The publication of John Smeds's doctoral thesis by Åbo Akademi is a timely one. The fiftieth anniversary of the appearance of Robert Graves's The White Goddess is due to be celebrated in 1998 and will, no doubt, give rise to a flurry of publications on the subject of the book and its influence on the poets and poetry of the latter half of the twentieth century. Fortunately Smeds's success is more than timely.

Critical opinion generally holds that Graves was first and foremost a love poet whose poetry dedicated to real life muses was, to some extent, justified by his loosely academic survey of matriarchal society which he referred to as his 'historical grammar of poetic myth'. His analysis of the goddess's reoccurrence in the myths, songs and poems of various civilisations posits that all 'true' writing reflects 'one story and one story only': that of the birth-death-devouring cycle of the king/poet/ bard and his queen/muse.

Surprisingly, few critics have tested Graves's pre-Goddess works for their consistency to his one-story theory and fewer still have gone on to examine his fictional prose for more than superficial references to proto-Goddess writings. This is where Smeds's work comes in. His analysis of Graves's use of myth is a significant departure from the majority of the existing criticism which tends to rely on biographical evidence to explain the waxing and waning presence of the Goddess in his poetry. Instead, he demonstrates how Graves's thesis evolves from early works ...