side of the coin, and hammers that will press the flans on the coin. Hammered coins are the outcome of the process.

They can be metaphorically seen as social practices: they have some kind of stability and their value is recognised. What then would be structure (applied from Giddens’ theory) in this case? It is neither the coin nor the hammer beating the coin. It is the die, or flan, lying in-between the hammer and the coin. It is what constrains the coin to have the shape of the flan (its structure). And it is what habilitates the coin (social practices) to have some stability and therefore influence the social system. But who shapes the flan and who holds the hammer? There lies the limit of the metaphor: in social systems there is no individual in particular who has the power to shape structure and to imprint it on practices like a hammerer would do. Rather, the social shape and the social hammer have a plasticity, a malleability, that concrete dies and hammers do not have. Hence, the Giddensian notion of “structure” is a kind of movement producing coins (social practices) through ever-changing shapes and changing hammers (forces that imprint shapes on practices). This underlines the recursive dynamics of structuration.

With this metaphor of coin and hammer, coin being an individual production (or practice) and hammer being a means of production, we see that the plural form is needed: individual productions are not identical coins, and there is no immutable hammer. Rather, as we may notice, coins (social practices) and hammers (forces at play) share a common feature: they shape each other through a broader movement. Structuration is this movement. We can now see how and why structure has the duality of this movement that is simultaneously habilitating and constraining the productions (social practices) of actors. Eventually, only actors have intentions (coins and hammers have no will of their own). The structuration is the movement through which intentions turn into (partly unforeseen) realisations and through which actual realisations turn into (renewed) intentions.

Used as a metaphor for the theory of structuration, the flans or dies, put between hammers and coins, symbolise production in its dual character: they are simultaneously the medium and the outcome of production. We can now come back to our topic and see that any institutionalisation is a “production”. Institutionalisation is the imprinting of specific structures (flans or dies) on the social practices (coins) by specific forces (hammers). Individuals use these practices (as they exchange coins), they are not themselves the coins, although they identify themselves with these practices.

This metaphor of coins and hammers, and dies (or flans) in-between, helps understand why structure is dual. It also helps to understand that no individual in particular, as well as no social system, has, in itself, a driving force that would (like a hammerer) determine the whole production. The duality of structure is here essential. It can only be dual if it is both medium and outcome. If structure would only be a medium or only an outcome, it would be something that is just “linking” two other things together. It would just “link” (in an obscure way) individual and society. This would be a return to dualism (individuals/society) rejected in Giddens’ theory. In order for something (like a structure) to have the duality claimed here, it must not depend on elements (actors, systems) that would act on it from the “outside”, like external forces. Rather, is must be constitutive of these elements, that is, of both. But as social systems have no will of their own, and are in the end a production of the interactions of individual actors (who are the only subjects, as only they have consciousness and will), the only logical conclusion is that social systems are just instantiations of the “structure”, which is also shaped by the only “subjects” of the story, namely the actors.

Structure (and the forces shaping it) actually cannot have an existence on its own, which would be separate from actors and determining their existence. This fictional space is typically imagined as “somewhere above us”, actually ressembling to a replacement of God by a secular force called the “social system”. The status of “structure” is a “non-thing”, it has no ontology. Neither systems nor structures have an ontology, only individual actors do.

Now that the duality of structure has become somewhat clearer, we can consider that the institutionalisation of children’s rights is like a production, with a dual character. A production that, following Giddens’ theory of structuration, has the dual character of being both a medium for social practices and an outcome of them. It is simultaneously an outcome of former practices and a medium for future practices. In this, the time-space, or dynamic aspect, of structuration (claimed by Giddens) is fully acknowledged.

The institutionalisation of children’s rights can be seen as a production in the making. Rights are given a specific form, which corresponds to the translation (imprinting) of certain forces into institutionalised social practices. What are the forces at play in the translation of certain claims into an institutionalised set of norms (children’s rights)? We have to identify these forces. The three dimensions of structure identified by Giddens (signification, domination and legitimation), and their interplay, are useful for this identification.

3.3 Three Dimensions of Structure

Giddens’ distinction between system and structure is made by insisting on the non-ontological character of structure:

According to the theory of structuration, an understanding of social systems as situated in time-space can be effected by regarding structure as non-temporal and non-spatial, as a virtual order of differences produced and reproduced in social interaction as its medium and outcome.

giddens, 1979: 3
The “virtual order of differences”, which Giddens refers to, is made up of the combinations between three dimensions of “structures”, namely signification, domination and legitimation, and their modalities in daily interactions (see diagram 3).
Diagram 3
Diagram 3

Dimensions of the duality of structure

Citation: The International Journal of Children's Rights 25, 3-4 (2017) ; 10.1163/15718182-02503014

giddens, 1984: 29

These dimensions are tied together: ‘Structures of signification always have to be grasped in connection with domination and legitimation’ (Giddens, 1984: 31). This is why, as suggested above, a symbolic world detached from the materiality of power – distributed along different kinds of capital (Bourdieu, 1992) – would be nonsense.

The connections between signification-domination-legitimation underline their interdependence and the element placed in first position only indicates analytical focus, and not causal determination: ‘Thus when we focus upon those institutional forms through which signification is organised, we are concerned with the analysis of symbolic orders and modes of discourse’ (Giddens, 1979: 107). The latter are always ‘interwoven with forms of domination and legitimation’ (ibid.). Inversely, political and economic institutions are the institutional forms through which domination is organised, and the law and modes of sanction represent the institutional forms through which legitimation takes place. This framework is very useful as it shows that the gc on css cannot be elaborated in a lawful (rights-based) way without, at the same time, conveying political, economic and symbolic orders. Children’s rights are not simply a law issue; law cannot be asocial. The selection of some rights as more important than others for css – identified in the introduction of this paper – reflects political, economic and symbolic orders that are hiding behind the apparently neutral “rights-based” approach. It is in this way that the gc is actually enacting hidden translations.

With this understanding of institutionalisation as a structuration process in which signification, domination and legitimation are interwoven, it becomes more apparent that formalised channels of communication (see diagram 1) will involve much more structured relations of power than informal channels. Actually, formalisation can be seen as the stabilisation of power relations. Hence, the predominantly informal consultations of children in street situations is a direct reflection of their absence of power.

Throughout this endeavour, I consider that the label “street children” can be seen as the outcome of a structuration process whereby changes in signification are linked to domination and legitimation (Giddens, 1979, 1984) and, that, simultaneously, the concept of “translations” (Hanson and Nieuwenhuys, 2013) highlights the centrality of semantics in the interplay between signification, domination and legitimation. That is to say that translations are core in the central notion of “structure”. This is the combined framework I will now use in the next section to analyse the labels, especially the new ones (css and scc) found in the gc.

4 The General Comment on Children in Street Situations

As has been suggested, translation should not be equated with institutionalisation. Rather, translation is part of the “processual” mechanism leading to institutionalisation. The fuel of the differential mechanism is the variability of people’s interpretations. Translations are located in people’s interpretations and they render institutionalisation unforeseen and unpredictable, as ‘a circular process whereby source and target languages constantly engage with one another giving rise to unforeseen complexities that produce a state of constant indeterminacy’ (Hanson and Nieuwenhuys, 2013: 19). Therefore “translations” cannot replace the theory of structuration. The concept of translations ‘is about what happens’ (op. cit.: 16) more than about what makes it happen. The latter can be captured by the structuration theory. The two concepts give added value to each other because they are complementary.

I see their relation as follows: if ‘Signification refers to structural features of social systems, drawn upon and reproduced by actors in the form of interpretative schemes’ (Giddens, 1979: 98), then the actors’ interpretations and translations of reality become central in the activation of the “virtual order of differences” (Giddens, 1979: 3). Inversely, the interpretations and translations made by all social actors, as different as they may be, also have some similarities because of dominant frames constraining the practical consciousness and the discursive consciousness of actors. Words and labels used in daily interaction are instantiations of these structured translations. Empirical presence of these frames can therefore be identified in the labels used to speak of, for and/or with “street children”. A review of the labels used for defining “street children” is then the methodological device to understand fundamental aspects of the institutionalisation of children’s rights. This is what I undertake now.

4.1 Labeling “Street Children”

Labels have tremendous impact on the (dis)respect of the rights of children. The aspects and dimensions of street situations are highlighted differently depending on the labels that are used to speak of “street children”. The semantics of these labels and the underlying dimensions that are currently considered when speaking of, and acting for and with, street children, hence shed light on hidden assumptions and implicit theories of action that condition acknowledgement of children’s agency and its limits in the street setting.

An extensive literature review (Aptekar and Stoecklin, 2014) helps identify four major descriptions of “street children” (see table 1).


