What happened to body-to-body sociability?

: This article aims to investigate how the body-to-body forms of sociability evolved from 1996 to 2009 simultaneously with the proliferation of ICTs in Europe and why this happened. The article also aims to find out how the socio-demographic profile of Europeans practising these forms developed in the same period of time. The analysis is based on two surveys carried out in Italy, France, the United Kingdom, Germany and Spain in 1996 (N=6,609) and 2009 (N=7,255). Results show that although the internal diffusion and frequency of the forms of communicative sociability changed, on the whole the amount of sociability has increased so slightly that it would be more appropriate to speak about real stability over the time. Secondly, results reveal that the possession of mobile phones and personal computers in 1996, and respectively the Internet in 2009, was especially associated with the increase in sociability. Lastly, the socio-demographic profile of the Europeans practising these forms of sociability changed between 1996 and 2009, although less than one might have expected.

This question has remained interesting to investigate because previous results are contradictory.
Their inconsistency depends on many reasons. First of all, various concepts, such as contact, social bond and relationship, are arbitrarily used synonymously with sociability, creating a lot of confusion because they refer to different notions. The term 'contact' puts the accent on the beginning of social interaction, the phatic dimension of the relation, to borrow Jakobson's definition (1960), and indicates merely that one knows how to trace the person in question. Hence, this term does not grasp the intensity and solidity of the relationship. The phrase 'social bond' stresses the outcome of meaningful social relationships that cements a close tie between individuals (Hirschi, 1969;Krohn & Massey, 1980). Finally, social relationship makes a reference to the social interaction that occurs between two or more individuals (Giddens, 2006). Using these words indifferently does not promote clarity of analysis and research. Secondly, sociability is associated with a wide range of practices, which have been investigated one by one and by using different methodologies. Thirdly, the inconsistency of the available results stems from the differences in research designs and the lack of standardized measures (Zhao, 2006). Diverse variables have been applied to explain the variability in people's sociability. As a result, there is a theoretical dissatisfaction which relates to a 'taken-for-granted and careless application of the concept of sociability.
The aim of this article is to investigate only the most common forms of body-to-body sociability. The most common forms of sociability were chosen based on a careful analysis of previous sociological studies. This analysis was followed by an empirical pretest of the questionnaire with 100 respondents which verified the most practised sociability forms. This study focuses on how these forms evolved from 1996 to 2009 -the time when ICTs truly proliferated in Europe -and on the main reasons for their evolution. The expression 'body-to-body sociability' is utilized by drawing upon a theoretical work published by Fortunati (2005). Fortunati has challenged the 'face-to-face' expression as a huge amount of research shows that we communicate not only with our words, gazes and facial expressions but with all our body, which is the main site of the non-verbal language (i.e. sweats, tremors, gestures, spatial positions and postures). Thus, it is paradoxical that communication studies continue to use the reductive metaphor 'face-to-face communication', while as Fortunati writes (2005: 1), body-to-body communication 'expresses more accurately all the richness of communication between copresent individuals'. Furthermore, the article aims to find out how the socio-demographic profile of Europeans practising these forms evolved simultaneously and elaborates reasons for this evolution. The study will look at how the possession of various media and newspaper reading are related to practising concrete forms of body-to-body sociability. Newspapers are investigated among ICTs since a recent study (Fortunati, Deuze, & de Luca, forthcoming) shows that print, online and free newspaper users are more likely than non-users to actively engage in socio-cultural forms of sociability.
From the above aims of this study it follows that the electronically mediated forms of sociability are not addressed here. Instead, while previous studies have mainly focused on one technological medium at a time (e.g. about the Internet, Rice et al., 2007; about the mobile, Campbell & Kwak, 2010), the aim of this study is to contribute to a broader picture by looking at the possession and use of different technologies in one and the same study. These technologies include television, mobile phones, personal computers and the Internet as well as newspaper reading. The most common forms of sociability will be summarized in an index that is envisaged to help us to analyse trends in people's sociability. The study is based on two telephone surveys carried out in Italy, France, the United Kingdom, Germany andSpain in 1996 (N=6,609) and2009 (N=7,255).
Finally, previous studies present an unsystematic exploitation of the concept and measures of sociability, which has negatively affected the consistency of the previous findings and the possibility of comparing them. Hence, in the following section the concept of sociability will be scrutinized with the purpose of clarifying the theoretical premises of our study. It will be shown that this study is grounded on the meaning of sociability, which emphasizes its concrete expressions.
Furthermore, it will be shown that the forms of sociability that are under investigation represent a fundamental part of the historical development of this concept in sociological discourse. The next section will also reveal the need to expand the concept of sociability in the light of other fundamental sociological categories, such as communication, mobility and labour.

