The Genius Plague
Page 9
I was interrupting his work yet again, but he sighed and waved me to roll my chair over to his. He reminded me a bit of my dad, though they looked nothing alike. My dad was tall and thin, with thick hair, while Andrew was short and heavy and mostly bald. Maybe it was just the intense way that he talked about his favorite subjects.
“So. Why do we monitor the world?” Andrew asked.
The question was so fundamental it took me aback. I tried to think of what answer he might be looking for. Because intelligence won wars. Because it was what the NSA did. “Um . . . I guess I thought we needed to be ready for any threats to our interests. So that if we had to fight, we would know what we were up against.” As I said it, I realized how that would apply to the drug lords. If it became necessary to fight them, we would want to do so with as little loss of life as possible, and that required precise intelligence of the enemy’s intentions and capabilities.
“That’s part of it,” Andrew said. “But not the whole part. You think you know where the serious threats are. Iran. Russia. China. But it’s not always as easy as just taking out a threat. In fact, it never is.”
Andrew stretched and leaned back in his chair. “Any time we interfere in a foreign situation, we affect change, and change is complicated. It has consequences that are hard to predict. If we weaken one nation in a region, we give a competing nation more power. If we take out one dictator, we enable other factions to take control. Every change has tendrils that run deep. In order to create the results we want, we need information. Thorough information, accurate information, about how power flows, what its sources are, and where it will go if we alter the current.”
“So . . . Colombian revolutionaries?”
“Part of the picture. It’s not just about collecting intelligence in case we have to fight them directly. It’s about keeping track of how they fight one another, and making sure none of those thugs ever gets enough power to take over a country or a real military and make itself a threat to its neighbors. That would upset the balance of power, which would make it more likely that we’d have to step in to maintain it. To maintain the balance, however, we need to know who the players are, how much power they have, and where they get it.”
“Okay,” I said. “So what do you want me to work on first?”
“Doesn’t much matter. The best thing is probably to pick a message, crunch on it for a day or two using any method you can think of, and then come back and show me what you’ve tried. We’re all about originality in this group, and these are messages everyone else has given up on. So you can’t do any harm.”
I chose at random a group of messages from the stack. There were twenty-four in the group, probable text emails traded over the course of three days between members of the FARC in La Uribe and members of the ELN in Quibdó. The messages were thought to be encrypted with the same method due to various mathematical similarities, but there was no way to be sure.
The shortest message was less than a kilobyte; the largest was over twenty. Reasonable sizes for text, but too small for much in the way of images or data. Using tools that Andrew had shown me on my new computer, I set about compiling some basic information about them. Frequency tables, showing how often each letter, number, or other symbol appeared. Bit patterns that repeated with statistically improbable frequency. One tool allowed me to enter guesses at words that might appear in the unciphered message, which it then used to look for matching patterns, much as I had done with my entrance exam. It came with a huge list of significant words—in English, Spanish, Portuguese, and several local languages—that had been previously used in communications by these two groups.
All of this work had been done before, with the results bundled with the files, but I did them again, both to familiarize myself with the tools, and to start to get a sense for the messages themselves. It was tempting to concentrate on the shortest message only, which was small enough that I could print it out on a single sheet of paper and stare at it. I knew, however, that the more text I had, the more likely I would be to find patterns that would help me crack it.
I stayed until almost midnight that night, long after everyone else had gone home. When I finally left, my eyes blurring and my stomach growling for the dinner I had missed, I was not one step closer to finding a solution. It wasn’t that I had found no patterns. There were too many patterns, most of them either coincidental or not consistent enough to be helpful.
That night in bed, my restless brain concocted endless arrangements of symbols, their meanings slipping aside every time I tried to look at them. I woke twice, convinced that I had solved the puzzle, only to realize that my brain was playing tricks on me. It was like playing hours of Tetris and then seeing the pieces still spinning endlessly behind my sleeping eyes.
The next day, I thought about pulling a different set of messages, but there seemed little point. A new set was unlikely to be any easier to crack—these were the indecipherables, after all. The trick was to think of ways to approach the problem that I hadn’t tried before. Half of my team were software specialists of one type or another, mobilizing banks of servers from the room next door to crunch through tremendous quantities of data. That wasn’t my skill set, and never would be, and besides, those avenues were already being explored. I had to get original.
Originality meant thinking out of the box, considering approaches no one had thought of before. I decided to list my most basic assumptions.
• The message is text in a written language.
• The text was written in a standard computing symbol set.
• The message makes sense.
There wasn’t much I could do about the last assumption. If the message was random atmospheric noise or a bad transistor, then I was wasting my time. The first one, also, seemed pretty basic. The second assumption was interesting, though—if the message encryption was based on some unusual character encoding, then considering the bytes as message elements might be a mistake.
I spent a few hours with this idea. Any message that came through a computer system was, of course, a series of zeros and ones, but that data was generally turned back into letters and numbers through a standard 8-bit encoding set such as UTF-8. There was no reason a creative person might not use a 5-bit system, for instance, or an 11-bit system. It would be unusual, but it could be done.
