The way I read the OGL 1.0a is that the Contributors are always the licensor (using your view a specific case might be WOTC being 'the Contributors' for the SRD 5.1). Thus, it seems that Licensees of the OGL 1.0a aren't sublicensing anything when posting the OGL 1.0a, but instead just posting 'the Contributors' offer.
What do you mean, in legal terms, by "posting the Contributors' offer"?
Also, when you refer to "the Contributors' offer", which Contributor are you referring to? There is no single offer made by all owners of the copyright in OGC. For instance, WotC made an offer in respect of the OGC it declared in its various SRDs some number of years ago (different numbers of years for different SRDs). But probably some publisher published something under the OGL, licensed by WotC, just last week (or last month or whenever), and that publisher made its offer at that point.
If so, then this has downstream implications on what deauthorization would accomplish (assuming they can deauthorize) and also, what a withdrawl of the offer would accomplish (assuming they can withdraw the offer).
What downstream consequences?
I can't work out what legal state of affairs you're setting out to describe, and so I can't work out what legal consequences you think might follow from it.
But frankly I don't understand why you and some other posters are making such heavy weather of relatively clear elements of the OGL.
W(otC) offers to license elements of its SRD (ie the ones it notifies as OGC) pursuant to the terms of the OGL v 1.0a. Parties who take up that offer enter into a contractual relationship with WotC. The contract includes conferral on them of a power to license the OGC, and derivative material of that OGC, to downstream parties. They are also contractually obliged to offer such a licence to all the world, in the terms of the OGL. (Under section 9, they have a power to pick and choose among authorised variants when they do this. In practice, this has not turned out to be very significant because there are only two variants and everyone seems to prefer to use v 1.0a because it provides clearer protection of Product Identity.)
Let A be a downstream licensee of W, who in turn licenses to B. At that point, B enters into a contract with A. Does B also enter into a contract with W? If B is using W's OGC (and using W's OGC is defined to include using material derivative of W's OGC, which might be A's OGC) then the answer is
perhaps - it might depend on B's state of mind vis-a-vis W's offer. On the other hand, if W has withdrawn its offer, then the answer presumably is No. Does it matter? It's not clear why it would, given that all that would flow from B's contract with W would be a licence in the terms of the OGL in respect of W's OGC, and B can already get that from A, in virtue of the powers that A gains from their licence from W.
Suppose, contrary to what I think is the only plausible interpretation of the OGL v 1.0a, W has a power to void - or "de-authorise" - all the licences that exist in respect of its OGC. I don't see how it would matter to the exercise of that power, and the consequences of its exercise, that some of those licences were granted directly by W and that others were granted by A by way of sub-licence.
But as I said, I don't see any plausible argument that W has such a power. As I've already posted (eg #2140) the real action, it seems to me, pertains to the subject-matter of the licences granted by W, and the way (if any) in which the endurance of that subject-matter depends upon W keeping its offer to license on foot. That doesn't depend either on how the licences in respect of that subject-matter were granted, as best I can see.