By Melissa Silmore (TPR'85)

“Kathleen is an amazing American. She has done so much for this country and the current war on terrorism. My security clearance prevents me from telling you even a fraction of what her research has meant.”
Major Ian McCulloh, U.S. Army

The small, round beam of a flashlight pinpoints a circle of words on a page in a darkened room. A 12-year-old, huddled in bed, gratefully surrenders herself to the futuristic world of sci-fi author Isaac Asimov. It had been another typical, long day—starting with frustration at Catholic school. Not that learning was frustrating, of course. Learning was exhilarating. And math was particularly fascinating. No, it was the way they continually discouraged her. “Kathleen, a girl shouldn’t waste her time on mathematics!”

Pueblo, Colo., of the 1960s is just a small town like Mayberry R.F.D. And it’s a poor town, especially her side, where many—her family included—live below the poverty line. Her father works for a local company, triple time, and in his spare time builds swimming pools with his tractor. Her mother helps with the pool business. Much of the family responsibility falls on her, the eldest of their four children.

After school that day, she had headed back to her grandmother’s small house, two blocks from her own, to do the family laundry. They didn’t have a washing machine at her own house. She loaded the clothing into the old contraption, using “that stick” to agitate the clothes up and down. To speed the mind-numbing work, she counted the stabs. Up, down. One, two. Up, down. Three, four. She had always been a counter, lining things up to tick off, enthralled with the numbers. She then fed the clothes through the ringer, cranking to squeeze the water out. One. Two. Three.

Her thoughts wandered back to their farm. They’d had to move here, to town, two years ago, after they were flooded out by the Arkansas River. They’d lost everything, even her baby kittens. She missed her animals. She missed her friends. On to the ironing. The small, windowless room was stifling. She’d constructed a small stand next to the ironing board where she could balance her beloved science fiction books. She was an avid, avid reader and it passed the time, but she had to be careful. She had to hide the book each time her grandmother came in, for no one believed she could both read and iron.

After hanging the clothes on the line, she walked to her own similarly small house in time to make dinner. Corned-beef hash that night. And then it was time for TV. She was expected to sit with the family after dinner. It was insisted upon. They’d paid good money for that little black and white, and it was the only time the family was all together. She was eager to do her math homework, but rules were rules.

In fact, her parents preferred that she focus more on her tasks stuffing envelopes and answering phones for the pool business than on schoolwork. Her mother joked that her eldest’s first words were, “I’m going back East to college.” Funny, considering no one in the family had finished college. Many hadn’t graduated from high school. Pencil and math text in hand, she tried her best to tune out the drone of the show.

Bedtime. She’d come upstairs, not to sleep, but to finish her math. And now, tucked in bed, flashlight in hand, she devours another Asimov book. It’s inspiring, this Foundation series, all about a system of social mathematics that can predict the future. She smiles. This book represents her destiny. She knows she can—she will—make this reality.

More than four decades later, the phone rings in the Carley household. It’s Christmas break, and the Carleys are gathered together—Kathleen, husband Rick, and their two daughters, Cassandra and Arianna. Kathleen had followed her dream to MIT, through loans, scholarships, and numerous student jobs. She’d been so hell-bent to get there, she’d purposely flunked the Colorado School of Mines’ scholarship test. She’d met Rick during her freshman year at MIT. Within two years, they were engaged. They married right after her graduation.

Unbelievably, she’d faced familiar hurdles along the way. There was the MIT professor who warned she’d never make it as a female mathematician and should find a different field, even as she was taking every artificial intelligence course offered. And there was the cousin at her engagement party, relieved that Kathleen could now forgo that bachelor’s degree for her “Mrs.” with Rick.

Instead, she’d gone on to Harvard for a PhD, following the goal she’d set years before—to combine social science and artificial intelligence. She studied under Harrison White, a pioneer in the use of mathematics in social network analysis, the mapping and measuring of relationships between people and groups. Rick, meanwhile, had stayed at MIT for his own doctorate in electrical engineering.

While visiting a relative at Carnegie Mellon soon after earning their degrees, the Carleys decided to pop in to their respective academic departments: Kathleen to social science, Rick to electrical and computer engineering. They walked out with invites to return and talk about possible faculty appointments. Kathleen couldn’t pass up the opportunity to work with Allen Newell, a world-renowned researcher in computer science and cognitive psychology. So, both she and her husband followed through on the invites and, sure enough, the young couple received Carnegie Mellon appointments and settled in Pittsburgh.

At Carnegie Mellon, Kathleen Carley founded CASOS, the Center for Computational Analysis of Social and Organizational Systems. She developed the multi-disciplinary center to bring together social scientists, organizational theorists, statisticians, and computer scientists—combining their skills to further understand and predict how and why groups behave the way they do.

CASOS developed sophisticated tools, including ORA, which employs mathematical algorithms that represent general social findings, like the tendency of friends to be similar. With vast quantities of data gathered from sources as varied as texts, news reports, interviews, blogs, and email, the system uses algorithms to analyze social networks. ORA has the unprecedented capability of keeping results up to date, tracking networks as they move through time. And, as Carley puts it, “it lets you look at not just who talks to whom, but the who, what, where, when, how, and why. It puts it all together.”