A first major distinction can be made between objective and subjective definitions. Definitions that can be classified as objective are the ones whereby the authors and the users claim to define “street children” by objective features, such as the length of stay in the street and the amount of family contact. This is the case for the dominant label used in the public space, “street children”: the criteria for a child to fall into this category are supposed to be easily assessed in an objective way. They are, namely: being separate from parents, without sufficient subsistence means, devoid of any protection, supervision or guidance by reliable adults, and surviving by living and/or working in the street. Children who share one or several of these features are supposed to be the “core” or “genuine” street children (Pare, 2003).

The “on/of” distinction made by Taçon also reserves the term “street children” to the ones whose family contacts are weak and who spend almost all their time in the street. The common feature of these definitions is that they make it relatively easy to put children, actually only very briefly seen on the street, into this category that speaks for itself: “street children”! … But this quick and easy labelling has tremendous and long-term implications: stigmatisation, discrimination and repression. The objective definition turns the child into an object: the child is symbolically associated with a space presented only as negative and problematic, therefore denying the child the status of subject and consequently reducing the child to an object on which interventions are carried out according to modalities disconnected from the child’s opinion.

The label, “street children”, erases the main problem, which is that these children are actually most exposed to multiple discriminations: being children, being poor and being deprived of basic rights. Their presence in the streets is a symptom of these discriminations, but it is turned into a cause of fear. It is the label itself which makes their presence in the streets a “cause” of a public problem and a “reason” publicly to intervene, inducing types of intervention that are not in line with a rights-based approach. The label “street children” confuses macrosocial causes with microsocial outcomes: the behaviours and experiences of children deprived of basic rights are seen as individual cases of a pathology and not anymore as the outcome of a collective construction of inequality.

The problem of these deprived children is distorted into a problem for the public. Hence, the situations of these children are largely ignored or silenced. Their multiple deprivations are minimised and public security is overemphasised. The latter justifies intervention, mainly repressive, with philanthropy as an alternative way towards a similar goal: pull children off the street. The perspectives of the children directly experiencing multiple deprivations are mostly “lost in translation”. The irony is that the children most exposed to the related inequalities are eventually also deprived from the possibility of voicing their own perspectives. This is the ultimate deprivation: silencing deprived children by labelling them “street children” and therefore reducing them to “the” problem.

Some alternative approaches give voice to these children and hence focus more on their subjective experience. These are the two later descriptions in table 1: “children in street situations” and “street-connected children”. Contrary to the objective definitions, these definitions do not lead to a priori categorisations of children as they require that children have to be listened to in order for practitioners to respond adequately. css and scc both emphasise children’s interactions and stress their subjective relationships to the street.

In the objective definition, “street child”, the child is associated with the object (street); he/she is not seen as having an opinion about it, and accordingly neither a status of subject. Those who have an opinion are in this case the ones pointing out the child as an object: the child is thus simply an object that is “of” or “on” the “street”. On the contrary, in the subjective definitions, the object is the street and the subject, the child who has a view about it. The child is thus granted a position of subject who, like any other actor, is legitimised to talk about the object “street”. It is an inversion of perspective, whereby the object is not anymore the “street child” but “street life”, allowing for children to become part of those who have an opinion on it and a right to be heard about it (art. 12, uncrc). In the next two sections, these subjective definitions are presented in detail, before stressing their similarities and differences.

4.2 Children in Street Situations

“Children in street situations” is the English translation of the French expression, “enfants en situation de rue”, coined by Stoecklin in his doctoral thesis manuscript (1998: 8), and its published version (Stoecklin, 2000a: 22).14 The label css casts light on the interactions that “out-of-place” children have with other social actors: ‘The emphasis should shift away from attempts to define street children towards analysis of their relationship with street environments’ (Ennew and Connolly, 1996: 131). Stoecklin added the plural form to highlight the diversity of “situations”, while drafting the sectoral policy on “Children in Street Situations” for the ngo Terre des hommes on the basis of an action-research (1998–2005) involving 12 countries.15 Since then, the expression has become widespread, well before the final strategy was eventually published (Terre des hommes, 2010), and it is today the one upheld by the un Committee for the Rights of the Child for the drafting of the General Comment on css (un crc, 2016).

Noticeably, a definition of css came rather late: “children for whom the street has become a primary consideration” (Stoecklin, 2008: 53).16 The reason for this timespan was reluctance to categorise children, as any definition actually ruins the endeavour of framing the issue in another way. The css perspective does not intend to define “children living in the streets”, but rather to point to the peculiarities experienced by ‘children for whom the street is the main living place, all the while pointing out the diversity of situations these children can experience’.17 It is also a label – all are labels – but this one disentangles children from street situations, stressing that the problem is not inherent to children but to the situations in which they have to make a living in the street. Therefore, it obliges one to address the situations, and not just to deal with the children experiencing them. The plural form not only emphasises the diversity of situations, it also takes account of the evolutive dynamics of the streets, as opposed to the static view of streets implicitly conveying the image of waves of uncontrolled children filling in dead-ends and submerging cities, as is especially portrayed by the Chinese words for “street children” – liulang ertong – (Stoecklin, 2000a; Aptekar and Stoecklin, 2014).

4.2.1 The “Child-Street System” (Riccardo Lucchini)

The perspective on css comes from an empirical approach, using the model “Child-Street System” (Lucchini, 2007),18 as a way to reconstruct the multidimensional relationships children have with their street environment (see diagram 4).

Diagram 4
Diagram 4

The “Child-Street System”

Citation: The International Journal of Children's Rights 25, 3-4 (2017) ; 10.1163/15718182-02503014

(Lucchini, 2007)
The “Child-Street System” is an analytical framework inductively constructed through empirical observations made in Rio de Janeiro (Lucchini, 1993), Montevideo and Mexico (Lucchini, 1996). It captures the diversity of relationships that different children have with the street space, depending on several dimensions making up the child’s subjective experience of street life. These dimensions make up the child’s subjective experience of street life:

The dimensions of the system are interlinked and a variation in each of them will influence all the others. This explains the diversity of the personal circumstances of children in a street situation even if the children share similar life histories. Amongst the factors that explain this diversity there are:

  1. (a)the modalities of the exit to the street;
  2. (b)the references and models of identification;
  3. (c)the symbolic and instrumental competencies;
  4. (d)the degree of insertion/participation of life on the street;
  5. (e)the movements between the different fields (street, family, parents, school, institution, etc.);
  6. (f)the necessities and motivations;
  7. (g)the modalities of the exit from the street;
  8. (h)the institutional experiences (ngo, police, courts, media, education system, work).
lucchini, 2007: 58–59
These elements make up the phases of what Lucchini calls the “street career” (Lucchini, 2007). This diachronic perspective on css, focusing on biographies, is opposed to the merely synchronic public opinion whereby children seen as hanging out on the street, at a point in time, are labelled “street children”. With the “Child-Street System”, on the contrary, it is the child’s own view that becomes central. Children who have street surroundings as their main environment integrate it as a major component of their life course, or “street career”. This dynamic aspect is of primary importance for intervention, as its adequacy depends on the identificiation of stages in a “street career” (Lucchini, 1996).

The “Child-Street System” has been extensively used by ngos in their intervention with css (see Section 5.1.). This tool accompanied and fostered the paradigmatic change caught in the wording itself: “children in street situations”. The spreading expression “children in street situations” indicates that changes are occurring in the scientific and programmatic conceptions of the phenomenon whereby children are seen in the overall context of their social and symbolic relations. A Weberian non-judgmental interpretative sociology of understanding has thus been favoured in this particular field. However, the review of the labels used around “street children” confirms that scientific evidence is lagging behind the institutional agenda. The structuration theory is quite useful to understand that, in itself, this situation is a proof of the dominance of economic institutions (see Section 6.1.2.)

4.3 Street-connected Children

A street-connected child ‘is understood as a child for whom the street is the central reference point – one which plays a significant role in his/her everyday life and identity’ (un ohchr, 2012: 10). Street connections draw attention ‘to choices children make in developing relationships on the street, alongside other connections they have with their families, neighbourhoods and schools’ (ibid.). This perspective emphasises relationships and hence the relative importance of the street compared to other spaces:

Most children have some connections with the street (for play, socialization, leisure and consumption) but are not reliant on public spaces for their development; they have stronger connections with family, school and peers in the community (ibid.).

It also recognises the identitary dimension of street life that can have long-lasting effects:

Taking a holistic approach that understands children as growing and developing within a series of inter-connected environments, the term “street connections” recognizes that the street may be a crucial point of reference for some children, even when they are not physically present there. Street connections can become vital to children’s everyday survival, their selection of coping strategies, and their identity development. (ibid.).

