Keywords

1 Introduction

Food forms a social bond. The social aspects and functions of food that reflect and symbolize this are well known. It goes without saying that among these, commensality is especially important.

Commensality means eating together with someone else. Eating, for the most part, is an activity which relates to the needs of individuals; maintaining bodily functions or responding to feelings of hunger. At the same time, however, the act of eating is often done together with someone, and it is even considered preferable to do so. This means that eating is thought to be a social act in principle. In this sense we could say that communal eating is one of the clearest manifestations of food’s social aspects, and furthermore a clear display of the kind of sociability that sets humans apart from animals. The Japanese anthropologist Ishige says that humans are animals with the shared habit of eating together (Ishige 2005:12).

In this paper I will look at commensality, but the focus will not be on commensality itself, rather on the nature of research into commensality. While most of us have a particular image of commensality, we have not discussed in any depth the diversity and complexity of its reality. This may be because the images associated with commensality are so natural and positive that we have stopped thinking about or unconsciously suppressed any negative ones. This has had a detrimental effect on the study of commensality. This is particularly the case when it comes to family meals, which are generally imagined to be a typical form of commensality. In this paper, therefore, I will focus on family meals, discussing how they have been researched and debated so far. I believe this will also contribute toward a reevaluation of the study of commensality in general.

For the purposes of this discussion, I will use case studies of Italian and Japanese family meals. The former is mostly based on a survey that I conducted in Italy, and the latter on previous studies accumulated in Japan. While dealing with evidence, this paper does not represent a comprehensive review of the data, and moreover does not engage in precise comparative studies between Italy and Japan. Its aim is to create a set of general principles that should precede these kinds of studies. Family meals have been reported through various cases in various locations, but the study of family meals is not at a stage where a proper comparison can be undertaken. The task of systematically describing them and considering their reliability still remains to be done. Contrasting two cases provisionally here enables us to understand that the discussion needs to be further refined.

In the following sections I first briefly outline the previous discourse and studies related to the issue of family commensality and commensality in general, and highlight the problems I have identified, with reference to my discussion of Italian cases. Then returning to the study of family meals, I explain how the a priori images of family meals that both researchers and subjects of research unconsciously assume are of particular concern. I term this phenomenon “family commensality myth.” As a way of countering and overcoming this problem, I introduce some unique research that has been conducted on Japanese family meals. This also provides significant suggestions for reconsidering the study of commensality in general. Besides it is necessary to add some points about the usage of the terms, “family meal” and “family,” in this paper. “Family meal” is used to refer to everyday meals, including meals at weekends, and excludes meals related to special events such as festivals and ceremonies. “Family” is a controversial term, but in this paper, whose focus is not family per se, I use it in a broader, less strict sense; that is, I use it in accordance with the relevant family’s consciousness of their own family unit.

2 Close Connection Between Family and Commensality

As mentioned above, the most commonly cited example of commensality is the family meal. When you think of eating with someone, family is the first thing that tends to come to mind. The same can also be said of family life. It is very common for commensality to be considered the most important element within family life and it is often assumed that there is a very close relationship between family and commensality.

For example, in Japanese we have the expression kazoku danran. Kazoku (家族) means family. Danran (団欒) refers to sitting in a circle and has the meaning of “gathering to spend enjoyable time together.” Kazoku danran, therefore, expresses the idea of the family gathering to spend quality time, and while this in itself does not explicitly refer to meals, the family mealtime is the image that is generally conjured up. In Italy too, there is a very close relationship between the family and meals, as I will describe later.

The strong association between family and eating has also been highlighted by anthropological research. For example, Nakane, a Japanese anthropologist, surveyed the ethnographic studies that had been previously conducted in different societies around the world, and found that, while no uniform definition exists, there are four major functions and elements that constitute family; meals, co-habitation, economic factors and blood ties (Nakane 1970). Nakane does not argue that meals are the common element of all families, but does focus on their importance. Carsten (2000), who provided a new approach to kinship research through the concepts of “substance” and “relatedness,” have recently highlighted the importance of shared food or commensality. They argue that daily accumulations of relations as well as norms and regulations are required for the consciousness of a family bond, and that in many societies sharing and eating food together is one of these key elements.

