[PHI 3000] Aristotle's regress argument

In the Metaphysics (1006a, 8-10), Aristotle writes:

It is impossible that there should be demonstration of absolutely everything; [for then] there would be an infinite regress, so that there would still be no demonstration.

Aristotle is making a reductio ad infinitum argument that can be reconstructed as follows:

1. For any proposition p, reasons can be given for/against p.
2. q and r are reasons for/against p.
3. If q and r are propositions, then reasons can be given for/against q and r.
4. q and r are propositions.
5. (Therefore) Reasons can be given for/against q and r.
6. s and t are reasons for/against q.
7. If s and t are propositions, then reasons can be given for/against s and t.
8. s and t are propositions.
9. (Therefore) Reasons can be given for/against s and t.
10. u and v are reasons for/against s.
11. If u and v are propositions, then reasons can be given for/against u and v.
12. u and v are propositions.
13. (Therefore) Reasons can be given for/against u and v.

And so on, ad infinitum.

What do you think about Aristotle's regress argument? Is there any way out?