The topic of this thesis is book reviewing in Sweden with a focus on three years: 1906, 1956 and 2006. By describing and analysing the critical reception of literature in these periods, the thesis aims to increase the knowledge and understanding of literary critical discourse and the function and power of the literary review in the past and today. The theoretical approach is inspired by Michel Foucault’s concepts of discourse and power. In particular, the thesis addresses the issue of norms in relation to literature and literary criticism by studying how criticism is written, by whom, on what, where and how much. The study, which is partly sociological and partly text analytical, is based on a systematically chosen sample of book reviews – altogether 200–300 reviews per year.

The thesis maps out the publication forms for literary criticism each year as well as the changes that the critical community has undergone. It shows a movement from the often extensive, yet occasional and seasonal, literary reception in 1906, to the establishment of separate cultural sections in the newspapers of 1956, and the signs in 2006 that the threat from the new media has led to a state of crisis in the daily criticism. Regarding the critical community, a major change was the gender distribution, which showed that there were almost exactly as many women as men reviewing and being reviewed in 2006.

A commonly held view is that contemporary criticism does not have the same high quality as it used to and that the space for literary criticism has been considerably reduced. One part of the study is therefore a quantitative comparison of the reviews and reviewers of each year at group level. The comparison shows, for instance, that the reviews in 2006 were indeed much shorter than a hundred years earlier, but the length of articles had already decreased in 1956. In fact, the average length per reviewed work was not shorter in 2006 than in 1956.

In the text analysis, norms of evaluation, recurring themes, the role of the critic and approaches to reading are examined as a discursive text system. The aim is to show how these norms embody orders of critical discourse and how some reading approaches were privileged in the reviews while others were more or less ruled out. The analysis also reveals that the critical discourse both influences and is influenced by the discourses of closely related disciplines, such as journalism, academic criticism, and – of course – literature itself.