The Anatomy of Neighbour Relations

Relations between neighbours represent informal social relations that constitute a part of everyday life for virtually all individuals in contemporary urban societies. While aspects of neighbour relations have been studied in connection with research on local communities, gentrification, and neighbourhood effects, little research focusing specifically on interactions between neighbours has been conducted and theoretical reflection on the central concepts remains lacking. This article attempts to fill this research gap by developing a theoretical framework for the sociological understanding and investigation of neighbour relations. The research material used consists of previous studies on neighbour relations. The method for the task was analysis of concepts coupled with theoretical reasoning based on a review and a synthesis of relevant previous research. Familiar concepts like ‘neighbour’, ‘neighbouring’, and ‘neighbourliness’ are scrutinised, deconstructed, and redefined, noting that there are often two perspectives on each: the experiences of residents and those of outside observers. Various sociological aspects of neighbour relations are discussed, including neighbour relations as an acquaintance relation, neighbour relations as the basis of a community, and the cultural content of the neighbour role. A model of levels of neighbourly interaction is presented.


Introduction
Neighbour relations are informal social relations that constitute a part of everyday life for virtually all members of contemporary urban societies. Looking at the existing sociological literature, a range of views exist regarding whether neighbour relations have been well-researched. Studies on local communities, of which there are many, often deal with relations between neighbours, while there is also a growing body of research on gentrification and neighbourhood effects. For example, the literature on British community studies (see Crow and Allan, 1994) deals to some extent with relations between neighbours, albeit focusing mainly on local communities. There are far fewer studies focusing directly on how neighbour relations are constructed and reproduced, the nature of neighbourly interaction, or the meaning of neighbour relations for people (Crow, 1997: 22;Stokoe, 2006: 1.10;Stokoe and Wallwork, 2003: 554). These remain, however, interesting and important questions in that they touch upon our mundane everyday existence and link to several key questions regarding the nature of post-or late modern societies. For example, in Robert Putnam's (2000: 105-107) best-selling Bowling Alone, the decline in neighbourly interaction is a frequently mentioned indicator of the erosion of social capital in US society.
This article attempts to fill this research gap by developing an analytically sharp theoretical framework for the sociological investigation of neighbour relations. The framework is developed through a close analysis of familiar terms used both in everyday conversation and research, such as 'neighbour', 'neighbourhood', 'neighbouring', and 'neighbourliness', as well as a review and synthesis of various ideas, concepts, and observations in the research literature. The article is written with the conviction that though various studies may have touched upon neighbour relations, a theoretical analysis of what neighbour relations represent in the sociological sense is lacking in the existing literature.

What is a neighbour?
Like many everyday terms, the concept of 'neighbour' is seemingly simple, but in a more detailed investigation proves to be tricky. Stokoe and Wallwork (2003) point out that in neighbour-related research, 'there is remarkably little commentary on defining "neighbour" or the experience of managing neighbour relations' (p. 552). On occasion, researchers have simply passed the problem of defining who is a neighbour to the people they research: 'Neighbors were people who respondents called "neighbor"' (Fischer, 1982: 98). If the definition of what 'neighbour' means is left to research subjects, whom I shall call members, it is possible -even likely -that one person's definition will differ from another's. Though members' conceptions of what constitutes a 'neighbour' are worth researching, we nevertheless need for many research purposes a more 'objective' conception, which I shall call an outsider's definition of 'neighbour'.
A good starting point for discussing the concept is the simple definition given by Philip Abrams in his posthumously published study Neighbours: The work of Philip Abrams (Bulmer, 1986). By neighbours, Abrams means 'simply people who live near each other' (Bulmer, 1986: 21;also Cheshire et al., 2019;Cockayne, 2012;Fischer, 2004;Grannis, 2009;Heberle, 1960;Keller, 1968). Thus, physical proximity represents the basic defining feature of the concept of 'neighbour'. People who move out of our neighbourhood but keep contact with us are for us 'former neighbours' or perhaps 'friends', not just neighbours (Grannis, 2009: 20;also Keller, 1968: 25-26). Being a neighbour is a different kind of relation from friendship; we need to make friends with the people who become our friends, whereas neighbours are neighbours just by being where they are (Keller, 1968: 24;also Rosenblum, 2016).
However, this simple definition is somewhat problematic. The people who live next door are our neighbours, as well as the people who live close by in the same neighbourhood. While this is clear in urban surroundings, like the English urban communities that Abrams researched, when we think of sparsely populated rural areas, for example, Lapland in Finland, the idea of what constitutes a neighbour becomes more complicated. In such circumstances, my neighbours may live kilometres away from me, but I may nevertheless still refer to them as neighbours as they are the people who live closest to me. The term 'neighbour' in everyday language refers to a spatial relation: neighbours are persons living at a closer distance to us than other people, whatever that distance may be. When used metaphorically, the term 'neighbour' refers to relative distance, as, for example, when talking about 'neighbouring' cities, countries or planets. The distance may not even be geographical, as we can also speak of neighbouring disciplines.
