We provide a commentary on current debates about the definition of dyslexia. We agree with others that dyslexia is best thought of as a dimensional disorder with the best established causal risk factor being a deficit in phonological processing. Dyslexia is particularly common in children from families with a history of dyslexia and in children with preschool language difficulties. We argue that definitions may differ depending upon their purpose. Traditional discrepancy definitions may be useful for research purposes, but when considering the provision of educational services discrepancy definitions are not useful since all children with reading difficulties require reading intervention regardless of their level of IQ.