It has long been claimed that regular inflection is explicitly guided by rules and innate structures, and studies of differences in the way that grammatical constraints apply to regular and irregular inflection have been put forth as supposed demonstrations of the formal distinctions made between them by the language processor. Do the production and interpretation patterns of plural forms in noun-noun compounds reveal the workings of innate constraints that govern morphological processing? This paper examines the arguments in a series of seven nuanced experiments and finds - contrary to the nativist claim - that the interpretative and productive patterns of irregular plurals in compounds are entirely consistent with those of regular plurals. We argue that a simple account based upon learning from the input provides an economical and compelling account of the data.