Compared to the css perspective, the scc label claims to put more emphasis on the subjective assessment and on the “bond” or the “connection” children have with the street:

The definition of street-connected children, whilst drawing upon much of the thinking that informed the children in street situations definition, adds a further dimension – some children may not be physically present in the street but still be influenced by their street life or career. For instance, a child may be in a welfare shelter but their personal and identity development affected by a number of years’ of experiences on the street. The street-connections definition recognises the role of the street and public space in children’s lives without constructing what that role is for them.

turgut, 2016: 3
There are several claims here that need to be discussed. First, the feeling of being street-connected, although not necessarily present in the street, is not really a “further dimension”, as the css perspective made clear in a similar point :

The child in a street situation is a child for whom the street has become the central point of his/her subjective world. This child can be de facto and objectively out of the street – in a home, prison, slum – but his/her thoughts are in the street, a “subjective world” shaped by interpersonal relations the child has entertained in the street, or in other places but connected to the street.19

ide, 2008
Secondly, the connectedness in the identitary aspect has also been stressed when the label css was discussed as holding for ‘children for whom the street has become a major consideration’ (ide, 2008).20 Finally, this approach does not specify, in place of the children themselves, the role that the street has for them. It sets forth an important question: is it possible, at all, to clearly circumscribe the influence of the street?

4.4 Similarities and Differences between css and scc

I will begin with the differences, as it is the best way to circumscribe the similarities. What differentiates css from scc is that the former focuses on situations whereas the latter focuses on the connections made by the child. Being “wider”, the css approach has a problem with definitions centred on children alone. The problem arises each time a definition of “these children” is asked. The scc definition is maybe trapped into the reduction imposed by the obligation to give a definition of “street children”. It is time to exit this definitional trap of the labelling process, and the door out is, in my view, a more comprehensive theory of “street situations”. In short: an approach, not a definition.

4.4.1 Careers and Profiles

The notion of “career” is a very good point to highlight what differentiate css and scc. Lucchini divides the street career into five phases :

  1. Progressive distancing: the child approaches the street in successive stages; s/he explores the urban space found between her/his house and the streets in the centre;
  2. The observed and the ludic street: the child still maintains a distance from the street;
  3. The alternating street: here the child takes on the world of the street with all its contradictions and almost always claims the status of “street child”. For her/him, the street is neither good nor bad, it is ambivalent;
  4. The rejected street: the child recognises that the street no longer meets her/his necessities;
  5. The exiting from the street, which includes many modalities (Lucchini, 2007: 59).

These stages of the street career are described as subjective connections to the street, that evolve according to the different dimensions of the “Child-Street System”: ‘Not all children pass through these stages in the same way. The stages are not something static, they are the result of a combination of many variables amongst the factors that we mentioned above’ (Lucchini, 2007: 59). Therefore, the connections to the street must be seen as a complex set of factors bound together in a systemic way. Yet, these factors are structured not only by the child’s own subjectivity but also by macrosocial variables (housing, economic and social policies, etc.). Consequently, the street as “the central reference point” of street-connected children cannot be seen only in subjective terms.

Profiles are actualised stages of the street career. Reifying these profiles is the error committed in a too strongly individualised child-centred approach. Profiles are only ideal-types (Weber, 1978) to which real identities can be compared. Identities are fluid combinations of predispositions and potentialities like (a) motivations, (b) roles and knowledge that allow individuals to interpret them, (c) norms and values, (d) cognitive competencies and (e) self-image (Hewitt, 1970). These components are related to possible behaviours. They are actualised when the child is able to identify a chance to realise some goals; at this point, the child creates an opportunity (Cusson, 1981). Identity is part of the child’s personal resources that are necessary for him to transform chances into opportunities. Hence opportunities, or capabilities (Sen, 1999), are always a combination of personal and social factors. This is consistent with Hewitt’s definition of identity. Consequently, the child’s capacity to give meaning to chances and to choose among a set of capabilities is also socially determined. Creating opportunities is not reducible to personal skills: it is linked to former transformations of chances into opportunities that have become institutionalised. We can see here that what we call the “social context” is in fact the oucome of agency; it only appears to be already there, but it was constructed.

Another difference is that css allows more fine-tuned typologies. In comparison, scc runs the risk of homogenising all children who, for quite different reasons actually, have the street as their central reference point. css requires attention to the meaning that all the actors involved, primarily the children but not only them, give to the situations experienced in the streets. Both definitions (css and scc) focus on the subjective perception of the street by the child, and are looking for children for whom the street is subjectively a prominent life setting. But css, in contrast to scc, also includes others’ points of view. We therefore see an important difference in the fact that while the label “Children in street situations” takes more account of children’s own subjective appraisals of street life, it is not solely based on children’s feelings. By comparison, the label, “street-connected children”, is much closer to a purely subjectivist view of street life.

4.4.2 Overweighing the Individualistic Perspective

The relevant call to curb the imbalance in power distribution within the institutional discourse (Becker, 1963) sometimes turns into a new imbalance in the opposite direction. There is a risk of overweighing the perspectives of “street children” in spite of taking into account all the different perspectives and their conflicting outcomes. Requirements to “listen to children”, “hear their voices”, has in some instances resulted in the tendency to reify them. The risk of essentialising the “core” or the “real street children” (Pare, 2003) is linked to the greater proximity of researchers and activists, mainly ngos, that resulted from at least two factors: the necessity for researchers to cooperate with ngos in order to gain access to their fieldwork, and the institutional setting whereby “street life” can be observed more openly in contexts where ngos are numerous and contacting children through outreach activities that are, if not encouraged, at least tolerated by the authorities.

Other factors also play an important role, such as the rights-based approach that speaks for outreach activities, research priorities in academia, access to public and private funding for action-research involving ngos and academia. This very brief evocation of the structural conditions of research conducted on “street children” is sufficient to show that “street situations” are made up of many factors that are actually influencing street life from the “outside”. The number and qualifications of street educators is just one example of how specific perspectives are built and circulate between actors outside and within the street. All these elements are indirectly constitutive of “street situations”.

Therefore, the label “street-connected children” contains the risk of focusing on children’s subjective meaning-making that would be analysed within a too narrow perspective on the “street”, not taking full account of the structural conditions that have built up this perspective itself. A broader view of “street life” requires a reflexive appraisal of research conditions. An equidistant positioning of the researcher in face of the different discourses on street life is the only way to be able to analyse the social construction of the phenomenon. And this positioning begins with the terms we are using to name the phenomenon. My claim is that “street situations” is an expression that favours the required equidistance as it clearly fosters a debate on what a “situation” is actually made of, and identification of the actors who are building up specific situations. It is the undetermined and undefined term “situation” itself that calls for a structural analysis. The latter is not so much implied with the label “street-connected children”, which reduces the situation mainly to the way it is perceived by children who feel connected to the street. Inclusion of other perspectives (authorities, media, ngos, academics, etc.) is thus mediated by children’s own subjective perspectives, which only account for specific perceptions of other actors and does not fully cover their own perspectives. In this way, children’s voices become a filter through which a researcher reconstructs “street life” in a rather partial way.

As we know, in any scientific approach, the interpretive “filter” should be an explicit theory which research uses to understand the actors and transcend their points of view. When no explicit theory is used, the researcher runs the risk of sympathetically being trapped into the subjects’ own filters. The distinction between the researcher’s theory playing the role of analytical filter (emic perspective) and the subjective points of view of the people observed (etic perspective) is a guarantee of scientific validity. This distinction is not clearly implied in the label “street connected-children”, with its concentration on children’s perspectives, by which the analysis of the researcher might be absorbed into the etic perspective. Any approach restricting understanding of street life to the subjective perspectives of “street children” is therefore questionable.

The css label is different from all the others, as it is not centred on characterising the children, but on the interdependence between actors and settings, hence allowing a socio-centric instead of an individualistic view. Seen against the background of the “child-centred” discourse, it is understandable that this reframing of the “street children issue” leaves some traces of an egocentric perspective. Hence, the contradiction between the css label and its definition centred on the child’s subjectivity (Stoecklin, 2008). This points to the fact that, as a radical move against common stereotypes, the views of the individual child have become the new point of Archimedes. The problem is that this individualistic approach hides major aspects of street situations, because it does not respond to questions around the structuration process of individual actions.

