Mastering language with technology

Language is something that is inherently human, though it is also something so incredibly complex that whole areas of study have been devoted to further understanding it.  Until recently, doing something such as deciphering a language of the past sometimes took teams of linguists months and even years to make the most basic progress.  Now, with the invention of computers, the process is becoming much easier.

The further one goes back, the more problems arise when it comes to piecing together an extinct language.  In many cases, the best a linguist can hope for is to use matching equivalents in our own world and trace them back, step by step, until there is some understandable correlation.  Since many of these languages use conventions that we no longer adhere to in our modern languages, this can be a massively complex process.

The new computer programs are using what we know of existing languages and applying the patterns and rules that linguists accept as truth today.  They then work backwards, determining which languages are related to which other languages and making more sense of the grammar, syntax and other linguistic elements. 

The process is somewhat complicated, but basically the computer uses a list of words to make deductive conclusions.  While normally linguists would use a small sampling of languages or perhaps go through each language comparison over a long period of time, these computers can examine hundreds of languages at once.  What was once a labor intensive process is now at the click of a button.

This isn't the first time such a program has been used, but the latest improvements are lowering the margin of error.  The latest testing shows that the programs being used today are running with about 85 percent accuracy.  Still, linguists hope to improve it even further, teaching the programs to use more and more linguistic rules when making their deductions.

On the other side of the coin, linguistic mastery in computers is looking to possibly change the power and reliability of the Internet.  By figuring out a way for computers to better make the jump from natural language systems to functional languages (i.e.: mathematics), one group hopes to develop a powerhouse search engine.  The goal is to get a computer search to recognize not just the words and grammar, but implied context and other human linguistic elements.

Language is one of mankind’s greatest gifts, but also one of the most complex.  It’s amazing to me that computers are helping us surpass our brains’ limitations to the point where they can do the workload of small armies of linguists.  By using these new technologies, we can both save ancient languages from the dangers of fading away forever and improve the link between man and machine for the future.

Language Globe courtesy of M. Adiptra via Wikicommons