Can we make isoglosses?

[Koen G]

If x is interchangeable with cks then I don't see how exactly that would point to the expletive on its own. To me it looks much like the p/b interchangeability. It would be used as genitive for bock, no matter whether expletive (for God's) or literal (for goat's)

I'm probably looking at this differently (as a linguist), but the two phenomena are not the same.

For p/b, the change depends on the dialect of the speaker. If they say /pok/, they write a [p]. If they say /bok/, they write a [b]. You write what you hear.
For [x] though, the question is not one of dialects but of spelling. Assume a dialect area where the word "bock" or "pock" is pronounced with a /k/-sound at the end. In that case a conventional spelling would be [pock], which is all fine. 

Now if you turn this into a genitive form, you say /poks/. The standard procedure would be to write [pocks]. BUT, a weird spelling option exists (from Latin), where you can represent /ks/ by [x]. To do so is a spelling choice, not a reflection of one's way of speaking. 

So my question is: when did medieval German scribes make this spelling choice, to represent a /ks/ sound in their native language by the "Latin" [x]. It certainly happened, but did it also happen for genitive forms? People like "word images" to stay consistent, so pock+s = pocks. If you write [pox], you infringe on the word image of "pock". 

So why is "pox" more likely for the expletive than for the animal genitive? Well, for one, I only recall seeing examples of the -x spelling with the expletive - but I have not studied this in depth yet. And second, I suspect that when the expletive form is used, people think less about the original word image [pock] or [Gott] or whatever. "Pox" just becomes its own word with its own word image, so the /x/ is more likely.

In short, writing an [x] in /poks/ or /boks/ is a spelling choice, and my hypothesis is that it is more likely to occur in the expletive usage. However, I might be wrong, and this should be tested. The research question is basically: did 15th century scribes ever spell genitive forms ending in -/ks/ with an /x/? If they did, the question "ingredient vs expletive" remains open. If they didn't, that is a very strong argument in favor of the expletive.