I apologize because clearly I didn't explain well my thought.
I know that the entire diagrams at p86 represent a situation hard to verify in a real game. But I focused on the right hand side of the diagram, where there are the bases marked with capital letters. Imagine a BG charged in front and on a flank only (not both flanks). One base of this BG
must turn to face the flank charge. If you look at p56 and follow, you'll find a such BG
must turn at least one base also if all its bases were in contact with enemy at the moment flank is charged. So, the only situation where a BG could not turn a base to face a flank attack is that described at p76, and anyway it is up to player to turn or not one base.
This bring to the conclusion the rules described at 2nd bullet of p86 points to a situation which practically doesn't exist in the game. So IMHO there are 2 possible situations:
- Authors are very scrupulous, or
- They thought to get a different result
It's clear my opinion could be influenced by some reminiscences of DBM (I don't play it since 5 years, but still I can remember the general structure of the game), but my feeling is that intention was to exclude or to limit overlap on a corner where there are 2 enemy BG, but this practically has not been achieved. IMHO, when 2 BG fight against one that face in two directions, there shouldn't be overlap in the corner between the two side contacted, if the 2 BG are contiguous, because this is too much penalizing for the BG facing in 2 directions and also because, if you think to a real situation, there is no room to claim a contribute from troops not in front contact with enemy; if you look at miniatures (imagine the situation of diagram at p86 left side) you can say the right front base of the left cavalry BG is in contact with the side of the base which is fighting to its front, but if you think a such situation in real life it shouldn't happen so (never experienced, but...).