The same thing has been pointed out in sociology, psychology, and other disciplines. For example, Kaufman, a sociologist who conducted interview research in France, showed in great detail how family meals form family connections, including the associated power relationships and conflicts (Kaufman 2005). Lupton, who argues that food and eating are emotional experiences that are central to creating an individual’s subjectivity, explains that “it is in the context of family that the social dimensions of eating and those of emotion are particularly tied together” (Lupton 1996:37).

It seems, therefore, that there is a natural relationship between family and commensality. However, is this true? Has this not been exaggerated? For example, despite the value we seem to place on this relationship, we do not always eat with family, and moreover do not even seem particularly concerned about that situation either. Moreover, if you look at societies other than your own, you may observe other forms of family meals. One researcher who conducted field research in Sundanese villages in Indonesia told me that families there rarely eat together on a daily basis and each member of the family is permitted to eat food in the kitchen at any time.Footnote 1 This suggests that the strong relationship between family and commensality has not yet been fully examined or validated. I would argue that this is one of the most important problems in the study of family meals.

3 Problems of Commensality Studies

Firstly, it is worth noting that there has been surprisingly little discussion on what commensality specifically refers to. The importance of commensality has already been pointed out by many scholars. For example, Simmel wrote that “persons who in no way share any special interest can gather together at common meal…There lies the immeasurable sociological significance of the meal” (Simmel 1997:131). However, the study of commensality itself has only begun in recent decades (Mennell et al. 1992). Even now, relatively little quantitative or qualitative research has been done about any commensal circles, and as a technical term there is still no clear definition of commensality (Fischler 2011; Sobal and Nelson 2003). Is commensality just a question of everyone eating at the same time in the same place, or does it include other elements, such as conversation? What about frequency? Is it OK if the food being taken is different, or does it have to be the same? As soon as we start trying to think about the definition and indicators of commensality, a variety of questions arise.

Indeed, the occasions and situations of commensality are generally very varied. Faced with this problem, Grignon (2001) looked at commensality in terms of frequency, the relationships between the people involved, and its character, and tried to categorize them into three axes of conflict: Domestic Commensality (e.g., meals at home) versus Institutional Commensality (e.g., at schools and hospitals); Everyday Commensality versus Exceptional Commensality (e.g., special occasions such as ceremonies); and Segregative Commensality (e.g., a club party attended only by its members) versus Transgressive Commensality (e.g., a banquet where a king served a meal for all social classes in the medieval era). This scheme provides lots of suggestions, but it is just an initial hypothesis, as Grignon himself conceded. There has not been any subsequent discussion on how to develop this hypothesis yet, and that might indicate the difficulties involved in studying commensality.

A similar situation has been also highlighted in the study of family meals (Mestdag and Glorieux 2009; Murcott 1997; Sobal and Nelson 2003), which is still the most studied area of the few commensality studies that exist. To sum up their arguments here, the first problem of family meal research so far is that the actual number of studies that have been conducted is still very small. The second, related to the first, is that the definition and indicators of family commensality are still unclear. For this reason, even if surveys are being conducted, the respective indexes are often different, preventing sufficient analysis and comparison from being performed. This becomes even more acute when sociocultural differences are also brought into consideration, as in the case I will discuss later.

The third problem is the difficulty of the survey itself and, in particular, the question of data reliability. Family meals are part of people’s private lives, so many are reluctant to participate in investigations. This makes it very difficult to collect data. Moreover, those respondents who do participate tend to display a subconscious bias in their answers that deviates from reality. For example, when asked about who they dine with, many say they eat with their family, despite regularly eating alone. The reason may be that they are unconsciously trying to convince themselves or others that they are observing the social rule of eating together with one’s family. I will consider why this happens later, but in any case, the survey results often reflect the opinions of the respondents rather than their actual reality. Researchers need to carefully examine the method of investigation, too.

