Naming is a topic dangerously near to my heart—near enough that if it moved slightly to the right I would sustain cardiac damage—so at the risk of clutching my chest and keeling over, I'm selecting this article I wrote about naming to be regifted during our holiday hiatus. This article ran just before Mirrodin Besieged previews, but it takes on issues of nomenclature and creative sustainability that will be relevant to any time in Magic's future. I wanted to single it out so that it might help generations of card namers to come, or at least to aid my own Future Self as he tracks it down for reference. (Hi, Future Doug! Now this article comes up twice in Google! Or in whatever replaces Google in 2024, when Duels of the Planeswalkers becomes self-aware.)
I hope you're enjoying a holiday season filled with something near to your own heart.
One card, one name. That's the rule. A card's name is its fingerprint—it's the distinguishing feature that separates one card in Oracle from another. Cards don't have serial numbers (well, in our Multiverse and Oracle databases, and in Gatherer, card records do have unique record numbers—but those numbers are not used to distinguish cards in the rules). Even "Erase (Not the Urza's Legacy One)" technically has its own name, thanks to the handy parenthetical.
That means it's vitally important that all the card names be different from one another.
Here's the thing. Magic has no end date. There is no limit to the size of your Library of Leng—I mean, there's no upper limit on the number of potential cards in this game. There are, at last count, over eleven thousand unique card names. By the end of 2011, there will be over twelve thousand. In time, there will be tens of thousands—even hundreds of thousands of distinct names.
Hold on. Room is ... spinning a little. Must ... grip desk ... for a few seconds.
Okay. No, I'm good. Thanks though.
Hundreds of thousands of uniquely named elements in a single game. Is that even possible? Has anyone run this experiment, ever? Did anyone stop to, you know, make sure, when Magic began, that it was even close to feasible to come up with distinct names for all those moving parts?
So, yeah—could someone forward me that email? Because I'd love to see it. You know, that email. That email where Richard Garfield conclusively proved—back in 1993, over seventeen years ago—that we would be able to name cards set after set, year after year, and on and on ad nauseam (already a Ad Nauseam, by the way)?
There was such an email, right? Because you're making that face, and it's kind of scaring me. I'm doing that nervous chuckling thing.
Names and Games
Okay. Breathe, Beyer. Inhale, exhale. No, there was no such email, and there was no such assurance that cool card names would continue to pour out of Imaginationland to meet the demands of an ever-growing game. But somebody must have figured this out already. Tons of games out there have large inventory lists. Let's check the rest of the gaming industry.
Indeed, there are games out there that boast huge collections of unique collectible items, all with unique names. But there's a scale problem. Most of these other games are talking item lists that number in the hundreds. I've seen games that proudly advertise their "dozens" of different types of monsters or spells. That's a far cry from eleven thousand.
Games like this have not solved my problem, at least not in the scale that Magic requires. If your game is designed to allow the player to encounter around twenty to forty hours worth of game play, then you only need twenty to forty hours worth of creatures and items, and therefore a few hundred uniquely named entities is probably enough. (In fact, more than that and you're taxing the ability of your artists to render all the 3D models by the time the game ships.) So games like this, games that have all of their distinct items built into them at launch, are not going to help us. These games are like movies—one-shot pieces of entertainment. You can play through them again, but they're not going to bring you any new names of things, so they're not going to teach us much about large numbers of names.
So we have to look at games that have a different model. We need to look at games that have eternally refreshed content, more like a television series than a movie. We need to check with entertainment that comes in several "seasons" of content over multiple years, like Magic.
Some games have randomized content and/or crafting systems that generate very large numbers of distinct items that nevertheless have unique names for each. They often use naming conventions to generate those unique names—systems with linguistic "atoms" and combinatorial rules that allow those atoms to be constructed for any given item. These tend to come down to long, aggregate phrases (a la "Frosty Swift Longsword +20 of Gorgon Slaying") or conjunctions of syllables. That's attractive, from a design standpoint—with flexible enough naming rules, these naming systems could produce as many unique names as a game could ever need. Even games with massive numbers of distinct elements could still guarantee distinct names for each of them.