The concept of sociability
The term sociality has two meanings, the first of which is the tendency to live in society understood both as an attitude of human beings to do so and as a psychological disposition to do good things for others (Amirou, 1989). The second meaning is the ensemble of intercurrent relations or interactions among the individuals who are part of a given society. However, another term, sociability, which refers to the capacity to socialize, intersects with sociality. Sociability has a positive meaning as benevolence towards others, but also a negative one as behaviour against the common and general good. Over time, the meanings of politeness, kindness, pleasantness and civilization have also converged in this tenable sociological concept (Gemelli & Malatesta, 1982: 11).
Several disciplines, one after another, have contributed to the sociability concept: from sociology to history, from anthropology to economy, and from psychology to ethnography. In German philosophy, Kant (1784Kant ( /2010, for example, introduced the notion of 'unsocial sociability' whereby human beings are inclined both to associate themselves with and to isolate themselves from others. Schleiermacher (1799Schleiermacher ( /1995, for one, considered that free sociability is recognized as a fundamental need of every educated human being. Scottish philosopher Adam Ferguson (1767) discovered the 'law of sociability', which says that living in society comes naturally to humankind and hence the state of nature is a social state.
In German sociology, the discussion about sociality was furthered by the sociology of forms through scholars such as Tönnies, Weber, Simmel and von Wiese. But in general, the study of sociability became the specific purpose of social morphology which had the mandate to investigate the forms of sociality. In the French world, the contribution of Gabriel Tarde (1893) was fundamental in examining the historical transformation of sociality forms, in particular of friendship and its change after the development of urbanization. A few years later, Bouglé (1902) broadened the idea of sociability by proposing that association and cooperation should be included in it as well. By studying pauperism and the life of the working classes, social economy focused on moral education and rationalization of leisure time. Social economy saw education and leisure time as areas where it was possible to impose a model of a controllable and ordered sociability (Gemelli & Malatesta, 1982: 19). Simmel's (1910Simmel's ( /1949 contribution to the elaboration of sociability is generally considered fundamental. But while Simmel reduced sociability mainly to individual interaction, Halbwachs (1933) investigated the forms of sociability by analysing those connected to the lifestyle and mentality of the working class. Some years later, Gurvitch (1938) identified the study of sociability forms as the starting point for understanding and investigating the entire society. Focusing on the structural characteristics of the forms of sociability, he pointed out that they do not belong to the sphere of the totally spontaneous or to the sphere of the totally organized; they do not identify themselves with stable collective units but they are rather particular forms, ways of being of these units. It is exactly their relative evanescence that has made, in general, their analysis elusive.
In those years in France, the experience of Les Annales developed a broad vision of everyday life practices, ritualization and ceremonies. Thereafter, in the forms of sociability, termed by Goffman (1971) as forms of focalized interaction, many social dimensions such as politics (strikes, forms of resistance), religion (ceremonies, feasts, charity institutions), market (markets, fairs, shops), prostitution, playfulness (carnivals, games, pubs, cafés, taverns), communication (direct or mediated by information and communication technologies), culture (from theatre to singing and dancing), family (visits, common events), as well as the overall maintenance of social relationships, are seen converging. It is inside these forms of sociability that men and women can also meet each other, choose their partner and date him/her.
To sum up, the study of sociability forms has been central to sociology as it fosters fundamental sociological themes such as social solidarity, integration or conflict and social cohesion. Over time, the term 'sociability' has superceded the term 'sociality', although the latter would have good reasons to be maintained, and has settled fundamentally on two meanings: the generic propensity to establish any type of social relationship and the multifarious concrete expressions of this propensity as a group, social network, association, community or mass (Forsé, 1980;Amirou, 1989). Although there has recently been a certain overlapping of the notion of sociability with that of social capital, social capital has a distinct and specific sociological meaning.
This study is instead focused on two features that form the basis of the values of social relationships: the forms of sociability and the frequency of their occurrence; it is therefore grounded on the second meaning of sociability, which emphasizes the concrete expressions of sociability.
To illustrate further the notion of sociability, we propose looking at sociability as being closely interconnected to three structural concepts: communication, mobility and labour. The linkage between sociability and communication is evident. Let us think, for example, about primary socialization processes: how would it be possible to accompany a child into a society without teaching him/her communication skills? On the other hand, communication is not only needed to elicit sociability, but sociability also promotes communication, as became apparent in another study carried out by Fortunati and Taipale (2012: 34). Fortunati and Taipale argue that as communication is an action with a low output of energy, to be effective and to last over time it has to be embedded in social activities, which imply a higher energy requirement. For example, going with our friends to a restaurant or to the cinema helps to keep 'communication' alive. In other words, we need to do things with others in order to nourish the communicative process.