For example, the team can gather timely data from a group’s email and text interaction, as well as articles and blogs that reference them. From this data and more, ORA’s algorithms are then able to identify the most prominent leader. This information, combined with other tools, can then help pinpoint the target’s location.

Carley’s network science techniques and software can be used to study diverse social issues and problems, from how beliefs spread through cities to evolving revolutions to drug and terrorist networks. People call from every discipline and field, from academia to U.S. Homeland Security. With Rick, she is exploring methods to assess a region’s capability for building weapons of mass destruction. And then, of course, there is the military.

The phone rings insistently through the Carley house.

“Kathleen? It’s Ian McCulloh. I’m in Bagram, Afghanistan, trying to map out terrorist networks. I’m having a problem. Can you help?”

McCulloh is a professor at West Point who had started teaching a social network analysis course in 2005, the first such course ever offered anywhere, according to the International Network for Social Network Analysis. He had 19 cadets working on social network research projects when he met Carley at an intelligence meeting and subsequently brought some of his work to her attention. They began a collaboration. One of their early network-analysis projects involved open source videos of sniper attack areas. In one kill zone, for example, the analysis detected a blue truck driving through multiple correlated videos, one time with a license plate visible. Identifying that license plate led to the eventual takeout of an insurgent cell. According to McCulloh, when the military in Iraq adopted the video-analysis technique, it resulted in an 80% decline in sniper attacks that year.

McCulloh was so pleased with the results and intrigued with Carley and her work that he applied to Carnegie Mellon to study full-time with her for his doctorate in Computation, Organizations, and Society. After earning his PhD in 2009, he returned to West Point, adding his newfound knowledge to his course curriculum. Evidently, the news spread quickly. He received an email from some prior students, then serving as intelligence officers in Afghanistan. They’d heard there were analytical tools—Carley’s tools—in use at West Point that were better than those actually in use in Afghanistan. Could McCulloh do something?

McCulloh flew over to work with the task force. Running some of their data through ORA, he identified insurgent targets in 15 minutes. These were targets the officers believed would have taken an analyst with decades of experience more than a week to identify. Even more important, ORA’s mathematical analysis resulted in verifiable outcomes. This was in contrast to current methods, which too often were based on hunches and weak connections, leading to improper arrests and botched results.

After finishing his initial analysis, McCulloh stayed in Afghanistan, training the other officers in understanding network analysis and how the underlying ORA system worked. He continued his analytical work while he taught. One day, while performing a key analysis, McCulloh encountered a difficult problem. He couldn’t convert the data files into a usable format for ORA. At wits end, he reluctantly made that phone call to Carley, at home with her family, celebrating Christmas.

McCulloh can’t believe how quickly she drops everything to spring into action, immediately devoting herself and the resources of her lab to the problem. Even though it’s Christmas break, she manages to enlist her fellow CASOS researchers, as well as necessary military personnel. Phone calls fly back and forth from Pennsylvania to Afghanistan for four days. McCulloh barely sleeps. Carley and her group are faced with the challenges of working without personal access to the classified data. Finally, the group develops a technique to translate the data. McCulloh feeds it into ORA and is able to identify a group of insurgents moving from Afghan village to village. They’re training people how to build IEDs, or improvised explosive devices—more commonly known as roadside bombs.

Months later, McCulloh hears that U.S. troops were able to find and neutralize the group. And without the insurgents’ bomb training, he learns, the level of sophistication in IED technology has fallen sharply, meaning fewer soldiers and civilians killed. He’s delighted, but only wishes that Carley and her colleagues could be compensated for their critical efforts. He knows it isn’t the first time or the last that they won’t be recognized. With or without explicit recognition, McCulloh sees the ORA software, and software developed from it, permeate the military and other U.S. agencies as more and more officers are trained and disseminate their knowledge.

Almost a year after McCulloh’s phone call to Carley, he is stationed in Iraq, spending a lonely Thanksgiving holiday. He and a colleague walk into an establishment frequented by U.S. personnel. The two spot a group of CIA analysts and sit down to chat, voices lowered. It’s an interesting conversation, and McCulloh can’t elaborate on it. Months later, he receives a call from a reporter for The Economist. The reporter’s anonymous source claims that Carley’s software may have been involved in finding Bin Laden. Can McCulloh comment? He replies he cannot.

Today, McCulloh is stationed back in the States. He had been training Iraqi intelligence officers to use the ORA software until he was identified by the insurgents and targeted three times for assassination. That’s when the U.S. military spirited him out of Iraq on a flight home. The assassination attempts were a sinister validation of the effectiveness of Carley’s technology. “She is probably doing more for the U.S. military in academic research than anyone else in the country,” says McCulloh.

Melissa Silmore (TPR’85) is a Pittsburgh-based freelance writer and a regular contributor to this magazine.

Related Links:
Social science: Web of War