Japanese and English differ in the idea of “ego”.

In English, an “ego” (= subject) is always responsible for the actions and therefore required. E.g. you would say “Somebody likes the cake” to express that somebody likes the cake. This requires an ego or an agent (= “somebody”) in order for the cake to be likable.

In Japanese, that is not the case. To say “Somebody likes the cake” you could say “だれかはケーキが好きです”. However, literally this means “When it comes to somebody, the cake is likable”. Furthermore, just saying “ケーキが好きです” (see Particle ∅) would be sufficient, which translates to just “The cake is likable”.

Implications §

This has implications on how you would “say” things in Japanese. For example: “ケーキを食べたい” (“[I] want to eat cake”) is much less common than “ケーキが食べたい” (“Cake is ‘desirable to eat’”).

It often comes down to choosing between Particle を (more focus on ego) and Particle が (less focus on ego).

Note §

From Unlocking Japanese:

The key to the apparent oddness of these forms of speech really lies in the Western, and particularly Anglo-Saxon, idea that only the human ego can actually “do” things. The Japanese (and even vestigially the Spanish) usage reflects a more “animist” notion that “doing” is something that takes place on many levels, often with “things” as the “doer”.

This [English] is a cultural prejudice, and the prejudice is: that there should always be a human agent, a human ego, at the forefront of every action.