Neighbouring countries are not only close to each other, but they also share a physical border (though planets and disciplines do not). The aspect of borders is central to neighbour relations between people. People erect social and symbolic boundaries, or 'fences' (Gullestad, 1986), between themselves and their neighbours. They do this in order to manage the social distance and proximity between themselves and others (e.g. Crow et al., 2002: 137;van Eijk, 2011: 1.4). People erect symbolic fences through messages they send each other. These messages are mediated not only through verbal interaction but also through the senses (Lewis, 2020: 107): 'tone of voice, facial expression, gestures, gaze, body language, placement of people in relation to doors and furniture, etc.'. (Gullestad, 1986: 55). The difficulty of balancing proximity and distance has preoccupied many sociologists. The boundary between friendly supportiveness, interference violating others' privacy, and total detachment is not clear-cut in neighbour relations; the friendly distance, considered ideal in neighbour relations, needs to be skilfully worked out (Crow et al., 2002: 137;also May and Muir, 2015). Neighbour relations involve a special kind of reciprocity: 'It is necessary to show some commitment to neighbourliness, but it is equally important not to give the impression of interfering, since interference compromises the other's privacy' (Crow et al., 2002: 140).

Neighbours in neighbourhoods
Van Eijk (2011) solves the problem regarding the neighbour concept by specifying that she is focusing on 'immediate neighbours, that is, those who live in adjacent dwellings immediately next, above, below, or opposite each other' (p. 1.4; also see Zang, 2007: 25). Focusing on next-door neighbours may be a reasonable solution in some research settings, but not always. Sociologists, as well as the people we investigate, often consider people who do not live next door but anyway quite close to each other as neighbours. Sometimes the concept of 'neighbour' is defined with reference to a bounded geographical location, and that may indeed be a sensible sociological usage of the term. For example, Morgan (2009: 11) defines neighbours as people living near each other in a neighbourhood.
What actually constitutes a neighbourhood? According to Abrams, it is an 'effectively defined territory or locality inhabited by neighbours' (Bulmer, 1986: 21). This is a problematic definition: is it really the case that all people in an 'effectively defined territory' are neighbours with each other? We are well advised to look at other definitions. According to Heberle (1960), a neighbourhood 'may or may not have physical boundaries and a name' (p. 3). A busy road or a canal may cut through an urban area, separating two different neighbourhoods, or a forest or some other empty space between two residential areas may effectively delineate distinct neighbourhoods. For Morgan (2009), a neighbourhood is a 'locality that has some kind of meaning for the participants' (p. 11). He does not specify what the meaning consists of. A minimum level of meaningfulness would surely be that both residents and observers identify the area and its residents as distinct from other areas and their residents.
Both Abrams and Morgan stress that different actors may draw the boundaries of a neighbourhood in different ways. A researcher, a statistician, or a planner may define 'neighbourhood' objectively based on geographical or architectural characteristics. Residents may define their neighbourhood differently, such as based on social definitions of 'us' and 'them' or by their time-space paths in the residential area (see also Burrell, 2016Burrell, : 1603. There are thus two perspectives on the concept of 'neighbourhood': the 'objective' conceptions of observers and the 'subjective' conceptions of members. Such an 'objective' definition is based on the built and natural environment of the residential area with its physical boundaries as well as its history as a community. Neighbourhood as a material and historical structure may be more enduring than residents who happen to live in it at a particular point of time (Burrell, 2016(Burrell, :1613. I refer to this conception as neighbourhood O . However, there is also the members' neighbourhood M , which is defined by the residents' imagined geography of a neighbourhood, determined by the places and people that residents encounter regularly in their residential environment (Rosenblum, 2016). Such a conception of neighbourhood raises issues of residents belonging (or not belonging) to and feeling at home in the neighbourhood (e.g. Watt, 2009 or May andMuir, 2015). Feelings of belonging to a place may depend on inclusion or exclusion in neighbour networks (Burrell, 2016;Elias and Scotson, 1994;Swann and Hughes, 2016). Those having a longer history and established neighbour networks in the area felt suspicion and resentment towards newcomers and excluded them from their neighbour networks. Consequently, both groups built their own geographically distinct neighbour networks. When social and symbolic fences limit neighbour contacts, the members' neighbourhood M may be different for different members (Burrell, 2016(Burrell, : 1603. Moreover, people use the neighbourhood differently, depending on their situation and background (Blokland, 2003), and these uses contribute to defining their neighbourhood M . For some, the neighbourhood is a place to sleep in, while for others it is the site of all daily life. Therefore, neighbourhood O and neighbourhood M need not coincide, and Abrams thinks that there is a necessary mismatch between the two.

