One frequent appeal by determinists in the free will debate, involves invoking certain facts about neuroscience to deny efficacy to the conscious subject. In order to do this, one of the things the determinist must say, is that sense impulses are somehow processed unconsciously into a coherent whole, before they are presented to the ‘conscious’ subject as an ’experience’, and that this processing (along with pre-conscious processing of decision-making activity), shows that we are entirely causally determined.
Aristotle’s argument in Physics II 8 can be summarized as follows:
Dogs typically develop teeth good for biting and chewing.
A typical result is not a coincidence.
So it’s not a coincidence that dogs develop teeth good for biting and chewing.
If the development is not coincidental, it must be “for something”.
So the dog’s development is “for something”. (that is, it is goal-directed)
The problem with this argument lies in premise 4. Aristotle’s use of “for something”, implies some conscious agent that has intended the thing to be the case. You make this implication clear yourself, by calling the development “goal directed”.