The other structural concept is mobility. To socialize with others and to be part of a society, it is necessary to get out of the house. Family relationships that exist within the four walls of the house are fundamental, yet not sufficient for a balanced development of the individual. Those who stay at home excessively long, such as housewives, may suffer from greater social isolation. We need to go out and move into public spaces in order to encounter friends or to meet new people. In this respect, sociability structurally involves mobility. The third concept is labour, and in particular care work, or, in more technical terms, reproductive labour. This is the work needed to reproduce the labour force, both in the sense of giving birth to children and in the shorter term of feeding and clothing, when not at work. Reproductive labour constitutes a large share of immaterial labour in which many different tasks, such as affect, love, sex, psychological support, knowledge sharing, entertainment and information, converge. All these tasks are, of course, conveyed by communication. Thus, it is labour which still involves women much more than men, which is the spine of the value production in the domestic sphere, and on which a broad feminist literature is available (Fortunati, 1981;Hochschild, 1983). Care labour grounds also on cooperation and organization and is hence a broader concept than mobility. In sum, to build concrete forms of sociability we need to work in coordination with others and to handle the organizational and logistic features of this work. In the light of these specifications it is clear that sociability is a process. It is a process that applies an intensifier logic: individuals feel more reassured and are more reassuring if they can practise any of the sociability forms together with another person.

Body-to-body forms of communicative sociability
To underline that sociability takes places in cooperation with others and necessitates movement as well as communication, as described above, this study talks about the body-to-body forms of communicative sociability. In this research we focus on the following forms of sociability, which have been chosen on the basis of an empirical pretest carried out as part of the surveys used here.
Visiting friends and relatives takes place between familiar 'faces' and is a response to the desire for physical proximity and intimacy. Gabriel Tarde (1899: 706-707) reconstructs how the rituality of visits to friends and relatives, including the offer of a gift, has developed in the course of time. In advanced societies, the high complexity of social life calls for mobility and thus, especially in urban environments, the sphere of interpersonal relationships is characterized by random encounters. People have the chance to meet up and socialize more randomly with close friends and family members than in spatially close-knit communities. The spatial displacement of family members, relatives and friends requires such a high level of geographical mobility that visits to friends and relatives have become the subject of travelling and tourism studies (Backer, 2007). The rarity of unplanned social encounters has resulted, according to some (Putnam, 2000), in a decrease in social capital, to borrow Bourdieu's expression.
The second dimension of sociability relates to going out to the cinema, theatre, opera, museums, libraries or exhibitions. Previous literature clearly indicates that the presence of one or more companion plays a great part in the visitor's motivation and satisfaction in places such as a museum. Impressions and experiences are shared with a companion, which may feel reassuring, especially for infrequent visitors (Debenedetti, 2003). The presence of a companion also generally makes a visit to a cultural attraction a sociable event.
The third aspect of sociability deals with going out to restaurants, pubs or dances. It is not surprising that this form of sociability has been much studied since the art of conversation was born around a laid table in court society (Craveri, 2005). Contemporary sociological literature on food consumption clearly points out that one of the main reasons that people like eating out is to socialize with friends and family (Finkelstein, 1989). Warde and Martens (2000: 205), for instance, show that it is not only the eating but the possibility of dancing afterwards that makes eating out with friends fun for young people in particular. Sociologists from Georg Simmel (1910Simmel ( /1949 to David Riesman (1993) all argue that such playfulness is an essential part of sociability.
The fourth form of sociability addressed here is going out with friends for a walk, going shopping or watching sporting events. This particular form of sociability reminds us of the flâneur figure: a gentleman exploring and experiencing the city by walking (Benjamin, 1935). Featherstone (1998) claims that this originally masculine phenomenon, which is said to be dying due to increased traffic and the lack of public spaces, has become feminized with the rise of shopping centres. De Certeau (1985) adds to the rhetoric of walking, seeing it as a style, a way of being and behaving on the part of dwellers. In fact, going out with friends is a common form of co-present sociability, which is said to take place typically after a work day (Lehtonen & Mäenpää, 1997). Sport spectating provides another arena for sociability. It is especially seen to serve the need for sociability in urban settings, where maintaining primary social ties is difficult due to distance and congested traffic (Melnick, 1993). Besides spectating sport, previous studies also deal with the sociability of doing sporting activities. Bourdieu (1991), for example, perceives sport as a field of struggles where economic and cultural capital operate as the main medium of distinction.