Defining the neighbour
Based on the above discussion, I shall first develop an observers' definition of residential neighbour (neighbour O ). Despite its everyday familiarity and its apparent simplicity, the term 'neighbour' is a polysemic and somewhat vague. Therefore, an explication of the term, based on what I consider our basic intuitions of the concept, is needed to be clear about what kind of social relation this article is talking about. Moreover, it opens the possibility of discussing one of the dimensions of the neighbour relation, that of recognition, leading to a second definition (neighbour M ). For simplicity, I will restrict my discussion to persons and households as neighbours and will exclude collective or corporate neighbours, such as pubs, shops, or schools. I will add now one aspect to physical proximity and living in the same residential environment. The people who we call neighbours have the capacity to affect our private lives, both positively and negatively (e.g. Allan, 1989;Rosenblum, 2016) According to Silverman (1986), neighbours are people who live close to the private space of one's home (p. 314; italics added).
Here is the definition: Ego and Alter are residential neighbours O if and only if (i) Ego and Alter live in the same residential neighbourhood O (ii) They live in dwellings that lie physically so close to each other that (iii) Ego's home life can affect Alter's home life, and vice versa.
'Home life' refers to all activities at the home and its environment: paid work, household work, recreation, and socialising. The influence that Ego and Alter may have on each other's life may be positive, like when neighbours are helping each other, or negative, like when neighbours are causing each other nuisance. The fact of living close to each other makes Ego and Alter interdependent (Blokland, 2003). They can contribute to creating collective goods in the neighbourhood, like peace, tidiness, and order, but they can just as easily contribute to collective bads, like neighbour disputes, feelings of insecurity, and neglect regarding the residential environment. Coming back to my Lapland example, this way of defining the neighbour relation makes calling people who live tens of kilometres away from each other neighbours metaphorical, as clause (iii) is hardly fulfilled. This is clearly an observer's definition, as it does not require that Ego recognises Alter as his or her neighbour, or vice versa. To be neighbours O , Ego and Alter simply have to be 'geographically available to each other' (Grannis, 2009: 10). In this definition, the word 'can' is italicised in order to highlight that neighbours need not affect each other's home life, but that they can potentially do so. The observer's definition does not require that there should be any intended interaction, nor compassion, between neighbours. Whether neighbours know each other, interact with each other, or feel sympathy or antipathy towards each other are contingent qualities of the neighbour relation, not its defining characteristics.
Recognised neighbours (neighbours M ) are a subset of all neighbours O , and arguably the most important one. Sometimes, people lament the fact that they do not know their neighbours, and sometimes they boast about having exceptionally good relations with their neighbours. All such statements are about people they recognise as neighbours. Some researchers approach the definition of 'neighbour' entirely from the members' perspective. Fischer's distinction between 'just neighbours' and 'real neighbours' is indicative. People who live near each other are 'just neighbours', but the concept can also have a 'weightier meaning' referring to 'an intimate social group composed of people who live nearby (what many might consider to be "real neighbors")' (Fischer, 2004: 131). However, real neighbours are not just recognised but also interacting in some deeper meaning than just saying hello. Rosenblum (2016) defines 'neighbour' explicitly from members' perspective: But for us personally and individually, who counts as our neighbor is not automatic with proximity. Location is not enough. We don't assign the status of neighbor to everyone residing in an officially designed neighborhood. -For neighbor is a matter of both location and personal knowledge, of recognition. (pp. 23-24) Rosenblum (2016) does not require that neighbours M should be an intimate group, as it is enough that they 'emerge from the background to affect the quality of our lives' (p. 24). Therefore, if the observer's definition refers to the potentiality of neighbours influencing each other's lives, Rosenblum stresses the actuality of such influence: By affecting the quality of life, Rosenblum (2016: 24) refers to contact with them, and, as a corollary, people next door are not neighbours M if there is no actual contact with them. This is perhaps too restricting. Merely recognising people in the neighbourhood as neighbours can sometimes be affecting the 'quality of our lives' in that it can contribute to a feeling of belonging and security in the neighbourhood (Henning and Lieberg, 1996). Even sounds that neighbours make can be experienced as positive (Lewis, 2020: 104). However, a recognition of neighbours appearing threatening or making inconsiderate noise can also diminish these feelings.
