Subtracting log-likelihoods within a model

[Martin R. Smith]

I'm implementing a model in RevBayes.

Tersely, I observe data x under the model M(θ) with parameter θ.  I know that the data x satisfy condition C.

I wish to perform inference under the conditional, which requires the calculation of 

π(x | θ, C) = Pr(x | θ) / Pr({X satisfies the conditional} | θ)

Both Pr(x | θ) and Pr({X satisfies the conditional} | θ) can be computed by sending certain parameters to dnPhyloCTMC() on different pre-computed observations.  I can multiply the probabilities (I assume) by clamping these osbervations to successive entries within a model, but I can't find a way to divide.  Is this possible within RevBayes?