The last and arguably biggest problem is the view of family meals held by the researchers themselves. I have already mentioned that there is a strong association between family and commensality as something positive and natural. Researchers also share this view, and this has had a detrimental effect on their understanding of the reality of family meals. For example, there is a tendency to accord a positive image to family meals as something fun and intimate, but in fact there are other various aspects which conflict with this view, such as strict seating arrangements and table manners, as well as occasional discord and arguments between family members. In previous studies, these negative aspects have been so barely noticeable or visible, that family meals have been imagined to be a universally ideal form of commensality. Arguably it is this overarching problem that has led to other ones listed above. Researchers strongly influenced by this image may not have had any significant impetus to pursue surveys of the family meal or discuss its definition and indicators—at least not until recent years when worries over the decline of family meals are growing as mentioned below.

Next, in order to further clarify this issue, I will present my case study of Italian family meals, focusing reflectively on how I have discussed it as a Japanese anthropologist. This is an attempt of thinking about family meals from a comparative cultural perspective, which will further elucidate the problems of family meal study.

4 Is the Italian Family a Commensality Group?

4.1 My Survey in an Italian Town

To be honest, the starting point for my interest in writing this paper was an ethnographic survey I conducted in an Italian town (hereafter named R). It is located in the area called I Castelli Romani, to the immediate southwest of Rome, and had a population of about 8,000 in 1986, when I conducted my first survey. For the further details on the town of R, please refer to my works (Udagawa 1989, 2015).

It is said that in Italy family is one of the most valuable things and Italian family bonds are very strong. This is often referred to as “familism,” including by Italians themselves. Family, therefore, has been one of the most important and controversial topics in Italian studies, especially since an American anthropologist Banfield highlighted this by coining the term “amoral familism” in his 1958 work, The Moral Basis of a Backward Society. There have since been numerous discussions on Italian families from diverse perspectives by both Italian and foreign scholars (Bagnasco 2006; Meloni 1997; Kertzer and Saller 1991; Barbagli and Kertzer 1992; Udagawa 1987).

Soon after starting my first field study in the town of R, I discovered for myself that Italian family ties were as strong as previous studies had pointed out. I also found that these ties were closely related to the habits surrounding family meals. Back then, people in this town usually had lunch with their family at home, as their lunch break was a few hours’ long. This gave workers enough time to return home. But that was not all. On Sundays I saw extended families eating together. People who were married and living apart from their parents visited their parental home to have lunch. They also had to visit their spouse’s parents, so they would adopt a policy of alternating their visits each week. If they lived far away, they would return to their parental home in the summer vacations and other holidays with their children.

Let me present the case of A (Fig. 1), an elderly lady I rented a room from during my first research (Udagawa 1989). At that time she had already lost her husband, and her four children had all married and moved out, so she was living alone. Two of her children lived in the same town, while the other two lived in neighboring ones. When they all gathered at A’s house on Sundays bringing their children, there were more than fifteen of them. What was of particular interest was that in this circle there were also two nieces (L&M) who had lost their parents, and a non-blood-related woman (N) with her daughter, who for some reason had broken ties with their blood relatives and had been looked after by A who lived nearby.

Fig. 1
figure 1

A’s family as her “commensality group” (Udagawa 2011: 585)

Full size image

Generally speaking, their actual range of commensality varied depending on each individual’s circumstances. But, at least on Sundays they would make every possible effort to ensure that the family and close relatives gathered to eat lunch together, even if they were not living together. In the weeks when this was not possible, they would visit their parent’s house on the Saturday evening (that is, the day before) and have dinner together. I have actually seen a number of such cases in the field. They often explicitly acknowledged those they were eating together with (especially on Sundays) as family, using possessive pronouns (for example, miei, which is equivalent to English “mine,” in the sense of “my family”). I therefore termed their family a “commensality group” (Udagawa 1989).