Here's how to create such a naming system:
- Create name-units that correspond to every possible property in the game. (I'm assuming items in the game have a number of "properties," which are player-visible characteristics of those items. For example, a property of some weapons could be the ability to deal ice damage, and you could name that property "Frosty.")
- Create rules for combining those name-units. (In essence, this is a grammar. For example, you could specify that all weapon modifiers, such as "Frosty," are listed before the weapon type, in alphabetical order. Maybe "of [Creature Type] Slaying" is a special case—those get listed after because it sounds cool that way.)
- Make sure that every valid combination of those properties results in a unique item. (If you create a good naming system, then two items with the same properties will end up with the same name. You want to make sure that's what's desired in all cases—you don't want two longswords that both have the same list of properties to somehow be different items, or else it defeats the purpose of the entire naming system.)
If you've done that, then every possible item (or creature, or whatever was being named) would have a unique name that described it.
Why is it that I am feeling this itchy problem not solved feeling?
The Namer's Craft
The trouble is, system-generated names don't measure up to hand-crafted names. Buying a "Frosty Swift Longsword +20 of Gorgon Slaying" from a merchant rips me out of the metaphor of being in a fantasy world, giving me the uncomfortable sensation that the world's underlying physics are embarrassingly exposed. It's like my great-aunt's bra strap is showing and we're all coughing politely into our fists and pretending like it's fine. "Yeah, it's cool—everybody here calls it a Frosty Swift Longsword +20 of Gorgon Slaying. All the time. Yep. Yyyyep. Ahem. Hey, don't give me that look, man!"
Maybe that's okay. Maybe awkward concatenation is the cost of name uniqueness. I like knowing the rules to things anyway. And Magic has certainly had its share of names that seem a bit ... let's say algorithmic. Lesser Gargadon and Greater Gargadon. Polymorph and Mass Polymorph. Ten-card cycles of Ravnica guildmages and signets.
But it's not a combinatorial system. A Magic card name is far from a recipe for specifying all the parts of the card. It's a handle created with Vorthosian love and care, specially built to merge the flavor of the spell, the tone of the setting, and the practical needs of specifying which card you mean into one easy-to-use package. It has survived a litany of potential Name Killers and other rules that could have prevented it from seeing print. It has avoided collisions with the thousands of other existing card names. A Magic card name takes work, but the overall effect and value to the game is worth the effort.
Can we sustain it?
The Tools in our Toolbox
If I live to be a hundred, then at the current pace of releasing new cards, Magic will have somewhere in the ballpark of sixty thousand unique cards. I figure it's safe to say that the moment of my death will represent the final time that I, personally, will ever have to worry about generating a Magic card name collision. Therefore, to quiet the demon-voices, I need to feel like we can generate another fifty-or-so thousand card names. Somehow.
There is hope. The facts are on our side. There are tools in our toolbox.
Magic has a tradition of modifiers. Although Richard Garfield may not have written some kind of naming sustainability document, he and the other initial creators of Magic did lay the groundwork for generative card naming. From the get-go many of Magic's creatures and spells have followed a pattern of including modifiers in their names, and those modifiers not only leave room for many new names but also establish flavor. The one Angel creature in Alpha, for example, could have been named simply "Angel," but it wasn't—it was named Serra Angel. There was only one Angel to name at the time, yet Serra Angel prophetically distinguished itself from the existence of any number of future Angels. It also established the term "Serra," which grew to become a huge aspect of the Magic storyline. Even humble creatures like Scathe Zombies and Scryb Sprites called out modifiers that left room for zillions of future names. Just by following this pattern of card naming, we leave open namespace for future cards.
Fantasy is vast. The fantasy genre gives us access to a deep well of potential terms and concepts to use in names. We can delve into time-lost folklores and scour ancient languages for imagery and expressions. With fantastic worlds serving as our excuse, we can whimsically re-contextualize and recombine real creatures into magical beasts, giving them wholly new names to suit. And Magic often goes beyond traditional Western European fantasy; we've tapped African, Japanese, Scandinavian, Romanian, Middle Eastern, and Native American cultures for inspiration. When you dive into a culture for setting inspiration, you also uncover priceless pieces of nomenclature.
