ABSTRACT
So far in this book we have examined various linguistic aspects of mutation in Welsh as if they were totally separate and have often found this separation difficult. It is clear, however, that to account for both the triggering of mutation and the resultant sound changes, we also need to examine various linguistic interfaces. In fact, the investigation of the interfaces between different levels of linguistic analysis has gained the attention of many scholars in recent years, some of whom have used evidence from ICM in Celtic languages to illustrate their arguments.
