The specificities of language are obvious – fall, fell, falls, falling, fallen, stand, stood, stands, standing, and so on. These are points of potential difficulty for children and second language learners alike. But there are universals too.
A four year old on his or her first day at school may be asked “Who do you want to play with?” or “Who do you want to play with you?” But in the underlying structures who relates to different elements in different positions. Competent speakers understand the two questions in correspondingly different ways. Although the difference is by only by one final word, the onus in any response is quite different. The child may respond appropriately in different ways. But the child is most unlikely to reach a full understanding of the structure here until four or five years later. But this is never taught. It is hard to see how it could be.
The relevant principles here, known as ‘case’, ‘tense’, and ‘control’, appear to be universal across languages. The simplest explanation is that they are specified by the human genome.
The linguistic research on which this thinking is based goes back two and a half thousand years to work in India. Along with astronomy, linguistics is one of the two oldest sciences.
Children with speech and language difficulties may need help with any of a number of parts of what is an intrinsically very complex system – not just for English, but for any language. On my understanding of the data, case, or, tense, or control, or two of these, or all three, can be problematic for some children.