English is vaster. According to oxforddictionaries.com, "The Second Edition of the 20-volume Oxford English Dictionary contains full entries for 171,476 words in current use, and 47,156 obsolete words ... . This suggests that there are, at the very least, a quarter of a million distinct English words, excluding inflections, and words from technical and regional vocabulary not covered by the OED." That is a big playground. Now, a lot of those are not going to be useful to us as card names or parts of card names (I'm unlikely to be able to use "refrigerator," and I am at a loss with what to do with something like "smoothly"), but that is a huge base of words from which to work. I am a particular fan of those 47,000-some-odd obsolete words, many of which can find new life gracing the name bar of a Magic card. Look at song titles and book titles, or the odd world of band names—even just using the regular English dictionary, people are still churning out unique names after thousands of years of naming.
Infinite planes = the vastest. Ultimately, the Multiverse itself is our best source of naming power. Every time we travel to a new world, we open up the potential of the unique locations, races, cultures, and history of that world. We go to Mirrodin, and we get access to Auriok This and Glissa's That. Stop by Dominaria and it earns us terms like Llanowar and Benalish and Keldon. Infinite planes means infinite proper nouns.
So take that, demon-voices.
Season to Taste
So, sure, "Lightning Bolt" is used up. So are eleven-thousand-whatever other card names—they're gone. And we're a game that's more like a long-running TV series; we're sort of in Season Seventeen. We might need to be careful with the supply of one-word-transitive-verb card names at this point, say.
But there is vast amounts of room to go. Really we've barely scratched the surface. The designers will make new cards, and when they do, the names will appear, too.
Letter of the Week
A couple of quickies this week, lofted in by the official Savor the Flavor Dirigible.
Dear Doug Beyer,
Regarding your article "Mail Bonding":
I wanted you to know that I agree with your stance on printing iconic spells (e.g. Day of Judgment) with no flavor text (just three simple words and lots of white space). I just think that when you do that, you should use a 72 point font and really fill the text box. ;-)
Noted! Just for you, Scott:
Dzulai asks a perceptive question about iconic creatures.
Dear Doug Beyer,
Since the beginning of Magic, Angels, Dragons and Demons have been the iconic creatures of the colors White, Red and Black respectively. Vampires has also been Black's other iconic creature before it was demoted to a tribal creature type.
Sphinx has recently been for Blue for quite some time now. Although, before, I thought Shapeshifter could have been a viable iconic creature for blue, pre-Lorwyn.
But with Green, from what I observed, there's no creature type that is being pushed as the color's iconic creature. Amongst the green-aligned creatures, I think that the most viable iconic creature candidate is the Wurm creature type. But the problem is, it is freely used at all rarity.
Therefore, I suggest that Wurm should be Green's iconic creature and should be restricted only to Rares and Mythic Rares. Just my two cents. :D
We have talked internally about this exact problem with green, and we've indeed discussed the promotion of Wurm (and Wurm's subsequent restriction from lower rarities) as one of the possible solutions. For a while we were playing around with Hydra as green's "rare iconic," but Hydras tend to have mechanical baggage that makes them less flexible than we'd like. (Many Hydras have X in their mana cost and/or use +1/+1 counters to represent their changeable size. That's flavorful, but it makes them not usable in every set. In contrast, Angels/Sphinxes/Demons/Dragons all fly, but otherwise they have a wide range of abilities that lets them adapt to the setting and mechanics at hand.)
Wurm is a strong candidate, and I personally love me a Wurm. (Snicker.) However, there's such a strong pattern of common and uncommon Wurms in Magic that it might be a "hard sell" to represent them as a special thing we only do at high rarities. Maybe that's the best plan, though, I don't know. There may be other creature types that could work instead, but nothing has stuck out as the perfect solution. I like baloth, although baloths are currently Beasts, which throws them in with a wide range of odd critters and blunts their unique coolness. There could be an entirely new type, like Behemoth or Colossus, but those terms already have terminological baggage from other cards (which aren't even all green), and they just don't have the weight of history behind them as Wurm or other types. It's an open issue that we continue to ponder. So, dear reader, what do you think?
Next week: Mirrodin Besieged previews begin, and I troll the internet.