The Modern Forgery Hypothesis

[JustAnotherTheory]

i've come across almost everything you can imagine: juillet, jullet, julet, juingnet, juinget, jungnet, jugnet, jules, jullé, julh, iuylol, even an extremely rare joingnet or joignet. Not a single jollet has ever been found, other than a lone "joilet" that I had to request from the BNF library. 

Aberil is much the same. There are loads of abril, abrel, aberel, abrillen, abrellen floating around, but actual usage of "aberil" is almost non-existant and highly unlikely to come across. 2 of the 3 aberils we have (aberillen) are actually direct copies of an extremely old poem (and are nestled within 700 pages of text). The only other places we've seen aberil are in farmer records, where the people writing were far less educated and more likely to phonetically write their dialect. 

One could make the exact same argument about the lobster, whose:

• legs are on its tail (of which only a few very rare cases exist),
  
• has four tail flaps (of which no examples have ever been found, only 3 or 5 flaps exist),
  
• scissors on its head (of which still no examples exist).
  

It's not unlikely that a forger would simply:

• make mistakes by accident, say, because of a copy error, or
  
• make deliberate mistakes in an effort to obfuscate the manuscript even more.
  

The fact that you have found no examples of aberil or jollet in manuscripts, points to one of these hypotheses. It could therefore be that they were simply copied wrong, deliberately or otherwise, as the vast majority of the imagery in the VMS.