Thus, the definition of 'neighbour' from members' perspective would be the following: Ego and Alter are residential neighbours M if and only if (i′) Ego and Alter recognise that (ii′) They live in the same residential neighbourhood M (iii′) In dwellings that lie physically so close to each other that (iv′) Ego's home life can affect Alter's home life, and vice versa In the observers' definition, the clause about recognition (i′) is irrelevant. Here it is central, and above it is assumed that it is mutual. However, this need not be so. Instead, it may be the case that Ego recognises Alter as a neighbour but Alter does not recognise Ego as one. This makes the members' conception of who their neighbours are subjective, and therefore, Alter's neighbour M may not be Ego's neighbour M .

The nature of neighbour relations
David Morgan writes that individuals 'have three sets of people within their social horizons' (Morgan, 2009: 12): intimates, strangers, and acquaintances. Friends, partners, and family are intimates. With them, we share and exchange intimate knowledge about ourselves, feel emotionally close, and share physical affection. At the other end of the spectrum are strangers, of whom we know, and with whom we share, very little. Between these extremes are acquaintances, people we know a bit about and with whom we share a little. A crucial distinguishing characteristic between acquaintances and strangers is the lack of interchangeability: an acquaintance is just that person, not anybody, whereas a stranger can be just anybody (Morgan, 2009: 16). According to Lofland (1989), these three different sets of people typically (though not exclusively) interact in different realms of urban life. The private realm of the home is a space for intimates 1 , while the public realm of the street is for strangers. The realm of 'acquaintances and neighbors' (Morgan, 2009: 19) is the parochial realm, with workplaces and neighbourhoods cited as examples. Therefore, the social horizons are partly, though not entirely, associated with different kinds of place.
The category of neighbour O crosscuts Morgan's categories of personal relations and Lofland's realms of urban life (Kusenbach, 2006: 280). Neighbours O can be strangers, intimates, or acquaintances. On one hand, neighbours we know little about, have no contact with, and perhaps even do not recognise as neighbours are strangers to us (neighbours M cannot be total strangers (Rosenblum, 2016: 40)). On the other hand, our relations with our neighbours may grow so close that they become intimates, rather than mere acquaintances. Keller (1968) points out that when neighbours become friends 'the friendship relation usurps the neighbor relation' (p. 25), meaning that the intimacy of the friendship relation overrides the level of intimacy accorded to a person as a 'just neighbour'. However, our neighbour friend does not cease to be a neighbour: she or he is a neighbour AND a friend (an intimate).
We see here two conceptual distinctions (also Rosenblum, 2016: 40): being a neighbour or not being one, and being an intimate, an acquaintance or a stranger. The two dimensions are sometimes confused because we often think of neighbours as only acquaintances -people we know by appearance and name, stop to chat with in passing, exchange small favours, and share some minor intimate information with. The research literature usually assumes that the neighbour relation is an acquaintance relation (e.g. Kusenbach, 2006;Morgan, 2009;Rosenblum, 2016). This assumption is also empirically supported, as studies asking people about their contacts with neighbours show that mostly these contacts are so weak that they should be classified as belonging to the acquaintance category (see Henning and Lieberg, 1996).

The neighbours' community?
From the observers' perspective, our neighbours are not only those who we wish to be in contact with, but everyone living close enough within the neighbourhood. Therefore, neighbour relations should not be seen 'solely as matters of individual choice' (Silverman, 1986: 314;also Rosenblum, 2016). There are limits to the extent that we can choose our neighbours O (e.g. Lewis, 2020: 96). What can be chosen is the neighbourhood, though that choice does not always guarantee that our neighbours are the kind we expect them to be. The wealthy have a good deal of choice regarding the neighbourhood in which they live, whereas the poor have little choice. However, the choices of the wealthy may also be constrained due to status reasons. According to Michel Pinçon and Monique Pinçon-Charlot, the residential choices of the French haute bourgeoisie are limited to a few central neighbourhoods in Paris (les beaux quartiers). Due to their socially conditioned choice of neighbourhood, the elite are more segregated than the poor, usually studied by segregation researchers (Pinçon-Charlot and Pinçon, 2018).
According to Silverman (1986), people living close to each other share a common fate: 'factors such as relocated bus line, gentrification or a good local school will affect all' (p. 314). Though all are affected by the same changes in services, redevelopment, road plans, and so on, all are not necessarily equally affected. A nearby football stadium can represent a great nuisance for some, whereas others welcome the short travel distance on a match day. Location can be important: those living close to a bar opening on the block may feel a loss of peace and quiet, whereas others living slightly further away may welcome the new meeting place. Though Silverman's shared fate is not equally true for all, neighbours are nevertheless people who potentially share locality-based interests and can potentially form a community furthering them.
