Reply to Moss and Tyler

Domain-Specific hypothesis, they seem at least as problematic for this theory as do many of the facts raised by Caramazza and Mahon for alternative accounts. However, the model is so underspecified that it is possible to develop explanations for most patterns of data. This flexibility significantly undermines the explanatory power of the account and essentially renders it unfalsifiable. The challenge for Caramazza and Mahon, therefore, is to specify what would count as evidence inconsistent with their hypothesis. More generally, we suggest that it is unlikely that any one of the existing theories (including our own) can accommodate all the ‘facts’, and that a more promising approach would be to develop integrative theories that combine insights from several different accounts (e.g. [10]).