It sounds nearly impossible, right? How could one language be implemented across cultural divides to encompass each culture’s unique linguistic needs?
Well, DARPA (the Defense Advanced Research Projects Agency) is giving it a go. But in this instance, they’re not trying to push a new language on anyone other than robots and machinery.
They are attempting to design a mathematical language to interpret everything the military sees, hears, or does. It would teach all sensors to think and process data before feeding it back, giving analysts only the information they really need instead of overwhelming them with the sheer volume of the data.
As the writers of one article at Wired.com write, “In other words, one mathematical formula has to teach machines how to create order out of the chaos of the world around them, and to use that common ontology to develop a “learning capacity and expected rate-of-learning.”’
And if that idea doesn’t take off, the Pentagon is working on something else: a universal language translator program. (Click here to read a whole article on it.) So far machine translation is faulty at best. It fails to capture the nuances of a language or implement grammar correctly. But with a $15 Million dollar budget set aside for the project, researchers are aiming for more than accurate machine translation of text. They’re also expecting their programming to be able to interpret speech, including rarely heard regional dialects, and poor pronunciation.
The program is called Boundless Operation Language Translation (BOLT). Officials say if this works, troops will be able to better communicate with local populations even when an interpreter is not readily available.