Recent attempts to make literary criticism more ‘rigorous’ have sometimes only succeeded in alienating it from many authors and readers. The discussions that follow are written in what I hope is plain and straightforward language. But I concede to the rigorists that plain and straightforward language can beg questions and gloss over problems, and I have tried to avoid this by being as fully explicit as possible about the standards, criteria, and methods which I favour. I believe that this is the only permissible way of arguing about literary questions at a time when there is so much disagreement about matters of principle.

The first four essays deal with general questions. In ‘The Definition of Literature’ I have attempted, not to fix on a specific formulation, but to decide what kind of definition is preferable. ‘On Liberty of Interpreting’ examines the much-canvassed question of the relevance or otherwise of authorial intentions to criticism. Its purpose is to bring out what seems true in both ‘intentionalism’ and ‘anti-intentionalism’, and to discard what in both seems false. ‘Evaluative Criticism, and Criticism without Evaluation’ argues that one widely favoured account of literary appraisal is in fact empty. ‘The Novel: a Critical Impasse?’ raises a general problem about prose fiction. It may serve as an introduction to the three following essays, which discuss some famous novels in detail. A contrasting kind of approach is used in the essays on Tennyson and Frost, which survey a writer's work as a whole.