4.4.3 Contextualisation

Profiles can be identified, like the assertive gamines and the submissive chupagruesos in Colombia (Aptekar, 1988), the beggar and the vagrant in China (Stoecklin, 2000a), or the six profiles in Chittagong, Bangladesh: the hero, the hardworker, the ambivalent, the survivor, the isolated and the dependent abused (Stoecklin, 2000b). But there is still an insufficient explanation of the structuration of these profiles and of the reasons why we find various numbers of profiles in different contexts. A better understanding of this would certainly allow better intervention strategies, as they may address micro and macro dynamics. We know much about numbers in specific contexts. Take, just for one example, the gender distribution of css in Africa: 91 per cent of street children are male in Kenya (Onyango, Suda and Orwa, 1991), over 90 per cent in India (Patel, 1990; Verma, 1999), 95 per cent in Zimbabwe (Muchini & Nyandiya-Bundt, 1991), 84 per cent in Angola (Moberly, 1999), 76 per cent in Ethiopia (Veale et al., 1993), and 70 per cent in Zambia (Mambwe, 1997).

But the link between these statistical data and the qualitative data of individual profiles is still lacking. What is the relationship between gender distribution and profile distribution? What is the correlation between the gender imbalance and the number of profiles? These are the kind of still unanswered, but crucial, questions to address in a global strategy for, and with, css. Without answers to such questions, the gender imbalance is only explained by cultural variables such as abuse and neglect:

The gender ratio is particularly salient because across cultures girls are more likely to be abandoned and abused than boys (Kabeberi, 1990; Korbin, 1981). Thus if being a street child is the result of neglect or abuse, we would expect a high proportion of girls to be on the streets, which in fact explains the differences in the ratios among developed and developing countries. In the developing world, abuse is a less common factor to becoming homeless, thus the high percentage of boys, while in the developed world abuse is the main factor of homelessness, thus a much higher proportion of girls (Schaffner, 1999), between a third to a half.

cauce et al., 2000; aptekar & stoecklin, 2014: 36
This may in fact be only a part of reality, based on “push factors”, whereas the weakness of “pull factors” for girls may be linked to public policies restricting their street life to only one profile. For instance, all the girls observed in Shanghai were beggars, no girl was found among the vagrants (Stoecklin, 2000a). This is linked to the one-child policy and the preference for boys, condemning unplanned girls to the sole profile accessible to them. This example clearly shows why a different framing is needed: not only statistical data, not only individual profiles, but an intersectional approach. A different framing is reached when individual profiles (typologies of street children) are replaced in the contextual constructions of the social order (typologies of street situations). This involves a theory of action that is able to account for the structuration of street situations.

4.4.4 Towards a Typology of Street Situations

The subjectivist definitions given for scc (“Children for whom the street is the central reference point”) and css (“Children for whom the street has become a primary consideration”) only constitute a starting point for a perspective that should illuminate the structural conditioning of the child’s subjectivity. Profile identification of scc or css is only the first step towards a rights-based strategy and the whole picture is best captured by the css perspective.

A situation does not exist in itself: it is constructed, along with the perceptions and interpretations of actors participating in direct and indirect ways, into a set of interactions which, at some point, acquires the status of a “situation” in people’s minds. In the same way, street situations are constructs made of the perspective of many actors, amongst whom are children currently labelled “street children”. This is why one should take into account the different perspectives in order to understand the specific views of children in street situations, as it is only when articulated with other perspectives that their own views and corresponding behaviour can be fully understood. The emphasis on the subjective view of “street-connected children” gives only a partial view of street situations.

Street profiles are typical ways of behaving, and css may transit from one profile to another along their street career. Street profiles are therefore closer to structural aspects than to purely individual characteristics. Thus, profiles can be seen as actualisations of the different phases of the street career. According to Lucchini, the street career is –

a central element that defines the place that the child occupies on the street. This place is almost always different from one child to the next because of the stage in which s/he is found as well as because of the stages s/he has already passed through.

lucchini, 2007: 59
Children pass through these stages according to their evolving identities. Lucchini (2007) and Stoecklin (2007) came to the conclusion that instead of a typology of “street children”, it would be better to elaborate a typology of “street situations”, because reducing children to typical profiles comes down to putting them into categories, which is still a top-down approach. A real step ahead for the respect of children’s rights should go in the opposite direction: rather than further individualising children and intervention schemes, which soon reaches its limits in terms of financial and human resources, a more general typology of street contexts should be elaborated. This can be done by linking the “street profiles” (or phases of the street career) with “street situations” whereby all stakeholders and institutions are considered and not only the children and their families. As long as the framing stays the same, intervention will remain top-down. A real participative strategy requires a change of frame. Researchers are therefore invited to propose new framings, and especially theories of action.

The shared feature between the two approaches – css and scc – is the reliance on the child’s subjective sense of belonging. A major difference, meanwhile, is that css is not relying just on the child’s own feelings; it further situates the child’s voiced experience within a set of dimensions (re)constructed by the sociologist, as in the the “Child-Street System” (diagram 4). To which extent this reconstruction relies on the children’s own words while being organised into the sociologist’s framework of understanding is a disputable matter, and indeed a central problem in grounded theory (Strauss and Corbin, 1990). At some point, the inductive approach should be confronted with a deductive application of a theory that is anterior to the observations made, and through this confrontation some new theoretical developments could occur. This is precisely what I propose to do in the next section.

5 Lost in Translation

The phenomenological perspective of css can be used to understand the different ways of experiencing street life in a particular city. It helps to identify major profiles, based on case histories. The constructionist perspective sees the actor as making choices and adopting strategies related to personal and social elements. This paradigmatic change replaces the socially constructed and essentialised “street child” by a consideration of several types of relationships to the street world that any child, as a social actor, may experience at some stage. The focus is on relationships, using the perspective of symbolic interactionism: the street is invested with differentiated activities and meanings, depending both on the children and on the contexts.

Even if “children in street situations” can be considered a broader approach, hence more comprehensive than “street connected children”, they are nevertheless both labels and as such they cannot fully convey the underlying complex set of issues that contributed to their formulation. Reassessing these issues sheds light on the discrepancy between the current use of the labels and the constructionist perspective that, as I argue, got almost entirely lost in translation. This helps highlight hidden assumptions and implicit theories of action that impede full acknowledgement of children’s agency and its limits in the street setting.

The multiple translations have several layers. These layers are not just to be seen as “stages” which children’s voices must undergo to reach formal approval and mention within the gc on css (see diagram 1). Layers are also, and foremost, the processes of meaning-making that are not exactly identical when “voices” are passing through these different levels. That is to say that translations are not the same: for instance, translations from children to academics do not have the same features as translations from academics to policy. This is because the actors involved in the respective interactions do not share similar frameworks. Hence it is necessary, when observing the relations between actors (ngos, academics for instance), to be aware of the semantics. The relations between actors are also relations between words.

5.1 Relations between ngos and Academics

The implementation of the “Child-Street System”, an academic perspective applied to intervention, will here serve as a good example to highlight the relations between ngos and academics. It is, of course, not the only possible example, and one should avoid generalisations based on a single case. However, this case contributes to an understanding of institutionalisation.

I first applied the analytical framework called “Child-Street System” (Lucchini, 1993) during my 14-months fieldwork (in the years 1993–1996) in the streets of Shanghai (Stoecklin, 2000a). I then adapted it as a tool for intervention or consultation with various ingos, local associations and networks (1998–2007),21 with the support of its conceptor, Riccardo Lucchini. During this decade, the “Child-Street System” was mainly used for bottom-up identification of the needs and capacities of children living in the streets and/or victims of trafficking. Comparisons among the different applications of the tool in several countries allowed identification of major achievements, and also limitations, in translations of children’s living rights as linked to processes that touch upon different levels of project-cycle management. There were profile-wise recommendations, as well as transversal recommendations for all profiles of children.

The purpose of the application of the “Child-Street System” within the intervention scheme of several ngos was to obtain a better image of children’s experiences of street life and consequently improve the ability to provide an appropriate response to their situations. In practice, their points of view could be better included into a rehabilitation work fostering child participation. By helping in the reconstitution of the child’s history, the “Child-Street System” made intervention more personalised. It is therefore a tool for the respect of children’s rights, as it helps in getting closer to the child’s subjective impressions expressed in his/her views (art. 12, uncrc) and measuring them in accordance with the child’s best interests (art. 3, uncrc). Lucchini’s “Child-Street System” proved useful to identify specific profiles of children in street situations. However, there were also obstacles, like the limited resources for staff training and the corresponding difficulty in dealing with a tool that requires time for cross-checkings. A brief account is made here about the use of the “Child-Street System” in the different phases of a project-cycle.