While meeting relatives and friends represents the informal part of social engagement, taking part in club, religious, trade union or political activities is its formal side (Coleman, 1990;Putnam, 2000;Campbell & Kwak, 2010). Putnam (2000) argues that it is the individualization of leisure time, especially the increased television viewing and the time spent on the Internet, that diminishes this kind of civic involvement. Other studies have, however, challenged this argument by bringing out other factors that modify or eliminate the association between the use of new media and the decline in social capital (Uslaner, 1998;Wellman et al., 2001;Katz & Rice, 2002;Sugiyama & Katz, 2003;Taipale, Oinas, & Salminen, forthcoming).

Are changes in sociability related to media?
Based on the above-reviewed literature and the following argumentation, the exact research questions and the related hypotheses can be formulated for the study. The first research question (RQ1) is: Did body-to-body forms of communicative sociability increase or decrease between 1996 and 2009? In recent sociological discourse, many scholars have argued for the increase of social isolation and decline of social capital (e.g. Fukuyama, 1999;Putnam, 2000;Pherson, Smith-Lovin, & Brashears, 2006). These studies have based their argumentations on different findings. For example, Putnam stresses that the decline results from inactivity in political participation and civic engagement; Fukuyama (1998)  According to the above-mentioned scholars, a great part of social interaction and communication that previously took place in the co-presence of others has become mediated by ICTs, and thus sociability has decreased (e.g. Putnam, 2000;Turkle, 2011). In fact, some early studies on the relationship between ICT use (such as the Internet) and physically proximate social interaction provided support for these claims. Internet use, for example, was seen to have a 'displacement effect' on physically proximate interactions and, hence, the time spent on the Internet was even considered as an asocial activity (e.g. Nie & Erbring, 2002;Nie, Hillygus, & Erbring, 2002). This particular discussion acts as a starting point for our second research question (RQ2): How does the possession of old and new media and the frequency of newspaper reading affect the body-to-body forms of communicative sociability?
Interestingly, many later studies have shown the opposite results. These studies, such as those carried out by DiMaggio et al. (2001) and Wellman et al. (2001), give strong reasons to believe that Internet users actually have larger social networks than non-users. In a more sophisticated statistical analysis, Zhao (2006, 13) shows that the type of online activities and the amount of time people devote to online activities are linked to offline social connection. It is more likely that people engage in larger social circles offline if they use the Internet for interpersonal contacts (e.g. email and chat) than if they use the Internet for solitary purposes (e.g. Web surfing). All these studies considered, findings show that there is not one but many ICT effects on sociability (e.g. Wellman et al., 2001: 451), and if each technology is investigated in a separate study the overall picture remains disjointed. We consider the studies which indicate that ICTs support and add to sociability to be the most reliable and up to date. Hence, in contrast to the general decrease in sociability between 1996 and 2009, we expect that: H2: The possession of new media and the frequency of newspaper reading are associated with the increase in body-to-body forms of communicative sociability.
Our last research question is (RQ3): Did the socio-demographic profile of the Europeans practising these forms of sociability change in these 13 years? To obtain answers to this question, a range of socio-demographic variables will be utilized. These variables include gender, age, family type, the degree of urbanization or respondents' place of residence, house size, mobility and quality of life. In fact, previous research also indicates that the impact of these factors on the forms of sociability has changed over time. For example, political participation and civic engagement have perhaps become less common among young people (Putnam, 2000;Pherson, Smith-Lovin, & Brashears, 2006), while going out with friends for a walk and shopping has feminized over time (Featherstone, 1998).
There has also been a trend towards smaller families, and one-person families have begun to move towards urban centres. Meanwhile, in most cities families with children are moving away from the urban centres and gravitating to the surrounding suburbs (European Environmental Agency, 2009: 27). Thus, it is plausible to believe that people's mobility and place of abode, as well as family type, had different effects on sociability between 1996 and 2009. Hence, we expect to find that: H3: The socio-demographic profile of Europeans regarding these sociability practices changed from 1996 to 2009.

Data and respondents
The data we present here belong to a unique research design composed of two nationally representative telephone surveys that were carried out in Italy, France, the United Kingdom, Germany and Spain (EU5). 2 The first survey was implemented in 1996 (N=6,609) and the second one in 2009 (N=7,255). This is a cross-national study which allows the sociability behaviours of the population of the five most affluent and populous European countries to be compared. The sampling procedure and the questionnaire were almost the same in both surveys, although it should not be considered as a true longitudinal study as the samples were not the same. In 2009, however, the questionnaire was adapted to the new technological situation. Questionnaires were pretested both in 1996 and 2009 with 100 participants in order to verify the appropriateness of the questions and the duration of the interviews. In this study we use weighted data which correct some distortions relating to age, education, ownership of a computer and access to the Internet. The respondents' profiles of the data sets are presented in Table 1. Both studies were funded by Telecom Italia.
