Oh, ok. Here goes... hoping my tex skills are up to the challenge:
First, let's start by writing down the chemical potential of water: this is the amount of energy required to add a molecule (or a mole) of water to a solution:
[tex]\mu_{W} =\mu^{(0)}_{W} + RT ln X_{W} + P\overline{V_{W}}[/tex]
Where [itex]\mu^{(0)}_{W}[/itex] is the chemical potential defined at STP, [itex]X_{W}[/itex] the mole fraction of water, [itex]\overline{V_{W}}[/itex] the partial molar volume, R the gas constant, P the hydrostatic pressure, T the temperature.
If we have two compartments separated by a water-permeable membrane, such that one has solute and the other does not, both compartments are at internal equilbrium, then [itex]\mu(1)_{W} = \mu(2)_{W}[/itex]. Substituting that big expression about for [itex]\mu_{W}[/itex], with the knowledge that X_W(2) is 1 (pure water) and that a hydrostatic pressure difference must exist to oppose the flow of water across the membrane, we get
[tex]\Pi\equiv[P(1) - P(2)] = -\frac{RT}{\overline{V_{W}}} lnX_{W}(1)[/tex]
Now, for a dilute solution ln(X) = X, and doing a few other manipulations of X into n you end up with the van't Hoff expression.
How's that?
Edit: oops, made an error in tex formatting.