I this talk I outline the evidence for the dynamic multi-functionality of the various parts of the brain, and use it to put pressure on the leading account of scientific explanation in the neurosciences, Craver-Bechtel componential mechanistic explanation. Then, via a case-study of Starbust Amacrine Cells, a motion-sensitive neuron in mammalian retina, and of Hippocampal place cells, in both of which cases specific cellular function seems to depend on the state of the overall network, I introduce the notion of enabling constraints as an alternative to componential mechanisms. I will suggest that function sometimes arises from the (dynamic) constraints placed on structures, and that this style of explanation can capture a wider array of neural phenomenon than componential mechanistic explanation.