An easy approach to the walls must be provided against: indeed they should be surrounded by uneven ground, and the roads leading to the gates should be winding and turn to the left from the gates. – Marcus Vitruvius Pollio
A wall is not a natural thing in the real world – it is something that must be built by humans. It is not exactly a natural thing in chess either; however, the strange thing about chess is that the separate and opposing actions of two players build the game – like a city built by two fighting antagonists. Thus the position that arises in a chess game is usually the result of a chaotic and organic process, rather than an intentional one.
I think there is some intellectual pleasure to discovering a “fortress” in an endgame. To see that all attempts by the attacker will be rebuffed. Normally a fortress occurs when the game is very simplified and the defender has only a small part of the board to defend. The attacker controls the majority of the board and usually has more material, but can do nothing with it because he cannot pierce the defender’s small encampments:
This is one of a large number of basic fortresses. Despite White's extra bishop, he cannot win. Any attempt to approach by the king will lead to stalemate, and otherwise he cannot break into Black's fortress. Here is a more unusual example, from a 1930 composition by G. Zahodyakin:
However, what we are concerned with here are not those – not infrequent – instances where the defender holds a fortress deep in the endgame. Instead, we will look at some examples of a fortress extending over the entire board. This is fairly rare in practical games; but check out the following position, which – according to GM Andrew Soltis in his The Art of Defense in Chess occurred in a junior tournament in Belgium, although he did not name the players:
Obviously this kind of scenario doesn't arise often in practical games. However, study composers, with their rich imagination, have dabbled in this kind of idea. For example, the following:
Vitaly Chekhover used this idea in a few of his compositions, such as the following well-known one:
But I like even more the following, also by Chekhover:
You might object that these extremely unusual positions are not applicable to practical chess, and only serve as entertainment. While I don't see what is wrong with entertainment, I also do not agree that these things are useless for your chess improvement. The concept of a fortress is crucial to practical chess, and this includes the middlegame. At the root is the understanding of the opponent's possibilities. There are a great many "semi-fortresses" which can arise just after the opening. See for example this position:
Now it is your turn to practice your stone laying. Build a chess wall to rival the Great Wall of China!