Partial Combinatory Algebra

A partial combinatory algebra is a Partial Applicative System ( A , β‹… ) 𝐴 β‹… (A,\cdot) equipped with a function ⟨ βˆ’ ⟩ : 𝖳𝖾𝗋𝗆 A ⁒ ( n + 1 ) β†’ 𝖳𝖾𝗋𝗆 A ⁒ ( n ) : delimited-⟨⟩ β†’ subscript 𝖳𝖾𝗋𝗆 𝐴 𝑛 1 subscript 𝖳𝖾𝗋𝗆 𝐴 𝑛 \langle-\rangle:\mathsf{Term}_{A}(n+1)\to\mathsf{Term}_{A}(n) called bracket abstraction such that:

Where ⟦ e ⟧ ρ \llbracket{e}\rrbracket\;\rho denotes the interpretation of a term with n 𝑛 n variables as a Partial Element of A 𝐴 A , and e ⁒ [ x ] 𝑒 delimited-[] π‘₯ e[x] denotes substitution of the final variable of e 𝑒 e with x π‘₯ x .

Note that this presentation of PCAs is different from the one often found in the literature, which encodes a sort of simultaneous bracket abstraction.

The bracket abstraction operation axiomatizes the S-M-N Theorem for A 𝐴 A , which essentially lets us form Closures.