**8. Discussion**

Cemeteries frequently are public spaces, sometimes walled or fenced-off and closed at night, but otherwise open to all. This does not preclude them from being ritual spaces, as outlined in the Introduction, with specific expectations of behavior during funerals, collective commemoration ceremonies and individual grave visits (Bachelor 2004; Francis et al. 2000, 2005; Kjærsgaard and Venbrux 2016; Schmied 2002). Of course, not every behavior at cemeteries is ritualized; cemeteries are also popular with tourists and flaneurs, who are nonetheless also expected to behave "properly" and can also be(come) "pilgrims" to certain graves (Kmec 2020; Rugg 2000, p. 264; Venbrux 2010)

We would like to suggest that the inner divisions of Cemetery Tongerseweg over time develop in accordance with Evans-Pritchard (1940) concept of "segmentary opposition". The dynamics implied in segmentary opposition make it a very suitable theoretical tool for a diachronic analysis of the cemetery as a ritual space. In the original context, as Evans-Pritchard points out, "The relation between tribes and between segments of a tribe which gives them political unity and distinction is one of opposition" (Evans-Pritchard 1940, pp. 282–83). A segment can thus be in opposition to another segment, but both segments in turn can join in opposition to a third one. The Calvinists and the Lutherans had each a section; they formed two segments in opposition to each other. This distinction, however, disappeared when they stood as Protestants in opposition to the Roman Catholics. Next, they became a unity as Christians (the hedge separating them was removed) vis-à-vis the Muslims (spatially separated from them) who began to enter the cemetery around that point in time. The weakening of the ranks of the Catholics in the cemetery might in part be resolved by Armenian Christians, with a new section or segment, moving in.

As the Armenian example shows, this may be a win-win situation. There is, nonetheless, the possibility of a "re-pillarization" of Dutch society unintentionally effected by the actions of cemetery planners and managers. Rather than integration (Krabben 2004), in actual fact, segregation appears to be taking place. What is more, the native Dutch opting for the cemetery decreases, whereas migrants are expected to follow a unilinear trajectory of pillarization as a means of emancipation.

As this article has shown, a cemetery is not only shaped by municipal councils, planners and managers, but it is also inventively customized and cocreated by its users. Our examples show different ways of articulating the position of one's family (or deceased family member) as "different": to overcome distinction by emphasis on high office and family unity across confessional divides (Figure 3) or to transcend distinction by emphasis on common values (Figure 5) and hybrid identities of migrant and host country culture (Figures 7 and 9a,b).

The role of service-providers should not be underestimated either, as the design of individual gravestones is to a large extent predetermined by the funeral industry, but that discussion is beyond our scope here (Schmitt et al. 2018). This cocreation of space is not only reflecting past and present changing social relations but also has a prospective dimension: the cemetery is an arena, in which people seek to achieve social recognition, visibility and respectability. It constitutes a means to display groupness (Brubaker 2001) but also social bonds across groups. In an increasingly plural society, cemeteries continue to provide "order" and may thereby contribute to a "re-pillarization" or segregation.

We propose to view Cemetery Tongerseweg not as one single space but as a multitude of overlapping (ritual) spaces, just as society is not uniform but a multitude of intermeshed imagined and interpretive communities. The early nineteenth French "egalitarian" ideology did not, contrary to popular assumptions, seek to dissolve all communities in one large citizenship but rather encouraged emancipation and identity claims—while simultaneously aiming at building an overarching rational society. We find similar dialectics at work today, where diversity and integration are promoted concurrently. What is happening in Cemetery Tongerseweg is that it creates, builds, evokes and represents as well as relativizes and backgrounds communities; the cemetery is thus an arena of and for society.

Cemetery Tongerseweg is thus a hybrid space, on the one hand, a rigorously rational, economically minded administrative infrastructure, and on the other hand, emotionally charged at an individual as well as collective level. To use (Tönnies [1887] 2001) ideal-types, the cemetery is clearly not just mirroring (rational) society but is also addressing (affective, normative) communities.

A model needs to be built to illustrate that and how ritual spaces and communities shape each other. The cemetery is the arena in which this is taking place; planners and users with their respective allies are cocreators of the ritual spaces in a cemetery. All of this goes beyond the opposition between the passive mirror analogy and the depiction of the cemetery as an idealized compensatory world that can be found in literature.

Building a model that gets the balance right is by no means easy. The required model needs to have explanatory value through idealization as well as enough detail to capture the complexities of cocreation. The following "Arena Model" of cemeteries as dynamic ritual spaces strikes in our view the desired balance and captures our discussion (Figure 10):

Understanding ritual space is a complicated matter because the actors and factors engaged in processes of cocreating a ritual space are diverse and complex; moreover, we should view cemeteries as an ensemble, a plurality of ritual spaces. As Figure 10 shows, the causality (if that is not the wrong word entirely) goes both ways, or rather runs in a circle: cemeteries as an ensemble of ritual spaces are reliant on pre-existing communities, but they also evoke, produce and maintain communities. And, to complicate things further, not only does the physical layout codetermine the ritual space at the disposal of its users, but the users respond and cocreate ritual space in placing a wide range of ritual markers (e.g., gravestones and semipermanent or transitory ritual props that can enhance, but also mitigate or even contradict, the communities created by the spatial layout).

**Figure 10.** The Arena Model of cemeteries as dynamic ritual spaces (model: C.J., infographic: T.K.).

All of this shows how different stakeholders and their allies cocreate ritual spaces in the cemetery by means of permanent, semi- or nonpermanent performative acts that range all the way from expansive gestures such as erecting walls or laying out paths to the smaller, less permanent performative acts such as the placement of a prayer bead.
