A different approach to languages

In the last session I ran of B2 Keep on the Borderlands, one of the players noted that a paucity of linguistic skills stopped the PCs from having any kind of non-combat interaction with the kobolds. Someone would have had to actively take the language, and the PCs present didn't have high enough Intelligence to get bonus languages. It's a picky and difficult thing, and monstrous languages are kind of a crapshoot where players make guesses at what might be useful.

At the same time, the ability to talk to monsters is one of the more interesting parts of D&D. As an RPG, players can strike any bargain or make up any trick they want; the monsters don't have to be bags of hit points that deal damage. So this is an area of serious potential for roleplaying if the language issue can be fixed.

There could be a solution closer than people think: alignment languages. I've written about them before, and I don't have the strong objections to them that many people seem to. They are often ridiculed in no small part because the Dungeon Masters Guide makes it clear that alignment languages are meant for relatively abstruse and philosophical discussion, with practical matters all but verboten. I think this is wrong-headed.

Alignment language could play a very interesting role as a lingua franca between humans and monsters. I think this might work best with a Holmes-style five point alignment where some monsters have different alignments, but continuing to take the Chainmail-style stance that alignments are basically "sides." This works well in Holmes because goblins, hobgoblins and kobolds are lawful evil while orcs, gnolls and bugbears are chaotic evil. That means a character can realistically talk to goblin-types but not orc-types, and vice versa.

The shift would be to make alignment languages fully functional tongues - LG being something like Latin, CE being akin to the Black Speech of Mordor, CG as Sindarin, etc. While they have a ceremonial use, they are widely known and more importantly are complete languages. The other change is to make them learnable by individuals not of that alignment. So a PC could speak the CG and CE languages - and it might be suspect - but it would certainly allow them to communicate with all CE creatures. Elves would get CE in addition to their alignment language, and dwarves get LE. This is more or less close to the creatures they can speak with per B/X.

This approach has two benefits. One is that there is a "level safety" to this; if kobolds are only encountered in this one cave, your 12th level magic-user won't have a wasted slot with "kobold" on his character sheet. Instead he'll have "Lawful Evil" or the in-campaign name of the same language. The second advantage is that it lets the referee play around with different monster archetypes without making things totally incompatible. What if your PC took Orc and Goblin as languages but all you run into are beast-men? Then the slot gets totally wasted. But if there's a Chaotic Evil language you can learn in addition to your own alignment tongue, you can talk with both orcs and beast-men.

Of course, this is something of a worldbuilding question - but so is the default D&D language situation. Having "Common" and various monster and alignment languages instead of pseudo-historical or historical human languages is a definite shift away from most fantasy works, especially post-Tolkien ones that revel in their worldbuilding. The best way to do this, IMO, is to turn these alignment languages into a worldbuilding tool instead of a hindrance. Figure out why there's a common language among kobolds, goblins and that evil empire.

The one other shift referees may want to consider is whether humans have the alignment language related to their own alignment, or the prevailing alignment of their home area. For instance, in a Lawful Good town - how would a Chaotic Evil fighter pick up the Black Speech? More likely he knows enough of Church Latin to get by. It's another layer of complication but once we take it up, it makes a bit more sense out of the whole concept.

On Reading and Writing Magic

First off, I want to give a hat tip to D. Vincent Baker for his The Seclusium of Orphone of the Three Visions for inspiring this particular article. It's one of a number of new releases on the Lamentations of the Flame Princess store that were made available today.

From the original edition of D&D onward (Holmes copies the text almost exactly), Read Magic has been a fundamental spell for reading scrolls and generally "magic." Without it, the instructions on scrolls, spellbooks, wands, and so on are incomprehensible. In Seclusium, Vincent discusses how - and this is an "in his campaign" moment - there is something personal of the wizard wrapped up in his own writing, explaining why magic writings even in an age of print as assumed by LotFP are still hand-written. That's one way to do it, but I've been thinking of another.

As I see it, the words of magic-user spells are inherently magical. Uttering them is enough to tap into unknown energies and change reality itself. From there, the words on a scroll or in a magic-user's memory are in a semi-stable form, where they've been encoded in some way so that the magic-user has to interpret either the contents of the scroll or his own memory to transform it into the actual words. This form is based on a personal cipher that only the magic-user himself knows, and disappears once it is actualized. That's why the memorization isn't permanent: the form requires the magic-user himself to complete it.

The level of a spellbook has to go even further. Since the magic-user must be able to memorize the spell without it disappearing from the book, it has to be an encoded form of the already semi-stable form. This means that the magic-user has two personal ciphers, one for memorization and scrolls, and the second unique to his spellbooks. It follows that these ciphers themselves are also magical, and that writing with them takes a certain number of pages, words, diagrams and so on that, when read, will store the semi-stable form of the spell into the magic-user's memory without casting it.

