As a social theory, boundary object (határtárgy) aims to provide a model of translation and understanding between groups that possess different types of knowledge. It can be applied to any cooperative situation where the involved persons or groups have no shared background to fall back on. Such collaboration creates boundary objects. This is typically applicable to participative museum work, in which museum objects often take on the role of boundary objects.

 
The term boundary object was introduced to the social science vocabulary by Susan Leigh Star and James Griesemer (1989) in the framework of a specific museum case study (Museum of Vertebrate Zoology). From a history of science point of view, the concept originates from symbolic interactionism, and has its foundations in science and technology studies (STS). From here, it migrated to actor network theory (ANT) and activity theory, followed by a series of other disciplines, and is still in use today (Wenger 1998). Cultural anthropology, however, has not employed the model to this day, even though within the field of cognitive anthropology distributed cognition has worked a lot with the similar role of cultural objects (see, for example, Hutchins 1995). This absence can probably be explained by the fact that ethnographies working with the boundary object model mostly originate from the fields of STS and ANT (cf. Vinck 2012) – fields that primarily focus on organisational issues. Literature on museum praxis has used the term with increasing frequency in the last few years (Pahl–Pollard 2010; Waller 2016: 198), because issues related to cooperation between groups with diverse backgrounds have become more prevalent in this domain.
 
 
The concept
 
When defining boundary objects, Star and Griesemer started out from the simultaneous presence of, and tension between, heterogeneity and cooperation in scientific work, and the ensuing need for translation between different worlds. Their initial observation is that collective scientific work does not require consensus between all participants. They presume that due to the diverse background of participants, scientific work necessitates constant coordination and translation. Cohesion in this heterogeneous group is created by a shared purpose, to which compromises, discussions, and debates are adjusted. In their concrete example, amateur collectors and museologists with an academic background are attempting to establish a shared model, i.e. to create something coherent out of radically different meanings (Star–Griesemer 1989: 392). This can be realised by standardising analytical methods and creating boundary objects.
 
Consequently, what we may call a boundary object is an object used simultaneously by multiple groups, which gives it the ability to mediate and translate between various types of knowledge. The concept is especially useful when studying everyday material culture and museum objects. A given thing (be it concrete, physical, or abstract, text-like, semiotic) can typically be considered a boundary object if it connects situations in manifold ways, and fulfils an integrating function on the interface of knowledge domains. Boundary objects are also called common information spaces or mediating objects.
 
In other words, the concept refers to things that act as interfaces or shared environments between diverse communities, each of which makes use of the object, yet interprets it in a very different way. Although participants need a shared aspect, it is not necessary to fully understand each other. In this process, boundary objects serve as points of orientation, mediation, and compromise; understanding is contingent upon the efficiency with which they serve communication and cooperation. Boundary objects play an organisational role in social interactions, giving a structure to the interactions themselves. Through their shared use, they instigate and facilitate cooperation. This makes it necessary for them to be obvious and publicly available. It is, however, not necessary to think of boundary objects as physical objects: they can be simultaneously concrete and abstract; they can take the form of information, discussions, interests, rules, plans, contracts, or even persons.
 
Boundary objects are shared segments of worlds that partially permeate each other; their ability to mediate during everyday communication and coordination results from the fact that they are present for every participant, i.e. all participants can relate to them one way or the other. As a research model, boundary object assesses tools by the way they mediate group activities.
 
 
Boundary objects in the museum
 
The theory is especially useful when interpreting museum objects, since it emphasises the coming together of persons and (not necessarily homogeneous) groups with diverse backgrounds. Boundary objects are often appreciated in museums, as during activities they serve as points of reference in the communication of groups with a different purpose, starting point, and background. In the museum context, boundary objects can be objects that a certain group considers distinctively exotic, or characteristic of the given culture (stereotypical).
 
In Star and Griesemer’s interpretation, there are four basic types of boundary objects: repositories, ideal types, coincident boundaries, and standardised forms (Star–Griesemer 1989: 410–411). This is rephrased by educational theorist Étienne Wegner as the four dimensions of boundary objects, i.e. modularity, abstraction, accommodation, and standardisation (Wenger 2000). In terms of modularity, boundary objects can serve as bases for dialogue (such as items in a library or a museum collection). Accommodation allows for an object to be used for a variety of tasks or purposes. Abstraction facilitates dialogue between different worlds. Standardisation defines a framework for interpretation.
 
Flexibility is a definitive aspect of the way boundary objects work: they possess a complexity that allows for a number of different, partial interpretations. This facilitates the use of boundary objects by diverse groups. At the same time, these objects are sturdy enough to keep their own identities instead of dissolving in any given interpretation.
 
Boundary objects can be simultaneously present for multiple groups, communities, or persons, creating the necessary conditions for connections to form between these actors. Possibilities for interpretation and use are, however, different for each participant, often changing from situation to situation. Nevertheless, the sheer presence of these objects is capable of structuring behaviours and interactions, and serves as a shared reference point in communication. The shared goal of fieldwork or the interpretation of objects in a museum setting may be a description, monograph, or exhibition, yet the ethnographer and members of the source community often participate in this process with radically different objectives, conceptual toolkits, and background knowledge. Depending on which participant interprets it, a photograph or a household object, for example, can simultaneously be a family heirloom, a binding link between generations, and also the representative of a certain object type, or the illustration of a given era or social milieu. It is this intermediary role that turns photographs and household objects into boundary objects.
 
Boundary objects are able to coordinate between different groups, individuals, actions, and communications without presupposing a shared knowledge or culture (not even one that would emerge as a result of cooperation). They fulfil their coordinating role through their ability to create networks. Boundary object, as a model, is ecological: it refers to the environment of objects, persons, and groups, i.e. it describes all participants as embedded in their respective environments. This is because boundary objects transcend the borders between different social realities.
 
