The current system for identifying and dealing with learning disabilities reflects, to a large extent, political rather than educational or scientific considerations, the authors point out. It is time to let educators rather than politicians determine what is best for children. THE FEDERAL definition of learning disabilities defines specific learning disabilities in a way that makes three key points.1 First, learning disabilities are intrinsic disorders in basic psychological processing - they are the individual. Second, learning disabilities are diagnosed in terms of a severe discrepancy between achievement and intellectual ability. Third, the discrepancy should be with respect to one or more of seven particular areas of psychological functioning: basic reading skill, reading comprehension, listening comprehension, oral expression, written expression, mathematics calculation, or mathematics reasoning. Given this definition, it is little wonder that our society's identification and handling of learning disabilities is so confused. All three aspects of the definition are wrong and represent politically expedient decisions rather than fundamental knowledge about psychology and education. In this article, we consider these three aspects of the federal definition of learning disabilities and point out why each is erroneous. Then we suggest what needs to be changed. The Nature of Learning Disabilities Some psychological and educational theorists believe that learning disabilities are intrinsic - that is, wholly inside the individual; others believe they are extrinsic - that is, a result of our educational system of labeling.2 Neither of these views holds up to even cursory examination. Imagine a child born in a preliterate society. The child is one who, at another time or in another place, would have difficulties with phonological processing. But the child lives and dies and never has any difficulty whatsoever relating to reading because there is no reading in the society. The concept of reading is, quite simply, meaningless. This same principle can hold at any time and any place. Is difficulty in mathematical calculation a disability? Perhaps in the 1990s it was, and perhaps even at the turn of the 21st century it still is. But how about in the year 2020? Or 2050? Will children in the developed world even be doing calculations? And, given that almost all calculations will probably be done by machine, if children do not do them well, will anyone care? One can vary not only time but also place. Let's stay with the turn of the 21st century without considering the future or the past. Suppose Joseph is born in an English-speaking country and has severe difficulties with phonological irregularities. He may end up being labeled as having a reading disability, because the English language has so many phonological irregularities. Such words as are tough to sound out because the letters can represent so many combinations of sounds. But suppose instead that he is Jose in a Spanish-speaking country. Spanish has an extremely regular pattern of correspondences between the written and spoken forms of the language, so Jose might possibly not be labeled as having a reading at all if he grew up speaking Spanish. So the first aspect of the federal definition, the idea of learning disabilities as wholly intrinsic, is not correct. But there's more. The choice of the seven categories specified in the federal definition is itself quite arbitrary. For example, the list does not include hunting. But a boy born into a society of hunters and gatherers who could do most of what he is required to do except hunt effectively might be labeled as having a hunting disability because the inability to hunt could have disastrous consequences for his future. And a hypothetical society that emphasized the great importance of music might label children who could not carry a tune or learn to play an instrument as having a musical disability. …