However, this is not a defining characteristic of neighbour relation, but one of its socially significant aspects. Research on neighbourhoods has to a large degree centred on the question of community. Local community can build on culture and sentiments, but also on shared interests -or interdependence. According to Norbert Elias' (1974) quite narrow conception of community, there is 'local community' to the extent that people's web of interdependencies is more local than supra-local. This definition of community makes the concept independent of any assumption of local identification or 'spirit'. According to Blokland (2003), neighbour relations are constituted by interdependence, in which 'people are bothered by the behaviour of others' or 'the presence of others is a condition of a certain event' (pp. 78-79), such as collective action to defend the neighbourhood. Like in the case of local community, the neighbourly spirit is not something that is inherent in neighbour relation, but rather something that has the potential to arise out of it.

Neighbourly interaction
Abrams defines neighbouring as the actual (observed) patterns of interaction between neighbours (Bulmer, 1986: 21). However, it should be added that neighbouring is interaction undertaken in the role of neighbour (Keller, 1968: 29). Some interactions between neighbours could have more to do with their roles as friends, relatives or lovers, rather than 'just neighbours'. This was the case in Young andWilmott's (2007 [1957]) classic 1950s community study of East London, where the researchers discovered that neighbours in Bethnal Green were often relatives or childhood friends. Keller (1968) distinguishes between interaction that is 'a predictable core element based on the neighbour role and additional nonpredictable elements reflecting the social and personal context in which neighboring typically takes place' (p. 30).
There is a certain ambiguity in the term 'neighbouring'. In contrast to Abrams, Morgan points out that it is generally a positive term used to refer to, for example, neighbours' 'willingness to carry out some chores for each other or adopting a general watchfulness on behalf of the other' (Morgan, 2009: 19; also Grannis, 2009;Kusenbach, 2006). However, not all interactions between neighbours are neighbourly involving consideration, mutual help, commitment and care (Bulmer, 1986: 21). Bridge et al. (2004) have aptly pointed out that there might be quite a lot of interaction between neighbours without there being a friendly relation between them: 'We could, for example, spend much of our time knocking a neighbour's door to complain about loud music'. To avoid ambiguity, it would be advisable to avoid the term 'neighbouring' and speak simply of 'neighbour interaction', which can be neighbourly or unneighbourly.
There are two possible perspectives regarding the term 'neighbourly', and these can, with some difficulty, be seen to represent the outsiders/members distinction. 'Neighbourliness' could mean Ego's positive attitude to her or his neighbour Alter, 'which result in positive action when a need arises, especially in times of crisis or emergency' (Mann, 1954: 164;also Bulmer, 1986). Mann points out that latent neighbourliness is also possible: there may be positive attitudes without much interaction (also May and Muir, 2015). Ego and Alter may be favourably disposed towards each other, but their daily time-space paths are different to the extent that they rarely meet. What is neighbourly may not require much. The residents May and Muir (2015) studied valued in neighbours 'the absence of negative attributes: noisiness, bad behaviour, messiness and so on' (p. 4.4).
Positive neighbourly interaction can be seen as an occasioned activity that arises from when the situation allows for it (Laurier et al., 2002). For example, in a neighbourhood with unlively neighbour relations, there might be a road plan coming that the residents are worried about. This brings them together and increases interaction between them. If they agree on the issue, and the interaction is rewarding, more interaction that is neighbourly is likely to follow. This may become permanent or only until the threat to the community is resolved. In the latter case, the ties that have become close and amiable fade out. Things may also change for the worse. The fight against the road plan may bring out differences of opinions and interests, creating conflict. A warm, neighbourly relation may turn into resentment, with lively interactions becoming sparser.
Should neighbourliness be seen as merely a question of attitudes? A neighbour relation where there is an asymmetry of attitudes is also possible. For example, Ego experiences Alter as an intrusive busybody, whereas Alter feels that she or he is just acting neighbourly. This may affect the quality of the relationship, but not necessarily. Ego may continue to act neighbourly, though perhaps unenthusiastically. This may be because that kind of behaviour is habitual and expected in the neighbourhood, and therefore there is social pressure towards it. Neighbourliness can thus also be understood as a quality of neighbours' interaction abstracted from their subjective orientation towards each other. This is similar to Rosenblum's (2016) argument that neighbourliness is not an emotion, but a practice: 'a set of hard-won, complicated habits' (p. 5) that support or increase neighbours' 'quality of life'.

