Published online by Cambridge University Press: 12 March 2014
A pair of r.e. sets A1, A2 are said to split an r.e. set A (written A1 Δ A2 = A) if A1 ∩ A2 = ∅ and A1 ∪ A2 = A. In the literature there are various results asserting certain splitting properties hold for all r.e. sets. For example Sacks' splitting theorem (cf. [So]) asserts that an r.e. nonrecursive set A may be split into a pair of Turing incomparable r.e. sets A1, A2, and Lachlan's splitting theorem [La5] asserts that A may be split into a pair of r.e. sets A1, A2 for which there exists an r.e. set B with B ⊕ A1, B ⊕ A2 <TA and with the infinum of the Turing degrees of B ⊕ A1 and B ⊕ A2 existing in the upper semilattice of r.e. degrees.
Two of the earliest observations establishing that splitting properties possessed by some r.e. sets are not possessed by others, are Lachlan's nondiamond theorem [La1] (and so, in particular, no complete r.e. set can be split nontrivially with degree theoretic inf 0) and the Yates-Lachlan construction of a minimal pair (a pair of nonzero r.e. degrees with inf 0). Other examples of this phenomenon, which are not obtained by interpreting degree theoretic results in the r.e. sets, are Lachlan's construction of nonmitotic r.e. sets [La2] (sets which cannot be split into a pair of r.e. sets of the same degree) and, later, Ladner's [Ld1, 2] and Ingrassia's [In] analysis of their degrees.
The authors would like to thank Mike Ingrassia, Iraj Kalantari, Jeff Remmel and Michael Stob for many helpful discussions.
Some of the work presented here was carried out whilst the first author held a visiting position at Western Illinois University, was announced in abstract, and was presented to the Second Biennial Cambridge Logic Meeting, Boston, 1983.