4.2 Comparative Discussion of Italian and Japanese Family Meals

This high tendency to eat together could broadly indicate one of the major characteristics of Italian families in general, but it is, of course, necessary to take into account other such aspects as regional diversity.Footnote 2 One further thing to note especially here, is that my research and discussion above was inevitably done in comparison to the Japanese family which I had personally experienced growing up in, albeit unconsciously. This is the very problem that I would like to highlight here.

Unlike in Italy, in Japan there are only a few occasions when members of a family who have left home get together to eat. Once the grandchildren have grown up, those occasions become even less frequent, with family gatherings only occurring once or twice a year at New Year’s and other events. As a result, relationships with relatives outside of the household (which in almost all cases means the nuclear family unitFootnote 3), such as cousins, nephews and nieces, uncles and aunts, tend to be more distant. If grandparents are not actually living with their grandchildren, it is rare that they even meet, let alone eat together. When compared with this kind of Japanese family, it seemed to me that Italians expend a huge amount of energy on, and build family relationships through, commensality.

To be clear, first of all, my argument is not that the Italian family is built solely by “eating together.” As Nakane points out, functional or structural elements of family in any society are not characterized by one single thing; Italians believe that it is important for their families not only to eat together, but also to live together or close to each other. After marriage, many Italians move away from their parents, but try to live as close as possible to their parental home. However, in Italy the greatest weight is placed on “eating together.” The same is true for Japanese families. In Japan, too, importance is accorded to the idea of families eating together, but the idea of “living together” is given greater weight. We can see this clearly in the Japanese word Kazoku (家族), which means family. Kazoku is a relatively new word from the late nineteenth century that was made by translating the English word “family.” Prior to that the Japanese word used to mean family was ie (家), which also translates as “house.” This means that the previous Japanese word ie had two meanings; family and house. Thus, a new word kazoku was invented with the exclusive meaning of family, using 家 (the character for ie). Kazoku (家族) literally translates as “people (族) of house (家).” This fact demonstrates that the Japanese consciousness of family has a strong association with the idea of co-habitation. It could be said that the Japanese family is a “cohabitant group” while the Italian family is a “commensality group.”

Now, it is clear that my point about the Italian family being a commensality group is an illustration of its characteristics when compared with a Japanese family. It is, therefore, important not to over-state these differences between Japan and Italy and not to essentialize the image of the Italian family as a commensality group and the Japanese family as a cohabitant group. Besides, I have to confess and stress that this argument I did above is based on research about Italian families, rather than their meals. In my research, as well as in most other previous studies, the actuality of their family meals has not yet been investigated in detail, and little has been said about which aspects of them could be called commensality. Nevertheless, if we only point out that the Italian family is a commensality group, and cease all further consideration, it indicates that we have been influenced by the notion of an a priori connection between family and commensality, and that we have confused both without sufficient examination.

5 Family Commensality Myth

This widespread and persistent idea can be called “family commensality myth,” and it can also be seen and discussed prominently in the recent debate over the decline of family meals.

The family meal in decline is a frequently cited issue these days, particularly in developed countries, and is regarded as a serious social problem. In Italy, too, people often deplore the decline of family commensality in recent times, and it is certainly evident to see. For example, in the case of A’s family, after A’s death in 1994, her eldest daughter (B in Fig. 1) became the center of commensality for a while, but the frequency of family meals gradually decreased. Each of A’s children then transitioned to family meals with their own children, and moreover,  in recent years there has been a decline in the kind of Sunday meals I previously described. Even in cases where people live in the same town as their parents, it is increasingly common to have lunch before going to visit the parental home.

It may be true that the number of family meals is decreasing almost everywhere in the world. Factors behind this trend include changes in lifestyles and working practices, the development of the food service industry and school lunch system, and so on. Families have less time to eat together, and on the other hand, the spread of fast food and the like has made it easier for people to eat alone.Footnote 4

However, recent research on whether family meals really are in decline have thrown up many doubts about this generally accepted view (Fischler 2011; Mestdag and Glorieux 2009; Murcott 1997; Sobal and Nelson 2003; Sobal et al. 2002). Researchers have pointed out that while we may have seen a decline in the last few years, it is not easy to say what the situation was like prior to that. The first reason for this is that there is very little evidence to properly reconstruct the past. Previously commensality has been considered and recorded most often in connection with special occasions such as ceremonies and festivals, so there are very few documents relating to everyday meals. Even in cases where records of some kind exist, it is difficult to compare on the same horizon as the current survey data, as their indicators are often different or unclear.

