In early D&D, there was the concept of an "alignment language."
The original "little brown book" D&D says only:
Law, Chaos and Neutrality also have common languages spoken by each respectively.
The Holmes basic rules, which come between "brown book" and Moldvay say:
Lawful good, lawful evil, chaotic good, chaotic evil, and neutrality also have common languages spoken by each respectively. One can attempt to communicate through the common tongue, language particular to a creature class, or one of the divisional languages (lawful good, etc.). While not understanding the language, creatures who speak a divisional tongue will recognize a hostile one and attack.
Moldvay D&D says:
Dungeons & Dragons, Basic Rules, page B11
Alignment Languages
Each alignment has a secret language of passwords, hand signals, and other body motions. Player characters and intelligent monsters will always know their alignment languages. They will also recognize when another alignment language is being spoken, but will not understand it. Alignment languages are not written down, nor may they be learned unless a character changes alignment. When this happens, the character forgets the old alignment language and starts using the new one immediately.
What?
So much of Dungeons and Dragons was very generic "sword and sorcery" fantasy that these rules always stuck out to me as an extremely sore thumb, or like one of those snails that has been infected by a fungus and begins to pulsate in random colors. What the heck is going on here?
So, alignment languages show up pretty early in D&D's history. Every intelligent being that is on the side of Law can communicate. Every intelligent being that has no particular feelings about Law can, too. If they experience a profound change in their feelings about this, they can no longer communicate with the people they used to. It isn't clear how much communication is possible in an alignment language, since Moldvay describes it as "passwords, hand signals, and other body motions."
In AD&D 1E, though, there is no such limitation:
Advanced Dungeons & Dragons, page 34
Character Languages
In addition to the common tongue, all intelligent creatures able to converse in speech use special languages particular to their alignment. These alignment languages are: Chaotic Evil, Chaotic Good, Chaotic Neutral, lawful Evil, Lawful Good, Lawful Neutral, Neutral Evil, Neutral Good, and Neutrality. The alignment of your character will dictate which language he or she speaks, for only one alignment dialect can be used by a character (cf. CHARACTER CLASSES, The Assassin). If a character changes alignment, the previously known longuage is no longer able to be spoken by him or her.
That cross-reference to the assassin class is there because assassins alone can learn the languages of other alignments. Probably not all of them, though, because now there are nine instead of three.
In AD&D 2E, the alignment language seems to have been silently dropped, and never spoken of again as far as I know.
When I was ten and played AD&D, the idea of alignment languages struck me as incomprehensible, and today even though I have come to love almost everything about basic and early-Advanced D&D, alignment languages still seem incomprehensible, weird, and unexplained.
What I want to know is this: how were (or are) alignment languages put to use, described, and conceived of in actual play ? So far, my use of them has always been "pretend they do not exist." I am curious as to how others view and address them.
Also: where did this bizarre idea come from? It smacks of being lifted from some fantasy book, but I can't recall ever reading anything like it.