Evaluate the statement: 'If for every $\epsilon > 0$ there exists a $\delta > 0$ such that $P(\epsilon, \delta)$, then there exists a $\delta > 0$ such that for every $\epsilon > 0$, $P(\epsilon, \delta)$.' Is this statement logically valid in general?