Insert Table 1 about here

Dependent variables
In this study we use seven measures that are indicative of the forms of body-to-body sociability, as well as an index composed of them, as dependent variables. The seven main measures are the following: 1) Inviting to own home or going to visit friends 2) Inviting to own home or going to visit relatives 3) Going out to the cinema, theatre, opera, museums, libraries or exhibitions 4) Going out to a restaurant, pub or bar, or going dancing 5) Going out with friends for a walk, going shopping or watching sporting events 6) Going out to take part in sporting activities 7) Going out to take part in club, religious, trade union or political activities.
The answering categories were: 'Several times a week, 'Once a week', 'Once/twice a month', 'Less often' and 'No/never'. These answers were grouped into two categories: the first one (including 'Several times a week' and 'Once a week') represents respondents who perform these forms of sociability at least once a week (weekly), and the second one (including 'Once/ twice a month', 'Less often' and the remaining category 'No/Never') represents respondents who perform them at most once or twice a month. Several socio-demographic variables were included in the research. Gender was measured by fixed categories: 'woman' and 'man'. The respondents' age was measured by years, and afterwards categorized into five groups (14-17, 18-24, 25-44, 45-64 and 65+). The typology of families was divided into singles, couples without children and couples with children, single-parent families and mixed families (all the other types of families). Education level was divided into three categories: low (primary and secondary school diploma), middle (high school diploma) and high (college/university degree or higher). As regards the degree of urbanization, we adopted OECD's proposal to distinguish three levels: essentially rural (up to 5,000 inhabitants), relatively rural (from 5,000 to 100,000 inhabitants), essentially urban (100,000 and more).
In 1996, we also found it useful to control the effects of house size 4 and people's mobility.
House size was measured in five categories: '10 to 49 square metres'; '50 to 99 square metres'; '100 to 120 square metres'; '121 to 199 square metres'; and '220 and more square metres'. As regards mobility, we use a specific measure of moving house, which along with the mobility as daily travelling and movement is another traditional form of mobility. We asked: 'How long have you personally lived at your current address?' This mobility was measured by years and afterwards categorized into four categories: 'Less than 5 years', '6 to 10 years', '11 to 20 years' and '21 or more years'. However, these questions were not asked in 2009. On the contrary, both in 1996 and in 2009 the quality of life was investigated and measured by asking 'Over the last two years, would you say that your quality of life has significantly got better or got worse?' The following answering categories were supplied: 'Much better', 'Better', 'No change', 'Worse' and 'Much worse'.

Data analysis
Our study is based on the analyses of frequency distributions and linear regressions. Frequencies are examined to obtain answers to RQ1, which focuses on the temporal change in the body-to-body forms of communicative sociability. The linear regression analysis with an entered procedure is used, in turn, to obtain answers to RQ2, which aims to find out whether the possession of new media and the frequency of newspaper reading affect the body-to-body forms of communicative sociability, and RQ3, which focuses on the temporal changes in the socio-demographic profile of Europeans practising these forms of body-to-body sociability.

Body-to-body sociability in 1996 and 2009
To test H1, we first explored how common the various communicative forms of body-to-body sociability were in 1996, and how spread these practices were 13 years later in 2009. Table 2 shows that visiting friends and relatives has become more infrequent over the years, while all other forms of sociability have become more evenly spread in everyday life.
Insert Table 2  There was only a connection with visiting relatives and taking part in sporting activities in 2009.

Changes in the frequency of communicative sociability
To further clarify temporal changes in the various forms of body-to-body sociability, attention is paid here to weekly and monthly sociability practices. In Table 3, all seven forms of communicative sociability are presented by splitting respondents' answers into two categories: the first presents the answers of respondents who perform these forms of sociability at least once a week, and the second of those who perform them at most once or twice a month.
Insert Table 3 about here   Table 3 shows that the forms of communicative sociability that were practised weekly have decreased systematically over time, except for the category of going out to the cinema, theatres and museums which has remained at the same level. At the same time, the results show that the majority of the forms of communicative sociability that were performed monthly increased during the studied 13 years. This is the case regarding going out to the cinema, theatres and museums, as well as going to restaurants, pubs or dances, which are all performed now more often than in the 1990s.
The share of people who go to clubs and take part in other civil society activities on a monthly basis has increased too. An increase can also be found in relation to going out for a walk, shopping and watching sporting events with a friend and taking part in sporting activities.
All in all, Table 3 shows that such forms of sociability, whose diffusion has increased over time in the social body (e.g. walking, shopping and watching sporting events; restaurants, pubs and dancing; cinema, theatre and museums, see Table 1), have, however, become more infrequent practices. This means that some sociability practices have actually become more widespread among the whole population, but at the same time they have become less frequent. Previously weekly forms of communicative sociability have become monthly.