From this perspective, the ciphers are "securing" the underlying spell in place. No magic-user knows another's ciphers; they are not just a secret, but something crafted as an apprentice that is as much a part of the MU as his arm. And that's where the Read Magic spell comes into play. Read Magic lets the caster read without casting, and what it's doing is the work of simultaneously deciphering the original and presenting a re-ciphered version to the magic-user as if it were written on his own scroll or spellbook. This allows, as per Holmes, for a MU who has read a scroll written by a different magic-user once to cast the spell from it without casting a second Read Magic.

One side effect of this is that magic-users who have Read Magic memorized has a 25% chance to spontaneously decode any nonmagical message written with a non-mechanical cipher (alphabet substitution, etc). Actually casting the spell and using it to read a mundane-encoded text is possible, but it's much too powerful; there is a 10% chance that doing so will cause a backlash that knocks the character unconscious for 2d6 turns.

On OD&D Alignment and Language

This post on RPGnet has had me thinking a bit about original D&D's alignment and the languages associated with them. Alignment languages seem to be a sticking point to many people, which I understand - in AD&D they don't quite make sense since people are so fragmented between the nine alignments. But for OD&D I like it.

First, let's look at the origin of alignment in Chainmail:

It is impossible to draw a distanct line between "good" and "evil" fantastic figures. Three categories are listed below as a general guide for the wargamer designing orders of battle involving fantastic creatures:

This is then followed by lists for "Law", "Neutral" and "Chaos" with some figures appearing in two lists. The description for neutrality makes it clear how the alignment in Chainmail worked:

Underlined Neutral figures have a slight pre-disposition for LAW. Neutral figures can be diced for to determine on which side they will fight, with ties meaning they remain neutral.

The underlined Neutrals, for the curious, are Elves and Rocs. Wizards appear for Law and Chaos, and Giants and Lycanthropes can go under Neutral or Chaos.

The reason I bring it up is that it makes fairly clear what alignment was at the dawn of D&D: it meant there were basically two sides, and there were neutrals who could go either way or stick to themselves, like Switzerland. This idea is simple and makes perfect sense of why alignments exist.

As for Law and Chaos, these are taken from two sources. One is Michael Moorcock's Elric novels, which are better known than the source that they borrowed from. Poul Anderson's Three Hearts and Three Lions was the first modern fantasy to create an explicit Law vs Chaos conflict. Moorcock expanded this, both in his cosmic scope and the length of the works treating it. In Chainmail, it's actually much closer to Anderson: these are sides in a conflict that is happening here and now.

Dungeons & Dragons makes it a bit deeper. More types can switch between columns - Men can be in any column, and lycanthropes are spread across (mainly because of Werebears, which can be lawful). Very little else is added; there are swords which have alignments, certain effects on clerics based on alignment, some bits given about characters in castles, etc. But the biggest come in the language section.

Law, Chaos and Neutrality also have common languages spoken by each respectively.

This mainly becomes important because of the sentence before it:

All other creatures and monsters which can speak have their own language, although some (20%) also know the common one.

So for 80% of creatures, there are two ways to speak with them: you can know their language, or if they are of a similar alignment, you can speak in alignment tongue to them. It's a bit risky, though, as we see soon after.

One can attempt to communicate through the common tongue, language particular to a creature class, or one of the divisional languages (law, etc.). While not understanding the language, creatures who speak a divisional tongue will recognize a hostile one and attack.

This is why alignment language is a good idea in OD&D. Three pages earlier, we had a list of all kinds of creatures that have various alignments - so for instance a Neutral PC could try and talk with the Minotaur in the Neutral alignment language, even though nobody in the group speaks Minotaur. Of course this could go horribly wrong, since some Minotaurs are Chaotic, and will attack if you speak in Neutral to them! Orcs, Ogres, Dragons, Chimerae, Giants and some Lycanthropes also vary between Neutral and Chaotic. Lawful PCs run the same risk talking to Centaurs, Werebears and Rocs, and - in an interesting twist - Elves.

All this makes OD&D alignment much more interesting than AD&D's alignment. Rather than an all-pervasive cosmic force or a way to categorize the outer planes, it's an immediate question of where you stand among the creatures in the world. A Neutral can get Giants on their side - but only some of them, while it takes a Lawful to converse with a Hippogriff. Chaos is bad news, as all the undead are there along with Evil High Priests and Balrogs - the big threats from OD&:D that really fell by the wayside in later editions.

I'm curious for the comments as to who's done much with alignment tongues in your games.

