The concept of dyslexia can be confusing to clinicians who encounter children with severe reading difficulties within their day-to-day practice, given the problems of definition. Usually, such reading difficulties are accompanied by emotional and/or behavioural adjustment problems, or more widespread problems at school, hence the child's referral into child mental health services. Understanding the nature of the reading difficulty offers the prospect of prioritising treatment alternatives. For instance, if a child has a specific reading difficulty that has resulted in much teasing at school, and high levels of parental disapproval, he/she may then manifest symptoms of anxiety that will in turn further hamper the acquisition of reading skills. Remedial teaching, albeit accompanied by work with parents to foster an understanding of their child's difficulties, becomes an urgent priority. Alternatively, it is possible to envisage a situation in which an emotionally abused child is too distressed to acquire reading skills commensurate with their cognitive potential. With the latter child, the underlying causes of their presenting reading difficulty are emotional in origin rather than reflecting a core difficulty or abnormality in their cognitive processing. Although remedial reading may well be necessary at some stage, the clinician should first concentrate on alleviating the abuse. These are, of course, extreme examples and children rarely present in such a clear-cut way. Nonetheless, it is important for the clinician to know whether there is a group of children with meaningfully divergent reading difficulties, such that they can be characterised as having a specific reading retardation. If there are such children, how would they be identified and distinguished from those readers regarded simply as generally poor readers; what is the prognosis for such children; and what remediation would be appropriate and helpful?

In recent years, limitations of the discrepancy definition of specific reading difficulties have led researchers to propose that dyslexia is best described as a core phonological deficit. With studies of the normal development of reading as a backdrop, this paper reviews the phonological processing deficit hypothesis of dyslexia. The extant evidence suggests that phonological difficulties in dyslexia persist throughout development from the preschool years into adulthood, despite compensation of reading deficits. Moreover, individual differences in the behavioural manifestation of dyslexia are consistent with differences in the severity of underlying phonological difficulties. Implications for the early identification of dyslexia and for interventions to prevent reading failure are discussed.

We argue against defining dyslexia as requiring by definition the presence of demonstrable phonological difficulty, and also against defining it in relation to exclusion criteria. Instead, we suggest that what matters is evaluating any child's set of reading subskills against age-related norms for those subskills. We also argue that in considering the causes of reading difficulty it is essential to distinguish between proximal cause (some abnormality in the information-processing system that the child is using to read) and distal cause (the reason for this system being abnormal). The same proximal cause (e.g. poor phonic knowledge) can be the product of different distal causes in different children (e.g. the distal cause might be constitutional, environmental, or educational).

Despite some disagreements among the respective authors, the previous two articles contain many thoughtful recommendations about how to conceptualise reading disability. My reactions will be organised around two issues: the role of intelligence in understanding reading disability and the implications that recent work on subtypes has for the phonological core deficit model.

Snowling and Coltheart and Jackson provide an excellent overview of the contemporary cognitive and linguistic approach to the description and definition of dyslexia. Both contributors consider dyslexia to be a disorder of written language. Snowling focuses on the origins of the difficulty in a central ‘phonological deficit’. Coltheart and Jackson emphasise the internal modular structure of the reading and spelling systems, suggesting that different components may be impaired in different individuals, resulting in distinctive (phonological and surface) patterns of dyslexia. Both conclude that future research might focus on implications for effective programmes of treatment.

There is much wisdom in these two essays on dyslexia, by Snowling and by Coltheart and Jackson, as well as a useful and informative summary of many key research findings. Given the fact that debates about the diagnosis and concept of dyslexia have been running for at least the last 40 years (see e.g. Rutter, 1969), and given the immense amount of high-quality empirical research that has been conducted (see e.g. Bryant & Bradley, 1985; Goswami, 1994; Rispens, van Yperen, & Yule, in press), it might be expected that most of the main controversies ought to have been resolved by now. But they have not, as is apparent from the contrasting positions adopted in this forum. It might be hoped that, at least, any well-informed dispassionate reviewer should be able to arbitrate and so decide which view is most valid. Unfortunately, even that is not readily possible. A key problem is that the two papers seem to be starting from different premises and, thereby, tackling a somewhat different issue.

This paper describes attempts to implement a ‘functional analysis’ approach to the treatment of severe self-injurious behaviour in a 4-year-old boy living at home. Although a degree of success is reported, problems in systematically applying assessment and intervention procedures are described. The paper discusses the successes and limitations of such an approach in the home environment and in a crisis situation.

This paper discusses the difficulties researchers are likely to encounter when assessing the self-concept of children with Down syndrome. It describes four potentially appropriate assessment methods: projective techniques, nonverbal and/or pictorial techniques, construct generation techniques, and topological techniques. The paper concludes that methods based on the Pictorial Scale of Perceived Competence and Social Acceptance (Harter & Pike, 1984) and personal construct psychology offer the most appropriate means for assessing the self-concept of children with Down syndrome.