Yep. Donnel doesn't say "printed", so you lose all keywords, both printed and gained, and even if you have multiple instances of the same keyword. Literally ALL keywords.

It's worth noting that not only is this the way "loses/gains all keywords" is handled in Conquest, it is also the way "loses/gains all keywords" was handled in AGoT 1.0.

The terminology "loses/gains all instances of <whatever keyword>" is what would be used if the single loses/gains effect is to "overpower" all other modifying effects. Otherwise, "all keywords" refers to a single instance of each keyword that exists in the game - not all instances necessary to ensure that the card will never/always have a functioning keyword.

Of course, FFG could rule differently for AGoT 2.0, but they have been pretty consistent about this across most games.

ktom is right, I was wrong (memory glitch i guess =_= )

“All keywords” means “one instance of each keyword”.
 

Hello Nate!

Somehow I remember asking you this, but I cannot find any logs, so I guess I'll have to ask again:

Sweet Donnel Hill
While Sweet Donnel Hill is defending, each attacking character loses all keywords.

Does 'all keywords' mean 'all instances of all keywords', or 'one instance of each keyword'?

So that if an atacking character has two instances of renown, will he lose both?

I think I remember you saying that if you wanted it to mean 'lose one instance of each keyword', you'd word it as 'lose each keyword'. Might be my mind playing tricks on me though...

Thanks Nate!

I’m trying to recall when I made such a ruling, but I don’t have any record of it.
 
In first edition and in Conquest we treated “all keywords” as “one instance of each keyword.”
 
I think it’s best to remain consistent with that interpretation here.
 
So that, if a character has stealth, loses all keywords, and is gaining stealth, it is
at +1 stealth (1 for printed, -1 for loses all, and +1 for the gain).
 
Thanks!
 
Nate

If a character bypass one of mine using stealth, and then I defend with Sweet Donnel Hill, can I also defend with my bypassed character?

No. Two reasons:

1. If the attacking character loses stealth, it does not retroactively "un-bypass" the defending character. Once a character is bypassed by stealth, it cannot be declared as a defender in that challenge, period. The attacker does not need to keep stealth (or remain in play, etc.) in order for stealth to keep working.

2. All defenders are declared at the same time. So even if #1 was not true, defending with SDH wouldn't help because by the time he is declared as a defender (so that the attacking characters lose their keywords), it would be too late to declare additional defenders, "retroactively un-bypassed" or not.

"the name of the keyword stealth is misleading, it should be called "deceive" or "distraction""

~mplain

cannot agree more