Historians and linguists have long wondered how ancient human languages might have sounded, but have had no conclusive data on which to base estimates. Now, linguistics researchers at the Universities of Cambridge and Oxford have developed a method to recreate through statistical analysis the sounds of languages as they were spoken up to 8,000 years ago.
According to a news release by Phys.org, the team of linguistics has submitted their research for pre-publication. The technique is based on the mathematical analysis of the recordings of sounds, or phonemes, of a variety of languages.
Using spectrographic analysis of these and historical recordings, the scientists were able to create a statistical linear model of how language sounds have changed in modern history. Using this linear model, the linguists then used computer modeling to predict how sounds might have sounded in the distant past.
So far, researchers have been able to trace common roots of words for numbers, since these have the same meaning in nearly every language. The lab behind this study has posted recordings of recreated sounds on their website. Interestingly, this same model could be used to predict how the sounds of languages might change in the future as well.
According to Dr. John Aston, from Cambridge’s Statistical Laboratory, this recreation technique is dependent on the physics of sounds such as spoken languages:
Sounds have shape. As a word is uttered it vibrates air, and the shape of this soundwave can be measured and turned into a series of numbers. Once we have these stats, and the stats of another spoken word, we can start asking how similar they are and what it would take to shift from one to another.
Many linguists believe that an early language called Proto-Indo-European (PIE) is believed to be the common ‘ancestor’ of hundreds of modern languages. This proto-language is theorized to have evolved in Eurasia sometime around 3500 BCE, although estimates vary.
There are no written records of this early language, and linguists so far have only been able to theorize what early languages sounded like by examining common features of contemporary living languages that are thought to stem from PIE.
Aside from helping linguists better understand the roots of modern languages, this same application of spectrographic analysis could help improve automatic translation technologies and even film dubbing. Thus, the next time Mel Gibson decides to make a film about early human civilizations, the characters’ speech might be a little closer to reality.