Prairie Dogs and Animal Languages

Prairie Dogs: America's Meerkats - Language is a short video talking about the language that prairie dogs in the southwestern United States speak. Through a series of short, chirping vocalizations, prairie dogs convey useful information about an approaching predator that seems to convey an avoidance pattern. They can distinguish not only the predator but the size, shape and the dominant color; this has been tested to the point where the same human with different shirts will be described differently. All of this is in a short pitched call a tenth of a second long, repeated several times. Other vocalizations don't have an accompanying behavior so it's not clear what they mean. It's a neat bit of nature and, as the video speculates at the end, this is probably only the beginning of our ability to decode meaning in the calls of animals.

This kind of thing is useful to think about in a game that has spells like Speak with Animals. In the video we learn that prairie dog language has rough semantic equivalents to nouns and adjectives, but not to verbs. I'm thinking that when speaking with animals, a few simple verb-like things can be conveyed such as "run", "hide", "attack", "eat" and maybe for some types of animals "take" but the possibilities should be fairly limited. And of course it would vary for some types of animal, so some might understand "climb" or "push" but others maybe not.

Of course what makes it interesting is that you have to actually describe the thing you're talking about, to some extent. Now, a giant rat would probably understand "orc" since he's seen it; using Speak with Animals to describe an orc to a ferret might be more challenging. What's interesting in the prairie dog video is that they never confuse dogs with coyotes, and they do invent words for shapes, so the cleric using Speak with Animals may actually wind up coining some new terms in his target's language!

Extrapolating a bit from the existence of the spell, it may be that people in the D&D world have actually done investigation, perhaps with teams of researchers, one conversing with an animal using Speak with Animals and the other listening without aid of the spell, transcribing the conversations, and using them as a sort of Rosetta Stone to decode the animal languages. So for instance, a druid may be genuinely able to speak with the black bears in his forest, or magic-users could send messages by speaking with ravens (who we already know are highly intelligent).

In game terms, this could allow clever characters to learn the language of an animal companion such as a ferret, which would then serve as a sort of scout around the dungeon in the vein of Beastmaster. A Magic-User being able to speak with his familiar cat, or a paladin with his warhorse, etc - the list of uses for this sort of thing is not a short one. Of course the limitations are real, so for instance the ferret or cat shouldn't be able to count, but knowing whether it's an ogre or a group of kobolds in the next room is certainly knowledge worth having. And it makes the fantasy world just a bit richer.

Goblin doesn't have a word for "friend."

One of the oft-neglected concepts in gaming is of the various languages spoken by player characters and monsters. At most, players who roll well for Intelligence write down a few languages out of the list and periodically remember that they have such a thing. This, of course, becomes an issue when the monsters surrender or the players attempt to parlay instead of fighting. Typically, it's handled relatively simply: PCs with the language get to talk with the monsters.

Personally, I like to have a bit more fun with the concept. The one I've run into most is goblin, which is a great stock humanoid type. The way I figure it, goblins are mostly a savage, uncivilized race who break down into small tribes unless actively enslaved by some higher force, like orcs or hobgoblins. This brings us to the question of what goblin language is actually like.

The first thing about goblin is the counting system. There's really no reason for goblins to have a significant counting system; the exact numbers aren't their concern as much as having a rough estimate. So they have words for one and two, which are pretty much universal, and for "some" (which may vary from goblin to goblin) and "many" (which also varies but is bigger than "some"). This is conveniently frustrating for their interrogators, for whom the difference between 5 and 9 goblins may be more significant. I would expect "some" to be based around the goblin's family or fighting unit, depending on the exact context.

Then there are other fun things. Goblins aren't nice folk. The way I figure it, they probably don't have a lot of words for making nice – as the title of this post says, there's no word for "friend." The closest would mean something more like "goblin of my tribe," with a different word for "goblin of another tribe." A human would mostly be referred to by whether they were a threat, or whether the goblin group could defeat them, or whether they were slavers. No concept of allies and alliances exists, and even attempts at diplomacy would involve threats or admission of weakness. Lofty concepts of "fairness," "equality," "justice" would be boiled down to a handful of ideas - "human nonsense" and "weakness."

Goblin language's richness is one that humans would not prefer - the word for what smells good probably includes a rat on a stick. Threats abound, as do vocabulary for hunting, killing, tunnelling and so on. War is present, but as a permanent condition of goblin society. There is no word for "peace" or even "truce." Likewise, what need is there for a distinction between "earn," "find," and "steal"? If goblins are primarily raiders, and secondarily scavengers, there is fundamentally no difference between them.

This is just a sketch; I think I'll write up a more complete (and definitive, possibly with "translations" for effect) article on goblin speech for the miscellany. Has anybody else done any work on this? Or have any input on what a goblin (or orc, or what you like) language should be like?