“I think she must have dosed you with drugs,” the Keeper of the Moonlight Prism said in chat that night.
“No way,” I said. “I’ve been a drug-study guinea pig before. You have to sign real medical consent forms. They have to tell you what they’re testing on you. Doctors and nurses have to be there.”
“When were you a drug-study guinea pig?” the Keeper asked.
“When I was poor in college, I volunteered at a research hospital on campus that was testing a new antiemetic med against common opioids—morphine, methadone, and fentanyl—administered in a closely monitored environment. In half the sessions, I took the drugs orally; half the sessions I was given the drugs IV. The study paid out twelve hundred dollars for six day-long sessions spread out over twelve weeks. I always joke that’s when I turned pro as a drug user.”
“I haven’t taken any of those,” the Keeper said.
“The fentanyl was amazing; the methadone was meh; the morphine made me puke my guts out, despite the presence of said new antiemetic; and then I got paid. So back to your original point, if they were giving me drugs, I believe I would know about it, and anyway, why would an advertising agency be administering drugs in the first place?”
“It’s just that you’re describing some pretty weird shit,” the Keeper replied. “Can you record your session tomorrow with your phone so I can hear it?”
“Not according to my NDA,” I said.
“Which you are currently violating as we speak,” the Keeper pointed out.
“True,” I admitted. “Maybe. We’ll see how smooth I can be.”
“What, the Queen of Sparkle Dungeon admitting she might be less than smooth?”
“I’m a fighter, not a spy. Well, I mean, I’m a fighter and a spellcaster. Well, I mean, I’m a fighter and a spellcaster and a house DJ—”
“Yes, you’re multi-classed, I get the point. Anyway, why would you go back? Isn’t it weird and maybe dangerous that you can’t remember what’s happened to you?”
“Maybe,” I said. “But isn’t it kind of awesome that she can produce that experience in another person simply by talking at them? Like, how weird would that be if the spells in the game actually mapped to real-world effects?”
“C’mon, you know that’s not possible,” the Keeper said.
Obviously the Keeper was right. Still, something about the whole thing felt fascinating, not dangerous. Of course, if they were planting hypnotic suggestions in me or something bizarre like that, fascinated might be exactly what I’d feel, which yes, could be dangerous. But the fact was, I did feel fascinated, undeniably so, which meant I was going back.
Anyone who truly cared could go to the SparkleWiki, look up the entries about the Sparkle Dungeon diva-casting protocol, and learn an array of semi-useful trivia that might make you a better player. For example, actual linguists who played the game made it a pastime to identify and catalog all the unique sounds that were used for spellcasting, teasing sequences apart to reveal the component phonemes (discrete units of sound within a word) in circulation. No one had ever discovered a true language behind the spells, but there were apparently detectable patterns and rules for how phonemes were distributed that hinted at an underlying structure of meaning, and kept people searching for it.
And some folks enjoyed putting forward theories about how speech recognition and machine learning were being utilized by the game engine to make spells “work” at scale in the first place—the game had to normalize across millions of players to recognize an accurate delivery of a spell, with tolerance for regional accents on vowel sounds, let’s say, or variances in sibilance that should all be considered accurate delivery.
And some of the more complicated sequences relied on pitch variance to determine accuracy—kind of like how the vocal track in a game of Rock Band works, where the game is judging how well you hit the intervals between notes—except in a spoken-word context where the starting pitch could be anything in any vocal register.
Then, on top of all that, the game engine seemed to award you some kind of unpredictable bonus based on style. I was good at the style aspect for sure. I always told people you should be so confident and natural that a spell sounds like a personal catchphrase.
But there was apparently a lot more to it than that. Timing and emphasis mattered a lot—not simply how loud you were, but how fine-tuned your velocity was on a syllable-by-syllable basis within a sequence. Which of course is how this thing called a “sentence” works in language, but with diva-casting, we didn’t have any literal meaning to help us remember how to pronounce this stuff.
Prior to my big first day at Jenning & Reece, I didn’t really give that theoretical or academic stuff much thought. I didn’t learn spells by studying a lexicon on SparkleWiki or even going through the tutorials in the game; it was just raw repetition in context that did the trick for me, and luckily I was a natural. That night, though, I scoured those SparkleWiki pages again, looking for clues that might situate me as I prepared for tomorrow morning’s session with Olivia Regan. Sadly I wasn’t sufficiently savvy about linguistics to pick up anything specific in one night that I thought could help me.
But qualitatively, I’d definitely noticed that Olivia’s enunciation of the new sequences had been quite remarkable. Clean and smooth even as it tended toward raw and aggressive; silky and smoky and at the same time jagged and forceful. When you speak, your intonation reinforces or subverts the words that you say; intonation is a distinct layer in the meaning that you deliver. Olivia had that piece nailed down tight. I used to think that the earliest implementation of voice spellcasting sounded like a parody of Klingon or something, just brute force barking to get the game engine to recognize what you were saying, but these new sequences were starting to sound positively persuasive.
Diva-casting had come a long way in the game since the original Sparkle Dungeon, and if Olivia was showing off the new evolution in technique for Sparkle Dungeon 5, I was starting to get hyped.