The index of communicative sociability was created as a measure that took into account all forms of sociability together, and also the frequency of their performance. The index, which ranges from 0 to 28, shows a very slight increase from 12.1 to 12.6 between 1996 and 2009. However, the application of the t-test for unpaired samples tells us that this difference, while small, is statistically significant (t=-4.87, df=11139.3, p<.001). Hence, although the internal diffusion and frequency of the forms of communicative sociability have changed, on the whole the amount of social sociability has increased so slightly that it would be more appropriate to speak about real stability over the time. This result leads us to reject H1 and to argue that sociability has not decreased simultaneously with the advent of the new media as proposed by earlier studies (Putnam, 2000;Pherson, Smith-Lovin, & Brashears, 2006). Nor do the results support the idea that sociability would have substantially increased.

Predictors of sociability indexes
Two linear regression models were executed, with the purpose of testing H2, dealing with the effect of new media on the communicative forms of sociability. The same models were also used to test H3, dealing with the socio-demographic profile of Europeans regarding the intensity of their communicative sociability practices and changes in the profile from 1996 to 2009. The results of these regression analyses are presented in Table 4.
Insert Table 4 about here In response to H2, Table 4 presents results that are supportive of the hypothesis. In the 1990s, both the computer and the mobile were supportive of the studied forms of sociability, and the infrequent reading of newspapers was related to a more limited sociability. When entering 2009, the support of computers and mobile phones for body-to-body sociability decayed, while the Internet took over as a platform of communication. In other words, the results support the previous studies and scholars who argued that use of ICTs has added to sociability instead of decreasing it (e.g. Hampton & Wellman, 2000;Wellman, Boase, & Chen, 2002;Wellman & Haythornthwaite, 2002;Rainie & Wellman, 2012). However, the results also imply that the role played by ICTs on sociability is conditional on the historical moment under investigation.
As regards H3, Table 4 illustrates that the socio-demographic profile of Europeans regarding the sociability practices changed from 1996 to 2009. In 1996, the intensity of social life in France and Germany was lower than in Italy, while in Spain it was more intense. Male respondents had more articulated social interactions than females, and singles enjoyed a more vivid sociability than the respondents living in all the other types of family. The more the age increased, the less intense was the social life practised. Those living in a house of between 121 and 199 square metres practised a more intense social life than those living in the standard apartment of 50 to 99 square metres. Also, the respondents who had stayed in the same house for 11-20 years received higher scores on the sociability index. Respondents living in relatively rural and essentially urban areas could count on a more intense social life than those living in essentially rural areas. Higher levels of education were also connected to higher scores on the sociability index. As regards activity, housewives were less likely while students were more likely to receive higher scores on the sociability index than workers. Also, the perceived change in the quality of life during the last three years was connected to the sociability index. The intensity of social life was higher for those who perceived that the quality of their life had become better or much better than for those who were thinking that no change had happened, or even that it had worsened. The ownership of a mobile phone and a computer -as we have already reported -was positively associated with sociability, and those who read print newspapers several times a week obtained higher scores on our sociability index than those who read newspapers once a week or more rarely.
The regression analysis for the 2009 data set reveals that only France still received lower scores for sociability than Italy. In 2009, as in 1996, males received higher scores than women, and people with high or medium levels of education received higher scores than those with a low level of education. As regards activity, students continued to enjoy higher levels of sociability than workers. As a new feature, pensioners received higher sociability scores than the reference group of workers. With regard to the degree of urbanization, in 2009 social life was concentrated more intensively in essentially urban environments. Regarding the quality of life changes, the results are more ambiguous than in 1996. Yet it seems that those who perceived that their quality of life had not changed during the last three years were less likely to receive higher scores on the sociability index than the reference group, while those who perceived that their quality of life had much/very much improved or worsened were more likely to receive higher sociability scores. With regard to the possession of new media, Internet access was linked to higher levels of sociability. Taken altogether, the results provide support for the approval of H3.

Discussion
What changed in sociability forms between 1996 and 2009? The decline in the visiting culture can be considered an important result of this study. This decline reveals a preference for a public sociability that takes place outside private, intimate and domestic settings. It may be that visiting others' homes is considered to require a lot of pre-planning, while public places, such as restaurants and coffee shops, offer an easier and ready-made setting for such informal gatherings. Maybe the decline in visiting is also explained by the fact that it requires a certain amount of organization and housework and reveals many personal habits, such as the heightened expectations of what the home should look like and how visits should be arranged with their peculiar ritualization, including the application of etiquette. In other words, visiting seems to be perceived as 'old-fashioned' because of its rigid rituality (if you go to visit a friend then you have to return the courtesy) and commitments related to housework and planning. Moreover, comparisons over time did not reveal any clear relationship between visiting behaviour and the technologies that were studied. While visiting friends was commonly associated with the use of a computer, mobile and the Internet both in 1996 and 2009, visiting relatives is less supported by these new technologies. It may be that the coordination for family meetings is carried out more often in co-presence with others compared with visiting friends.
