I'll give you the benefit of the doubt that this was not intentional, but you are misrepresenting my argument here a little bit. I do not claim that there is an in-fiction explanation for "why cover and shields don't work with barkskin until they suddenly do work with barkskin and override it." My claim is that there is an in-fiction explanation for the mechanics of barkskin, which as @Harzel eloquently put it, is that barkskin offers a separate "layer" of defense to normal AC, and that cover and shields apply to the latter layer, not the former.It's not a matter of handwaving... we all know we can just ignore it all and not put any thought into it. The question has always been in these last few posts for myself and Charlaquin whether there's an actual logical explanation for what is happening in the fiction to explain why cover and shields do not work with barkskin... until they suddenly do work with barkskin and override it. They claimed there was... I said there wasn't.
That math is accurate. I only disagree with your view of what fiction that math reflects.And I've noticed that no one has tried to counteract or counterclaim the math I put up explaining how shields, DEX, and cover don't work and have no effect at one AC point but then suddenly pop into reality when enough of them get combined together and reach a new AC point.
I don't follow?The only explanation has been to remove the math and bonuses of what protections actually have in their entirety, and wait until after a hit has been declared to then explain away why stuff now suddenly "works":
D&D 5e does not distinguish between "hard" and "accurate" attacks. Every attack that is accurate is hard. Every attack that is hard as accurate. Both force and precision are represented by to-hit bonus. However, AC does not always represent both evasiveness and toughness. In particular, an inanimate object's AC only represents toughness, as an inanimate objects has no ability to evade. My argument is that the AC granted by Barkskin is akin to the AC of an inanimate object.If the druid only has a shield, then the druid has an AC of 16 and every hit is only of the "really hard" type that gets through barkskin... and the shield never not once comes into play (because if it did, it should have contributed to the AC of the druid.) But since the spell somehow keeps a shield from ever getting in the way of any attack and instead they are all really hard... it doesn't.
But if the druid has the shield and and is behind 3/4ths cover (and now the AC of 17)... then now all of a sudden the shield does start getting in the way and helps protect the druid. And now every hit somehow changes over to only the "accurate" ones that get by the shield deflecting, and the cover blocking. But no more worrying about the really hard attacks.... those don't happen anymore since the druid moved behind cover. Funny that... outside of cover every attack is hard, which is why shields don't work.. but behind cover every attack is now accurate and thus the shields have an affect on the AC. And never the twin shall meet. We will never an accurate attack out of cover, and never a hard one behind it.
In order to hit hard enough to penetrate barkskin, an attack roll must be forceful enough to exceed AC 16. We can imagine this is the same AC an oak tree might have. If between any armor the druid might have, whatever the druid's Dexterity bonus may be, whatever cover the druid may be behind, and any shield the druid may be using, if that number is still below 16? It still takes an attack roll of 16 to harm them, because their skin is like the bark of an oak, and an oak has 16 AC representing the force required to pierce that bark. However, if between the druid's armor, Dex mod, cover, and shield, their AC is greater than 16? Well, it's possible that an attack roll that would be high enough to damage an oak tree might still not harm the druid. Because unlike the oak tree the druid can dodge.
Put another way, hand an oak tree a shield, and its AC doesn't increase. Strap a breastplate to it, its AC doesn't increase. Stand in front of it, its AC doesn't increase. Build a wall around it with a slit wide enough for a sword to fit through, its AC doesn't increase. Barkskin makes a druid's skin work like that. But to get to the druid's skin, you have to get past the druid's armor, shield, and cover, and since a druid can move, that offers a separate layer of defense that must be overcome before you can compare your attack roll to the AC of their oaklike skin. It just so happens, because D&D combines accuracy and force into one number, if the armor/dex/cover/shield layer adds up to more than 16, any attack that can beat that can also beat the AC of the bark layer. If it adds up to less than 16, any attack that can beat the bark layer's AC can also beat it.
This is not the kind of debate where one position can be "proven" or "disproven." We are each presenting our cases for interested parties to consider and draw their own conclusions.But of course, you're right... none of this actually matters in the slightest, but like I said originally I just enjoy debating the attempts to align fiction to mechanics in this case. Which as far as I'm concerned, I've still succeeded in proving that they ARE none.