And the second problem is; people tend to idealize past family meals and miss their reality easily. To consider this point, I would like to share the results of an interesting survey of Japanese family meals for urban housewives conducted in the late 1990s (Iwamura 2009). The survey consisted of three stages: in the first phase, the participants answered a questionnaire; they then submitted daily written and photographic records of their meals; and then finally they took an interview. According to Iwamura, who conducted this survey, many of the participants replied on the questionnaire that they ate together as a family, but in the second phase of the research it was revealed that this was not actually the case. When the discrepancy was checked in the interview, the respondents answered that “it just happened to be the case on that day,” or “it was an exception.” Iwamura noted that participants didn’t seem to see a problem with the discrepancy.

This suggests that the respondents were unconsciously reconstructing reality because they were so strongly influenced by the idea of a connection between family meals and commensality. I think that this way of thinking also pervades the recent discourse concerning the contemporary decline of family meals. That is, we tend to apply this idea to the family meals of the past, idealizing and romanticizing them. As a result, it is often thought that the contemporary family meals are on the wane by comparison to the idealized picture of the past. If this is the case, we must reconsider carefully whether the recent discourse reflects reality. In fact, the argument that family meals are in decline has not yet been proven clearly in recent studies, as mentioned above. They also cast doubt on whether the real situation of family commensality has ever existed (Murcott 1997).

Thus, we must strongly recognize that our society as a whole, including both researchers and respondents, is greatly affected by “family commensality myth.” Why and how this idea has become popular is not the purpose of this paper. But I note here very briefly that it is closely related to the idea of “modern family”; a new family image which was born during the process of modernization and has been the basis of today’s family. According to “modern family” theory, during the industrial revolution and establishment of the modern state the family was reformed as an intimate relationship or place for breeding good workers and citizens (Ochiai 2000). The bourgeoisie which emerged around the same time, embodied and spread this new family with their lifestyle and thinking; a family consisting of parents and children, fixed gender roles, and happy homes filled with love. What is interesting as part of this process of change, is that the dining table became regarded as one of the most important symbols of family intimacy. Eating together around the table became the most central image of this new family, and thus “family commensality myth” was born.

It can be said that “modern family” has generated “family commensality,” or at least has strengthened its value. In fact, the discourse of the current decline of family meals is often equated with the decline of the family itself. A lot of research is still needed on this subject. However, in any case, it has become clear that the link between family or family meals and commensality is arguable intrinsically.

6 Ishige’s Research on Japanese Family Meals

Then, how can we proceed with research on family commensality without drawing facile links between family meals and commensality? It is first of all necessary to investigate the reality of family meals more precisely. The reason why families eat together is not only because they value commensality; indeed, the main reason may be one of convenience. If everyone eats food together that has just been prepared, it can be eaten before it gets cold, and the cooking and clearing away can all be finished in one go. Moreover, when we consider that the home is the place where people receive their most regular supply of food, we can see it as the source of nutritional sustenance for each individual, and a place for distributing and sharing food. Considering the presence of children, eating together with the family functions as an act of feeding and child-raising, and is often a good occasion for education. In short, there are various meanings, functions and reasons associated with family meals. Until now we have overlooked these or reduced them to the single aspect of happy family circle commensality.

However, as I have already noted, trying to understand the actual situation of family meals is not easy. Here I would like to introduce an intriguing case study of Japanese family meals that might provide useful suggestions for further consideration.