The other sociability practices, such as walking, shopping and watching sporting events, going to restaurants, pubs and dancing, cinema, theatre and museums, have actually become more common, but at the expense of their frequency. Compared to the past, more and more people experience different forms of social practices and enjoy a more varied sociability today. However, what was previously considered a weekly practice has now became monthly. Hence, it can be concluded that the social body is more broadly socialized, and perhaps also homogenized, in respect of sociability style. At the same time, the social body is less intensively socialized on the whole as these social practices have become more infrequent.
With regard to the impact of ICTs on body-to-body sociability, the results demonstrate that much of the micro-coordination of sociability practices conveyed through other technologies in 1996 was taken over by the Internet on moving into 2009. Those who had access to the Internet were more likely to be sociable. These results, however, do not reveal the precise role played by different ICTs in respect of various forms of body-to-body sociability. Based on these results, it can only be supposed that sometimes ICTs facilitate and at other times they supplement the sociability process.
But what the social situations are where the facilitative and supplementary functions of ICTs are active in reality still needs to be studied further.
In addition, the results of the study call for the importance of historicizing the results dealing with the relations between ICTs and sociability. Firstly, although ICTs seem to add to sociability, this phenomenon is clearly conditional on the historical moment under investigation. Secondly, the results show that it is not possible to talk about ICT in general and there is a need to distinguish one medium from another. This is most reflected by the results which show that in 1996 it was the computer and the mobile phone, and in 2009 the Internet, that predicted the highest level of sociability.
How should we understand the changes in the socio-demographic profile of Europeans practising different forms of sociability? The comparison of the social profiles of Europeans regarding the intensity of their communicative sociability practices in 1996 and 2009 highlights some important results. First of all, our results show that the studied forms of body-to-body sociability are shaped in divergent ways by different countries. While in 1996 the southern countries such as Spain and Italy were confirmed to be more sociable than the northern countries (Germany, France and the UK), in 2009 the only certainty was that France continued to be less sociable than Italy.
Second, ageing seems to play against the studied body-to-body forms of sociability, which according to our results remained the prerogative of younger people between 1996 and 2009. In fact, it can be argued that along with health, sociability is a major source of wealth for the younger generation. Third, compared with people with low education, those with medium and high levels of education enjoyed a richer life on a sociable level. However, this did not change during the studied 13 years. Fourth, sociability is still a more masculine than feminine asset. At first glance, this might seem paradoxical as women are typically seen as the 'immaterial workers' of the sociability of the family; women keep alive the communication with relatives and friends (e.g. Wellman, 2001) and organize visits or going out with them. However, when it comes to personal sociability practices, where playfulness, time and income matter, women appear to be in a disadvantaged position. Fifth, some important changes also emerge regarding respondents' activity. Students have continued to enjoy the privilege of being most sociable, while housewives and house husbands suffer less social isolation than in the past. Retired people appear now to be more sociable than workers. This is in line with the trend that pensioners are healthier than in the past and are encouraged to engage in active lifestyles. Sixth, family typology shows that singles have lost their more fortunate contact with sociability over time, probably also for economic reasons. For example, widowed, divorced or separated people may have had more economic restrictions in practising the public forms of bodyto-body sociality in 2009 than13 years earlier. Seventh, people living in a house of between 121 and 199 square metres, which is bigger than a standard apartment, seemed to live a more intense social life in 1996 than people living in all other house sizes. Those living in smaller and larger houses did not differ from each other with regard to the amount of sociability. In the case of the largest houses, with an area of 200 square metres or more, it may be that the higher amount of disposable money makes people more reserved towards others. Eighth, although implying mobility, to be performed sociability needs a certain stability of the settlement, probably because creating one's own social networks and the related rituality requires quite a long time. Ninth, with regard to the degree of urbanization, in 2009 social life was revealed to be concentrated in urban living environments.
Compared with rural environments, the intensity of social life had become more similar in relatively urban areas. Tenth, as to the quality of life, those who did not perceive changes during the last three years enjoyed a less intense social life for the main reason that those who do not improve their life lose the capacity to entertain social relationships. Lastly, with regard to the possession of new media, only Internet access remained associated with higher levels of sociability.

Conclusions
This study contributes to the debate on the relationship between ICT and sociability in several ways.