5.1.1 In the Planning Phase

The tool helped in collecting information on the child and its environment, in order to define the profiles and to work out a strategy adapted to each profile. The tool has proven to be heuristic in overcoming the reductionism contained in the dichotomy between “children of the street” and “children on the street” (Taçon, 1985). The change of denomination (“children in street situations”) entailed a paradigmatic change: these children are seen as persons and subjects of rights, not as problems or “objects” of intervention. Their abilities and competencies can therefore be considered as valid resources for participation at any stage of the project cycle, from situation analysis to capitalisation. However, it remains a challenge to have strong involvement of children in street situations in the different stages of programmes that are designed to help them.

5.1.2 In the Implementation Phase

The tool served for monitoring and collecting further information on significant changes in the child’s life, which is especially important for referrals. One must understand the child’s motivation and respond to it in a timely way and with the appropriate activities, matched to the child’s abilities and expectations. It is necessary to integrate the skills which the child has developed, even in the street, in a positive way by transposing them constructively into the project whenever this is possible. It was striking to see how rapidly social workers in such diverse contexts became familiar with the tool.

However, some difficulties remained with the cultural translation of the concepts and the systemic logic behind the model. Generally speaking, further capacity-building would be required to narrow the gap between recommendations drawn from action-research and institutional capacities of ngos to develop a more bottom-up line of intervention. In the different settings, the knowledge gained with this tool has had some practical impact on the adaptation of the project’s components. The tool made it possible to translate specific know-how into concrete action adapted to the various contexts. It does not lead to the building of a “one-fits-all” model of action, but takes the diversity of situations and children’s profiles into account for targeted interventions.

Nevertheless, intervention did not always follow the profile-wise recommendations as it was difficult to adapt the structure to the profiles. This raises the question of the contradiction existing between recommendations made from a qualitative situation analysis, and structural constraints which are at times very hard to adapt, not only to the real needs and competences of children, but moreover to a real rights-based perspective.

5.1.3 In the Evaluation Phase

Eventually, the css tool served for impact assessment, through analysis of the overall level of empowerment of the children (acceptance by society, changes in the environment induced by the project). Qualitative indicators to assess the level of empowerment are still needed, but the elaboration of qualitative indicators depends on a theory of action, and this was not built into the “Child-Street System”.

Actually, evaluations never went so far as to officially include a critical view of the relationship between ngos and academics engaged in action-research. This in itself is also an indicator of processes that impinge on a truly transparent evaluation of the whole intervention strategy. Hints of this can be found in formal and informal exchanges between the staff, but these details cannot be accounted for within the scope of this paper. They have just to be mentioned for awareness of the inevitable tensions surrounding any research-action, and to situate more interestingly some major issues in the process of institutionalisation in the following ex-post assessment that I now make.

5.1.4 Ex-post Assessment

The academic field is only very partially conveyed, according to the internal variations or restructurations of ngos. The example of Terre des hommes is revealing. The label “Children in Street Situations” (css) has circulated within the ngo and its partners, mainly according to structural changes within the organisation. The css label was most visible across internal and public documents when the position of Resource-Person was designed, during the period 1998–2005, to “capitalise” on projects and to draft a Sectorial Strategy under the same label (css). The Resource-Person argued for the css label, according to considerations derived from his PhD and periodic consultations with Riccardo Lucchini, considered a pre-eminent researcher in the field of “street children”, and conceiver of the “Child-Street System” (see diagram 4).

This framework has become a methodological device within a dozen css projects led by Terre des hommes and partners. However, both the css label and framework slightly faded away a few years after the Resource-Person left the organisation, leaving just hints of traces nowadays. Some new slogans appeared, even ones that are far from the css perspective, like for instance, “Invest in a street child”.22 Even though a sectorial strategy on css was adopted in 2010, the “css projets” are now included into the more general strategy coined “child protection”. This move from situations (to be first of all understood) to protection (to be first of all guaranteed) is a clear indicator that policy dictates (mainly through principles around fund allocations) and sets the agenda for intervention (beginning with labelling and categorising children).

There is no clear-cut theory of action or development playing a greater role. This example shows how institutional incentives and limits are conducive to changes in perspectives and labellings of “street children”, including step-backs. This sheds light on hidden translations that make definitions change. The changes of definitions regarding “street children” tell us more about those who use the labels than about those labelled. There are changes across different organisations, adopting and rejecting each others’ labels and definitions. But there are also changes within organisations, as in the case of Terre des hommes and other organisations like, for instance, unicef. In 1985, unicef adopted Peter Taçon’s distinction between children on the street and children of the street, and a year later, despite the large audience and adoption of this on/of division, unicef came back to an indistinct use of “street children”, when it included them into the very large category of “children in especially difficult circumstances” (cedc).23 There is a back and forth movement, where distinctions are made within a category (street children become on/of the street) which are then followed by groupings of former categories into an even more general category (“street children” under the cedc). This seems to be relatively independent of new scientific evidence that would speak for alternative groupings and/or labels. Rather, both levels of variations (across and within organisations) are indicative of a positioning strategy on the intervention market in which children themselves are hardly heard (Poretti et al., 2014).

5.2 Relations between ngos, Academics and the Committee for the Rights of the Child

The previous section illustrates the relations between ngos and academics and more generally it can also help to come to a closer understanding of how power is exerted concretely among ngos. The recognition of their contributions by the un Committee for the Rights of the Child is part of the game. It gives legitimacy to ngos on the intervention market. The example of Terre des hommes indicates that, in the absence of a clear-cut theory of action, it is the field of intervention itself (the Committee being part of it) that sets the agenda for intervention, with academics playing only a marginal role. Changes in labels therefore also include step-back movements. A similar “back-to-protection” movement seems to be filtering through the call of the un Committee’s gc on css. The wording css is officially used, but this does not yet guarantee that the profound implications of this approach will be given due weight.

As a matter of fact, the interplay of organisational actors presides over cross-topical variations. It can be noticed, in this regard, that the “street child” has been replaced by the “child victim of violence” in the iconography of victimhood during the past two decades (Poretti et al., 2014). But also inner-topical changes are made through the evolution of inter-agency relations that impinge on the terminology and corresponding approaches. These changes have therefore a strong political dimension. The scope of this paper does not allow for an extensive description of the relations among ngos and with the un Committee for the Rights of the Child. I have just briefly mentioned (see diagram 1) the formal and informal channels of communication, participation and representation, and underlined that these are only a configurational representation of power relationships. It is now time to scrutinise these relationships, and this requires that we first understand two important notions, the “paradox of institutionalisation” and the “problematic of representational power” (Stammers, 2013), and then their relation.

5.2.1 The “Paradox of Institutionalisation”

The notion of “paradox of institutionalisation” deals with the ‘extent to which emancipatory struggles can be sustained through processes of institutionalisation’ (Stammers, 2013: 282–283). This notion can be mobilised to stress that global recommendations, as institutionalised in the present gc on css, can hardly take account of the complexity and heterogeneity of street situations which, to a large extent, get “lost in translation”. Hence, the discrepancy between the labels used in the General Comment – css and scc – and the underlying protectionist view. The “paradox of institutionalisation” is a matter of conflictual influence: any struggle, sooner or later, faces the necessity to compromise, to some extent, with other views involved in the process. The question therefore is not whether or not struggles are mitigated in the processes of institutionalisation, but rather how much, or to which extent, they have to compromise.

5.2.2 The “problematic of representational power”

In order to respond to this question, another useful notion, also suggested by Stammers (2013) is the “problematic of representational power” (Baxi, 2000). Stammers points to this issue of ‘democratically constructed channels of communication, participation or representation between ngos and the people they claim to speak for’ (Stammers, 2013: 283). He points to the fact that the lack of such channels, ‘poses extremely serious problems, which is what Baxi’s notion of the problematic of representational power tries to capture’ (ibid.). Stammers illustrates this with the issue of child participation, collocating the paper from White and Choudhury (2007), about Amra, a children’s organisation of and for street and working children in Bangladesh. These authors suggest that, ‘… as the emphasis on child participation grew in the development community, ngos involved with children’s rights felt increasingly pressurised to “produce” children’s participation’ (Stammers, 2013: 284). Hence, according to them, the ambiguity of children’s participation fostered by adults is that it is simultaneously challenging and reproducing power: ‘development agency staff commonly claim the right to determine both what counts as “participation” (and what kinds of participation count) and how it is represented’ (White and Choudhury, 2007: 536).