This case study was part of a research project entitled “Life history about family meals,” conducted in 1983–4 by a team led by Ishige from the National Museum of Ethnology in Japan (Ishige and Inoue 1991). Its aim was to clarify how Japanese family meals changed in the twentieth century through interview surveys. The interviewees were 284 men and women aged 70 or older at the time who lived their lives during the twentieth century. In the interviews, the researchers asked each interviewee to answer a set of very detailed questions in relation to each stage of their lives. The questions included with whom and in which room they ate, how the seating was arranged, what kinds of tables and tableware were used, what the atmosphere was like, whether there was any conversation, and so on. No other research in the field of Japanese family meals can compare with this study, in terms of both quality and quantity. This study ultimately revealed how a happy family circle was formed in Japan during the twentieth century.

Let’s take a closer look. In contemporary Japan, most homes use a Western-style table as their dining table. However, until the 1970s, a low, round (or occasionally square) dining table named as chabudai was used, and before that, a  small tray table individualized for one person, named zen, was popular. In other words, Japanese family meals in the twentieth century can be roughly divided into three eras depending on the type of table used: the era of the Western-style table, the era of the chabudai, and the era of the zen (Fig. 2).

Fig. 2
figure 2

Transition of dining table types in Japan in 1900s (Ishige and Inoue 1991:69). The vertical axis represents the number of cases (total of 284 cases surveyed). □ = Zen, ■ = Chabudai, ▲ = Western-style table

Full size image

In the era of the zen (it is difficult to identify its origin, but it is thought that it spread widely in early modern times), at family mealtimes each person would eat from tableware on their own individual zen. One of its popular types was a box-type one known as hako-zen (hako means box), which had a place to put away tableware. After eating, therefore, each person would wash their tableware and stow it in the hako-zen. What is particularly interesting is that in meals with the zen a family (not all families) would eat in different locations; the father would eat alone in the tatami room (that is, the highest-ranking room), and the mother and children would sit on the wooden floor in the kitchen. The content of the meal was also different often, with the father enjoying more lavish dishes. Furthermore, the mother was not permitted to eat until the father had finished his meal. So strictly speaking, in terms of both time and space, it is hard to describe this as family commensality. Family members were also required to observe strict table manners. Conversation was actively discouraged during the meal. This was clearly very far from the atmosphere of a happy family circle.

Around 1920, with the shift to the chabudai from the zen, the form of the whole family sitting around the chabudai table to eat began to emerge (Fig. 3). That is to say, everyone now shared the same table, so they began to share the same eating time. The chabudai was actually developed around 1900, when modernization began in Japan.Footnote 5 Back then some thinkers even heralded the invention of the chabudai, at which the whole family could eat together, as modernization and democratization of the home (Ishige 2005:159). Unlike with the zen tray tables, where everyone would start the meal with their own individual dishes, with the chabudai the practice emerged of placing dishes in the middle of the chabudai for everyone to share. In this sense, the chabudai enabled family commensality to materialize further.

Fig. 3
figure 3

Scene of meal at the chabudai. A screenshot of the Japanese film Meshi in 1951

Full size image

However, Ishige’ research points out how the mealtime atmosphere was not so quick to change. The father’s meal continued to be more lavish than everyone else’s, and there was still little conversation or enjoyment. It is said that conversation during meals only became desirable after the Second World War with the spread of the Western-style dining table. This is how I remember growing up. My family was a nuclear one, comprised of me, my parents and my younger brother. In 1972 we rebuilt our house, modernized the dinning-kitchen room and purchased a dining table. Before we had a chabudai in the tatami room. Kneeling at the chabudai to eat meals required good manners, particularly of children, and conversation was considered bad manners. In addition, before mealtimes, we used to serve a cup of tea or a simple meal as an offering in front of the photo of my deceased grandparents, which created an even more ritualistic atmosphere. I recall that things did not change immediately after switching to the Western-style table, but at least we were no longer kneeling, and there was an increased sense of freedom.

7 Discussion

The detailed facts about Japanese mealtimes that Ishige’s study revealed are very interesting and valuable, but here I would like to discuss some points that have implications for future family meal research.