First of all, the study makes a theoretical contribution. It sheds light on the relations between the concept of sociability and the notions of communication, mobility and labour. The last three notions appear relevant when trying to better understand the social significance of the changes in the level of sociability. More specifically, physically proximate social interaction seems to call for more mobility -from home to public spaces -as well as communication and concerted actions with people other than family members and close friends. The fact that sociability, as one form of reproductive labour, has demonstrated noticeable stability over time may mean that the global economic system does not need or want a more socialized labour force in co-presence. In fact, a more developed sociability could bring about unwanted consequences in the everyday life world as it would be difficult to control it. At the same time it may generate solidarity and social cohesion that oppose the interest of the ruling classes. Hence, other matrices of sociability, such as mediated forms of communication and sociability or online entertainment businesses, are more likely to produce value and be more compatible with the needs of social control.
All in all, our results show that the five largest countries in Europe are quite impervious to change.
The studied Europeans did not become either richer or poorer in co-present sociability between 1996 and 2009. The focus on sociability moved from visiting friends and relatives towards communicative sociability that takes places in the public sphere. It would thus be tempting to ask whether sociability is becoming a less intimate, family-centred and home-based practice.
Secondly, this study makes an empirical contribution to the existing body of literature. It shows changes that have occurred in the internal structure of the forms of body-to-body sociability, in their broader distribution in the social body and in their more sporadic frequency. Even the take-up of revolutionary ICTs, such as the mobile phone, the computer and the Internet, has not dramatically affected the overall volume of sociability. Paradoxically, the stability of the intensity of social life at the European level is a finding that gives much food for thought. Is it just one more sign of the slowness of a true social change, or is it the sign that thrusts and counter-thrusts have eliminated each other? Further research is needed to rigorously tackle these questions.
The study also underlines the fact that the effects of ICTs on sociability are bound to time. Between 1996 and 2009, the Internet replaced and became a more prosocial tool than the mobile phone and personal computer. In other words, those who were socially more active in 2009 were also more often connected to the Web than others. This probably means that the Internet serves to maintain an acceptable level of sociability and that without the Internet some forms of body-to-body sociability would have probably decreased. But of course, the Internet itself has opened many possibilities for communicating in a computer-mediated manner, yet its diffusion may also have had the retroactive effects of desocialization, whose understanding requires further research. By these effects we refer to what, for instance, Turkle (2011) describes in her book Alone Together: even if communication on the screen increases and intensifies, people may remain or even become more unconnected in terms of co-present interactions. Also, Kant's expression of unsocial sociability gives the sense of the retroactive effects that the practice of mediated communication can generate.
Despite the several theoretical and empirical contributions it offers, this study suffers from some shortcomings. Firstly, the body-to-body forms of the communicative sociality we analysed do not include any 'irregular' forms of social interaction such as prostitution. This is a serious limitation since these forms are increasing as well as blurring the boundary between licit and illicit. Secondly, the study lacks the qualitative aspects of the studied social practices. We do not know, for example, how much attention people give to these practices, the emotional temperature characterizing them and the intensity of the participation. Thirdly, the article was too short to discuss the social functions of sociability practices and whether these functions underwent changes between 1996 and 2009. It was also too short to appropriately investigate the differences in cultural meanings that the same practice of sociability presents in various countries (e.g. what going to a restaurant means in Italy is not perhaps the same in the UK) (Bourdieu, 1998). Finally, it remains unknown to us with whom the studied people practised the investigated forms of sociality and whether there were temporal changes in this regard.
Finally, further research is needed to investigate more carefully family-based sociability. In so doing, it would be possible to verify whether this very important form of sociability has remained stable or has decreased over time. In this connection, it should also be further investigated why there is a decline in visits to relatives and friends. Another topic which deserves more detailed examination is the sociability in workplaces. Finally, the relationship between ICTs and sociability should be further investigated to capture its concrete terms and meanings. 3) including various measures of the use of these media, of which there are actually very many (e.g. number of calls made/received, sent/received SMSs, the duration of the calls), would make the analyses themselves less manageable; 4) the possession of devices correlates with actual usage. 4 The question regarding house size, which was investigated only in 1996, had to be structured in a different way in the UK ('How many living and bedrooms do you have?') and in all the other countries ('What is the surface area of your home?'), since this variable required information which is given in a different way in various cultures. In particular, while in all the Continental countries people were accustomed to stating the dimensions of their house in square metres (sqm), in the UK people were accustomed to describing the dimensions of their house using the number of rooms. In order to have a consistent variable, we estimated the dimensions of a room at an average of 16 square metres so that it was possible to discern a unitary answer for all the respondents. However, this variable must be treated with great caution.  36.0 25.9 -10.1*** 6.6 29.6 +23.0*** Take part in club, religious, trade union or political activities 18.4 13.8 -4.6*** 12.6 31.5 +18.9*** Note: * = p<0.05; ** = p<0.01; *** = p<0.001. The significance of the change was assessed with the chisquare test with Yates' correction.