Boundary objects simultaneously connect and separate participants. They unite their conclusions and contributions despite their different perspectives. Therefore they are also able to mediate, build bridges, and create connections between different participative forms, since coordination is more important than the establishment of a shared interpretation. In this respect, it is worth referring to the concept of friction, introduced by American anthropologist Anna Lowenhaupt Tsing (2005), according to which cooperation is not consensus, but rather a set of productive misunderstandings (Tsing 2005: 247). This can be the most productive when natural or social objects are being created.
 
Characteristic and classical examples of boundary objects include work environments, libraries, museums, maps, and scientific work. All of these places and processes are based on a shared use of objects by a number of groups with diverse backgrounds. As shared objects, they are loosely structured, i.e. they allow for a variety of interpretations, but when used by individuals, they become highly organised. They are powerful enough to induce a shift in perspectives, thereby creating coherence between different social realities. Boundary objects are physically stable, so that they can be used in a variety of contexts without losing identity, yet they are flexible in interpretation, permitting diverse local interpretations and applications. Scientific work, for example, is heterogeneous; performed, and contributed to, by a number of different groups, which makes cooperation inevitable. Consensus, however, is not a precondition for cooperation. The boundary object model can also contain relationships and conjunctions that are unintended constituents in the cooperative process. In turn, the reorganisation of groups facilitates the formation of new boundary objects (and translations).
 
 
Boundary objects in a broader context
 
The concept of the boundary object can also be extended to the encounter of different cultures or intercultural analyses. British anthropologist Marilyn Strathern, for example, uses this notion to define cultural studies as a boundary discipline with reference to its interpretative method (Strathern 2004: 45). In many ways, this extension likens the boundary object to the reciprocal contact zone, introduced by James Clifford (Clifford 1997: 204). In other words, if the space of cultural encounter can be interpreted as an object – which is not excluded from the original definition of the boundary object – then the contact zone is a special type of boundary object that derives from a hierarchical difference between participants.
 
The boundary object is typically and fundamentally an epistemological concept: it is our related knowledge that is of relevance to the model. Although there are many overlaps with the notion of the intermediary object (used primarily in actor network theory), this idea is pertinent to an ontological model: the intermediary object is present in the world, independently of our knowledge. ANT’s intermediary objects are therefore different from boundary objects (Vinck 2012: 93). Although they both originate from the sociology of knowledge, boundary objects come to make sense in the intersections of different social worlds (groups of activity). The question is how to synchronise them. As an integral part of ANT, intermediary objects describe the network of relationships between groups and individuals: the description follows intermediary objects. Their roles are, therefore, open-ended: they can work between and also within groups. They are, in a way, the manifestations of translation, and it is exactly their materiality that makes them open (Vinck ibid.: 96).
 
Using Thomas Kuhn’s theory of paradigms (2002 [1962]), American historian of science Peter Galison has created the model of the trading zone, in which he declares that the boundary object is basically a temporal segment of the trading zone (Galison 2010: 46). In his analysis, he applies the concept of the trading zone to scientific objects that acquire a new meaning compared to the former one as they are passed on. His approach is fundamentally inspired by Marcel Mauss’ anthropological theory of exchange (Ibid.: 35). Mauss also places the emphasis on coordination instead of sameness or agreement in signification. Thin description is sufficient in the case of exchange (Ibid.: 36); there is no need for thick description in the Geertzian sense. Thick description (interpretation) is basically the same as consensus, while the thinness of interpretation equals exchange. A few words in common are sufficient for coordination; there is no need to share a whole language. Trading zone-like languages (pidgin or creole) are rather heterogeneous (Ibid.: 42). Although there is a clear similarity between the boundary object and the trading zone model, it is important to remember that the trading zone (Galison 1997) is an extension of the linguistic metaphor of translation, while the boundary object presupposes that cooperating partners do not fully understand each other.
 
 
Extended use of the concept
 
The concept of the boundary object becomes meaningful within a networked theory of objects, where objects and their social roles are defined by the relationships between participants. There are, however, multiple networked theories that differ from each other in many respects, such as in their definition of objects (e.g. the object is defined slightly differently by actor network theory and assemblage theory). Therefore boundary objects have a different meaning within these approaches. The concept is particularly useful for the analysis and interpretation of complex social situations and interactions, because it uses the domain of communication to find points of convergence that groups with divergent or even contradictory backgrounds, interests, or goals can rely on. These groups may not come from a shared social reality, but collective work, goals, and tasks necessitate some degree of integration between the different types of knowledge each group possesses. For successful cooperation, they constantly have to reconsider the meanings behind the tools and methods that they use.
 
Some of the most relevant issues of the topic are the efficacy of boundary objects, their conscious selection or creation, the existence of a proper context for their use, and the power relations immanent in their creation or selection. Indeed, boundary object is not a merely descriptive model; it can be applied directly. Instead of merely identifying and analysing the boundary objects that are already in operation and cultural use, it can also be applied to projects and collaboration, moreover, museum exhibitions or collections can be consciously planned with boundary objects in mind.
 
Although boundary object as a social science model was introduced in relation to a museum case study (Star–Griesemer 1989), it had to go a rather roundabout way before anthropology and museums started using it again (Pahl–Pollard 2010; Poehls 2010; Waller 2016; Wilhelm 2014). It seems, however, that it is worth experimenting with this approach within the museum domain, where cooperation, participation, and collaboration are increasingly gaining ground. Museum work involves a wide range of subjects, as well as a wide range of collaboration in both everyday tasks and larger-scale projects. Boundary objects feature in the relationships between experts with different backgrounds, and between museum and source communities.

Gábor Wilhelm