First, it has been demonstrated that the “family commensality” norm materialized during the process of modernization in Japan, although more research is needed.Footnote 6 This means that family commensality is not a natural phenomenon, but a historical one. And of particular note is how Ishige’s study has revealed and highlighted this. I have already mentioned several times that family meal studies face many difficulties even at the data gathering stage, as people’s strong unconscious commitment to family commensality can easily distort their perception of reality. Despite being based on memories of the elderly, however, Ishige’s research has succeeded in alleviating this problem by actively focusing on material and concrete aspects of family meals, such as tables, tableware, rooms, and so on. Such a thorough focus on material culture could represent an effective way of freeing ourselves from our assumptions and reconstructing a more detailed and objective reality. This is also useful for research in other societies. For example, even in Italy, there are documentary records which state that family members rarely ate together in the past because of a lack of dishes and chairs (Scarpellini 2012). Ishige's work may provide a new model for future family meal research also in a methodological sense.

Next, what is of even further interest in his research is that it reminds us of the importance of reconsidering what and how family meals exactly are. I have already stated that the meaning and function of family meals is not just to foster intimacy and bonds through eating together. In Ishige’s research, it emerges clearly that importance was placed also on manners and the hierarchy of family relationships, especially in the era of the zen tray table. Mealtime was a place where power relations between family members were reflected and reproduced. For children in particular, meals were a kind of a training ground for becoming well-behaved and educated. This is very similar to the situation described in Norbert Elias’ The Civilizing Process, which charts the changes in dietary etiquette from the Renaissance to the modern era in Europe. In adittion, gender roles are another important aspect of this, often with women (as mothers, wives) being required to prepare and serve the food even now. Therefore, the family meal is also a place where various discord among familial members often comes to the surface, as, for example, Kaufman (2005) highlighted in his French survey. Now these diverse functions and meanings of family meals need to be carefully researched and analyzed.

This also suggests that it is inappropriate to think about all everyday meals (breakfast, lunch, dinner, including Sunday ones) in a uniform way, and that each has its own functions, meanings, and even forms. I previously described the Italian family as a commensality group, but in fact this relates mainly to the Sunday meals, not to their other meals (Udagawa 1992). On weekdays, the commensality range at home is generally limited to the family members living together, and there are also many meals which are eaten separately due to the provision of school meals, work commitments or the time different family members return home. In particular, there is almost no expectation of eating breakfast together. Each family member attends to their own breakfast for personal health reasons and as a means of supplying energy. Regarding evening meals, the recent trend is for people to use them rather than lunch for family commensality, but also eating with friends is often observed in the evenings, as before. When I did the first research in R town in 1980’s, men in particular were more likely to have evening meals with friends than family.Footnote 7 This situation still needs further examination, but if we release ourselves from the myth of “family commensality,” we can see that there are various forms of meals, including ones where families do not eat together without it representing any particular problem.

Thus, we finally arrive at a stage of fundamentally rethinking the concept of family commensality. As mentioned before, many researchers have already stated that the definition of commensality is still unclear, and that this is one of the serious obstacles to further research. For example, in the era of the zen that Ishige described above, Japanese family meals took place under the same roof, but they were often eaten in different rooms and at different times. We still have no clear judgment as to whether this form can be considered as eating together, but I think we should not do so easily. Not only in this case, but in other cases as well, even if the family seems to be eating together, the details are very diverse, such as how to use tables and tableware, etiquette, and whether or not there is conversation. It will be further diversified if we include whether or not the foods eaten there are the same and whether or not they are made at home. We should first look into much more details of family meals, both ethnographically and historically, before easily defining indicators and definitions.

Moreover, if family commensality itself is a historical philosophy or value created in modernization, as previously discussed, it is problematic to consider this term as one of fundamental tenets of the discussion. Looking back at Ishige’s research again, it can be said that in Japan family commensality was an aspiration during the twentieth century, but it remains questionable whether it has been fully realized. This leads us to further questions as to whether the term “commensality” itself might be considered as just eating together, or an idealized word that incorporates a certain value; happiness, harmony, solidarity, sharing, and so on. In the latter case, commensality should only be used as a historical and cultural concept, or at least as a word that describes aspects of the value or atmosphere of eating together, and not as a technical term that indicates just eating with someone. I have used the term “commensality” so far without being clearly aware of the implications of these two dimensions, but in future a stricter distinction needs to be drawn between them.