Levels of neighbourly interaction
Proximity, the defining characteristic of neighbour relation, does not in itself entail interaction between neighbours, it only creates an opportunity for it (Burrell, 2016(Burrell, : 1609(Burrell, -1610. In his book From the Ground Up, Rick Grannis (2009) presents a stage model of neighbouring, where the perspective is shifted gradually from mere presence in the same residential area to the possibility of contact, to actual contact and co-operation, and to the sharing of norms and values. Each stage logically precedes the next stage, though the idea is not that neighbour relations would tend to develop according to the model, but rather that when a relation has reached the most advanced stage, it has gone through all the others. Below are Grannis' definitions of the different stages of neighbour relation (I have substituted his actors i and j with Ego and Alter, as well as term 'neighbouring relation' with 'neighbour relation'): A stage 1 neighbour relation exists between Ego and Alter if they are geographically available to each other (Grannis, 2009: 20).
A stage 2 neighbour relation exists between Ego and Alter when they unintentionally encounter each other and thus have the opportunity to acknowledge each other's presence, observe each other and initiate conversation (Grannis, 2009: 22).
A stage 3 neighbour relation exists between residents Ego and Alter, if they have intentionally initiated contact (Grannis, 2009: 25).
A stage 4 neighbour relation exists between residents Ego and Alter if they engage in one or more activities, indicating mutual trust or a realisation of shared norms and values (Grannis, 2009: 26).
Only stages 3 and 4 represent what above has been called neighbouring, that is, interaction between neighbours. Being 'geographically available' is one of the defining features of being a neighbour O . Grannis' stage 2 neighbour relation refers to the time-space paths of neighbours: in order for interaction to happen, there has to be a possibility of the parties in question meeting. However, from the perspective adopted here, stages 1 and 2 are not neighbouring, but rather conditions for it to happen. It is slightly awkward that stage 3, where people initiate contact, is followed by one where they trust each other and share norms and values. Grannis appears to be making fine distinctions between the conditions of interaction between neighbours, but does not do this in case of neighbouring itself. In fact, stage 4 is, in a way, illogical in that it does not fit the process that Grannis dissects with his distinctions between stages. Stages 1-3, however, can be seen as necessary conditions for any deeper interaction between neighbours.
Grannis' analysis can be complemented by distinctions regarding the different kinds of contacts developed by Henning and Lieberg in their study of neighbouring in a Swedish neighbourhood. They distinguish between three levels of contact between neighbours (Henning and Lieberg, 1996: 11 and 18). At the level of acknowledge-contacts, there were neighbours to whom the respondents said hello when seeing them in the neighbourhood. The next level of contact was greeting-contacts, referring to contact with people in the neighbourhood who the respondents not only said hello to, but also stopped to talk with. The third level was helping-contacts, who the interviewees had with neighbours with whom they exchanged small services, like taking care of mail or looking after flowers. The authors also identified absent contacts on the borderline of weak contacts: people the interviewees could recognise by face, but had no contact with (Henning and Lieberg, 1996: 17).
'Neighbourliness' is not a term that Grannis himself uses at all. However, this concept nonetheless enters the model in stage 4, although Grannis (2009: 25-27) is careful to emphasise that even a stage 4 relationship does not have to be intimate, affective, or friendly. What he is interested is community action -whether neighbours are capable of acting collectively towards the common good. He argues that such action does not require that neighbours are friends with each other, only that they trust each other, with this trust dependent on shared norms and values. Grannis' model covers quite well the various aspects of neighbour relation. A neighbour is a person who lives close to us, has the opportunity to encounter us, may interact with us, and with whom we might work together for the betterment of our neighbourhood.
Grannis' model, complemented with Henning and Lieberg's ideas, can be used to consider the conditions and levels of neighbour relations:

Conditions of neighbouring
(a) Ego and Alter have to be geographically available to each other. (b) Ego's and Alter's time-space paths have to be such that they have a possibility of meeting.

Levels of neighbouring
(i) When meeting, Ego and Alter signal that they recognise each other as living in the same neighbourhood. (ii) When meeting, Ego or Alter initiate conversation with each other. (iii) Ego and Alter exchange services (like watering the other's flowers when she is away) or favours (like lending tools). (iv) Ego and Alter act together towards a common goal in the neighbourhood in their role as neighbours (organising a garden party, signing a petition against a planning decision, etc.).
Levels (iii) and (iv) are close to Grannis' stage 4 neighbouring, although the idea of sharing values seems quite strong in connection with acquaintance relations. However, Grannis does stress that Ego and Alter act in a way that indicates mutual trust and shared values. Therefore, trust and shared values are inferred from co-operative behaviour: the fact that Ego and Alter are jointly preparing a garden party for the residents of the house is evidence of trust and shared values. However, the sharing of values may not need to be especially deep in such a scenario. The minimal agreement is that a garden party is worthy of the effort put into it by volunteers.

