The analysis of sound symbolism in poetry is one of the more promising applications of computational methods. This paper proposes using database software with spreadsheet capabilities to give maximum versatility in the examination of consonant alliteration. In this case the database is drawn from a 10th century anthology of classical Japanese verse called theKokinshû. In recent years scholars have pointed out a few obvious examples of sound symbolism inKokinshû poetry. This study attempts to show that with these few notable exceptions, poetry of the period seems to have striven toward a balance in sound, avoiding techniques such as word initial alliteration which might call attention to itself.