We will later see that the analytic sets correspond to the sets definable by means of Σ11 formulas, that is formulas in the language of second order arithmetic that have one existential function quantifier.
Therefore, we will denote the analytic subsets of X also by
In a course on computability theory one learns that there are effectively inseparable disjoint computably enumerable sets. i.e. disjoint c.e. sets W,Z⊆N for which no recursive set A exists with W⊆A and A∩Z=∅.
In contrast to this, disjoint analytic sets can always be separated by a Borel set - they are Borel separable.
Proof
Let f:NN→A and g:NN→B be continuous surjections.
We argue by contradiction. The key idea is: if A and B are Borel inseparable, then, for some i,j∈N, A⟨i⟩=f(N⟨i⟩) and B⟨j⟩=g(N⟨j⟩) are Borel inseparable.
This follows from the following observation:
If the sets Rm,n separate the sets Pm,Qn (for each m,n), then R=⋃m⋂nRm,n separates the sets P=⋃mPm,Q=⋃nQn.
By using this repeatedly, we can construct sequences α,β∈NN such that for all n,
Aα↾n and Bβ↾n are Borel inseparable, where
Then we have f(α)∈A and g(β)∈B, and since A and B are disjoint, f(α)=g(β). Let U,V be disjoint open sets such that f(α)∈U, g(β)∈V. Since f and g are continuous, there exists N such that f(Nα↾N)⊆U, g(Nβ↾N)⊆V, hence U separates Aα↾N and Bβ↾N, contradiction.
The Separation Theorem yields a nice characterization of the Borel sets.
Sets whose complement is analytic are called co-analytic. Analogous to the levels of the Borel hierarchy, the co-analytic subsets of a Polish space X are denoted by
we obtain a regular Souslin scheme Q=(Qσ) with AQ=AP. Moreover, if the Pσ are from a class Γ, and Γ is closed under finite intersections, then the Qσ are also from Γ. In particular, any analytic set can be obtained from a regular Souslin scheme of closed sets via the Souslin operation.