Cultures of neighbouring
'Being a neighbour is a role, not an existence', stressed Abrams (Bulmer, 1986: 21). In encounters with the people that live close to us, we are expected to play the role of neighbour; that is, to behave in a manner that is expected of neighbours locally. Schiefloe (1990) maintains that the role of the neighbour is 'a part of a culturally defined system, which by means of norms and sanctions controls neighbourly behaviour' (p. 99). Abrams disagrees: in contrast to 'co-workers and co-religionists, their [neighbours] interactions are distinctly not norm-governed' (Bulmer, 1986: 15). What Abrams probably means here is that formal rules for neighbourly interaction do not exist, in contrast to rules of workplace and congregation. That neighbour relations are devoid of formal regulations is, however, not entirely true. In societies, there are numerous legal regulations that govern the behaviour of neighbours towards each other, including laws about respect of privacy, property law, and tenancy contracts, while apartment houses usually have their own rules and regulations for their residents. Noting this, Rosenblum (2016: 5-6) nevertheless emphasises the informal nature of neighbour relations.
An informal culture of neighbouring that exists outside the formal rules that frame neighbour relations can be argued to consist of conceptions and norms about neighbours' rights and obligations, such as the following: (i) The mutual help and co-operation that are customarily required or considered normal between neighbours; (ii) The customarily required or expected sociability towards neighbours; (iii) The appropriate extent of intimate information shared with neighbours; (iv) The extent to which neighbours are allowed to interfere with one's or one's household members' actions.
To comply with the norms and conventions is to act as a 'good' neighbour. However, cultures are not immutable, clearly defined, and necessarily shared by all in a neighbourhood, society, or community. The culture of neighbouring may be the conception of 'what anybody would do here' (Rosenblum, 2016) regarding those established in the neighbourhood, with outsiders coming to live in the neighbourhood having to learn to comply with these conventions (cf. Elias and Scotson, 1994).
One overarching theme of the literature on neighbour relations is that of the thinning out of obligations attached to the neighbour role with the modernisation and individualisation of societies. Being a neighbour in traditional European rural, as well as pre-modern urban, society was different, bringing to its incumbent various customary obligations (Cockayne, 2012;Heberle, 1960). Historically, 'neighbour' has not only referred to persons who reside close to each other, but also to persons who are 'socially related by specific rights and duties established by custom and, in part, by law -because of the proximity of the dwellings' (Heberle, 1960: 3). Heberle refers here to Max Weber 1978(1921, who in his Economy and Society discussed neighbour units in terms of 'that somber economic brotherhood practiced in case of need' (p. 363). Neighbours were obliged to help each other when needed, and they were also entitled to a certain degree of hospitality: 'where a child is baptized, or married, the neighbors must be invited' (Heberle, 1960: 4).
Moreover, the obligations attached to neighbour relations had little to do with emotional ties: 'Neighbors will do certain things for each other, whether they like each other or not' (Heberle, 1960: 9). There is evidence showing that this kind of obliging neighbourhood custom does survive in some parts of the world, even in the 21st century (see, for example, Heil, 2014). In his fieldwork in the Bosnian countryside, Henig found that the traditional neighbourhood community komšiluk was still very much alive. The neighbours were engaged in a ritual pattern of visits, called sijelo, whereby each member of the neighbourhood community was expected to be a good host, as well as a good visitor. Furthermore, this exchange of hospitability happened even though some of the participants had bitter memories of their neighbours' behaviour during the Bosnian war. (Henig, 2012: 8-13.) In modern societies, the 'somber economic brotherhood' of shared poverty does not support the kinds of obliging neighbour relations that Heberle was writing about. In the past, neighbours relied on help from each other in many kinds of situations. By contrast, in modern societies there are specialised private and public institutions and organisations designed to provide help in these situations, such as childbirth, death, sickness, and accidents. The norms of the mutual exchange of labour services and hospitality, partly dictated by law and partly by custom, have been relaxed. Both establishing neighbour relations and their nature have now become more a matter of choice. The less neighbours need each other, the less clearly and rigidly defined the role of the neighbour and the associated norms have become (Keller, 1968: 23). However, at the very end of his article, Heberle (1960) points out that 'even in the city a minimum of neighbourhood custom survives along with legal regulations' (p. 11).
