This is not my area of expertise but as far as I understand we refer to a conditioned sound change if a phoneme when in a certain environment becomes another phoneme - there is a certain condition that propells the change. On the other hand we refer to an unconditioned sound change if all phonemes of, for example, two different kinds are merged into one phoneme - there is no condition as it happens in all instances, and ultimately there is a reduction in the number of phonemes as one is substituted for another.
