“The term ‘citizenship’ may have different connotations; in some countries it may merely indicate a judicial relationship between the citizen and the State, whereas other countries may refer to the term as the social role of coexisting in society (Torres, 2006; Roth, 2007). Eurydice uses the term ‘Responsible Citizenship’ and concludes that the term ‘citizenship’ is mentioned in all the guiding documents for the school all over Europe. This study (Eurydice, 2005) examines e. g. how ‘citizenship’ is approached in official school curricula and if/how teachers may be supported on citizenship. It also gives example of how teachers themselves describe their practice in this respect. However, it does not go into details about how the citizenship is observed to be put into practice in the teacher-pupil relation in the classroom. It does conclude that the matter needs to be studied further, though. In Sandström Kjellin and Stier (2008a) it was found, from five national case studies that those teachers who communicated an attitude of citizenship in their own classroom (the ‘attitudes and values’ aspect of citizenship education) managed to engage their pupils more in the lesson contents than those teachers who merely taught about citizenship. It was concluded in the study that teacher education needs to focus more on horizontal classroom dialogue if goals for citizenship education are to be reached. In a European collaboration project under the Comenius 2.1 action (Sandström Kjellin & Stier, 2008b) it was found that even in countries where the political goal is in line with the thoughts of Gutmann (1987): to teach children to judge critically, the pupils at an upper secondary school were not aware of, and consequently not critical of, the way they were treated at school (Gustafsson, 2008). A comparative multiple case study (Sandström Kjellin, Stier, Einarson, Davies and Asunta, 2007) showed that, according to pupils’ understanding, the focusing of citizenship education varied a lot in three European countries. In England, the teenagers seemed to be well-informed about ‘political literacy’; however, they did not seem to be accustomed to confident relationships with adults. In Sweden on the other hand, the situation was quite the opposite; the teenagers did not seem very well-informed about ‘political literacy, but they gave the impression of being used to open and confident dialogues with teachers, which corresponds to ‘the attitudes and values’ aspect of citizenship education. In Finland the pupils were altogether very taciturn and it was hard to perform dialogues with them. – In short, the conditions for citizenship education seemed to vary a lot. In the light of the last described study it is interesting to find out also how teachers and future teachers in the three countries look upon this. The aim of the present study is to find out how teachers and student teachers in three European countries (England, Finland and Sweden) understand and describe their job/future job as teachers as regards ‘the attitudes and values’ aspect of citizenship education.

Thirty-six focus group dialogues have been performed and 180 teachers/student teachers have participated in the study; in each country, six focus group dialogues with groups of teachers, and six focus group dialogues with groups of student teachers have been performed. The data will be analyzed using Colnerud’s distinction of moral education. With respect to the level of verticality/ horizontality’Horizontal communication’ means that the two parties are perceived as equal, and ‘vertical communication’ means that the relationship between the two is unequal. Any adult-child relation is by necessity to some extent vertical (Janson 2002 ); hence, provided by their institutionalised role, teachers have a posititional advantage compared to their pupils. Yet, by being ‘omnipresent’ and receptive in classroom discussions teachers can allow for horizontal communication (endnote 1) in teacher-pupils interaction patterns, Colnerud’s (2004) singles out three types of moral education – i.e. ‘moral instruction’, ‘moral conversation’ and ‘moral interaction’. Moral instruction means instructing pupils’ value-based ‘rights and ‘wrongs’. To Colnerud, it has to do with actions; such instruction is necessary for instance, when pupils harm each other or themselves. Moral instruction becomes problematic when instructions take the shape of language of force, scolding or violation, which typically lead to temporary adaptation, but also encourage pupils to use similar language to each other. Another problem is that moral instruction can cause problems of ‘timely obedience’. Moral conversation refers to joint teacher-pupil explorations of moral problems originating in the classroom or in society as such. Colnerud sees moral conversations as essential for pupils’ ability to develop and formulate both theirs and other people’s moral standpoints and perspectives. A problem with moral conversation stems from the asymmetrical teacher-pupil relationship. Teachers must find a productive balance between a classroom situation where values are ‘passed on’ to the pupils (and where they may not reflect upon them) and one where the pupils themselves are responsible for scrutinizing and arguing for adoption of their values. In an earlier study, Colnerud (1995) found that Swedish teachers find it difficult to know to what degree different opinions should be allowed in moral conversations. Therefore she says, teachers must be able to articulate their personal standpoints on value issues as well as open up for investigations of different standpoints. Finally, moral interaction refers to teacher-pupil interplay, where the teacher treats pupils respectfully, and pupils feel that he or she listens to them and has confidence in their abilities. Dialogues were conducted over a three-month period. Using the same manual, data was collected in England, Sweden and Finland. Dialogues lasted 20-60 minutes, were tape-recorded, transcribed and translated afterwards. Transcriptions of focus group dialogues can be made with different detail precision (Wibeck 2000). For the purposes here, where focus was the contents of the teachers/student teachers’ statements, less detailed transcriptions sufficed. However, the researchers documented their impressions of the dialogue sessions – accounting for prosodic cues, silence, apparent contradictions in statements etc - in the conversations. Study participants also completed a written evaluation following the dialogues where they were asked to state the extent (‘completely’/’to some extent’/’not so much’ or ’not at all’) to which they had stated their opinions in the conversation. Evaluations were then summarized quantitatively. Dialogues will be analyzed qualitatively. First a preliminary categorization and count of retorts will be made. Thereafter the researchers from England and Finland will be given the opportunity to add comments and suggest changes. A final analysis will then be made by all the authors together. endnote (1): ’Horizontal communication’ means that the two parties are perceived as equal, and ‘vertical communication’ means that the relationship between the two is unequal. Any adult-child relation is by necessity to some extent vertical (Janson 2002 ); hence, provided by their institutionalised role, teachers have a posititional advantage compared to their pupils. Yet, by being ‘omnipresent’ and receptive in classroom discussions teachers can allow for horizontal communication

Expected findings are that differences will appear in the interviews both between teachers and student teachers, and also between countries as regards the implications of citizenship education. A possible outcome of the study is increased awareness among researchers and practitioners as regards the performance of citizenship education.