There are many ways to organize formal proofs of typed formulas, e.g. natural
deduction, the sequent calculus, or its dual, tableauxtableaux
proof, or Hilbert style systems to name a few. We choose a sequent
calculus presented in a top-down fashion (as with tableaux). We call this a
refinement logicrefinement logic (RL). The choice is
motivated by the advantages sequents provide for automation and
display.18 Here is what a simple proof looks like for
,
;
only the relevant hypotheses are
mentioned and only the first time they are generated.
![]() |
by ![]() |
|
1. |
![]() |
by
![]() |
1.1 |
![]() |
by
![]() |
1.1.1 |
![]() |
by
![]() |
1.1.1.1
![]() |
by ![]() |
|
1.1.1.2
![]() |
by ![]() |
|
1.1.2 |
![]() |
by ![]() |
![]() |
|
| | |
1.
![]() |
|
| | |
1.1
![]() |
|
![]() |
|
1.1.1
![]() ![]() |
|
![]() |
|
1.1.1.1
![]() ![]() |