Research the crypt function for details on how UNIX password encryption worked historically, and the shadow system for when they moved that out of /etc/passwd completely.

UNIX as its now known never stored passwords in plaintext, that would be preposterous. /etc/passwd must be world-readable, they must be protected in some way. They didn't just encrypt the passwords, they encrypted them irretrievably. Not even the operating system can tell what the hashes are supposed to mean. Instead, when you login, it takes a hash of what you typed and compares the result to see if it's identical to the hash stored in /etc/passwd. If they match, you login.

There turned out to be vulnerabilities in letting everyone see all the hashes. If you happen to have the same password as someone else, you might notice the identical hash, something they fixed with a random salt which obscures the hashes from being checked quite so easily. Still, however, you can't go backwards from a hash, but you can check a thousand strings from a dictionary and all 256 of their salts to see if any of them become that same hash. They took measures to make crypt() too unwieldy to do that quickly, but advances in computing soon made it not unwieldy enough, and the password hashes were split out into a "shadow" file, which is only readable by root.

The old-fashioned UNIX crypt() algorithm is is mostly obsolete, now, but has been extended to allow other kinds of encryption in the same sort of stored hash.

As for echoing back to the screen, UNIX terminal control is also about as old as UNIX itself -- what else would they control them with back then? I suspect the ability to turn off echo predates UNIX, even.