We propose to ascertain if there is an effective procedure that determines the validity of an arbitrary first order $\forall\exists$-sentence in the upper semi-lattice of the $\Sigma^0_2$ enumeration degrees. It is known that the $\forall$-fragment of the first order theory of the $\Sigma^0_2$ enumeration degrees is decidable. Slaman and Woodin (1996) proved that first order theory of the $\Sigma^0_2$ enumeration degrees is un-decidable and more recently, Kent (unpublished) proved that the $\forall\exists\forall$-fragment of this theory is un-decidable.
This left open the question of whether the $\forall\exists$-fragment is decidable. Answering this question is equivalent to finding an effective procedure that determines, when we are given an arbitrary finite lattice $P$ and finite lattices $Q_0$, ..., $Q_n$ which extend $P$, if there is an embedding of $P$ into the $\Sigma^0_2$ enumeration degrees which cannot be extended to an embedding of any of the $Q_i$. Lempp, Slaman and Sorbi (to appear) have shown that the sub-problem where $n = 0$ is decidable.
In order to fully answer the question, we propose to answer the above problem in the following three sub-cases:
Case 1: Let $P$ be an arbitrary finite anti-chain and each $Q_i$ a one point extension of $P$.
Case 2: Let $P$ be an arbitrary finite lattice and each $Q_i$ a one point extension of $P$.
Case 3: Let $P$ be an arbitrary finite anti-chain and each $Q_i$ an arbitrary extension of $P$.
To answer the first case, more research needs to be performed on the properties of Ahmad pairs. In particular, we propose to determine the maximal size of an anti-chain where each two distinct elements form an Ahmad pair. Also, we propose to determine if the join of an Ahmad pair is non-trivially bounded from above. Cases 2 and 3 are extensions of case 1. It is believed that answering these questions will lead to an affirmative solution of the question of whether the $\forall\exists$-fragment is decidable.
Call for proposal
See other projects for this call