This is an interesting question. Often, folks think of mechanics and rules when considering adversarial GMs, but you can certainly have adversarial GM behavior in the social pillar. The one example above about saving the king and then the king turning tyrant, is just one. It could be a unique or logical development based on all the other elements of the campaign. Or, it could just be another in a long line of "players never win/win" choices by the GM. Whether its tolerable or not depends entirely on the context of the campaign and the GM's campaign reveals in total, and not just in one specific example.I am wondering if my gming style wouldn't be seen as adversarial... I am running an Eberron campaign pitched from start as more Zolaesque than Vernian tragedy. Yet a player pointed out that I did allow them to be proactive with their actions and succeed in their goal, but they feel the campaign let them often choose the lesser evil to save the world. Even the bad guy are arguably more nobly motivated than the good guys they work for. So they can do what they want, including saving the world, and get to shine doing so, but the world will still be equally sucking after their intervention (up to the point where they endeavour to change the nature of the world, an epic-level task). I am not delighting in making players miserable, they keep returning so they must have fun overall, but this in passing remark that anything that can go wrong will go wrong and sour their success (as in "if we save the king from an assassination plot we'll be heroes but we bet he'll become so paranoid about assassins that he'll become a tyran on a police state and even worde king tha. he was before...") makes me wonder whether it's seen as adversarial or just a tone disconnect (prompting me to hasten the point when they can embark on setting changing (and campaign concluding) events).