Claude Fischer, who summarised studies on contemporary US society, has charted what this minimum level is. Fischer (2004) suggests that there are two norms associated with the neighbour role: The first is: Be ready to assist your neighbor at those times when physical proximity is important -either in an emergency, or when the assistance costs little and it would be silly to force the other person to go long distances for it (for example, in the United States, loaning a cup of sugar). The second norm is simple: Don't be offensive. Good neighbors don't disturb or offend the people next door (or lower the property value of their dwelling). (p. 136) According to McGahan (1972), a good neighbour is 'someone who is friendly, but not a friend, someone willing to chat with you, but at the same time someone who does not intrude your privacy' (p. 402). These American accounts of what kind of behaviour we expect of our neighbours probably resonate quite well with the conceptions of the minimal requirements of the neighbour role in many other modern societies. We expect some kind of neighbourliness, some level of attention and consideration, however minimal, from neighbours. In addition, we are also ourselves obliged to behave neighbourly towards our neighbours. Respect of privacy is a vital part of the expected consideration. We need to be able to manage our relations with our neighbours so that an appropriate balance between social nearness and distance is achieved (e.g. Boyce, 2006;May and Muir, 2015;Rosenblum, 2016).
What maintaining an appropriate balance consists of may vary across different times and places. The cultural differences in neighbouring may be due to various bases of difference, including ethnic group, locality, generation, social class, and religion. The content of being a good or bad neighbour, that is, performing the role of 'neighbour' either according to or against social expectations, is likely to be subject to some (but perhaps not very large) socio-cultural variation, both historically and across cultures and subcultures (see Morgan, 2009 andKusenbach, 2006).

Conclusion
The article aims to develop a theoretical framework that would contribute to a better sociological understanding of neighbour relations. That has involved discussing and redefining a number of central concepts in the topic (neighbour, neighbourhood, neighbouring, and neighbourliness), as well as surveying and synthesising various sociological ideas about neighbour relations. The proximity of living in the same space is a basic defining characteristic of a neighbour relation but it is not sociologically particularly important as such. It creates a social and material possibility for social contact, for cooperative as well as conflictual social action (in the Weberian sense) by actors finding themselves in the context of a neighbouring culture. Such a culture suggests them roles and norms to follow or not to follow. The article is for the most part about the 'anatomy' of neighbour relations, that is, about what such relations consist of. Their 'physiology', the way neighbour relations happen and function, is not in the same way at the focus of this article.
I have argued that neighbour relations require special kinds of activity involving regulating the social distance between the parties. Such activity is a subtle matter utilising verbal and non-verbal means of communication. It is also occasioned activity, meaning that in ordinary circumstances interaction with neighbours happens in the background of social life. When help is needed or disturbances happen, the neighbour relation enters the front stage. Both neighbour help and disturbances link with the prevailing cultural norms and practices concerning the neighbour role. In contemporary European societies, these are not in the same way formal and obliging as in the past. The informal norms of neighbouring in these societies concern a limited reciprocal help between neighbours and respect of each other's privacy.
Neighbour relations are created and maintained in micro-level interaction between residents in a neighbourhood. This has implications for the ways of investigating them. Survey evidence is appropriate in investigating the frequency of contacts with neighbours, the prevalence of neighbour disputes, the differences in neighbour relations in different groups and residential environments, and so on. Such knowledge is important, for example, in evaluating general statements about the diminishing importance of neighbouring in modern societies. However, the ways how neighbouring happens and is patterned in different environments can be studied better with less structured methods than the survey (see Schiefloe, 1990: 99 andKusenbach, 2006: 283).
When the nature and experience of neighbouring in is investigated, more ethnographic methods than survey are needed. Approaching neighbour relations ethnographically requires that researchers participate in the life of the neighbourhood they are researching. By becoming quasi-members of the neighbourhood, the researchers get a grip of people's practices of creating and maintaining neighbour relations and the cultural norms that support them. As important as this is investigating the ideas and practices through which residents draw boundaries between those who do belong to their neighbour networks and those who do not. Cross-cultural research along these lines would be most interesting.
Across the industrialised world, national and local governments are struggling with problems concerning providing care for ageing populations, loneliness, exclusion, and urban disorder. In an era of fiscal problems regarding care and ideological shifts towards less public involvement in the provision of services, calls for solidarity and neighbourly support between citizens have become increasingly common. Europe's rising immigration levels will bring to our neighbourhoods strangers with different habits and cultures of neighbouring, potentially leading to new kinds of neighbour disputes. While professional mediators can be used to resolve and prevent these issues, it would be more sustainable and efficient if residents themselves could address their problems. In times like these, Fischer's 'real neighbours', who form neighbourhood communities, are in great demand. In many languages, the word 'neighbour' carries a double meaning. It signifies not just a person who lives close to us in a residential space, but also a fellow human being towards whom we should feel compassion.