I have only realized this in light of the recent discussion of how there is a considerable chance civ7 is going to limit the freedom od map generation script, so that there is always "new world" to explore. Civ switching is merely a symptom of a deeper underlying idea which is quite revolutionary (and very controversial) undertaking. In fact, devs have almost explicitly said it, but only now I truly "got" implications.
Previous civ games, and almost? all 4X games in fact (and most strategy games in general?) have been sandboxes. You freely choose the civ, the map script, and then off you go. The game consists of a world with civs thrown there and broad static rules (not really changing with eras), according to which civs follow snowballing accumulating growth to victory. There is no broader structure beyond what occurs as a result of AI or human civs individually do.
The upside of this approach is obviously freedom. The downsides are not so obvious. Total sandbox is frequently uneven and anticlimactic in all video game genres, as there is nothing commanding the "story" from beginning to an end. In 4X games this manifests as their endless problems with snowballing and steadily decreasing enjoyment the longer you play. After all, at a certain point victors and losers are clear; why continue if you are just going to repeat minutiate of more and more micro (which is less and less meaningful) in the static slog till the anticlimactic end? Additionally, it is very contrary to the real history, with its very dramatic twists and turns.
There has been a thread on this forum where we discussed how to solve the endgame problem. The general take I personally had from it was that in order to do that the game must at its very core be committed to some sort of structure commanding buildup and release of tension, some deeper logic which messes with the usual static snowballing accumulation. Such structure may emerge spontaneously from game mechanics, but this is very hard to achieve for very complex games and civ6 has also failed at this task, having all those "rise and fall" mechanics which together were clearly insufficient.
So civ7 is based precisely on this intuition and devs taking the risk going with it all the way. The game is going to have a clear beginning, middle and the end; very strong instead of soft segmentation into three acts, which radically mess with the players and the game. Each of those three acts is supposed to have somewhat dofferent mechanics and tell its own story and they all together come into one great story (history) which has shifts and turns, quantum leaps and collapses instead of endless static growth of the usual 4X games. Game rules change to acommodate different needs of beginning, middle and the end and the players (civs) themselves change to reflect that.
But why are civs changing, couldn't they have stayed the same? It is worth noting that the new approach is also explicitly more historical in its aims: more historically narrow civilisations, era transitions resembling historical shifts, and I suppose mechanics reflecting each period more specifically. If we have an actual attempt to emulate ancient era defined by bronze and iron and pantheons and then attacked by barbarians, then we can't have Brazil there - it breaks the historical narrative. Historical eras change dramatically, so it makes sense for the same to happen to their protagonists.
Time will show how it's all going to work out, but personally I am mostly excited by Firaxis taking this massive undertaking (though I would like option to stay with the same civs) - it's actually very novel approach to 4X games. Sid Meier's Alpha Centauri had a narrative structure as its characteristic, but Civ7 goes way further than that. It's an attempt to reflect how different eras of history make society functio n differently, and an attempt to strike directly at the 4X problem of static snowballing till boring endgame.
