This paper mainly throws light on a phenomenon that might have escaped family researchers' attention. The pattern of "the husband works outside, the wife stays at home" is pointed out to be a feature of the modern family as a result of industrialization. However, according to the results of my survey in 2001 in Kyoto city, informants consider this pattern to be a traditional division of role between the spouses, which creates a contradiction between the position of family sociology and common people's image about the pre-industrial times. Therefore, to find the reasons for what causes the "traditionalization" of the modern family in common peoples' image becomes an important issue for family sociology. From the interviews, I pointed out three phenomena as possible reasons. Firstly, one can explain it as a reaction to the increasing number of "career woman" (working woman) since the 1980's. Women who do not stay at home, but work like men, may create a new form against "the husband works outside, the wife stays at home" pattern with the result that the latter becomes "traditional". But findings show that it may not be the only reason. From the interviews, it turns out that the informants confuse two different types of divisions of role in their mind that might be a further reason for "traditionalizing". One type refers to the division of the housework, while the other one refers to the question of who is the breadwinner in the family. The former one has not gone through radical change since the pre-industrial times, while the latter one has. In pre-industrial Japan, women used to work in the fields, shops or workshops helping their husbands, besides doing the housework. Due to industrialization, women do not work with their husband together any more, but they stay at home, running housework issues mainly. The informants in my survey unconsciously focus on the fact that the housework has always been the job of the women, therefore they consider the pattern of "the husband works outside, the wife stays at home" traditional -- which actually refers to the question of who is the breadwinner. From the interviews, one can see a third reason too. Although peasants, craftsmen and merchants constituted the vast majority of the Japanese society in the past, the informants in my survey focus on the warriors' small stratum, while talking about the past. In the warriors' families, wives did not work with their husbands the way wives in other families did. By focusing on the warriors' stratum instead of that of the peasants etc., the informants regard it as a traditional pattern. This kind of "traditionalization" can be named 'spontaneous traditionalization', which does not belong to any categories of Hobsbawm's typology, which mainly focuses on intentionally invented traditions. Therefore, I consider that "spontaneous traditionalization" could be added to Hobsbawm's typology as a further category.