Skip to main content
4 events
when toggle format what by license comment
Dec 1, 2016 at 7:08 vote accept fhyve
Nov 30, 2016 at 22:03 comment added Noah Schweber But $V[H]$ satisfies "The unique real satisfying $\varphi$ does not contain $k$" - note that since the ordinals of $V[G]$ and $V[H]$ agree (being those of $V$), this is really the same sentence as the opposite one satisfied by $V[G]$. And that's a contradiction. So we can define the real $\nu$ names as "The set of all $k$ such that some condition forces $k\in\nu$" (this uses that the forcing relation is appropriately definable), and so the real named by $\nu$ in fact exists in $V$ already.
Nov 30, 2016 at 22:02 comment added Noah Schweber +1. For the OP, here's an outline of how this is shown: suppose $\nu$ was a name for an OD real in the generic extension. (That is, it is forced that $\nu$ is defined by some formula $\varphi$ with ordinal parameters.) Suppose $p$ is some condition forcing $k\in \nu$ ($k$ some natural number). Then $p$ forces "The unique real satisfying $\varphi$ contains $k$." Now suppose $q$ were any other condition which forced $k\not\in \nu$. By homogeneity, we can find generics $G\ni p, H\ni q$ yielding the same forcing extension: $V[G]=V[H]$. (cont'd)
Nov 30, 2016 at 20:54 history answered Andreas Blass CC BY-SA 3.0