5.2.3 Relation between the Paradox of Institutionalisation and the Power of Representational Power

The “representational power” of organisations that are ‘speaking on behalf of the oppressed and those whose rights are being violated or threatened’ (Stammers, 2013: 283) is closely linked to the “paradox of institutionalisation”:

If the key strategic question identified by the paradox is the extent to which emancipatory struggles can be sustained through processes of institutionalisation, then a fundamental question arises of how various aspects of social relations might be reconstructed and democratised.

stammers, 2013: 282–283
The filtering effect of institutionalisation can be seen precisely in the choices and translations that are made within the drafting of the gc on css. Hence, we should consider the gc not only as the instrument it claims to be, namely guidelines that serve as recommendations to State parties in their relations to css, but also as an instance of representational power, whereby certain perspectives are prioritised in the international development agenda, while others fail to be recognised and still others are not even given visibility.

The “paradox of institutionalisation” is rather acute in the case of the gc on css: the emancipatory struggle conveyed by the empirical and theoretical perspective on css (Lucchini, 2007, Stoecklin, 2007, Aptekar and Stoecklin, 2014) is not even mentioned in the outline of the gc taking on the label css. This demonstrates a particular case of “representational power”, whereby critical academic voices may be put aside even when their works have been foundational for the official wording. This is a particular case of the more general process of institutionalisation, which can be highlighted with Giddens’ theory of structuration. As the elements of the theory have been presented in Section 3, it will be easier now for the reader to understand its application to the specific case discussed here, and hence to see its relevance.

5.3 Obvious and Hidden Translations

Within the gc on css, there is a risk that the css perspective gets “lost in translations”. How is this paradox possible ? This is what I intend to discuss in the present section. I will begin with the concept of “translations” and with a differentiation between obvious and hidden translations.

I use the concept “obvious translations” to designate processes whereby some claims are translated (or not) into an institutionalised frame (here the gc on css). In our case, obvious translations can be seen with the selection of specific articles of the uncrc that are considered to be more relevant than others to address the issue of css.

I use “hidden translations” to speak of the processes where the specific features or aspects that are prioritised form an underlying discourse on the issue itself. This is so because the selected aspects are in a way connected, their interplay produces a kind of hidden discourse which becomes a “regime of truth” (Foucault, 2012). Therefore, “hidden translations” are captured in the assumptions, or the underlying discourse, hiding behind the interplay of the elements that have been selected and given priority.

I will now highlight the obvious and hidden translations that can be found in the process of the drafting of the gc on css. Is there a true commitment to css or is it only a label, taken at face value? It is too early to tell. This will be possible only when the gc on css is adopted. But the drafting process itself is a very important moment to observe: how the call is formulated and what are the articles of the uncrc that are prioritised? All this tells a lot about the structuration process of the gc and hence about institutionalisation of children’s rights.

5.3.1 Obvious Translations

I have already mentioned (in the introduction of this paper) the selection of articles (15, 20 and 27) of the uncrc that are considered key for the gc on css. I will now show that this is an obvious translation of a “back-to-protection” movement. The crc call for submissions on its General Comment on Children in Street Situations states that:

Submissions should address one or more of the following questions:

  1. 1.The right to freedom of association and assembly (crc article 15)
    1. a.What are the realities of association and assembly for street-connected children and how are their rights under article 15 violated?
    2. b.What are good practices/solutions in realising street-connected children’s rights under article 15?
    3. c.What are some concrete examples of legitimate and illegitimate restrictions on street-connected children’s rights under article 15?
  2. 2.The right to special protection and assistance for children deprived of a family environment (article 20)
    1. a.How can States guarantee children’s right to special care and protection by offering them better choices so they are not forced to depend on their street connections for their survival and/or development? (including good practices/solutions which address the care and protection needs of street-connected children).
  3. 3.The right to an adequate standard of living (article 27)
    1. a.What are the most effective measures that States Parties can take to implement the right to an adequate standard of living for children in street situations or at risk of being so, including assistance to parents and others responsible for the child?
  4. 4.Developing rights-based, holistic, long-term strategies to prevent children developing strong street connections and to support children in street situations
    1. a.What should be included in rights-based, holistic, long term strategies to prevent street children developing strong street connections and to support children in street situations?
    2. b.What are criteria and/or indicators for an intervention to be considered a “good practice” in relation to preventing children developing strong street connections and/or supporting children in street situations?
    3. c.What are the key recommendations the Committee should be making to States in this regard?24
un crc, 2016
The contributions to be made on the three articles considered key articles for css, are recalled along with questions framing the issues:

Article 15, uncrc shall be highlighted with an overview, good practices and legitimate restrictions on this right. Some other framing would have been possible, for instance, the distinction between the physical space to associate and assemble and the civil or political space where the association and assembly right would be accounted for and translated into formal claims. The restrictions seen as legitimate are based on an out-dated vision of assemblies: their location is supposed to be on the streets, and their limitation would be legitimate if they pose a threat to public order. But what about the social networks, the digital assemblies, and how they might become a political force? Rounding up “street children” is more difficult if they meet on the net … We see here that new forms of assembly should be considered for their potential for enhanced child participation. This requires a necessary debate about “association” (what goals, what places) and the new frontiers reached through the digital world. In a way, street assemblies are old-fashioned, and those who do not have access to digital assemblies will actually endure even more repression than before. Hence, there is here an implied aspect of domestication: good agency and good assembly, so it would be said, goes in the direction of domesticated use (digital tools, social networks) that are necessarily softening the claims and creating an imbalance between those who have the necessary social, cultural and economical forms of capital (Bourdieu, 1992) to use these tools and those who are deprived of it.

Article 20, uncrc is here interpreted in a way that makes an opposition between street and protection, with the family unmentioned and thus implied as being among the “better choices”, whereas it is widely known that domestic violence is one of the major factors expelling children to the streets. Domestication is also implied in the specific perspective based on the opposition of public and private space. The obligations of State Parties towards children without primary or proxy caregivers, the types of special care and assistance (aternative care, foster care, adoption, shelters), which is typically a public management of private spaces, could also be problematised. The symmetrisation of children, parents, and institutions that is observed in developed countries (Payet and Purenne, 2015) is certainly different from what happens in developing countries where access to rights still depends heavily on economic capital. Hence, the “better choices” should also be linked to article 4 uncrc (all measures).

Article 27, uncrc is presented without suggesting the implications of “an adequate standard of living” in terms of policies to prevent deprivation, inequalities and measure of redress and redistribution, which are necessarily linked to art. 4, uncrc (all measures). The rise of individual rights tends to heighten the competition between parents and children who are engaged in empowerment-processes by the authorities in charge of child protection (Becquemin and Robin, 2015). However, this is a reality in France, and probably in other developed countries. In the countries where the majority of css live, deprivation is much more linked to collective discriminations. Hence, the individualisation of rights, which can be a neoliberal perspective on the uncrc, may further jeopardise the claims for collective rights (Liebel, 2010). Hence, parental responsibility for securing ‘the conditions of living necessary for the child’s development’ (art. 27, uncrc), with State Parties having the subsidiary role of assisting parents, may actually be interpreted in a neoliberal way. And in many instances it is; for instance, the “Bolsa Escola” programme in Brasil, a conditional cash transfer to help mothers in poor households sending their children to school on a regular basis (de Janvry et al., 2006).

This critical appraisal of the framing given to the three key articles (15, 20 and 27) in the draft of the gc on css makes the quasi-absence of article 4, uncrc quite apparent. The fact that article 4, uncrc is only seen as secondary is intriguing, as the street situations are heavily linked to economic, social and cultural rights. Hence, the second paragraph of art. 4, uncrc on legislative and administrative measures for the implementation of rights should be especially underlined :

With regard to economic, social and cultural rights, State Parties shall undertake such measures to the maximum extent of their available resources and, where needed, within the framework of international co-operation.

Art. 4, uncrc
In place of a strong incentive to State Parties, most of whom have a growing imbalance in income, to use ‘to the maximum extent (of) their available resources’ to protect the rights of the child, the framing of articles of the uncrc that are seen as key articles is actually conducive to recommendations that will mainly focus on redirecting children in street situations to mainstream society. It looks like the explanation for “strong street connections” would be the poor implementation of the three “key articles”: a low standard of living (art. 27), a lack of alternative care (art. 20) and a lack of assembly in proper places (art. 15). Inversely, preventing strong street connections equates with developing a strategy that assembles these children in proper places, if possible in alternative care and, even better, by reintegrating them in the family. This sounds like the same old story of removal from the street in the protectionist approach.

Moreover, a fourth aspect is given a specific importance, namely Section 6 of the outline: ‘Developing rights-based, holistic, long term strategies to prevent children developing strong street connections and to support children in street situations’ (un crc, 2016). This formulation includes the underlying assumption that “strong street connections” are detrimental to css. Research does not support this view, as it shows that connections in the street are very diverse and this heterogeneity cannot be subsumed into something that would only be negative and contrary to the child’s best interests.

