Government agencies earn failing grades in foreign languages

It’s the same old story — very few people in key government agencies are proficient in important world languages.  When will this change?!  And, can technology really help fill in the gaps?

Amber Corrin, “DOD, DHS earn failing grades in foreign language,” Federal Computer Week, September 10, 2010:

Do you speak Arabic, Dari, Farsi, Pashto or Urdu? No? Neither do an alarmingly high percentage of troops deployed to the Middle East and southwest Asia, home to languages that local residents but few U.S. warfighters speak.

Far too few, according to the Government Accountability Office, which recently released a report that underscores the woeful foreign-language capabilities of the Defense and Homeland Security departments.

It’s more than just a matter of inconvenience. During a recent congressional hearing, Sen. Daniel Akaka (D-Hawaii) said the lack of foreign-language skills could be a threat to national security.

Industry has stepped up to try to fill the skills gap by developing technologies that they tout as solutions to the language barrier. But when it comes to rebuilding war-torn towns and villages, training local police forces, and administering humanitarian aid, will a handheld device or database solve the problem?

“It is a Band-Aid,” said Larry Korb, a senior fellow at the Center for American Progress think tank. “But 10 years ago, did we ever think we’d be invading Afghanistan? Did we think we’d be in Pakistan?”

In other words: Training takes time — more time than, say, a tool that scans and translates documents for intelligence purposes.

Unfortunately, there is no quick fix for a problem that is more deeply rooted than it appears. The issue is more than just a dearth of speakers of these languages; there are systematic flaws behind the language gap.

According to the GAO report, DOD’s strategy for alleviating the language problem has holes. Funding plans are splintered, and needs, requirements and existing capabilities are poorly understood “due to the lack of an agreed-upon way to assess and validate these skills,” the report states.

GAO auditors said DOD’s efforts to meet the language requirements “had yielded some results but had not closed the persistent gaps in foreign language-proficient staff and reflected, in part, a lack of a comprehensive, strategic approach.”

The report also hammers home the implications of failing to address the language shortfalls.

“The lack of foreign language capability at some agencies, including DOD and [the State Department], [has] resulted in backlogs in translation of intelligence documents and other information and adversely affected agency operations and hindered U.S. military, law enforcement, intelligence, counterterrorism and diplomatic efforts,” the report states.

That information backlog is the target of industry systems such as NovoDynamics’ optical character-recognition tool, used in the military’s Document and Media Exploitation program for identifying, collecting and analyzing enemy documents and media.

“It’s easy to collect the dots. Connecting the dots is the challenge,” said Mike Yeagley, director of global government sales at NovoDynamics. “The solution is going to be a hybrid between expert linguists and technology that triages and translates enormous amounts of data. It’s not realistic to expect humans to do all the translating.”

Korb agreed that technology can provide tools for bridging the language gap. “You have to take advantage of the technology in the short term,” he said. “In the long term, you need the Defense Language Institute, education and training. But in the short term, technologies like optical character recognition help keep the problem from getting worse.”

Of course, translation technologies are far from perfect.

In a 2008 white paper titled “Can Machine Translation Really Help the Department of Defense?” Nicholas Bemish, senior human language technology adviser at the Defense Intelligence Agency, considers some of the challenges posed by technology, including DOD’s notorious cultural issues with implementing new tools.

“Researchers will say that finding the correct algorithm and developing the tool is significant, but I have to submit that talking someone into using something new and ‘foreign’ can be even more difficult,” Bemish wrote.

His argument is similar to GAO’s when it comes to assessing the effectiveness of machine translation. “We don’t typically measure the results against bottom-line costs and how much we may save in overall manpower numbers,” he wrote. “A success within DOD can be as simple as a soldier using [a machine translation] tool to break down the cultural barrier when talking to locals on patrol and possibly defusing a dangerous situation.”

That potentially life-saving scenario illustrates why it’s vital for DOD to improve its foreign-language capabilities, with or without cutting-edge technology.

Advertisements

Leave a comment

Filed under articles

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s