By the end of the morning, however, I had discarded this notion. There were portions of the message that, when ASCII-encoded, had distinct patterns, such as:
5 5 5 6 6 7 7 7 8 8 8 9 9 0 0 : : ; ; = > ? @ B C D C B A ? > = < ;
The sequences ran straight through the ASCII “alphabet” in order, sometimes repeating, sometimes skipping a symbol, but always rising and then falling again. This pretty much clinched that it was an 8-bit encoding scheme, because that pattern couldn’t be coincidental. It raised a serious question, however, about the likelihood that this was a message at all. What possible communication could contain steadily rising and falling letters through the alphabet? It just seemed like noise.
One of my goals for the day was to actually eat a meal at the right time, so I took a break and wandered down to the cafeteria. The NSA cafeterias were enormous, serving thousands of people, all of them strangers to me. I went through the line and bought a chicken salad wrap and a lemonade. Looking for a place to sit, I was surprised to see a face I recognized. An Army captain with a USCYBERCOM sleeve patch. It was Scaggs, who had debriefed me after I was arrested for hacking, and he was eating alone.
I put my tray down across from him and sat. “Captain,” I said. “Mind if I join you?”
He looked startled to see me, but I couldn’t tell if that was surprise at having someone join his table, or surprise at seeing me with an NSA badge. He recovered quickly, though, and offered me his hand. “Mike Scaggs,” he said. “Glad to see that was all just a misunderstanding.”
“Me, too,” I said. “You had me worried.”
“Come on, now. I’m not exactly intimidating as an interrogator.”
/>
I admitted he wasn’t, and we laughed. I took a bite of my wrap, which was surprisingly good for cafeteria food. I had been expecting something closer to what they served at my high school.
“So what’s it like working for the Major?” Scaggs asked.
“For who?” I thought he must have me confused with someone else.
“Major Muniz,” he said. “The one who rescued you from counterespionage. I thought you must be working for her.”
“Oh. I am. She’s a major? She’s always dressed casually; I thought she was a civilian.”
“Ah, no.” Scaggs looked embarrassed. “She’s a civilian. ‘The Major’ is just what everyone calls her. It’s kind of a nickname. She’s pretty wellknown around here, has a lot of clout, even with the muckety-mucks up on Mahogany Row.”
“I see.” I had never heard anyone on her team call her Major. “Why do they call her that?”
Scaggs chuckled. “Uh . . . I’m pretty sure it’s short for ‘Major Pain in the Ass.’ She has a reputation for being tough and stubborn and getting her way most of the time. I guess it’s not exactly a compliment, but I think people have called her that for a long time. I’ve even heard her refer to herself that way.”
I nodded, taking all this in. “She’s always been pretty nice to me,” I said. “Though she’s often not around, either off meeting with other people or out of the country.”
“I didn’t mean to insult her,” Scaggs said. “I know some people hate her, but I don’t have anything against her.”
“No worries.”
“I mean, she’s been here since forever. She would have been an agent here back during the Cold War. Can you imagine? I’m sure she’s hit some serious chauvinism over the years and had to develop a forceful personality to overcome it. They say she once briefed President Nixon.”
“Not possible,” I said. “Nixon resigned in, what, 1974? That would make her seventy years old, minimum.”
Scaggs shrugged. “I know she’s turned down retirement. She must be getting up there.”
“Besides, people don’t brief the president when they’re new hires. I barely know my way to the cafeteria.”
“Yeah, she was probably a lot smarter than you, though.”
I grinned. “I don’t know. I bet she didn’t hack her way into someone else’s account on her first day.”
The lemonade was too sour for my tastes. I tore open four sugar packets from the holder on the table and poured them in, ignoring Scaggs’s raised eyebrow while I stirred. “How long have you been here?” I asked him.
“Going on ten years, now. Which is unusual—normally us military types get yanked around to a new assignment every two years or so. Just so happens my assignments have all been in this building.”
“You like it here?”
“Yeah, pretty well. I remember my first month, though. Turned around every which way, not understanding half of what anybody says to you.”
“I’ve mostly just been staring at messages from South America that are impossible to decipher and failing to decipher them,” I said. “Doesn’t take much skill, really.”
“Those messages from the Ligados, then?” Scaggs said.
I frowned. I recognized the word as Portuguese for “connected” or “plugged in” or as a slang expression for “in the know.” I wasn’t aware of any group of people called the Ligados, though. “Who’s that?”
“Don’t follow the news? It’s the terrorist group that took credit for that tourist boat massacre back in February. They apparently hit a nature preserve last night. Left no one alive.”
I hadn’t heard about that at all. “I’ve been working a lot. I haven’t been watching much of the news.”
“I just saw it on CNN last night.”
“My brother was on that tourist boat,” I said. “He was one of the survivors.”
Scaggs gave a long whistle. “Wow,” he said. “I’m glad he’s okay.”
But I wasn’t listening to him. The sound of Scaggs’s whistle was still playing through my mind. A long, even sound, rising and then falling. I jumped out of my seat, banging my leg against the table and almost spilling my lemonade. “That’s it!” I said.