The claims to ‘support children in street situations’ and at the same time preventing them from ‘developing strong street connections’ are paradoxical. They might be logical only on a purely idealistic level, whereby removal from the street is associated with cutting negative ties and supporting positive ties in new settings. But this idealistic view is not confirmed empirically, as the experience of street life is not purely a negative one; street connections are subjective ties that are inscribed in a child’s experience with lasting effects, and they are both barriers and resources for the child’s integration into other settings (Aptekar and Stoecklin, 2014; Turgut, 2016).

In other words, street connections cannot be reduced to negative and traumatising ties that could disappear if only the right support for css were provided. This idealistic view is not consistent with real life experiences of css, documented over the years (Aptekar and Stoecklin, 2014). A much more contrasted view of the complexity of street life is needed in order for a gc to have some real impact. This view should include the subjective experience and symbolic meaning of street life. This is about how the child appropriates the external world. Meanwhile, we see that the content of Section 6 of the gc’s outline is only based on the mechanisms that are supposed to change this external world:

Universal ratification; monitoring, accountability and redress; non-discrimination principle; development and strengthening of comprehensive, child rights-based child protection systems; structural support for quality service provision; implementation at municipal level; financial resources; addressing stigmatization and discrimination of children in street situations; criteria for good practices; data collection and research.

un crc 2016
The child is not at all included in the definition and/or appraisal of this external world. So it is likely that, unless significant weight is given to child consultations, the gc on css will end up in another deterministic approach whereby street life is seen as a product and not a construct. Hence, css will also have been “processed” as products of deprivation, victimisation, abuse, etc. – not really empowering them, because their street experience and connections are not properly recognised.

With the reduction of street connections to something dangerous, Section 6 of the gc’s outline gives a pessimistic coloration to children’s agency, as if this could only evolve towards further crime and abuse in the street environment. This reduction, and the focus on articles 15, 20 and 27, makes street life appear only through the lenses of deprivation: children look deprived of family, of minimum standards of living and of freedom of association. Whilst this might be true for a number of css, it is by no means the case for all of them. This back-to-protection movement is a visible translation: the discrepancy between the agency of css depicted in research and the protecting focus in the gc is quite obvious. As a consequence, we are here confronted by a specific (re)framing of the issue, an instantiation of how changes in “signification” (Giddens, 1979) are made alongside “translations” (Hanson and Nieuwenhuys, 2013). These obvious translations highlight the institutionalisation of children’s rights. But the latter also occurs through hidden translations, to which we turn now.

5.3.2 Hidden Translations

While the gc is on “Children in Street Situations” and not on “Street-Connected Children”, it is, however, not clear whether the Committee fully considers the implications of the css perspective or if the css label is only taken at face value. The mixing of css and scc within the call is a sign that the latter is probably the case: scc and street connections are used when speaking of association (art. 15) and family deprivation (art. 20), and css is used when standard of living (art. 27) is at stake. The inclusion of “concrete examples of legitimate and illegitimate restrictions on street-connected children’s rights under article 15” reinforces the impression that street connectedness is something that is not only bad for the children, but also may be dangerous for society at large. The same suspicion appears with the call, under article 20, for “special care and protection by offering them better choices so they are not forced to depend on their street connections for their survival and/or development”. When it comes to standard of living (art. 27), the call suddenly turns to css: “implement the right to an adequate standard of living for children in street situations, or at risk of being so, including assistance to parents and others responsible for the child”.

It looks as if the standard of living would be the main factor alleviating street situations. But, as we know, poverty is only a strong predictor but not a cause for street life in itself. The focus on “push factors” (standard of living) and the implied suspicion towards street-connectedness are indicators that the “pull-factors” (both addressed by scc and css) are treated as something children should avoid. These apparently tiny nuances are quite important: when pull-factors are implicitly depicted as wrong choices, children’s real connections (and not idealised ones) are not fully acknowledged, and children’s agency is actually denied. The gc on “Children in Street Situations” does not provide a definition or an explanation for the wording, and only a brief perspective on css is made in a former un document :

Terminology has continued to evolve to recognize children as social actors whose lives are not circumscribed by the street. Human Rights Council resolution 16/12 refers to children working and/or living on the street, and the Committee on the Rights of the Child has adopted the term “children in street situations”, recognizing that children engage in numerous activities on the street and that if there is a “problem” it is not the child, but rather the situations in which s/he finds her/himself.25

un ohchr, 2012: 10
Only hints of the perspective that presided over the wording “children in street situations” appear in this statement, namely that they are ‘social actors whose lives are not circumscribed by the street’ (op. cit.) and that the problem ‘is not the child, but the situations’ (ibid.). It looks like the perspective on css developed for years (Lucchini, 1993, 2007; Stoecklin, 2000a, 2007; Aptekar and Stoecklin, 2014) has, to a large extent, been lost in translation. How does this happen ? A closer look at the “channels of transmission” of academic works to the Committee provides an answer. It appears that this transmission is not only framed by formal procedures, like General Days of Discussions and contributions to General Comments; it may also depend on the personal willingness and academic references and networks of the members of the Committee themselves.
In the present case, during his second term of Office, Jean Zermatten, acting as Chair of the un Committee for the Rights of the Child, strongly advocated for the use of the label “Children in street situations” within the Committee. This terminology was then adopted as the official heading for the Committee’s Concluding Observations: ‘Since 2009, the agreed heading for the cobs is Children in street situations (see the new guidelines to reporting – CRC/C/58/rev.2) under Special Protection Measures’ (Zermatten, 2011: 4). Later on, during the expert consultation on children in street situations in November 2011, he further advocated for the css label to become the official terminology used in the present gc, based on the topical publications (Stoecklin, 2008a, 2008b, Rizzini et al., 2007), including the website26 of the ngo (the International Institute for the Rights of the Child) that he was directing at the time. His presentation ended with the following conclusion:

This is why the Committee, in view of the continuous use of different pejorative terminologies (such as street children) and because categorization can be conducive to discrimination, on the basis of a “rights based approach” decided to choose the new wording “children in street situations”, which puts the accent on the idea that the child has of his/her own situation. In fact the problem is no more the child her/himself but the situations in which s/he may find her/himself.

zermatten, 2011: 3–4
So, the Committee’s 2009 agreed heading and these publications provided strong legitimacy for the css label. However, the shift from child to situations, which is a dramatic change of focus, may not have been pursued in all its consequences, leaving the impression that the css perspective is mainly seen as just another label or, reversely, that the css label has been detached from the sociological interactionist perspective it stemmed from. This particular example of “translation” from the academic world into a specific policy shows interesting aspects of the paradox of institutionalisation (Stammers, 2013): although specific publications had gained greater accessibility to the Committee’s attention in 2009, and hence to the larger community present at the expert consultation in 2011, this is not enough to make sure that people would retain more than just hints of what is behind the new label used (css). Even with privileged channeling, a scientific perspective still has great difficulty to translate completely into policy.

This shows that not only are children’s views “translated” with specific limitations (we could also say that they are “mistranslated”) into policy, but also that academic perspectives suffer from specific deformations. Academic perspectives are not only a source of informed policy, they are also a mediating body between children and the States in charge of implementing policies regarding all the children under their jurisdiction. The advocacy of ngos and the (alleged) neutral academic discourse share the same fate, even though an ideologic position is a core component of the former and banned from the latter: they are reduced to the position that is considered acceptable within policy. In this process, the accepted academic discourse has more to lose than the accepted ideology advocated by ngos. The shift of focus, stating that problems are within situations and not within children, requires a fully developed theory of what “situations” involve. As it stands, this nice new wording only helps to depart from the image of a “street child” as victim and/or delinquent, but it is not sufficient to really replace it with the consequent new paradigm of the child as an actor.

Another translation, although less obvious, can also be seen when considering the articles of the uncrc that are not mentioned in the outline of the gc on css. We should not only focus on the selected articles but also identify those which have not been prioritised. What are the articles that are not selected? There are 10 articles: Art. 1 (definition of age), Art. 9 (the right not to be separated from parents), Art. 10 (family reunification), Art. 11 (illicit transfer and non-return of children abroad), Art. 14 (freedom of thought, conscience and religion), Art. 21 (adoption), Art. 22 (seeking refugee status), Art. 25 (periodic review of the treatment of placed children), Art. 30 (ethnic, religious or linguistic minorities), and Art. 39 (physical and psychological recovery and social reintegration).

