Would someone mind confirming (or refuting) the following...
Proposition. Let $K$ be a field and let $f \in K[X]$. Suppose $g \in K[X]$ is an irreducible factor of $f$. Let $L$ be a splitting field for $f$ over $K$. Then Aut$(L/K)$ permutes the roots of $g$.
Proof. Let $x \in L$ be a root of $g$ and let $\sigma \in$ Aut$(L/K)$. As $g \in K[X]$ we have $\sigma g = g$. Clearly $\sigma(x)$ is a root of $\sigma g$ and so the result follows. //
Many thanks :)
This is an expansion of the comment of LUXun.
There is a result that states that splitting fields are unique up to isomorphism.
Put it more precisely, it says the following
In other words, $\sigma$ is an extension of $\phi$.
In this case you apply this with $E_{1} = E_{2} = L$, $F = K(\alpha)$, $\bar F = K(\beta)$. The fact that $\alpha$ and $\beta$ are two roots of the irreducible polynomial $g \in K[x]$ yields the isomorphism $\phi$ $$ K(\alpha) \cong K[x]/(g) \cong K(\beta) $$ that carries $\alpha$ to $\beta$. By the theorem, $\phi$ extends to an element $\sigma \in Aut(L/K)$ that extends $\phi$, and thus carries $\alpha$ to $\beta$.