Scaggs was gaping at me. “Sorry,” I said. “Gotta go.”
I took off toward the hallway at a jog, leaving my tray and an astonished Scaggs behind me. I reached my desk and crashed into my seat. I mistyped my password twice in my haste, finally taking a deep breath and getting it right the third time.
I brought up the indecipherable. It had to be. I selected several of the phrases with rising and falling letters and graphed them as numbers. They were 8-bit words, but they weren’t ASCII-encoded at all. They were musical notes.
I was certain I was right, but even so, it took a lot of time to prove it. An hour later, I was in the NSA library, hunting through their vast linguistics section. It was my first time there, and it was overwhelming, an incredible collection of books on anything an NSA analyst might possibly need to research, including perhaps every book, doctoral thesis, and academic paper written on the study of languages in the world. After a long hunt, and with the help of one of the library staff, I made it back to my desk with a book called Aspectos da Fonologia do Johurá, written in Portuguese by an American linguist with an organization called the Summer Institute of Linguistics.
An hour later, I had it. I leaned back in my chair and shouted “Eureka!” as loud as I could. I’d like to say I was just caught up in the moment, but really, I’ve just always wanted to say that. Watching Shaunessy Brennan jump two inches out of her seat made it all worthwhile.
“What?” she practically snarled.
“I cracked it,” I said.
Andrew and Shaunessy and a few of the others gathered around my computer, and I showed them what I’d found. “The message isn’t encrypted at all,” I said. “Not really. It’s an encoding of the sound of a whistle.”
“A whistle,” Shaunessy said.
“There’s a tiny people group living along a tributary of the Amazon called the Johurá,” I said. “Maybe three, four hundred people tops, who speak this language that’s so weird and difficult to learn that it was decades after they were first discovered before anyone cracked it. It only has three vowels and eight consonants—the fewest phonemes of any language in the world. It’s tonal, like Chinese, only more so. The meaning in the language is communicated through tones more than the phonemes themselves. The word for ‘one’ and the word for ‘two’ are the same word, only with a different tone. It’s all like that.”
“And there’s a written form of this language?” Andrew said. “You’re saying somebody’s using it to pass messages like the Navajo Code Talkers?”
“Not quite,” I said. “One implication of such a tonal language is that it can be whistled. You can actually communicate effectively without the phonemes at all, just by whistling the tones. A native speaker can infer what the phonemes should be by context. That’s what this message is. Johurá whistle talk.”
“So?” Shaunessy said. “What does it say?”
“I have no idea,” I said. “I don’t have enough information. I picked out a few words, enough to be sure I was on the right track, but there’s no dictionary.”
“Does anybody at the NSA speak the language?”
I chuckled. “Nobody speaks this language. A few hundred Johurá along the Maici River, a missionary family that lives with them, and two linguists. Two. In the world.”
Andrew shook his head and grinned. “Then I guess we’re going to need to find one of them.”
CHAPTER 10
Later that evening, when most of the team had gone home for the night, Melody found me and dropped a carton of Kung Pao chicken on my desk. “Brains need food,” she said.
The hot chicken smelled delicious, and my stomach growled, reminding me that I hadn’t eaten since my abandoned lunch with Captain Scaggs. She handed me a plastic fork, and I dug in, while she sat in Shaunessy’s chair and ate from a
carton of her own.
“Andrew tells me you cracked one of the Ligados messages,” she said.
There was that word again. “Is that who it was from?” I asked. “The Ligados?”
She nodded. “We’ve been inundated with indecipherables from them. It’s starting to be a big deal. They’re growing in influence, taking hostages in Colombia and Brazil and demanding responses from their governments. They seem to have a lot of connections to existing groups—FARC, ELN—but we can’t pin down who they are. We’re intercepting all their messages, but we can’t read them. The DIRNSA wants to know what they’re saying.” The DIRNSA was the Director of the NSA, which she pronounced like a single word, dernza. “So do we know what the message says yet?”
“No. I did manage to turn it into sound, though.”
I hit play on my computer, and a series of whistles played out over the speakers. “I had to guess at the actual pitches,” I said. “I figured a 2000 hertz midpoint and a 100 hertz step between each unique value. It might not be accurate, but it seemed reasonable . . . ish.”
“Nice,” she said. “If we can find someone who speaks the language, that might be enough for them.”
“The problem is, almost nobody speaks this language. It’s not taught in any school; there’s no alphabet or grammar for it that I’ve been able to track down. Though my resources are pretty much the library and Google, so if you’ve got any better ideas, let me know.”
“One step ahead of you,” Melody said. She lobbed a thick manila envelope at me, which I caught. The label on the front read Summer Institute of Linguistics.
“What’s this?” I said.
“Research on the Johurá from SIL International. They’re stationed in Dallas, so I had an agent from our San Antonio office drive there and fax me everything they had.”
I was astonished. San Antonio to Dallas was a five-hour drive. Some poor junior agent had spent all day on the road just to acquire this document and secure fax it to us. Deciphering these messages was apparently more important than I realized. I was also starting to see how “the Major” might have earned her nickname.