Abstract:We investigate the failure of a local-global principle with regard to "containment of number fields"; i.e., we are interested in pairs of number fields $(K_1,K_2)$ such that $K_2$ is not a subfield of any algebraic conjugate $K_1^\sigma$ of $K_1$, but the splitting type of any single rational prime $p$ unramified in $K_1$ and in $K_2$ is such that it cannot rule out the containment $K_2\subseteq K_1^\sigma$. Examples of such situations arise naturally, but not exclusively, via the well-studied concept of arithmetically equivalent number fields. We give some systematic constructions yielding "fake subfields" (in the above sense) which are not induced by arithmetic equivalence. This may also be interpreted as a failure of a certain local-global principle related to zeta functions of number fields.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the "false sub - fields" phenomenon in number fields. Specifically, the author has studied a case of the failure of the local - global principle. This failure involves a pair of number fields \(K_1\) and \(K_2\), where \(K_2\) is not a sub - field of any algebraic conjugate sub - field \(K_1^{\sigma}\) of \(K_1\), but the splitting type of a single rational prime \(p\) in \(K_1\) and \(K_2\) cannot rule out the possibility of \(K_2\subseteq K_1^{\sigma}\).
### Specific description of the problem
1. **Background and motivation**:
- Two number fields \(K_1\) and \(K_2\) are called arithmetically equivalent if they have the same zeta function.
- This means that all rational primes \(p\) unramified in \(K_1\) and \(K_2\) have the same splitting pattern in these two fields, that is, their Frobenius elements have the same cycle type under these two permutation actions.
- Arithmetically equivalent number fields play important roles in some arithmetic problems, such as the Davenport - Lewis - Schinzel problem, the study of pairs of Kronecker conjugate polynomials, etc.
2. **Core problem**:
- The paper studies a more general situation: for two number fields \(K_1\) and \(K_2\), it is impossible to rule out the possibility that \(K_2\) is a sub - field of \(K_1\) only through local investigations (i.e., the splitting types of unramified primes).
- Specifically, the author has defined the concept of "local sub - fields" and further explored the existence of "false sub - fields" and their construction methods.
3. **Main contributions**:
- Some systematic construction methods are provided to generate "false sub - fields" that are not caused by arithmetic equivalence.
- These constructions can be interpreted as the failure of a certain local - global principle related to the zeta function of number fields.
### Main results
- **Theorem 1.1**:
- For all odd primes \(p\), the following conclusions hold:
a) Every number field \(K\) of degree \(p + 2\) with a symmetric Galois group \(S_{p+2}\) can be a false sub - field of some number field \(F\).
b) There exist infinitely many solvable number fields of degree \(2p\) as false sub - fields of some number fields \(F\).
- All of these examples can be chosen not to be caused by arithmetic equivalence.
Through these studies, the author has revealed some new, non - trivial phenomena in number field theory and provided new perspectives and tools for further research.