We can consider these unselected articles as an illustration of the “paradox of institutionalisation”: they actually tell a lot about what is not considered “emancipatory struggles” that are good or suitable enough to be ‘sustained through processes of institutionalisation’ (Stammers, 2013). It is, for instance, very surprising, and hence quite telling, that the periodic review of the treatment of placed children (art. 25) is not considered a priority for css, when we know that many so-called shelters for street children fail to meet the basic requirements and international standards of institutional care. The absence of these articles in the gc on css may actually have the effect of lessening the attention to what are actually very important issues, such as, inter alia, ageing out (art. 1), detention and deportation (art. 9), unaccompanied minors (art. 10 and 11), recovery and reintegration (art. 39).

So, the absence of these ten articles tells a lot about the absence of real efforts – and thus the complicit absence of will to remedy for it – with regard to options that are, for most of them, probably more concrete and also more expensive than the selected articles, especially those given priority (arts. 15, 20 and 27). The selection of some articles and the deselection of others raises a critical question about emancipation struggles as evoked by Stammers (op. cit.): emancipate from what, really? What would people and especially css really struggle for? An emancipation or rather a domestication of these children ? An emancipation from mainstream society, or from street life, or from the conditions that are conducive to street life? These things are not exactly the same …

6 A Field of Tension

The links that are made among the different articles of the uncrc that are prioritised, and, conversely, the links among the unselected articles, tell something about the (mostly hidden) underlying discourse. The drafters of the gc on css are located in a field of tension. They must still resist the tendency towards a back-to-protection movement mentioned before, and an urge for capacitation, with the consequences of putting the responsibility, and the blame for not meeting expectations, on parents and children themselves. They must also resist the neo-liberal colouration of the discourse on capacitation (a specific view of “empowerment”) that is linked to a particular kind of domination and legitimation. In the case of the gc on css, domination is legal-rational but also charismatic (sensationalist discourses of ngos) and traditional (the traditional representations of childhood). There is a combination of several kinds of domination. As we know, there is nowhere in reality such a thing as an ideal-type in its purity. These hidden translations can only be seen through a theory of structuration. The Giddensian approach offers very relevant insights for a theory of situated agency. Applied to children’s rights, it highlights the duality of the structure, in its discursive instantiations, which links subjects (children) and objects (rights) in given contexts. This offers insights into the institutionalisation of children’s rights that can be further developed with explicit theories of action (Stoecklin, 2013) highlighting the situated views of children.


  • Aptekar L. , Street children of Cali (Durham: Duke University Press, 1988).

  • Aptekar L. , Stoecklin D. , “Children in Particularly Difficult Circumstances” in Berry J.W. , Dasen P.R. and Saraswathi T.S. (eds.), Handbook of Cross-Cultural Psychology (2), 377412 (Boston: Allyn & Bacon, 1997).

    • Search Google Scholar
    • Export Citation
  • Aptekar L. , Stoecklin D. , Street Children and Homeless Youth: a cross-cultural perspective (Dordrecht, Heidelberg, New York, London: Springer Edition, 2014).

    • Search Google Scholar
    • Export Citation
  • Becker H.S. , Outsiders: Studies in the Sociology of Deviance (New-York: The Free Press, 1963).

  • Becquemin M. , Robin P. , “L’enfant, sa famille et les institutions qui le protègent. Enjeux et effets d’une symétrisation des forces?” in Payet J.-P. , Purenne A. (eds.), Tous égaux! Les institutions à l’ère de la symétrie, 5779 (Paris: L’Harmattan, 2015).

    • Search Google Scholar
    • Export Citation
  • Boltanski L. , De la critique. Précis de sociologie de l’émancipation (Paris: Gallimard, nrf essais, 2009).

  • Bonny Y. , “Les recompositions contemporaines des symétries et asymmetries à l’aune d’une sociologie des institutions” in Payet J.-P. and Purenne A. (eds.), Tous égaux! Les institutions à l’ère de la symétrie, 3356 (Paris: L’Harmattan, 2015).

    • Search Google Scholar
    • Export Citation
  • Bourdieu P. , The logic of practice (Stanford University Press, 1992).

  • Cauce A. , Paradise M. , Ginzler A. , Embry L. , Morgan C. , Lohr Y. and Theofelis J. , “The characteristics and mental health of homeless adolescents: Age and gender differences”, Journal of Emotional and Behavioral Disorders , 2000 (8(4)), 230239.

    • Search Google Scholar
    • Export Citation
  • Cusson M. , Délinquants, pourquoi? (Paris: Colin, 1981).

  • De Benitez S. , State of the World’s Street Children: Research (Street Children. Series 2) (London: Consortium for Street Children, 2011).

    • Search Google Scholar
    • Export Citation
  • de Janvry A. , Finan F. , Sadoulet E. , Evaluating Brazil’s Bolsa Escola Program: Impact on Schooling and Municipal Roles (University of California at Berkeley, 2006):

    • Search Google Scholar
    • Export Citation
  • Declaration on the Rights of the Child. Proclaimed by General Assembly Resolution 1386(XIV) of 20 November 1969:

  • Elias N. , The society of individuals (London and New York: Continuum, 1991).

  • Ennew J. , Connolly M. , “Introduction: children out of place”, Childhood, a global journal of child research , 1996 (3(2)), 131145.

    • Search Google Scholar
    • Export Citation
  • Foucault M. , Du gouvernement des vivants – Cours au Collège de France. 1979–1980 (edition établie par Senellart M. , sous la direction de Ewald F. et Fontana A. ) (Paris: ehess – Gallimard-Seuil, 2012).

    • Search Google Scholar
    • Export Citation
  • Giddens A. , The constitution of society: Outline of the theory of structuration (Berkeley: University of California Press, 1984).

  • Giddens A. , Central Problems in Social Theory. Action, Structure and Contradiction in Social Analysis (London: The MacMillan Press Ltd, 1979).

    • Search Google Scholar
    • Export Citation
  • Goodale M. , Merry S.E. , The Practice of Human Rights: Tracking Law Between the Global and the Local (Cambrigde: Cambridge University Press, 2007).

    • Search Google Scholar
    • Export Citation
  • Hanson K. , Nieuwenhuys O. , Reconceptualizing Children’s Rights in International Development. Living Rights, Social Justice, Translations (Cambridge: Cambridge University Press, 2013).

    • Search Google Scholar
    • Export Citation
  • Hewitt J.H. , Social stratification and deviant behaviour (New-York: Random House, 1970).

  • ide (2008), Institut international des Droits de l’Enfant:

    • Search Google Scholar
    • Export Citation
  • ingo, Inter-ngo Programme for Street Children and Street Youth. Summary Proceedings of Sub-regional Seminar for the Mediterranean on Street Children, Marseilles, 24–27 October (Geneva: International Catholic Child Bureau, 1983).

    • Search Google Scholar
    • Export Citation
  • Kabeberi J. , Child custody, care and maintenance (Nairobi: Oxford University Press, 1990).

  • Korbin J. , Child abuse and neglect: Cross cultural perspectives (Berkeley: University of California Press, 1981).

  • Liebel M. , Enfants, Droits et Citoyenneté. Faire émerger la perspective des enfants sur leurs droits (in collaboration with Pierrine Robin et Iven Saadi) (Paris: L’Harmattan, 2010).

    • Search Google Scholar
    • Export Citation
  • Lucchini R. , Enfant de la rue. Identité, sociabilité, drogue (Genève/Paris: Droz, 1993).

  • Lucchini R. , Sociologie de la survie. L’enfant dans la rue (Paris: Presses Universitaires de France, 1996).

  • Lucchini R. , “‘Street children’: Deconstruction of a category” in Rizzini I. , Mandel Butler U. and Stoecklin D. (eds), Life on the Streets. Children and Adolescents on the Streets: Inevitable Trajectories? 2007, 4975. Sion: Institut International des Droits de l’enfant.

    • Search Google Scholar
    • Export Citation
  • Mambwe A. , “Stress and coping strategies amongst street children in Lusaka”, Journal of Psychology in Africa, South of the Sahara, the Caribbean and Afro-Latin America 1997 (2), 2433.

    • Search Google Scholar
    • Export Citation
  • Merry S.E. , Human Rights and Gender Violence: Translating International Law into Local Justice (University of Chicago Press, 2006).

  • Moberly C. , “Creating policies which address the ‘voluntary separation’ of children in Angola” in Schrader A. and Veale A. (eds.), Prevention of street migration: Resource pack 1999, 36 50 (London: Consortium for Street Children).

    • Search Google Scholar
    • Export Citation
  • Muchini B. , Nyandiya-Bundt S. , Struggling to survive: A study of street children in Zimbabwe (Harare: unicef, 1991).

  • Nizet J. , La sociologie de Anthony Giddens (Paris: La Découverte, 2007).

  • Onyango P. , Suda C. and Orwa K. , A report on the Nairobi case study on children in especially difficult circumstances (Florence: unicef, 1991).