So, if not only the forms of family meals but also the values for it may differ depending on the times and societies, it may be too soon to conclude that having fewer opportunities for families to eat together will lead to less family commensality, as with my previous discussion comparing the cases of Italy and Japan. Certainly in Japan, there are fewer opportunities than in Italy, and there are some societies (like the Sundanese mentioned at the beginning of this paper) where families do not share meals on a daily basis. However, this does not mean we can automatically claim there is little or no family commensality ideas in those societies. It would be more appropriate to say that they have their own forms for manifesting these ideas. We do not get to the answer, until we can explain what kind of situation is or was considered for them as commensality and how that importance is or was felt. What we can say now with certainty is that the forms and values of family meals vary greatly. Therefore, we need to reinvestigate them in more detail, especially focusing on how each era and society has defined and valued them on its own.

8 Conclusion

As a matter of fact, I’ve only just begun considering this issue of family commensality, and especially in the case of Italy, I have not yet done enough research on their family meals themselves. Therefore, the discussions in this paper leave many points to be verified, including Japanese cases, and I intend to tackle them in the future. Thus, in conclusion, aside from the discussions on these specific cases, I would like to briefly summarize the discussions so far in terms of making the appropriate progress in family meal study in general, and point out some further possibilities.

Firstly, when considering family meals, we need to extricate ourselves from “family commensality myth,” which appears to be a product of the modern age. From this point of view, it becomes clear that family meals have and, in the past, had a greater variety of forms and meanings than we previously imagined. When describing and discussing such diverse situations, we also need to pay close attention to the use of basic terms such as “family meal,” “commensality,” “eating together,” and so on. This is the first step in gaining a nuanced understanding of family meals, including their various aspects that have tended to be overlooked. I would add that this approach to family meals may also help reveal complex family relationships in greater detail and make a major contribution to family theory, too.

From this standpoint we can open the doors to a wider debate about other kinds of commensality or eating together. As mentioned at the beginning, in any society around the world, commensality occurs on a daily basis not just within the family, but also between neighbors, friends, coworkers, etc. In fact, what left the greatest impression on me with the survey in Italy was not only the family commensality, but also the fact that people so frequently ate with friends especially in the evenings (Udagawa 2015). Thinking about these other occasions of eating together is also important in thinking more deeply about family meals. Whatever the occasion of eating together we are trying to understand, we need to look at the other kinds of meals in that society and consider it with them as a whole.Footnote 8 This will also lead us to a deeper consideration of the social power and function of food and eating.

There are still more problems and issues remaining, but finally I would like to point out that the most important and difficult one is about the idea and sense of commensality or eating together. For example, in some societies, it might not be considered commensality when the participants at the meal do not eat the same food, while in others, cooking and preparing meals together might be considered as an essential part of commensality. If so, we must reconsider that the concept commensality itself might need to be examined without limiting it to the act of just eating. Furthermore, we must also take into account that the form and sense of eating together may have changed more radically in recent years. For example, Traphagan and Brown (2002) have revealed that today’s proliferation of fast-food establishments provides opportunities for new intergenerational commensality and intimacy in Japan. There is also an interesting experiment to explore the possibility of “remote commensality,” such as using a technology probe to enjoy the feeling of eating together even when away (Grevet et al. 2012). And now this “remote commensality” on Internet is unexpectedly and rapidly spreading among people who have been forced to stay home due to the effects of the coronavirus. It would be interesting and also necessary to examine how this phenomenon changes our real eating behavior and its feeling.

At first glance the concept of commensality or eating together appears very basic and natural, so it may not even seem worthy of detailed discussion. However, like Simmel's words quoted at the beginning, commensality or eating together is a place or occasion where everyone is able to connect to various aspects of society in more diverse ways than we may have expected and imagined. In this sense, it is a topic which should be given further consideration, particularly in our modern society where eating has become such a big issue.