I understand that the strength of HF differs significantly depending on the concentration. I understand the basic theory behind this - homoassociation - and stabilization of the highly negative charge dense fluoride anion through hydrogen "bonds"/intermolecular attractions with undissociated solute $\ce{HF}$ molecules.
However, I am unclear on the Wikipedia explanation of homoassociation: http://en.wikipedia.org/wiki/Homoassociation
Regarding the equations - I cannot tell if the authors meant to refer to pure, liquid $\ce{HF}$ or a solution of HF. I'm thinking that the authors are referring to liquid $\ce{HF}$ because auto-ionization of HF is mentioned. I'm thinking that auto-ionization would only be significant in liquid $\ce{HF}$. Or in a concentrated solution.
Also, regarding the second equation - the authors make it look like the fluoride anion is forming an actual, intramolecular, bond with the $\ce{HF}$ molecule. I think the authors are implying the formation of an intermolecular hydrogen bond, correct?
On the other hand the bifluoride anion seems to be a real molecular entity (http://en.wikipedia.org/wiki/Bifluoride).
This only muddies the water further. I.e. which of the following best represents how the fluoride anion is stabilized, causing concentrated solutions of $\ce{HF}$ to ionize further than one would expect from $\ce{K_a}$ values?
$\ce{HF}$ ... $\ce{F^-}$ (hydrogen bonding)
or
$\ce{HF_2^-}$ (formation of bifluoride anion)
Also, any further links to resources about homoassociation would be appreciated.
ETA: Also, can't the fluoride anion form hydrogen bonds with water? The fact that concentrated HF is stronger than dilute HF than what mere $\ce{Ka}$ values suggest implies that H-bonding with water isn't as effective as H-bonding with HF molecules. Why? Is water less polar? On the other hand water can form a maximum of two hydrogen bonds with the fluoride anion, while a HF molecule can only form a maximum of one hydrogen bond with a fluoride anion.