Your encryption analogy doesn't really work either. Encrypted text encodes the key used to encrypt it. This is why it takes 2^N computational steps to recover t

Your encryption analogy doesn't really work either. Encrypted text encodes the key used to encrypt it. This is why it takes 2^N computational steps to recover the plain text, where N is the key size rather than where N is the text size. It's just that the computational cost of decrypting with sufficiently large keys is intractable. But human languages have much more exposed regularity than ciphertext from good encryption algorithms, so the computational cost is much lower. The "key", i.e. world model, in case of human languages is plausibly much bigger, but since human languages have significant exposed regularity, it is still tractable to compute, with the trade off that you need much more text to do it.