The examples I have chosen illustrate the problems of making abridged translations. There are no set rules or guidelines for this. As Klingberg said, his recommendations are just that - recommendations. They are not rules. They seem to work for certain kind of texts and a translator could use them as a starting point, but the actual shortening depends much on the type of the text, its function and the translator's judgement.
The function of the text and the readership seem to be the primary factors in deciding how to shorten the text. However, the ratio of certain kind of events may be increased or decreased considerably in the translation, if only these factors are taken into account.
The "selection on the basis of function" can easily become censorship and drawing a line between these two can be difficult. It is probably advisable to introduce changes, in function or otherwise, only when necessary, and the validity of the reasons for these changes should be considered very carefully.
The content and style change inevitably when making a shortened translation. Klingberg's recommendation of deleting or translating only entire paragraphs and sentences is based on the request of not changing the content or style. As these cannot remain unaltered, it is probably better to give priority to the function of the text and rewrite parts of the text when required.
It was interesting to note that the existing theory about the subject seems to be a by-product of studies in translation of children's literature, even though many of the abridged translation are made for adults and are non-fiction.
In this study, I have concentrated on abridged translations of fiction. It would be interesting to see whether abridgement methods used for non-fiction are different. In non-fiction, the priorities are probably different - the accuracy of content would probably be the primary concern and style secondary. Also, rewriting and editing may be more usual in non-fiction. These aspects would require longer and more detailed study.